(^,/f;^
]Jmagnet.
changes in r will be correspondingly small. If r is extremely large
the changes in direction and intensity for small changes in r may
become negligible in magnetic measurements. We call this a uniform
field.
11. Magnetic moment. — Suppose a magnet is suspended or
pivoted at its center of gravity so that it is free to oscillate in a hori-
zontal plane in a uniform horizontal field,/, figure 2. The force on
the north-seeking pole (+m) is +/m and on the pole (— m) it is —fm,
MAGNETIC OBSERVATORY MANUAL
[Ch.
where m is the pole strength of either pole of the magnet. As these
two forces are equal and oppositely directed, the clockwise turning
moment or couple, i', will be
L'=jm (2Zsin 6)
=2ml (f sin 6)
in which d is the angle between the magnetic axis and the direction
of the field,/. The quantity, 2ml, is constant for a particular magnet
so long as its magnetization does not change, and is called the magnetic
moment of the magnet. That is.
U=JMsme.
(7)
12. It is thus possible to measure the magnetic moment of a
magnet in terms of/, 6, and L\ When 6 is zero, sin 6=0, and the
couple is zero. Hence a magnet freely suspended in a uniform field
will come to rest with its magnetic axis parallel to the field. When
^=90°, that is, when the magnet is at right angles to the field, sin d
= 1 and
L'=fM (8)
and
M=j. (9)
If the uniform field has unit intensity, then M=L' , which means
that the magnetic moment may be defined as the couple (dyne-
centimeters) which will maintain the magnet at right angles to a
L
^ 1
' '
S\ -m
—
+ m»mN
h ^
I
H
Figure 3.— Relation between length
of magnet and pole distance;
shaded end is north-seeking
end of magnet.
Figure 4.— Resolution of a magnetic field
into two components.
uniform field of unit strength. It is expressed in dyne-centimeters
per unit field, that is, — — ^—r or simply the cgs unit of magnetic
moment.
13. Referring to figure 3, experiments show that, for bar magnets,
the pole distance, I, and the length of the magnet, L, are related as
follows:
Z«0.4Z. (10)
1.] FORCES BETWEEN MAGNETS 5
14. Resolution of fields, — A field has both magnitude and direc-
tion and may be resolved into components. The resultant of two
fields acting at a point may be evaluated graphically as in figure 4.
Let H=0.3 cgs unit at and directed along ON. Let P=0.4 cgs unit,
directed along OE. The resultant field, /=0. 5, is directed along OA.
A small compass needle placed at would come to rest with its axis
directed along OA. The components of / along OE and ON are
OE=OA cos e and ON=OA sin d.
15. Field at a point P in terms of magnetic moment. — Sub-
stituting M, the magnetic moment of the magnet, for 2 ml in equations
(5) and (6), we have the following approximate relations:
/«^, parallel to >SiV (11)
for the field of a bar magnet at a distance r from the center of the
magnet along its magnetic axis produced, and
/«^, parallel to iV^ (12)
for the field at a distance r from the center of the magnet along the
perpendicular bisector of the magnetic axis. Equations (11) and
(12) serve reasonably well for estimating the field of a magnet so long
as I is small compared to r. For more precise values of these fields,
corrections must be applied for distribution, as explained in para-
graph 37.^
16. Estimation of the field from a nomogram, — Corresponding
values of/, M, and r in equation (11) are shown graphically in figure
137. The variation of the field with distance, r, is shown in figures
138 and 139.
DEFLECTIONS
17. Gauss' first position, — Deflector end-on, approximate A posi-
tion (see par. 32). In figure 5, let Ms be a short magnet pivoted at
the point and free to turn in a horizontal plane. If H is the only
field acting on Ms the magnet will come to rest with its axis parallel to
H, that is in the direction of OA. Now place a second magnet, Ma, in
a fixed position at P so that its magnetic axis is on the perpendicular to
H through 0, and with its center at a distance r from 0. Ma produces
a field/ in the direction OB; the resultant of/ and H is OC. Ms will
turn through an angle u, coming to rest with its magnetic axis in the
direction OC of the resultant field. From equation (7), the clockwise
couple on Ms due to H is HMs sin u and the counter-clockwise couple
on Ms due to the field/, of the deflector is /Ms sin {90°— u) =fMs cos u.
1 The aim and scope of this manual preclude showing explicitly the permeability of air in each of the many
equations that need it under the standard conventions regarding the nature of pole strength. For full
dimensional coherence, one may read into the equations the factor fxcts accompanying and qualifying every
symbol that denotes a field component as derived from the pole strength or moment of the magnet that
produces the field. The equations affected include those for deflection experiments (e. g. eq. 41), so what
we really measure in deflections is, according to this approach, the induction B rather than the magnetic
intensity H.
6
MAGNETIC OBSERVATORY MANUAL
[Ch.
When equilibrium is established and the magnet is at rest these couples
are equal and
HMs sin u=JMs cos u
j^_sin u
H cos u
=tan u
(13)
(14)
but
whence
H=
f
/
H
tan u
2 Ma
2 Ma
r^ tan u
(15)
(16)
B
-^0°-u
^^_^^----^
t n
^^^
u
-^
r
H
p ^
. Ma
Figure 5.— Magnetic deflection, first position of Gauss.
Example: Let Ms=472 cgs; f=30 cm; and ^^"=0.175 cgs=175007.
Then
0.035 ^ ^^^
tan u^ ^ =0.200
U.l (o
^«11° 19'
IS. Gauss* second position, — Deflector broadside, approximate
B position. In this position the deflector axis is at right angles to H
1-1
FORCES BETWEEN MAGNETS
with its center at P, PO being directed along H (see fig. 6) . The field
of Ma at is directed opposite to Ma (see fig. 1) and
(17)
•a iH5
Figure 6,— Magnetic deflection, second position of Gauss.
Example as in paragraph 17:
472
/=
tan u
30^
1750
17500
=1750t
= 0.100
Hence, u^^° 43' (approximately one-half the deflection in the A
position) .
19. Lamont's first position, — A position, deflector end-on. In
this position the deflector is placed (fig. 7) so that its magnetic axis is
Figure 7.— Magnetic deflection, first position of Lamont.
always perpendicular to the suspended magnet when the latter comes
to rest along the resultant. The couple Ms due to H is HMs sin u and
that due to/ is /Ms sin v, but z; is 90° so that
HMs sin u=fMs
(18)
8 MAGNETIC OBSERVATORY MANUAL [Ch.
and
■^=sin u (19)
H=-X- (20)
sm u
H^^^ (21)
r' sin 1^
Equation (21), with some minor correction factors, is standard in mag-
netometer deflections. Equations (16) and (17) are basic in variom-
eter scale-value deflections where a permanent magnet is used as the
deflector.
ANALYSIS OF THE FIELD OF A BAR MAGNET
20. Procedure for short magnets, — Just as a field may be
resolved into components, so the magnetic moment may be resolved,
enabling us to evaluate the field of the magnet at any point P, or any
component thereof. In figure 8, let M represent the magnetic moment
of the magnet. We require the principal components of the field, at a
point P on a line OP making an angle d with the magnetic axis of the
magnet and at a distance r from 0. The various components are de-
scribed below.
21. The radial component, fr, is the component of/ at P in the direc-
tion of OP. Let Mr = the component of the magnetic moment, M, in
the direction of OP. Then
Mr=M cos e (22)
2 Mr
_2McosJ ^23)
22. The tangential component, fe, is the component normal to OP
at P. The component of the magnetic moment normal to OP is
Mfl=M sin e (24)
Mb
h=-,.
M^ (25)
1.] FORCES BETWEEN MAGNETS 9
23. The total field, f, at the point P, is the resultant of/r and/^.
=[-73- COS ej +[^ sin ^J
=^(4cos2 0+sin2(9)
= ^'[4 cos2 0+(l_cos2 (9)]
=^(3cos2^+l)
/=i!l73cos2 0+l. (26)
24. The parallel component, f\\, is the component of/ parallel to the
magnetic axis of M. Consider PJ as made up of PA~JA. By
construction,
PA=Jr cos d
JA=FC=RF sin e=fe sin (9
yii=PA— JA=/^ cos d—fe sin 6
Substituting values of/^ and/e,
/il^(^-73- cos 0j cos ^-[jj sm ^J sm
2M 2. M ' 2 a
= — 3- cos^ 6 3- sm^ 6
M
=^ (2 cos2 ^-sin2 d)
M
=^(2cos2 0-l+cos2(9)
M
=-^(3cos2 0-l)
=^(cos2 (9-0.333). (27)
210111—53 — —2
10 MAGNETIC OBSERVATORY MANUAL [Ch.
25. The 'perpendicular component, fj_, is the component at P, per-
pendicular to the magnetic axis of M. Consider the perpendicular
component, /j.> as made up of PB-{-BG:
PB=jr sin e
BG=RC=RF cos d=fe cos 6
J^=PB+BG=Jr sin B+Je cos d
2M . . . , M . ,
cos 6 sm d-\ — ?- sm 6 cos 6
M
=-y (2 sin 6 cos ^+sin 6 cos 6)
M
=-^ (3 sin d cos ^) (28)
26. We now list the five equations developed above, together with
values of the respective components (see fig. 8) based on assumed con-
stants as follows: M= 10,000 cgs; r=:100 cm; ^=40°. We have, then
Radial /r=^ cos 6 15327
M
Tangential /e=-^ sin ^ 6437
Total /^My'3cos2 0+1 16627
Parallel J\\=^ (3 cos^ 6-1) 76O7
Perpendicular /x= — 3- sin 6 cos 6 14777
27. Other axes and components. — It is often convenient to compute
a component of the total field,/, parallel to some other direction. In
figure 8, suppose the chosen direction is PQ. Then the component
of / parallel to PQ is PD, and the component perpendicular to PQ
is DF. If PQ is the direction of magnetic north, PD will be denoted
by /at and DF by /^. If PQ is in the direction of a suspended magnet,
PD=Jj, and DF=jn. If PD and DF are parallel respectively to the
X- and F-axes, then PD=J^ and DF=fy.
28. Components of the field for small values of 6. — It is interesting
to note the values of the various components for small values of 6
and for values of 6 near 90°. This is of special importance in con-
nection with placing the deflector in the orientation tests described
in chapter 12. The values of these components are given in table 1.
1.]
rOBCES BETWEEN MAGNETS
11
TABLE 1. — Components of the field of a short bar magnet.
[Computed for a Uniform Distance of 100 cm, With M=iooo, for Angles Close to 0° or 90^]
d
fr
fe
/
/ii
A
o
y
7
T
7
7
200. 00
0.00
200. 00
200. 00
0. 00
1
199. 97
1. 75
199. 98
199. 91
5.23
2
199. 88
3.49
199. 91
199. 63
10.46
3
199. 73
5. 23
199. 79
199. 18
15.68
4
199. 51
6. 98
199. 63
198. 54
20.88
5
199. 24
8. 72
199. 43
197. 72
26.05
85
17.43
99. 62
101. 13
97.72
26.05
86
13.95
99. 76
100. 73
98. 54
20.88
87
10.47
99. 86
100. 41
99. 18
15.68
88
6.98
99.94
100. 18
99.63
10. 46
89
3.49
99.98
100. 05
99.91
5. 23
90
0.00
100. 00
100. 00
100. 00
0.00
a 4
^fe ^
G F C
Figure 8.— Components of the field of a magnet at any point P.
29. Direction oj total field, j, at a point, P. — Consider figure 8.
Let the angle RPF=(f)=the angle between the total field vector, /,
and the radial field vector,/,.
tan =
2M
(29)
cos 6
sin 6
2 cos e
and for small angles
=-^ B.
(30)
(31)
12
MAGNETIC OBSERVATORY MANUAL
[Ch.
30. Distribution coefficients and distribution factors, — Equa-
tions (11) and (12) are based on the assumption that the distance, r,
is great compared to the pole distance. In practice this is not always
true so that in calculating the couple due to the interaction of two
magnets we must apply certain factors to the equations for the couple
based on (11) and (12) in order to obtain greater precision. Bergen,
Schmidt, Hazard, Hartnell, and others have developed special solu-
tions for these factors as they apply to pairs of magnets of different
lengths and different orientations with respect to each other. The
correction factors take the form
1 + -2 + -4+---'
in which P and Q are called the first and second distribution
coefficients.
31. Figure 9 represents three mutually perpendicular planes. All
axes pass through the center of a suspended magnet, Mg, the X and Y
^.
Fieure 9.— Deflector positions.
axes being horizontal and the Z axis vertical. The deflector is
designated Ma. (Ma and Ms are also used to designate the magnetic
moments of these magnets.)
32. Four positions are considered. The axis of the suspended mag-
net lies along the X axis for all positions.
A Position: The center of the deflector, Ma, is on the F-axis, and
the magnetic axis of Ma is parallel to the F-axis;
B Position: The center of the deflector, Ma, is on the X-axis,
the magnetic axis of Ma is parallel to the F-axis;
C Position: The center of the deflector, Ma, is on the Z-axis,
the magnetic axis of Ma is parallel to the F-axis;
and
and
1.]
FORCES BETWEEN MAGNETS
IS
D Position: The center of the deflector, Ma, is anywhere in the
YZ plane and the magnetic axis of Ma is parallel to the Z-axis.
33. It should be noted that, although the positions of Ma and M,
are described above relative to the vertical Z axis and the horizontal
X and Y axes, the axes may be rotated in any direction. The posi-
tion of Ma relative to Ms is considered in each case, rather than the
positions of the two magnets relative to a fixed set of orthogonal axes.
34. Table 2 gives the formulas for calculating the distribution
coefficients in terms of the pole distances of the pair of magnets and
also gives the form in which the distribution factors should be applied
for the four positions. In the computations in this manual, the pole
distance of a magnet is taken as 0.4 of the length of the magnet (see
par. 13).
35. Table 21 gives the approximate values of P and Q in terms of
the over-all lengths of the deflector and suspended magnet for various
combinations of La up to 20 cm and Lg up to 12 cm. This table
together with the nomogram, figure 146, will be found very helpful in
making a quick and reasonably accurate estimate of the value of the
distribution factor. With magnets of odd shapes, the pole distance
may be known to be other than 0.4 of the magnet length; in such
cases, enter the table with a nominal value of L equal to 2.5 /, irre-
spective of the actual length of the magnet, so that / will be 0.4 of
the nominal L. (For table 21 see pp. 227-228.)
TABLE 2. — Summary of form,ulas A, B, C, and D Positions.
Posi-
tion
P
Q
Factor
A
2Z„2_3/,2
4^
^=i+^+f?
B
-- |/a2+6/,2
1 c 4c
B=l + '^ + ^
C
-^aj+is')
~{ij+L'y
c^=i+^^+t
D
,,/35 . , 5\ 5,,
« = l + ^"+f?
If
Afa = deflector or its magnetic moment;
Af, = suspended magnet or its magnetic moment;
la = po\e distance of the deflector;
Z8 = pole distance of the suspended magnet;
^ = height of the center of the deflector above center of Ma',
r= distance between centers of Ma and iV/,;
e = angular elevation of center of Ma above (or below) center of M,
(D-position) .
(i = horizontal distance for D-position, r^ = h^ + d^, and tan e=-T-
14 MAGNETIC OBSERVATORY MANUAL [Ch.
36. In Lament's first position, the deflector is always in the A
position. The couple, Li, tending to turn the suspended magnet out
of the magnetic meridian is,
L.='-^{l+^+^} (32)
The restoring couple, L2, due to the uniform field, H, acting on Ms, is
L2=HMs sin u. (33)
When Ms comes to rest these couples are equal and
™sin«=^^(l+^+^) (34)
and
37. In equation (34), if r is quite large compared to the pole distance,
la, of the deflector, and if the pole distance, Is, of the suspended magnet
is very small, this relation may be written
H^mu=-^{^1 + ^J- (36)
And from equation (19),
f=H sin u
Hence,
/=^"(l+f> (37)
Equation (37) will yield a reasonably precise value of the field at a
point P along the magnetic axis of Ma produced. Similarly, for the
B position, the field in the direction of the perpendicular bisector of
Ms would be
38. Special values for Pa and Qa. — In the equation Pa=211—31^,
by letting Pa =0 and solving for f, we find ^= 1 .225. Likewise, in the
45 I
equation ^^ = 3^*— 15Z^/f+-^ It, by setting ^^ = and solving for fj
two positive values are found; ^=2.143 for the usual case that la is
greater than Is; ^=.641 for the rare case when /« is less than Ig. The
long and short magnets of a magnetometer are usually designed with
the ratio of these lengths such that Pa or Qa will be practically zero.
1-]
FORCES BETWEEN MAGNETS
15
Table 3 summarizes these ratios in form for convenient use. Note
that in solving for Pa = or ^^ = 0, la is taken as 0.4 La.
TABLE 3.— Special values for Pa and Qa.
For Pa =
For Qa = and la>la
(a) la =
(b) ls =
(C) La =
(d) Ls =
(e) Qa=-
(f) Qa=-
1. 225 I,
0. 816 la
1. 225 L,
0. 816 La
-4. 500 la*
-0. 115 La*
Condition
1 La la
i L-ls
la=0.4:La
(g) la== 2.USh
(h) Is^ 0. 467 la
(i) La= 2. 143 L,
(j) L,= 0. 467 L„
(k) P4= + l. 35Z„2
(1) P^ = +0. 216L„2
Condition
\ La la
i Lru
la = 0.4:La
Examples
For Pa--
=
FovQa
=
Casel
Case II
Equation
Case I
Case II
Equation
when La^
Ls =
Qa =
cm
4. 9
4.
-66.
cm
9. 28
7. 58
-854.
(d)
(f)
when La =
Ls^
Pa-
cm
4.9
2. 29
+ 51.9
cm
9.28
4.33
18. 6
(j)
(1)
CHAPTER 2. THE EARTH'S MAGNETIC FIELD
(c) The vertical intensity, Z,
ZENITH
39. The magnetic elements, — The direction of the earth's mag-
netic field at a point on the surface of the earth is the direction taken
by a freely suspended magnetic needle (free to turn in any direction
in space).
(a) The magnitude of the field is called the total intensity and is
indicated by F in figure 10.
(b) The horizontal intensity, H, is the projection of F on the hori-
zontal plane.
is the projection of F on the vertical.
(d) The true north component, X,
is the projection of H on the true
north direction.
(e) The true east component, Y, is
the projection of H on the true east
direction.
(f) The magnetic declination, D, is
the angle between H and X.
(g) The magnetic dip or inclination,
I, is the angle between H and F.
(h) The magnetic meridian plane
is the vertical plane through F con-
taining H, F, and Z.
(i) The magnetic meridian will be
used in this manual to denote the
horizontal line through a specified
point, in the direction of H.
( j ) The magnetic prime vertical plane
is^the plane perpendicular to H.
40. The principal magnetic elements are D, H, and /. A knowledge
of their direction and magnitude at any point enables one to compute
other desired components.
41. The relation between these various elements is shown in the
vector diagram, figure 10. Note also these primary equations:
Z=H tan / F'=H'+Z' X=H cos D
Z=F sin / Y=H sin D
H=F cos I H'=X'-}-Y'
42. Approximate values of the magnetic elements for any part of
the world may be scaled from world isomagnetic charts^ with sufiicient
accuracy for use in designing magnetic observatory instruments for
particular regions.
43. General pattern approximated by dipole field. — The
accrued results of past measurements show that the earth's magnetic
Figure 10. — Components of the earth's magnetic
field.
1 Those issued by the U. S. Hydrographic Office, Navy Department, are a standard source. For most
areas, adequate values may likewise be obtained from the Smithsonian Physical Tables (see item 14 of
bibliography on p. 228).
16
EARTH S FIELD
17
field is such as could be roughly accounted for by supposing a bar
magnet at its center with a magnetic moment, M, of about 8.1X10^^
cgs. The field of such a magnet would have a configuration depending
on its length — that is, on the separation of its poles. However, it
is found that the shorter the magnet the better the fit. As the poles
are brought closer together and their strength simultaneously increased
so as to preserve a constant magnetic moment, the field pattern at
remote points is but little changed, and with the pole spacing very
small in comparison to the distance, the field approaches a limiting
pattern which we term the field of a magnetic dipole, taking the pole
distance as an infinitesimal quantity. The field of a centered dipole
is the simplest over-all approximation to the earth's field. Note
further that the dipole field would be indistinguishable from one
produced by uniform magnetization (parallel to an axis) of the entire
earth, or of any smaller portion thereof occupying the volume within
a concentric spherical surface. A suitable distribution of electric
current, flowing along circular paths in a metallic core, could also
yield such a field.
44. The axis of the centered dipole that most nearly duplicates the
earth's field is known as the earth's magnetic axis. It pierces the sur-
face of the earth at two points known as the geomagnetic poles, where
the dipole field would be perpendicular to the surface.^ The places
where the actual field is perpendicular to the surface are known as
the earth's magnetic poles and do not coincide with the geomagnetic
poles, since the dipole field is only a rough approximation to the actual
field. The currently adopted positions of the two kinds of poles are
as follows:
POLE
LOCATION
Latitude
Longitude
North magnetic
73° N.
68° S.
78°. 5 N.
78°. 5 S.
100° w.
144° E.
69° W.
111° E.
South magnetic
North geomagnetic
South geomagnetic
45. Earth's field at various latitudes, — In figure 11 the circle
represents the surface of the earth and the small bar magnet repre-
sents the hypothetical magnetized portion at the center. Since the
north-seeking end of a compass needle is called the North end, the
magnetically active core of the earth behaves as though its geograph-
ically northernmost part were magnetically a South pole.
46. The approximate relative magnitude and direction of the
components Je, Jj, and /, at certain points on the circumference of
the circle of radius r, where r is very great compared to the length
of the magnet, are shown in figure 11, for various values of 0. The
angle is the angle between the magnetic axis and the radius through
the chosen point on the circumference. If we consider the plane of
the circle as a section through the earth and consider that the mag-
netic axis lies in the plane of this section, then the components of the
earth's field, at the surface, may be computed approximately, using
equations (23) and (25) of chapter 1 (p. 8), with M=8.1X102^ cgs.
2 Chapman and Bartels, Geomagnetism, vol. II, pp. 644-45, 1951 (see item 2 of bibliography).
18
MAGNETIC OBSERVATORY MANUAL
[Ch.
The magnetic latitude at a point is the angle whose tangent is
ji tan /, where / is the actual dip at the point. The geomagnetic
latitude is the angle whose tangent is K tan /', where /' is the dip
computed from the simple dipole approximation. It can be shown
(see eq. 30) that the geomagnetic latitude is a coordinate like ordinary
latitude but reckoned in relation to the geomagnetic rather than to
the geographic poles.
Example: Suppose 0=112°. 5, corresponding to a geomagnetic
latitude of 22°. 5 N, taking the radius of the earth to be 6.37X10^ cm,
Z=-fr =
2M
cos 0=0.241 cgs
H=fe=^- sin 0=0.290 cgs
F=^lH'+Z'=0.d77 cgs
7=tan-^^=39°.6
47. The correspondence between /, /r, /e, , and F, H, Z, / is as
follows:
Field of Magnet
Earth's Field
Name
Symbol
Name
Symbol
Radial component
Tangential component
Total field
/r
fe
f
Reversed vertical intensity-
Horizontal intensity
Total intensity
-Z
H
F
Direction of/
Direction of F
Angle between / and fe
90° -<^
Inclination (or dip)
I
The negative sign is required in the relation/;.= —Z because Z is taken
positive inward and/^ is positive outward.
48. Equation of a line of force. — It has been shown by Chapman
and Bartels,^ and others, that a line of force due to a magnetic dipole
is the locus of the equation
r=C sin^e
(39)
in which r= distance from a dipole magnet to a point P,
0=the angle r makes with the magnetic axis of the dipole,
and (7= a constant for the particular line of force.
49. It is obvious from equation (39) that a line of force (any line
of force) must pass through the dipole since r=0 when 0=0°. When
0=90°, sin^ 0=1 and r=0. By assigning arbitrary values to C (such
as 1, 2, 3, etc.), and 5°, 10°, 15°, etc. to 0, the corresponding values
of r may be calculated from equation (39). Figure 11 shows the
pattern of the lines of force about a dipole for various values of C.
3 S, Chapman an4 J. Partels, Geomagnetism, p. 11 (see item 2 of bibliography on p. 228),
2.1
eaeth's field
19
50. To draw a line of force which shall pass through a given point
P whose polar coordinates are known with respect to the dipole and
its magnetic axis, first solve equation (39) for C using the given values
of r and B. Then using this value of C and various values of B, solve
for the corresponding values of r. Figure 11 shows the lines of force
Figure 11.— The earth's magnetic field at various latitudes.
passing through several points on the circumference of a circle of
radius 10 units for values of (9-0°, 22^.5, 45°, 67°.5, and 90°, and
for the corresponding angles in the other quadrants.
51. Changes in the earth's magnetic field. — The earth's mag-
netic field is always changing in direction or intensity or both. These
changes are described briefly in the following paragraphs.^
52. Daily variation. — There is usually a fairly systematic departure
of the magnetic field from its daily mean value. This repeats itself
* Further details are given by A. K. Ludy and H. H. Howe, Magnetism of the Earth (see item 13 of bib-
liography on p. 228).
20 MAGNETIC OBSERVATORY MANIJAL
(though with somewhat variable form and ampHtude) day after day.
The amount of departure depends upon the time of day, the season,
the magnetic latitude, and other factors not wholly understood. This
systematic change is called daily variation.
53. Irregular disturbance and magnetic storms. — Superimposed on
the regular daily variation, there are usually irregular changes.
When they become v^ery large, we say there is a magnetic storm.
These storms are associated with sunspots, and are characterized by
auroral displays and pronounced disturbances to radio-wave trans-
mission and wire telegraphy. A magnetic storm may last many
hours or even several days, and the more severe ones occur all over
the earth at the same time.
54. Secular and annual change. — In general, the average value of a
magnetic element changes from one year to the next, and the change
usually continues in one direction for many years. This is called the
secular change. The amount in one year is called the annual change.
t
CHAPTER 3. THE MAGNETIC OBSERVATORY
55. Selection of an observatory site, — The primary objective of
the operation of a magnetic observatory is the measurement of the
continuous variations in the earth's magnetism in the particular
region where the observatory is situated.
56. The site should be free from local disturbances (irregularities
of the magnetic field), either natural or artificial. This can be ascer-
tained only by a preliminary magnetic survey of a large area sur-
rounding the proposed site, followed by a more detailed survey of the
immediate site, say over an area of 50 meters radius. Such a survey
should consist of magnetic observations made at points uniformly
spaced in the form of a grid over the immediate area with, say, 10
meters between stations. It is usually sufficient to test for irregu-
larities in magnetic declination only. Anomalies in vertical intensity
may be determiaed by a magnetic field balance. A quartz horizontal
magnetometer, QHM, is probably the most satisfactory instrument
for the determination of anomalies in horizontal intensity. In gen-
eral, if tests in the immediate vicinity of the proposed site show that
there are gradients in declination, anywhere in the area, as great as
10' between stations 10 meters apart, or as great as 2O7 in horizon-
tal or vertical intensity in the same distance, the site should be con-
sidered unsatisfactory. AJso, the mean value of the measured
element should be about the same as the mean value for the site,
as scaled from the latest available isomagnetic chart.
57. Buildings and equipment. — The minimum requirements for
a magnetic observatory are as follows:
(a) Satisfactory site, as described in paragraph 56.
(b) Magnetic variation building, including instrument piers.
(c) Absolute observatory building, including instrument piers.
(d) Magnetograph, consisting of H, D, and Z variometers, a photographic
recorder, time-marking mechanisms, and facilities and equipment
for scale-value and orientation observations.
(e) Office and photographic darkroom,
(f) Magnetometer (or equivalent) for measurement of H and D.
(g) Earth inductor (or equivalent) for measurement of /.
(h) Mean-time chronometer.
(i) Time marking clock or chronometer.
(j) True meridian line or true azimuth from declination pier in absolute
observatory to a permanent azimuth mark. (Two permanent azimuth
marks recommended.)
(k) True azimuth line established in the variation room.
(1) Miscellaneous equipment for processing records.
58. Figure 12a shows a plan view of a variation building, designed
for operation of one or two complete magnetographs. For other
types of magnetographs the pier plan would require alteration but
the basic requirements would be practically the same as described
here. ^ The building and piers are constructed of tested, nonmagnetic
materials throughout. The piers should have enlarged bases extend-
ing well below the ground surface and should rest on natural, undis-
turbed formation. Within the variation room and attached perma-
nently to the walls are special shelves of heavy construction for sup-
porting deflector holders used in orientation tests and scale-value
21
22
MAGNETIC OBSERVATORY MANUAL
[Ch.
are
observations by the magnetic method. The variation rooms
thermally insulated and are suitable for photographic recording.
59. The absolute building is a small, nonmagnetic structure shown
m figure 12b. Asjn the variation building, the bases of the piers
je|0|!iii*A « Bu|uiOJj |o*d*y-
H B B
a H B B R
1 — r
J_L_L_i_J^
o
•
r '1°
m
q::p
g \ ^
^8 B Ta
0,91 •pms.oo.s
a a a a
Ir
^ I
S3
oniA
.»-,l »,•-.! tJOOQ iO|D|UU»A
•a pj
* 5
« Q
■•
s
should rest on natural, undisturbed formation, well below the ground
surface. In routine operations the magnetometer is operated on the
central pier and the earth inductor and its galvanometer are operated
on the other piers. The windows of this structure are so arranged
as to prov^ide an unobstructed v4ew of one (or two) well established
true azimuth marks.
3.1
BUILDINGS AND EQUIPMENT
23
60. The three magnetic variometers shown in figure 13 are con-
structed to respond to variations in magnetic declination, horizontal
intensity, and vertical intensity. A recorder, a light source, and a time-
flashing lamp with auxiliary reflector complete the magneto graph
■f" MASONITE , SOFT
2 X 6" FLOOR JOISTS
NOTCHED 1^' ON SILL
CONCRETE POSTS
STORM OOOR 2'e" X6'8'X|i" 5 PANEL
12 MarbK CubM
IxS.i
SCottofl PocKin^
Section through Piers
Figure 12b.— Magnetic absolute building.
assembly. If scale values are to be determined by the electrical
method, each variometer must be provided with a Helmholtz-Gaugain
coil (not shown). The various components of the magnetograph are
described in more detail in chapters 5 and 7-9. Descriptions of mag-
netometers and earth inductors with detailed directions for their use
are contained in ''Directions for Magnetic Measurements," by D. L.
Hazard.
24
MAGNETIC OBSERVATORY MANUAL
^ 2
i ^
1 «
^ I s
■" O
13 a
2 2
.2f 'z
CHAPTER 4. INSTRUMENTAL CONSTANTS AND
CORRECTIONS
REFINEMENTS CONTRIBUTING TO ACCURACY
61. Quality of instruments, — In order to assure the successful
accomplishment of accurate measurements of the earth's magnetic
field, it is first necessary to be sure that the available instruments are
of satisfactory quality and are so designed as to be capable of measure-
ments with the required precision. Quality here
refers not only to proper design and fine workman-
ship on the part of the instrument maker but also
to freedom from hidden defects which could and often
do introduce the necessity for index corrections that
theoretically should not exist. Such hidden defects
may often be traced to the presence of minute quan-
tities of impurities in the materials of which the
instrument is made, such as to cause erratic or spuri-
ous deflection of the suspended magnet or to distort
the earth's field in the vicinity of the instrument.
62. Test for magnetic parts, — Before any meas-
urements or observations are made for the determi-
nation of any of the constants of a magnetometer,
earth inductor, or other instrument used directly in
measuring a component of the earth's field, the parts
of the instrument should be tested carefully for mag-
netic properties. Any part found to show objec-
tionable traces of magnetism should be rejected; if
that part is near the moving magnet or coil of the
assembled instrument, any detectable magnetic effect
should be sufficient cause for rejection.
63. To test a specimen for magnetism, it is usu-
ally sufficient to place it near one of the magnets of
an astatic moving-magnet system (magnetism tester) ^magn^et^'^'^S'^of
such as that shown in figure 14. The two magnets magnetism tester.
A and B having equal magnetic moments (10 to 20 cgs units) are
mounted about 10 cm apart in the same plane on a stiff brass rod,
but with their moments oppositely directed. The system is equipped
with a mirror, M, and is suspended by a fine filament, Q, such as a
magnetometer suspension fiber. A lamp and scale are provided for
visual observation of the deflections of the magnets. A copper damp-
ing box for the lower magnet is necessary to reduce undesirable oscil-
lations, and a cylindrical shield, with window, not shown in the illus-
tration, provides protection against spurious deflections due to air
currents. As long as the magnetism tester is not located too near
direct-current machinery or cables, it will prove to be quite satis-
210111—53 3 25
26 MAGNETIC OBSERVATORY MANUAL [Ch.
factory for use even in a machine shop for testing parts of magnetic
instruments. The specimen to be tested is placed near one of the
magnets of the tester, and the deflection (if any) noted while the speci-
men is at rest. It is important that the specimen remain at rest while
observing the deflection of the magnets, since secondary magnetic
fields will be set up by eddy currents induced in the specimen if it is
moved about in the field of the magnets. Specimens should be tested
in several positions.
64. Although the magnetism tester as described does not yield
quantitative results, it can be made very sensitive by properly balanc-
ing the fields of the two magnets, and a little practice with the tester
will enable any observer readily to become skilled at finding traces of
magnetic materials in instrument parts.
65. Instrumental constants to be determined. — For measure-
ments of magnetic declination, the scale value of the scale in the
telescope (or magnet) must be determined as explained in paragraphs
66 and 67. The equations used in the computation of H and M by
the method of oscillations and deflections involve the moment of
inertia, induction factor, temperature coefficients, deflection dis-
tances, and distribution coefficients. These instrumental constants
must be determined as explained in paragraphs 69-145.
SCALE VALUE
66. Scale in the telescope. — This scale value is determined as
follows: Focus the telescope on a well-defined, distant object and
determine the difference in circle reading corresponding to a difference
of, say, 20 divisions on the scale. The ratio of the difference in circle
readings to the difference in scale readings is the scale value, in minutes
of arc per scale division. Repeat several times and take the mean.
67. Scale in the magnet, — Proceed as follows: When magnetic
conditions are quiet, suspend the magnet with a filament having a
small torsion factor, and when the magnet is at rest, with the vertical
cross line of the telescope bisecting some even scale division, observe
the circle reading. Rotate the magnetometer through precisely 10
scale divisions to the right, repeat the circle reading; then 20 divisions
to the left, etc., making several determinations. As in paragraph 66
the ratio of the change in circle reading to the change in scale reading
is the scale value in minutes of arc per scale division. Note that
there will be a slight restraining effect due to torsion of the filament,
but if a fine filament is used, the effect will be neghgible for such small
angular displacements. The operation may be simplified and the
effects of change in magnetic declination and torsion entirely elim-
inated if the suspended magnet is held in a fixed position relative to
the ground by a nonmagnetic stop mounted directly on the pier and
independent of the magnetometer.
MAGNETOMETER EQUATIONS
68. As shown by Hazard,^ the two fundamental equations showing
the relation between the horizontal intensity, H, of the earth's mag-
Dir. for Mag. Meas. (see item 4 of bibliography).
4.] CONSTANTS AND CORRECTIONS 27
netic field, and the magnetic moment, M, of the long magnet of a
magnetometer, are
HM= f^ jj^ . (40)
Hi(■+$+^)(-|f)].^^.
in which M= magnetic moment of the long magnet at temperature t;
7iL= moment of inertia of the long magnet and its stirrup at
temperature t';
7^= observed time of one oscillation (half-period) in chro-
nometer seconds;
c?=rate of chronometer in seconds per day;
A-= angle through which magnet turns when torsion head
is turned through an angle /;
/= angle through which torsion head is turned to deflect
the suspended magnet through the angle h;
iLt= induction factor of the long magnet;
^= observed temperature of long magnet during deflection
observations ;
^' = observed temperature of long magnet during oscillation
observations ;
g= temperature coefficient of magnetic moment of the long
magnet;
r= deflection distance at temperature t;
16= deflection angle for particular value of r;
and Pa and Qa are distribution coefficients.
69. The moment of inertia, induction factor, temperature coeffi-
cient, deflection distances, and distribution coefficients must be
determined by special observations or otherwise evaluated before
TT
equations (40) and (41) can be used in the computation of ffllf and y^-
The time of one oscillation, its correction for rate of chronometer, the
torsion factor, and the deflection angles are determined directly at
the time horizontal-intensity observations are made. For all prac-
tical purposes, the distribution coefficients, induction factor, and
temperature coefficient are taken as constants at all temperatures at
which observations are likely to be made, but deflection distances and
moment of inertia of the long magnet change appreciably with
temperature, and corrections must be applied accordingly to these
terms. The torsion factor of the suspension filament is usually deter-
mined just before or just after each set of oscillation observations.
The rate of the chronometer is, of course, an independent determina-
tion not involving the magnetometer itself.
70. It is customary to combine some of the factors in equation (41)
and treat them as a single constant for a particular value of r at a
particular temperature. Thus, equation (41 ) may be written
H C (42)
in which
M sin u
c=i(.-#)(.+$+2.).
28 MAGNETIC OBSERVATORY MANUAL [Ch.
The constant C may be computed for various values of r and these
values corrected for temperature as explained in paragraph 454,
appendix I (p. 193).
MOMENT OF INERTIA
71. Basic relations, — The time of one oscillation, T, of a magnet
oscillating as a torsion pendulum under the influence of the horizontal
component, H, of the earth's magnetic field is given by
V^
M
in which K is the moment of inertia of the magnet and its stirrup and
M is the magnetic moment of the magnet. The moment of inertia
of the magnet and its stirrup cannot be readily calculated from the
dimensions and mass; it must be determined experimentally by the
kinetic method. That is, the period of the system is determined with
and without a special inertia weight mounted on the magnet system.
The added mass is usually in the form of a precision-ground, solid,
right circular cylinder of bronze or brass, or in the form of a cylindrical
brass ring. In either case the material must be homogeneous and non-
magnetic.
72. The moment of inertia of a solid, right circular cylinder, oscil-
lating about a transverse diameter through its center, is
in which T^=the mass of the cylinder in grams;
l=ihe length in cm;
c^=the diameter in cm.
73. For a cylindrical ring oscillating about a longitudinal axis
(axis through its center normal to its bounding planes), the formula is
K,=^{dl+dl) (46)
in which di and d2 are the inner and outer diameters.
74. It is important that the inertia weight be constructed with
great precision and that the dimensions be determined for several
uniformly distributed positions. The temperature at which these
measurements are made should be specified. In the United States,
the mass and dimensions are usually determined at the National
Bureau of Standards.
75. With the inertia weight added to the system, the time of one
oscillation is
4.] CONSTANTS AND CORRECTIONS 29
in which Ki is the moment of inertia of the inertia weight calculated
from its dimensions. Squaring equations (44) and (47) and dividing.
T' K ^ K
and
K=y^ (48)
- — 1
76. Equations (44), (47), and (48) are based on the assumption
that K^ Ki, H, and M are constants and remain constant under all
conditions and that no other forces are acting on the oscillating
system. However, during a set of oscillations both H and the
temperature may change. A change in temperature changes the
dimensions of the magnet and the inertia weight, thus affecting
both K and Ki. The magnetic moment of the magnet also changes
with temperature. The directive force of torsion of the suspension
filament affects the period and also changes with the load. The
chronometer may not have a uniform rate during a series of observa-
tions but it is usually assumed that the rate is the same for adjacent
loaded and unloaded sets, and therefore does not affect the ratio,
T
~- Induction and damping effects may be considered as equal for
loaded and unloaded sets and therefore may be neglected.
77. The moment of inertia, Kx, of the inertia weight and the
observed times of oscillation, T and Ti, must be reduced to standard
conditions in order to evaluate the moment of inertia, X, of the
magnet at those same standard conditions. Temperatures are
usually referred to 20° C and changes in iJ to a mean value for the
elapsed time or simply to the H base-line value.
78. For the unloaded magnet, the relation between the true time
of one oscillation (T) and the observed time T is,
(^'=^'5^^[l+(20-0«][l+(20-Q2a]-g (49)
and for a loaded set,
(ri)'=rfg^^^[l+(20-(,)?][l+(20-<02a,]g (50)
in which (T)=the corrected time of one oscillation (unloaded);
r=the observed time of one oscillation (unloaded);
(7'i)=the corrected time of one oscillation (loaded);
Ti = the observed time of one oscillation (loaded);
^=the observed temperature of the magnet in unloaded
oscillations;
fi=the observed temperature of the magnet and inertia
weight in loaded oscillations;
2= the temperature coefficient of magnetic moment of
the magnet;
a = coefficient of thermal expansion (linear) of the magnet;
«! = combined coefficient of thermal expansion (linear) of
the inertia weight and magnet;
30
MAGNETIC OBSERVATORY MANUAL
[Ch.
H and Hx are the average values of the horizontal intensity during
a set of unloaded and loaded oscillations, respectively; and Bh is the
horizontal-intensity base-line value.
79. The correction factors in equations (49) and (50) are explained
in appendix I, and their method of application is shown in the sample
sets of observations, figures 15, 16, and 17.
UNLOADED
Oscillations for Moment of Inertia
station, Cheltenham, Md., Pier 6
Magnetometer No. 5064.
Magnet No. RIC 6 Inertia ring or bar No. A
Chron. corr'n on 75 M. Time, +0 10.0 Daily rate (jjj^^ gaTniSg)' +^-^ ^/^^^
Date, Fri., July 21, 1960
Observer, RL V
Chron. No. 6666
Number
of oscilla-
tions
100
106
IIG
116
120
126
130
136
140
146
Chronometer time
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
12 15
16
12
21.7
41.8
02.0
22.0
42.2
02.3
22.3
42.6
02.6
22.6
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
12 22
23
03.6
23 8
43.8
03.9
24.0
44.1
04.3
H.4
44.6
25 04.6
U
Temp, t
OC
26.5
Extreme scale readings
17.1
44-0
43.0
40.3
Remarks
60=^12 18 42.7
0= 16 21.7
100-
5 21.0
= 22 03.7
Time of
100 oscil.
6 41.9
42.0
41.8
41.9
41.8
41.8
42.0
41.9
42.0
6 41-9
Means
26.60
41.90
FORMULA: (T)2=T2^^^^ [l+{20-t)q][l+(20-t)2a] -^
Torsion observations
Tors, circle
293
113
203
Scale
div.
28.9
26.8
32.7
29.5
div.
31. S
27.0
35.1
30.8
Mean
div.
SO. 10
26.40
33.90
30.16
Diffs.
3.70
7.50
3 76
Mean h=S.74 =4'.94
Magnetometer scale value, 1'. 32 /div.
(20-0
Log
Time of 1 oscil.
Log T
Log r2
5400
5400 -ft
Log a+(20-t)q)
Log (1+(20-O2a)
Log (T)2
4. 01 900
0.60 41 18
1.20 824
+40
-52
-6
+339
1.21 146
Computed by RRB Checked by RL V Abstracted by AM
Figure 15.— Observations for moment of inertia, unloaded magnet suspension system.
4.]
CONSTANTS AND CORRECTIONS
31
LOADED
Oscillations for Moment of Inertia
station, Cheltenham, Md., Pier 6
Magnetometer No. S05J^
Magnet No. RlC-6 Ineitia ring or bar No. A
Chron. corr'n on 75 M. Time, +0 10.0 Daily rate (j^f^J. gaSg)' +^-^ ^^^^^
Date, Fri., July 21, 1950
Observer, RL V
Chron. No. 6565
Number
of oscilla-
tions
Chronometer time
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
50
12 52
56.4
17.0
37.7
58.4
19.1
39.8
00.5
21.2
41.8
02.5
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
50.0
10.7
31.4
52. S
13.0
33.8
54.2
15.1
55.5
12 58 56.4
58
Temp, t
°c
26.8
26.
Extreme scale readings
10. 1
12.7
15.3
div.
50.0.
46.0.
Remarks
^ = 12 52 23.1
0= 48 56.4
3 26.7
60= 55 49.8
Time of
60 oscil.
53.6
53.7
53.7
53.9
53.9
54.0
53.7
53.9
53.7
53.9
Means
26.
6 53.80
FORMULA: (T)2=r2
5^ [l+(20-t)q][l+(20-t)2a] —
5400
Torsion observations
Tors, circle
Scale
Mean
div.
div.
div.
203
29. 5
31. 3
SO. 40
293
24.9
26.9
25.90
lis
SS.8
36.8
35.30
20s
29.4
SI. 8
30.60
Difls.
4-50
9.40
4.70
Mean h=4.65 =6'.L
Magnetometer scale value, 7'.5^/div.
(20-0
L0£
Time of 1 oscil.
Log T
Log r2
5400
5400 -/i
Log (l + (20-0e)
Log {\-[-{2Q-t)2a)
Log (T)2
6.89 667
0.83 8639
1.67 728
+49
-54
-8
+347
1.68 062
Computed by RRB Checked by RLV Abstracted by AM
Figure 16. — Observations for moment of intertia, loaded magnet suspension system.
80. Directions for determination of moment of inertia, —
(a) Read paragraphs 71-78, and paragraphs 447-452, appendix I.
(b) Determine the time of one oscillation of the magnet (unloaded)
as in oscillations for horizontal intensity. (See fig. 15.) Mark
this set unloaded.
(c) Observe one set of torsion observations (unloaded).
32
MAGNETIC OBSERVATORY MANUAL
[Ch.
(d) Mount the inertia weight on the magnet system, center and
level it carejully, and repeat oscillation observations as in (b) (see fig.
16). Mark this set loaded.
(e) Make one set of torsion observations (loaded) .
(f) Observe the temperature at the beginning, middle, and end
of each set, also the extreme scale readings at these times.
(g) Repeat (d), loaded.
(h) Repeat (a), 2 sets, unloaded, etc., until eight sets of loaded
and eight sets of unloaded oscillations have been completed.
(i) If a magnetograph is in operation at the observatory or in the
vicinity, place special time marks on the magnetogram at the begin-
ning and end of each set of oscillations. If a magnetograph is not
in operation, or if H cannot be determined independently while
Computations for Moment of Inertia
Station, Cheltenham, Md.
Magnet No. RIC-6
Observer, RL V
Inertia ring or bar No. A
Date 1950
Set
75 M. Time
H ordinate
(/Imm)
Log (Ti) 2 loaded
Log (T)2
unloaded
h m
mm
Jul. 21....
lU
12 15/25
54-7
0.00 339
1.21
145
2L
49/59
55 9
347
1.68 062
3L
13 06/16
57.0
353
1.68 090
4U
29/39
60 7
376
1.21
104
Sept. 11...
5U
14 54/64
54-7
339
1.21
115
6L
15 12/19
56.3
349
1.68 078
12^..
7L
10 43/52
41.6
258
1.68 070
8U
11 01/11
40.3
250
1.21
115
9U
16/26
47.9
297
1.21
109
lOL
12 12/22
43.9
272
1.68 059
IIL
25/32
50.9
316
1.68 069
12U
50/61
48.8
303
1.21
101
13U
13 10/20
52.7
327
1.21
119
14L
26/36
51.3
318
1.68 085
15L
41/61
54.3
337
1.68 083
16U
14 13/23
55.5
332
1.21
128
Fre\.Hs.v. = SH
2.607/mm
Log-|
L=o.434|^ft_
H Bh
Sums 596
136
Frel.H hlv. = Bh
18199y
0.0000 62 h^u^
Means 1.68 074
1.21
117
a (magnet)
0.0000 11
0.46
957
, /inertia bar\
" Wring ;
0.0000 18
a-\-a'
0.0000 14
2.94
829
[Tl]2 ^
1.94
829
For UNLOADED oscillations:
0.0000 10
0.0000 12
[T]2
Log (l+2a)
For LOADED oscillations:
Log (l+2ai)
(b): LogiiTi.so
0.28
965
031
2.47
FORMULA: K2o=^^f^
(c): LogK2o:(b)-(a)
2.18
066
(d): L0g7r2
(e): Log,r2/f2o:(c) + (d)
0.99
3.17
430
496
Remarks:
log KiMfrom NBS, Nov. 1934 2.47
31
K2o=151.6
r^K2,=
1496.1
Scaled by RRB
Computed by RRB
Checked by /ei^F
Figure 17.— Computations for moment of inertia.
Scalings checked by RLV
Abstracted by AM
4.] CONSTANTS AND CORRECTIONS 33
TT
oscillation observations are in progress, ignore the factor ^- in equa-
tions (49) and (50).
81. Computation of moment of inertia of the magnet, —
(a) From the magnetogram, scale the mean ordinates, hmm, in
millimeters, for the elapsed times of each of the loaded and unloaded
sets, fourth column on figure 17.
(b) Calculate the conversion factor, 0.434 p-= 0.434 ToVvt?!^
0.000062 (see fig. 17).
TJ
(c) Calculate log -^~ for each set. (For example, the first line on
TT
figure 17: log ^-=0.000062 A^^=0.00339.)
TT
(d) Transfer all values of log ^~ from the fifth column on figure 17
to the appropriate boxes on the observing forms illustrated by figures
15 and 16.
(e) Calculate log Ki from the dimensions and mass of the inertia
weight and reduce this value to 20° C (=log K^ 20)- (See paragraph
448, page 187.)
(f) Calculate log {TY and log (Ti)^ as in oscillation observations for
horizontal intensity, using log (1-J-2q:) =0.000010 for unloaded sets
and log (l + 2ai) =0.000012 for the loaded.
(g) Transfer values of log {TY and log {TiY to the proper boxes
on figure 17 and complete the computation of log tt^ ^20-
TEMPERATURE COEFFICIENTS
82. Temperature coefficient of magnetic moment, — The mag-
netic moment of a magnet decreases with increase in the temperature
of the magnet. If the magnet has been properly heat-treated and
stabilized, the process is reversible, that is, the magnetic moment is
fully restored when the magnet is brought back to its original temper-
ature, provided the original increase in temperature is not too great.
The rate of change of the magnetic moment with temperature is
almost but not quite uniform over ordinary temperature ranges, say
from 0° to 40° C. For all practical purposes, the relation between
the magnetic moment. Mi, at a standard or reference temperature, ^i,
and the magnetic moment, M2, at a temperature, ^2, is given by
M2=Mi -Ml CL {U -U) (51)
=MAl-{U-U)q\ (52)
^MAl^{k-t2)q\ (53)
in which 2 is the mean temperature coefficient of the magnetic moment
over that range of temperature. (For a more precise representation
of this relation see appendix II.)
From equation (53),
M,-M, _ M,-M,
34 MAGNETIC OBSERVATORY MANUAL [Ch.
Note that g is defined so as to have a positive value in the usual case
(M diminished by a temperature rise). It is necessary to have due
regard for signs in the above equations; thus, if ti = 20° and ^2=25°,
then ^1-^2= -5°.
83. The temperature coefficient, q, may be determined by deflection
observations, using a magnetometer. The deflecting magnet whose
temperature coefficient of magnetic moment is under test is enclosed
by a temperature bath, figure 18, and the deflection angle, u, is ob-
served at several temperatures ranging from 0° C to about 40° C.
The magnetic moment, M, at each temperature is calculated from
the equation,
M smu ^ ^
in which H, the horizontal intensity at the time u, is observed, and
a constant depending primarily upon the distance r. (See p. 8.)
Then
M=^^ (56)
and log M=log i7+log sin i^+colog C. (57)
Since the change in the deflection angle, Au, even for a large value
of M and a large change of temperature, will be only a few minutes
of arc, any changes in H and D during a set of observations must not
be neglected in the calculation of q. It is necessary to observe D and
H independently and simultaneously with observations of u or to scale
these values from a magnetogram. After the values of log M have
been derived for different temperatures, q is calculated from equation
(54), or more conveniently by logarithms from equation (53). (See
eq. 363, app. I.) Thus,
log M2=log Mi+log [l + (^i -t2) q]
and by equation (366)
log Mo«log Mi+iU -k) log (1+2) (58)
and
log(l + g)^ ''^g^--;'^g^- (59)
log Ml— log M2.
(60)
Suppose log (l+g)=0.000103, then from tables, 1 + 2= 1 000237;
2=0.000237. We may also write
2=2.30 log (1 + 2) approximately (61)
= 2.30X0.000103 = 0.000237,
4.]
CONSTANTS AND CORRECTIONS
35
84. Directions for determination of temperature coefficient
of magnetic moment, — On a magnetically quiet day, set up a mag-
netometer as for determination of horizontal intensity, short magnet
suspended, deflection bar attached.
85. Place the long magnet (or other magnet to be tested) centrally
in the inner chamber of the temperature bath holder, figure 18, and
mount the holder (with magnet) on the bar at such a distance, say
r=30 cm, that a deflection of 15° to 30° is obtained. Place ther-
mometers, fitted with cork stoppers, C, in each end of the magnet
chamber, B, and see that the thermometer bulbs are in contact with
the ends of the magnet so that the magnet may not creep or move
during the tests. Remove the holder (with magnet) to a safe distance.
Figure 18. — Temperature bath for magnetometer magnet.
Observations for Temperature Coefficient
station, Cheltenham; Pier 7
Mgr. No. 37, Scale value, 1.38' per div.
Deflection distance, 30 cm.
Magnet No. Colombia I Dimensions, 5.0 cm.
Date, Thursday, September 20, 1951
Observer, J. B. Townshend
Therm. Nos., 33246 (E.),33257A (W.)
Material, Oersted steel
i
Column
. (1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
75 M.
time
Temperatm-e
observed
Circle reading
Scale reading
Middle
Line
minus
mean
East
West
A
B
Left
Right
Mean
h TO
°C
°C
o / ,/
/ //
div.
div.
div.
div.
1
12 48
Away
303 30 05
30 15
49.0
51.0
50.00
0.00
2
12 54
-0.2
+0.7
287 48 20
48 20
49.3
50.7
50.00
0.00
3
13 03
+ 5.7
6.9
49.2
49.8
49.50
+0.50
4
09
12.2
13.0
48.5
49.4
48.95
+1.05
5
16
21.1
21.4
47.0
47.8
47.40
+2.60
6
22
28.1
28.2
46.6
47.4
47.00
+3.00
7
30
34.9
34.9
49.6
50.9
50.25
-0.25
8
39
42.3
42.2
50.2
50.9
50.55
-0.45
9
40
42.2
42.1
50.1
50.9
50.50
-0.50
10
45
34.9
34.6
50.9
51.1
51.00
-1.00
11
49
28.1
28.1
51.2
51.4
51.30
-1.30
12
53
20.8
21.1
51.8
52.0
51.90
-1.90
13
13 58
12.0
12.4
51.6
52.8
52.20
-2.20
14
14 03
5.3
6.5
51.3
52.3
51.80
-1.80
15
10
-0.5
+0.5
287 48 20
48 20
49.9
50.5
50.20
-0.20
16
14 14
Away
303 30 35
30 40
49.0
51.0
50.00
0.00
17
18
19
20
Notes
(a) Thermometer corrections; apply to observed temperatures:
East: 0°.0, West: 0°.0;
(b) Record corrected mean temperatures in last column of figure 21;
(c) Middle division of scale: 50.00;
(d) This form is designed for use at observatories where declination is West, and where "increasing declina-
tion" means increasing West declination. If this form is used in areas where declination is East, proper
corrections must be made for algebraic signs.
Figure 19.— Observations for temperature coefficient of magnetic moment.
36
MA.GNETIC OBSERVATORY MANUAL
[Ch.
86. Reset the horizontal circle so that the index line of the suspended
magnet bisects the central division of the telescope scale. Observe and
record the circle and scale readings, right and left, line 1, figure 19.
This is the first away position. Note and record the time of the
observation of the scale readings and place a special time mark on the
magnetogram at this same moment.
Temperature Coefficient of Magnetic Moment
Computation of Deflection Angles
station, Cheltenham; Pier 7
Magnetometer No. 37
Mgr. Scale Value, r.38 per div.
Date, Thursday, September 20, 1951
Magnet No., Colombia I
Middle div. of Scale, 50.00
Column
(l)
(2)
(3)
(4)
(5)
(6)
(7)
Scaled ordinates
Scale
Corrected
circle reading
(3+4+.5)
Line
Mean circle
reading
middle
minus
means
B
ordinate
Deflection
u
dmm
hmm
o /
,
,
o /
o /
1
+26.3
+15.4
303 30. 17
0.00
+26.3
303 56. 47
Away
2
25.9
+9.7
287 48. 33
0.00
+25.9
288 14. 23
15 42. 17
3
26.9
+11.8
+0.69
+26.9
15.92
40.48
4
27.5
+9.9
+1.45
+27.5
17.28
39.12
5
27.0
+9.8
+3.59
+27.0
18.92
37.48
6
28.8
+10.1
+4.14
+28.8
21.27
35.13
7
32.0
-3.8
-0.34
+32.0
19.99
36.41
8
32.9
-18.6
-0.76
+32.9
20.47
35.93
9
32.8
-17.5
-0.69
+32.8
20.44
35.96
10
32.9
-7.0
-1.38
+32.9
19.85
36.55
11
31.9
-4.5
-1.79
+31.9
18.44
37.96
12
31.8
0.0
-2.62
+31.8
17.51
38.89
13
30.8
+3.2
-3.04
+30.8
16.09
40.31
14
29.2
+7.9
-2.48
+29.2
15.05
41.35
15
26.4
+12.9
287 48. 33
-0.28
+26.4
288 14. 45
15 41.95
16
+25.7
+17.0
303 30. 62
0.00
+25.7
303 56.32
Away
17
Mean
away.
303 56.40
18
19
20
Notes
(a) Columns (1) and (2) : H and D ordinates scaled from magnetogram. Use same sign as in scaling
ordinates for base-line- values.
(b) Column (3) : Mean circle readings from figure 19.
(c) Column (4) : Conversion of ordinates in last column of figure 19 to minutes of arc.
(d) Scale value of Z>-vario meter: 1.00' per mm.
(e) Column (5): D-ordinate from column (1) converted to minutes of arc; these ordinates have the same
sign as the corresponding ordinates in column (1).
(f) Column (6): Corrected circle readings: Sum of columns 3, 4, and 5.
(g) Column (7) : Mean circle reading, line 17, minus corrected circle readings from column (6).
Figure 20.— Computation of deflection angles, temperature coefficient of magnetic moment.
87. Replace the holder (with deflecting magnet) on the bar, adjust
the circle so that the suspended magnet is centered on the scale as in
paragraph 86. Fill the temperature bath with cracked ice and water.
Allow 5 to 10 minutes for the magnet to reach the temperature of the
bath. Re-center the suspended magnet by adjusting the horizontal
circle, then observe the scale readings, temperature by both ther-
mometers, time, and circle reading, and again place a special time
mark on the magnetogram (line 2, fig. 19).
88. Raise the temperature 5° to 10° by addition of warm water to
the bath. Stir continuously, avoiding mechanical disturbance of the
magnetometer. After the temperature has been held constant for
about 3 minutes, observe and record the scale readings, the tem-
perature by both thermometers, and the time, and again make a
4.]
CONSTANTS AND CORRECTIONS
37
special mark on the magnetogram. Record these readings on line 3,
figure 19. Do not change the circle adjustment for this observation.
All of the changes, Au, in the angle u are to he read directly jrom the
scale in the telescope.
89. Continue deflection observations as in paragraph 88 as the
temperature is increased by regular steps of 5° or 10° up to about
40° C, then reverse the process by reducing the temperature by
uniform steps to 0° C. Observe and record the circle reading for
this last test (line 15, fig. 19).
90. Remove the temperature bath holder, with deflector, and
repeat the away observations as in paragraph 86 (line 16, fig. 19).
91. Computations, — Scale and check the H and D ordinates for
the times given in column 1, figure 19, using the same sign as in
scaling ordinates for base-line values (see par. 434, p. 178).
92. Convert the I) scalings in mm to minutes of arc (col. 5, fig. 20).
93. Calculate the corrected circle readings, that is, the circle
readings corrected to central division of the telescope scale (col. 6,
fig. 20).
94. Calculate the deflection angles, u, (Mean circle reading away,
line 17, minus corrected circle readings, from column 6), and record
the results in column 7.
95. Convert the H scalings to gammas, apply the H base-line
values to get the values of H for each deflection observation (cols.
1 and 2, fig. 21).
Computations for Temperature Coefficient of Magnetic Moment
Date, Thursday, September 20, 1951 from 12h 48°> to 14i> 14°^ (75 M. T.)
Magnet, Colombia I Mgr. 37
Prel. H s. V. 2.55 7/mm Prel. H blv. 18281 y Prel. Mgr. log C=5.86000
Column
(1)
(2)
(3)
(4)
(5)
(6)
Mean cor-
Line
H ord hy
Hy
log Hc«,
log sin u
logM
rected temp.
1
+39
18320
9. 26 293
2
-f25
306
259
9. 43 240
2.83 499
+0.25
3
+30
311
271
165
436
6.30
4
+25
306
259
103
362
12.60
5
+25
306
259
9. 43 029
288
21.25
6
+26
307
262
9. 42 923
185
28.15
7
-10
271
176
981
157
34.90
8
-47
234
088
959
047
42.25
9
-44
237
095
960
055
42.15
10
-18
263
157
9. 42 987
144
34.75
11
-11
270
174
9. 43 051
225
28.10
12
281
200
093
293
20.95
13
+8
289
219
157
376
12.20
14
+20
301
247
204
451
+5.90
15
+33
314
278
9.43 231
2.83 509
0.00
16
+43
18324
9. 26 302
Computation of l
OG (l+g)
(Data scaled from graph of t
vs log M; See fig. 22)
h= 0°. C
log Mi=2.83 504
Formulas:
f2=40°.0
log M2=2. 83 092
log M=log H+log sin u+co
logC
t
2-^1=40°
diff. .00 412
log (1+?)=
=^^«f^^=. 000103
(7=. 000237
9=2.30 log (1+?)
ogM
Af
Fo
r Ae=40°,
AM-=6. 5 cgs.
AM=2.30 MoA log M
Figure 21.— Computations of temperature coefficient of magnetic moment.
38
MAGNETIC OBSERVATORY MANUAL
[Ch.
96. Tabulate log H and log sin u, figure 21, and calculate log M
from equation (57)..
97. Plot log M against observed temperatures, figure 22, and
draw a straight line through the points in such a way that it will
represent the approximate mean rate of change of M with tem-
perature over the observed range. Scale the values of t and log M
from this graph for high and low values of t and calculate log (l+g)
and 2 from equation (60).
98. If the calculated points, figure 22, show a wide scattering or
if the magnetic moment does not return to its original value at 0° C
after the complete 40° temperature cycle, this indicates lack of care
in the observing program, errors of observation (or computation), or
that the magnet has not been properly heat-treated or stabilized.
2.83 500
2.83 000
10 20 30 40 50
Figure 22. — Variation of magnetic moment (log M) with temperature (Cent.)
99. Note that the method described above provides for corrections
for changes in declination and horizontal intensity. For example,
figure 20 shows a change of 3.2 minutes in declination between lines
6 and 7. This is more than half of the change in u due to the change
in magnetic moment for the full range of 40° in temperature. There
was also a change of 36 7 in ^ during this same interval. Without
taking into consideration these changes in H and D, the experimental
results would have little significance, since in this case, the changes
in scale readings due to changes in H and D were greater than those
due to changes in magnetic moment.
100. In this method it has been assumed that the deflection dis-
tance, r, does not change during the operation, and log C is taken as a
constant. Therefore the change in log H-\-\og sin u is identical to
the change in log ^f+log sin i^ + colog C so that C may be given any
value and the same results will be obtained for log (I+2) and g.
However, it is interesting to know how the value of the magnetic
4.] CONSTANTS AND CORRECTIONS 39
moment changes with temperature. The change in the magnetic
moment, M, may be computed for the full range of t as follows:
AM=2.3MX A(log M). For the example given, figure 22, AM=2.3X
684X0.00416 = 6.5 cgs. In this case the magnetic moment was
reduced by 6.5 parts in 684 for a change of temperature of 40° C,
or 1 part in 4200 per °C ( = 0.024% per °C).
101. Proposed alternate method for determination of q. — In
this method the magnet, Ma, to be tested is placed in a temperature
bath, north end of Ma to the south, at such a distance north or south
of a properly oriented H variometer that temperature compensation
of the variometer is approximately effected. The temperature bath
is mounted on a separate, rigid, nonmagnetic stand, independent
of the H variometer and at the proper elevation. Reduce the torsion
in the H fiber until the H spot returns to its original ordinate. Insu-
late the variometer against temperature changes by a nonmagnetic,
nonconducting shield or cover. Take Ma through several temperature
cycles, say 0° to 50° to 0°, and allow the H spot to record con-
tinuously.
102. For each temperature step of 5° or 10°, observe H inde-
pendently, or operate an independent magnetograph so that point
scalings of H can be made.
103. Let Ma be the magnet under test; and let
Ml = the magnetic moment of Ma at a standard temperature
of 0° C;
M2=the magnetic moment of Ma at another temperature, say
50° C;
AM= change in moment due to a change of temperature
=M2—Mi
ti = the standard temperature, 0° C;
#2= the other temperature, 50° C;
A^=the change in temperature = f 2—^1
A/= change in field at recording magnet due to AM;
r= deflection distance in cm;
2= average temperature coefficient of magnetic moment of
Ma between 0° C and 50° C (that is, between U and ^2)-
Then, from equation (11),^ page 5,
2M
By differentiating,
or
and
J^~ y.3
dJ=^^dM
A/=^^AM
AM=i r'A/. (62)
[Ch.
or
Combining (62) and (63),
40 MAGNETIC OBSERVATORY MANUAL
But, from equation (53),
Mi-M2=M,q (t2-ti)
AM=Miq i—At).
1
2
and
104. Example: Suppose:
Mi = 100 cgs; r=12 cm; g=0.00024;
then AM=100X0.00024X(-50)
= — 1.2 cgs (= change in M due to increase of
50° in temperature) ;
r^Af=—MiqAt
r^Af
(63)
(64)
A/=^AM=^(-1.2)
= — 1407 (change in F due to decrease in M).
If the H scale value of the H vaxiometer used in the tests is 3^ per
mm, then the deflection of the H spot on the magnetogram for
140
— A^=50° C will be —=—4:7 mm, a deflection of such magnitude
o
that it may be scaled directly with sufficient accuracy.
Temperature Coefficient of Magnetic Moment
(H Variometer Method)"
Magnet No. 20; Solid cylinder, 5 x 25 mm; i^lnico II; Ma=100 cgs
Scale value of standard H variometer 3.00 7/mm; (Variometer No. 5)
Scale value of variometer used in test 3.00 7/mm; (Variometer No. 7)
Deflection distance, r 12.6 cm
Time
75MT
Temperature
H Scalings,
No. 7
H Scalings,
No. 5
AHn
7
AH,
7
AHi-AIh
7
A
B
Mean
mm
y
mm
7
14:50
15:00
0.0
50.
0.0
50.
0.0
50.
15
62
45
186
20
25
60
75
Y4Y
"15"
"im'
A/=A//7-A//5 = 1267 = 0.00126 cgs
r^Af _ 2000X0.00126 _
^ 2MAt 2X100X(-50) """"^^
» Observations hypothetical.
Figure 23. — Suggested form for observations of temperature coefficient of magnetic
moment. H variometer method.
4.] CONSTANTS AND CORRECTIONS 41
105. Provided the H variometer recording magnet is properly
oriented, it will not be necessary to take into consideration changes
in declination during the tests if magnetic conditions are quiet.
However, corrections should be made for natural changes in H during
the tests even though these changes are small. Figure 23 gives a
hypothetical determination of q by this method.
106. The method is especially useful for determination of 5 for
very small magnets since small deflection distances may be used,
provided the dimension ratios of Ma and Ms are such that the distri-
bution coefficients are small.
107. Sensitivity of temperature-coefficient observations; mag-
netometer method, — ^Magnitude of Au: In determining the tem-
perature coefficient of the magnetic moment, it is desirable to know
in advance about what value of Ait may be expected from a particular
magnet, deflection distance, temperature range, and estimated tem-
perature coefficient. In equation (55) we have,
H C
DifTerentiate,
M sin u
log sin 16= log C+ log M— log iJ (65)
cos u du=-^rj dM—jjdH (66)
sin u M H
J sin u/dM dH\ ,^_.
d^= ("iT zj) (67)
cosu\ M H / ^
But, since
and
^ , /AM A^\
AM=-MqAt
-q^At
(68)
(69)
M
Substituting in (68),
Au (radians) = — tan ul q^M,-\ — jt)
' and
A^ (minutes) = — 3438 tani^UA^+-^\ (70)
108. li H remains constant during the test, equation (70) becomes
(A^) ' = — 3438 tan '?^ (g A^) (7 1 )
and we may expect, for example, the following results, based on
equation (71):
If ^=15°; 5=0.00025; and A^=40°, then Ai/=9'.7;
if u=30°; 2=0.00025; and Ai=40°, then At^=20'.9.
210111—53 4
42 MAGNETIC OBSERVATORY MANUAL [Ch.
109. Errors of scale reductions: In this method it is assumed that
the value of u as represented by the converted scale readings is
identical to the angle through which the circle must be turned to
bring the index of the suspended magnet to the central division of the
scale. In other words, it is assumed that the suspended magnet
remains fixed relative to the ground while the circle and deflector are
rotated through the small angle, Au. (Equation (55) is based on the
assumption that the deflector is always at right angles to the sus-
pended magnet.) It can be shown that the error introduced into
the calculated value of Au by following this procedure is never greater
than 3 seconds of arc for values of u up to 40° provided the scale
deflection does not exceed 20^ In this extreme case, the error in
Au would not be greater than 3'' in 20' (3 parts in 1200), which is
entirely negligible in this work. Likewise the effect of torsion in the
fiber is negligible for an angle of 20'. For example: If 90° torsion
produced a deflection of 5' (an exceptionally large magnetometer
fiber), 20' torsion would produce a deflection of about 0.02'.
INDUCTION FACTOR
110. Induction and magnetic moment. — When a permanent
magnet is placed in a weak magnetic field its magnetic moment, M,
is temporarily changed by induction, the amount of the change being
proportional to the magnitude of the component of the applied field
parallel to the magnetic axis of the magnet. The magnetic moment
will be increased if the applied magnetizing field is in the same direc-
tion as the initial magnetic moment of the magnet, and it will be
reduced if the magnetizing field is reversed. For a well stabilized
magnet and for sufficiently small magnetizing fields, the magnitude
of the change is repeatable and is the same for the field direct and
reversed. The change in the magnetic moment, AM, caused by a unit
magnetizing field is called the induction factor, fi. For any other field,
X, AM—ixX. The ratio of the change in M to the original moment,
for a unit magnetizing field is called the induction coefficient, h. That
is T7^~^- Combining these expressions we have iJL=hM.
111. The induction /ac^or, fjL, is approximately constant for a given
magnet regardless of its magnetic moment. It depends primarily
upon the volume or mass of the magnet, its dimension ratio, and its
''magnetic hardness," the latter in turn being rather sensitive to
temperature changes.^ The induction coefficient, h, changes if M is
altered, being almost inversely proportional to M for the ranges of
magnetization used in geomagnetic work.
112. In this work we are concerned only with the inductive effect
of the earth's field on the magnetic moments of permanent magnets
which may be used in geomagnetic measurements. For this reason
we shall be dealing with comparatively weak fields.
113. If a magnet, Ma, figure 24, whose induction factor is to be
determined, is mounted near a suspended magnet, Mg, so that the
magnetic axis of Ma is vertical and remains in a vertical plane which is
perpendicular to Ms, the latter will be deflected through an angle u,
since there will be a component, /j,, of the field of Ma, normal to Ms.
2 David G. Knapp, Reversible susceptibility and the induction factor used in geomagnetism, U. S. Coast
and Geodetic Survey Special Publication No. 301 (in press, 1953).
4.]
CONSTANTS AND CORRECTIONS
43
The relation between the magnetic moment of the deflector and the
angle u, is given by,
M smu ^ ^
in which H is the horizontal intensity and C is a constant depending
upon the relative positions of the magnets and the distribution
coefficients for that position. If a uniform vertical field, X, is now
Figure 24.— Relative positions of magnets for determination of induction coefficient.
established around the deflector in the same direction as the magnetic
moment of the deflector, the magnetic moment of the latter will be
temporarily increased by an amount, AM=hMX= fxX, and the
suspended magnet will be turned through an additional small angle,
Au, due to the increased magnetic moment (and field) of M«. This
uniform vertical field, X, may be either the vertical component of the
earth's field or an artificially created field.
114. When the north end of the defiector is down and the magnet-
izing field is directed vertically downward, then.
H
C
M+AM sin {u^Au)
H ^ C
M+hMX sin (u+Au)
H
C
M {l-\-hX) sm{u+Au)
(73)
(75)
44 MAGNETIC OBSERVATORY MANUAL [Ch.
Upon reversal of the magnetizing field, X,
—Ji = ^ (74)
Dividing (74) by (73), gives,
l-^hX_sm (u-^Au) _t€LB. It + tan Au
1—hX sin (u—Au) tan it— tan Au
Expanding and collecting,
, 1^ tan Alt .„_.
X tan u
Equation (76) will also apply to the case where the N end of the
deflector is up and the field is directed up, then down. Equation (76)
gives the induction coefficient in terms of the measured quantities,
X, u, and Au. The magnetic moment, M, of the deflector is deter-
mined by deflecting a suspended magnetometer magnet, and the
induction factor, fx, is calculated from the relation, ^i=hM. Two
methods for the determination of jx will be described.
115. Lamont's method;^ induction coefficient, — In this method
the magnet whose induction coefficient is to be determined is mounted
on a special holder attached to the deflection bar of a magnetometer.
The axis of the deflector is vertical and is maintained in a vertical
plane at right angles to the axis of the suspended magnet and at fixed
distances above and below the horizontal plane through the latter.
(See fig. 25.) The deflector. Ma, may be attached to the holder with
the north end of Ma up or down. (The reversal is made about a
horizontal axis parallel to the magnetometer deflection bar.) Its
magnetic moment is increased slightly by the inductive action of the
vertical component of the earth's field, when the north end is down,
and decreased by practically the same amount when the north end
is up. Thus the applied magnetizing field is the vertical component
of the earth's field, +Z or — Z, relative to the direction of the magnetic
moment of the deflector, and u is the angle through which Ms would
be deflected if Z could be reduced to zero. Substituting Z for X in
equation (76) gives
, 1 tsinAu .„„.
h^yy- (77)
Z tan u
One advantage of this method is that a fairly constant vertical field
may be used throughout the test. Since Au is a very small angle the
observations must be made with extreme care in order to obtain
reliable results.
116. It has been shown by Hartnell * and confirmed experimentally,^
figure 26, that for a constant horizontal distance, the defiection angle,
u, will be a maximum when the vertical distance between deflector
3 J. Lamont, Handbuch des Erdmagnetismus (see item 12 of bibliography).
^ Distribution Coefficients of Magnets, pp. 9-10 (see item 3 of bibliography) .
» Terr. Mag., 34, 243, 1929.
4.]
CONSTANTS AND CORRECTIONS
45
and suspended magnet is just one-half the horizontal distance between
them. Since the controlling error in this work lies in the determina-
tion of the small angle, Au, the deflector should be set for a position
Figure 25,
-D. T. M. C. I. W.6 Induction-coefficient apparatus (Lament's method),
witli modified magnet liolder.
which will give a reasonably large angle for u, whereupon Au will be
correspondingly large. Figure 27 gives a typical set of induction
observations made by Lament's method.
117. Directions for determination oj induction coefficients by LamonVs
method. — Set up a magnetometer, level the instrument, and suspend
Department of Terrestrial Magnetism, Carnegie Institution of Washington.
46
MAGNETIC OBSERVATORY MANUAL
[Ch.
the short magnet. Adjust in azimuth until the index of the magnet
is on the central division of the telescope scale. Attach the horizontal
deflection bar and mount the induction apparatus on the bar, on the
east side of the instrument at a horizontal deflection distance, r, of
15 to 25 cm. (Call the suspended magnet Ms and the deflector Ma-)
The induction coefficient of Ma is to be determined.
-+12 S
c
/ ^ \
— -T 1 1 1
+ +
.r^ 00
Deflection angle
• — 4
^-^ /
\ V /
- \ \ ;
^^ Deflector
elevation in cm
-30 -10
1 , 1 . 1 .
+ 10
I.I.
+30
1
Figure 26.— Variation of deflection angle with position of deflector. Curves show changes of deflection
angles for various horizontal deflection distances: a=22.5 cm, 6 = 26.25 cm, c=30 cm, d=40 cm. Curve
e is the locus of the maxima of the other curves.
118. Mount Ma, N end down, on the apparatus so that its center
will be at an elevation of approximately Jr above the horizontal
plane through Ms- In this position the jy component of the field
of Ma at Ms will deflect Ms counter-clockwise looking down.
119. Readjust the horizontal circle until the index on Ms is in the
center of the telescope scale when Ms is at rest. Read the horizontal
circle. This is the reading on line 1, figure 27. Mg has been deflected
through the angle NOF, figure 28.
120. Reverse Ma so that its A^ end is up and above the bar. Ms
will now be deflected clockwise, looking down. Adjust the instrument
in azimuth again until the index of Ms is on the central division of the
telescope scale when Ms is at rest; the deflection angle is now NOC,
figure 28. Record the circle reading on line 2, figure 27.
121. Continue observations for the remaining positions on both
4.]
CONSTANTS AND CORRECTIONS
47
east and west sides as indicated in figure 27 and record all observations
in their correct order on the form. Figure 24 shows the angles and
positions for line 4 of figure 27.
Induction Coefficient
(Lamont's method)
Place, Cheltenham Magnetic Observatory, Pier 7.
Magnetometer, 37
Magnet, 37L (Tungsten Steel); Magnetic Moment, 874
Temperature of Magnet, 27°.8 C; Mass of Magnet, 27 gm.
OBSERVATIONS
Date, March 22, 1938.
Observer, J. H. Nelson
Apparatus, Lamont's (modified).
Length of Magnet, 9.28 cm.
Line
Position of deflector
Side Holder N end of Af<,
East
East
East
East
West
West
West
West
Up
Up
Down
Down
Down
Down
Up
Up
Down
Up
Down
Up
Up
Down
Up
Down
Horizontal circle
95 58 40
117 05 30
117 12 50
96 02 30
117 11 30
95 59 10
96 06 10
117 15 20
B
59 00
(1)
05 40
(2)
13 10
(3)
02 40
(4)
11 50
(5)
59 20
(6)
06 30
(7)
15 40
(8)
Mean
95 58 50
117 05 35
117 13 00
96 02 35
117 11 40
95 59 15
96 06 20
117 15 30
COMPUTATIONS
Line
Position of Mc
East
West
East
West
Operation
(3)-(l)
(8) -(6)
Mean: 2 u+2Au =
(2) -(4)
(5) -(7)
Mean: 2u—2Au =
4tt = (ll) + (14)
u
4 Att=(ll)-(14)
Au
Angles
21
14
10
21
16
15
21
15
12
21
03
00
21
05
20
21 04 10
42 19 22
10 34 50
11 02
02 46
Time, 75 Meridian.
Began, U^ sO"*.
End, 14 44.
Vertical Intensity (Z), 0.54000.
colog Z
log sin Au
log cot u
log /*
log M
logM
h
0. 2676
6. 9057
0. 7286
7. 9019
2. 9415
0. 8434
6.97
0. 00798
^ ^. .1 tan Au , ,^
Equations: h=— — ; n=hM
Z tan u
Figure 27. — Observations for Induction Coefficient (Lamont's method).
122. Computation of h and jjl. (Refer to fig. 27.)
Line 9. — Reading 3 minus reading 1, gives one value of 2u-{-2Au.
Line 10.— Reading 8 minus reading 6, gives another value of 2u-\-2Au.
Line 11. — Take the mean. It is the largest angle, AOF, shown in
figure 28 and represents the double deflection angle for the condition
that Ma is used with the A^ end down, and its magnetic moment
increased by the inductive action of the vertical component, Z, of
the earth's field.
Line 14. — In like manner the angle 2u—2Au, COD, is the smaller of
the large angles, in figure 28, and represents the double deflection
angle for the condition that Ma is used with its A^ end up, and its
magnetic moment decreased by induction.
Lines 11 and 14. — Solve these equations simultaneously to give values
for both u and Au.
Determine the magnetic moment of Ala by the method described in
paragraph 319, page 129, or other suitable method.
48
MAGNETIC OBSERVATORY MANUAL
[Ch.
Scale the vertical intensity, Z, from the magnetogram, or use the mean
value for the month.
Compute h from equation (77) and ju from the relation, n=h M.
123. Repeat the whole operation using a slightly different value
for horizontal distance or vertical distance or both.
124. Induction coefficient.
Nelson's method J — In this meth-
od the applied field is produced by a
solenoid energized by a few dry cells.
Figure 29 shows an observatory
magnetometer upon which are
mounted a special deflection bar
and two solenoids. These solenoids
are identical and are suspended from
the deflection bar at equal distances
from the suspended magnet. The
coils are connected in series so that
the current (and the interior field
produced) in each is the same. The
terminals are so connected that the
fields in both coils will be directed
upward or downward at the same
time. By adjustment of the sole-
noids along the bar the components
of the solenoid fields at the center
of the suspended magnet of the
magnetometer will be practically
equal and opposite as indicated by no deflection of the suspended
magnet, Ms, when the coils are energized. A precision milliammeter
in the circuit indicates the current.
125. The axial magnetic field at a point on the axis of a solenoid of
finite length ^ is,
Figure 28. — Deflection angles in the measure-
ment of the induction coefficient of a perma-
nent magnet.
Xs = 0.1 ni (47r
0)1
C02)
(78)
in which coi and aj2 are the solid angles described by drawing elements
from the chosen point to the last turn of the wire at each end of the
solenoid.^
By referring to figure 30, coi and C02 may be evaluated in terms of the
coil dimensions and the position on the axis of point P,
or.
X, = 0.2Trni
Xs=i Cs
+ x
_V^^+6+^y V'^+S-^J
(79)
7 J. H. Nelson, Terr. Mag., 43, 159, 1938.
* Nelson, op. cit., p. 161.
> It may be noted that the formula, 0.4 ir n i, for the axial field in an infinite solenoid may be obtained
from equation (78) by making wi and w2 equal to zero, or from equation (79) by making I infinite.
4.]
CONSTANTS AND CORRECTIONS
49
Figure 29.— Observatory magnetometer equipped with apparatus for measuring induction coefficient
(Nelson's method).
2 "^ ^^ 2
Figure 30. — Diagram of solenoid, induction coefficient apparatus
50 MAGNETIC OBSERVATORY MANUAL [Ch.
in which
Xs = axial magnetic field on axis at point P, distance x cm from
coil center;
n = turns of wire per cm length of solenoid ;
i = current in amperes;
I = length of solenoid in cm;
h =coil radius in cm;
X = distance from coil center to point P on axis; x is plus in
the direction of the field Xs]
X
Cs =^=field per ampere at P (the coil factor).
C.=0.2Trn
I, I
2 + ^ , 2
_V*^+(l+^y V^^+G-^y_
(80)
126. The average value of the field, Xs, over the length of the
magnet is related to the average of the coil factor, Cs, over the same
distance by
Xs=i Cs.
When working with one particular solenoid it is convenient and prac-
tical to plot the coil factor, Cs, as a function of x, as illustrated in
figure 31. The solenoid is the one described in paragraph 128.
127. The average value of the coil factor, Cs, over the length of the
magnet to be tested, may be scaled approximately from the curve of
figure 31, using the method of equal areas commonly employed in
scaliQg averages.^^
128. The coils used originally are still in use and have the following
specifications:
Total number of turns in each coil 825
Over-all length of each coil (=1) 23.15 cm
Turns of wire per cm of length of coil {=n) 35.64
Radius of each coil (=b) 3.135 cm.
Figure 31 shows values of the coil factor Cs for this coil plotted as a
function of x. Sufficiently accurate values of Cg or Cs may be scaled
from the curve.
Example: Suppose the current is 35.0 milliamperes, the length of the
magnet to be tested (deflector) is 9.28 cm, and the deflector is centered
in the coil. Scale the average of the Cs curve from 2:=— 4.64 to
x=4-4.64. The scaled average is 43.0 oersteds per ampere. The
average field Xs over the length of the magnet is then,
Xs=i'Cs
=0.035X43.0 = 1.50 oersteds.
'» A solenoid will produce approximately the same field as a magnet at all points at great distances, r, in
any direction, e, if (a) the distance r is very large compared to the dimensions of the solenoid and magnet,
(b) the solenoid and the magnet are centered at the same point, and (c) the solenoid axis coincides with
the magnetic axis of the magnet. The equivalent magnetic moment of the solenoid under such conditions
is,
M»0,1 irnilb^
in the notation of figure 30,
4.]
CONSTANTS AND CORRECTIONS
51
129. Directions for determination of induction coefficient; Nelson's
method. — Set up the apparatus as shown in figure 29, using a deflection
distance of 25 to 30 cm for each coil. See that the coils hang vertically
and are free from pendular motion. Suspend the short magnet, Ms
in the magnetometer and bring it to rest in line with the central division
of the telescope scale.
130. Connect the east coil in series with two dry cells, a milliam-
meter, rheostat, and a reversing switch, figure 32. Keep the lead-in
wires and the ammeter well away from the magnetometer.
-I 1 1 1
1 1 1 1 1 1 1
1 1 1 1 1 1 1
1 1 1 1-
-10
45
-5 1
1 5
10
45 —
—
—
0>
E
—
-
(0
1
(A
i2
40
. 40 —
J)
O
35 /
\ 35 —
o
1
2
— 30/
\j
"" /-lO
i/ 1 1 1
-5 1
1 1 1 1 1 1 1
1 5
1 1 1 1 II
10 \ ~
1 1 \l
I
Distance, x, along coil axis measured from coil center, in centimeters
Figure 31.— Relation between coil factor and distance along axis from center of coil.
131. Close the reversing switch and note the direction of the deflec-
tion of Ms. If the scale reading increases, the field within the coil is
directed upward. Mark this position of the switch up. If the scale
reading decreases mark this position down. Repeat this operation
with the west coil only in the circuit and see that it is so connected that
its field will be directed as indicated by the switch. Note that when
the field is directed downward in the east coil the deflection will be in
the direction of decreasing scale readings but for the west coil the
scale readings will increase.
132. Now connect the coils in series, leaving all other connections
unchanged. Close the switch to up. If the suspended magnet is
deflected, adjust one of the coils along the bar until the deflection is
reduced to zero for both up and down positions of the switch. Adjust
the horizontal circle so that the suspended magnet is in line with the
center of the telescope scale. Observe and record several deflections
for both up and down positions of the switch. Even though these
52
MAGNETIC OBSERVATORY MANUAL
[Ch.
deflections are as small as 0.1 division they should be recorded together
with ammeter and horizontal circle readings; lines 1, 2, 3, and 7
(deflector away), figure 33.
133. Without disturbing the coils or leads, mount the deflector, Ma,
whose induction coeflftcient is to be determined, centrally within the
east coil, north end down. With no current in the coils readjust the
circle until Ms is again in line with the central division of the telescope.
3f^
Down (j Up
-@ — 'I'I'h
AAAAAAAA/^ AAAAAAA/ —
Figure 32. — Wiring diagram for induction coefficient apparatus (Nelson's method), showing solenoids
Ci and C2; milliammeter, MA; rheostats, Ri and R2, for fine and coarse adjustment; reversing switch
and battery.
Observe and record the circle reading; line 7, deflector in coil, and the
scale readings in the telescope, line 9.
134. Energize the coils, field directed upward, using suflacient current
to produce a deflection of several scale divisions. Bring the magnet
to rest, observe and record the scale readings and the ammeter reading.
Repeat for current reversed. Repeat these operations several times;
lines 10-14.
135. Repeat the operation described in paragraph 133, and record
scale readings and circle readings; line 15 and line 8 (deflector in coil).
136. Remove the deflector and repeat the operation described in
paragraph 132, no current in the coil and deflector away; line 8 (de-
flector away) and lines 4, 5, and 6.
137. Measure the magnetic moment of the deflector by the method
described in paragraph 319, page 129.
138. Calculate h and /z, using the form of figure 34. Record all
pertinent data such as kind and size of deflector, magnetic moment of
deflector, horizontal and vertical distances of deflector relative to Mg.
139. Repeat the whole operation, changing the current slightly for
the second set.
140. Additional notes on induction coefficients. — Observations are
taken with deflector east, below the bar, north end down.
141. The deflection u, caused by the magnet alone, will be in the
direction of increasing circle readings, that is, clockwise, and will be
taken as positive (+).
4.]
CONSTANTS AND CORRECTIONS
53
station, Cheltenham, Pier 7
Magnetometer No. S7
Induction apparatus No. J
Induction Coefficient
(Nelson's Method)
East coil No. 1
Date, I'ueiiday, August 22, 1960
Observer, R. L. Viets
Magnet No. RIC-14
Deflections for Balance Correction
Magnet away
Before
After
Direc-
tion of
field
Up
Down
Up
Up
Down
Up
Current
in coil
ampereo
0.250
0.250
0.250
0.251
0.250
0.250
Mean current 0.250
Scale readings
Left
50.1
49.8
60.2
49.7
49.6
49.7
Right
60.4
50.2
50.5
50.1
50.8
Mean
60.26
50.00
50.35
60.16
49.85
60.26
Correc-
tion to
2Am
(U)-(D)
div.
0.25
0.S5
0.80
0.40
Correction for balance -\-0.82
Description of
Magnet
Material:
Alnico II
Chill cast
Density 7.1
Dimensions:
0.62 cm diam.
4.9 cm long
Mass: 10.6 g.
Moment:
784 c. g. s.
Solid y Cylindrical \/
Hollow Octagonal
Other
Deflections for Angle u. No Current
Before
After
75 M. time
14 07
14 16
Magnet away
47 4i
47 41
227 42
227 41
Mean
47 42
47 41
Mean ui 47 42
Magnet in coil, east side,
north end down
58 33
58 33
238 34
238 34
Mean
68 34
58 S4
Mean U2 58 S4
u = U2—Ui = 10 52
Deflections for 2Au
Magnet in coil, east side, north end down
75 M. time
h m
14 10
14 IS
Direc-
tion of
field
None
Up
Down
Up
Down
Up
None
Current
in coil
amperes
None
0.260
0.260
0.249
0.249
0.248
None
Mean current 249
Scale readings
Left
div
49.1
62.3
46.6
52.3
46.3
62.3
Right
61.0
64.0
47.3
54.0
47.4
54.0
51.3
Mean
div.
50.05
53.15
46.90
53.15
46.85
53.15
50.06
2Au un-
corrected
(U)-(D)
6.26
6.26
6.30
6.30
Mean (U) - (D) (2Aw uncorrected) 6.28
Approx.
horizontal
deflection
distance
of magnet
250
Approx.
vertical
deflection
distance
of magnet
170
Remarks:
Temp. 27° C.
Computed by RL V
Checked by CB
Figure 33.— Observations for induction coefficient.
Abstracted by WW
54
MAGNETIC OBSERVATORY MANUAL
[Ch.
Computations fob Induction Coefficient
(Nelson's Method)
Station, Cheltenham, Pier 7
Mgr. s. v., 1.S8' per div.
Date, August 22, 1950
Induction apparatus No. /
Observer, R. L. Viets
East coil No. 1
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18).
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
Magnet No
Magnet length
Magnetic moment- .
Coil length
Coil diameter
Total turns
Coil turns per cm...
Average Coil factor _
Current in coil
log i
.(i)
XM)
-d)
-in)
XC\)
-(0
-(10)+(11)
log O^-
logX
Average field of coil
over length of magnet 'xl.
Mean 2A^i uncorrected (U) — (D)
Correction lor balance (U)-(D)
2Aii corrected (14) -(15)
2A«
A?i
. = (20) +(21) +(22)
log tan Am
log cot tt
colog X,
log ft
log M
log/iM=logM = (23)+(24)
h induction coefficient
II induction factor
RIC-14
4.9
784
28.15
6.27
8'45
35.6J,
43.1
0.249
9.396
1.634
1.030
10.7
6 28
0.32
5.96
8.' 22
4' 07"
10° 52'
7.078
0.717
8.970
6.765
2.894
9.659
0.000 58
0.456
c. g.
cm
cm
amp.
div
div.
div.
e.g.
cm
cm
amp.
c. g. s.
div
div.
div.
cm
c. g. s.
cm
cm
amp.
c. g. s.
div.
div.
div.
X.=i C.
h=
1 tan Am
Computed by RL V
X, tan u
Checked by CB
ti=hM
Abstracted by WW
Figure 34. — Computations for induction coefficient.
142. Under the conditions of paragraph 141, the deflection Au,
caused by the applied field must be taken as positive (+) when that
field is directed downward, that is, in the same direction as the magnetic
moment of the deflector, since the magnetic moment of Ma will be
increased by a small amount AM.
143. Note that decreasing scale readings in the telescope (scale in
telescope) correspond to increasing circle readings. Therefore when
the magnetizing field is applied, decreasing scale readings indicate an
increase in the magnetic moment of the deflector or an increase in the
applied magnetizing field, /j,, acting on Mg.
144. If the deflection, Au, for field up or field down, with magnet
away, is in the same direction as the corresponding deflections when
the deflector is within the coil, the correction for coil balance is
negative.
145. Correction for lack of balance of the coils. — This correction is
obtained while the suspended magnet is in the magnetic meridian but
is applied to the deflection angle 2Au, but 2Au is measured while Ms
is deflected through an angle u. The effect of the thus unbalanced
field on the deflection angle varies from a minimum when Ms is in the
magnetic meridian to a maximum (infinite) when the defiection, u,
4.]
CONSTANTS AND COKRECTIONS
55
reaches 90°, at which point Ms becomes unstable. The relation
between the true correction, c, for unbalance and the observed cor-
rection, x, is given by c= ? u being the deflection angle when the
cos 111
deflector is within the coil. The correction to x is negligible for values
of u up to 30°. Table 4 gives the factors to be applied to x to obtain
c for different values of u. These factors are always positive ( + ).
TABLE 4.— Correction factors
u
Factor
o
1.00
10
1.02
20
1.06
30
1. 15
50
1. 56
70
2.92
SUMMARY OF CONSTANTS
146. The constants of a magnetometer should be summarized in
convenient form for field or observatory use as shown in figure 35.
Critical tables should be prepared in all cases where their use will
result in conservation of time and labor involved in routine compu-
tations, no matter how simple these computations may be. For
TT — —
example, in figure 35, any value of log -^r between 5.79587 and 5.83934
will yield the same value of log ( 1 +/a^ j to an accuracy of ±5 in the
sixth decimal of logs.
ADJUSTMENT OF LOG C
147. Magnetometer, — Comparison observations usually indicate
that the value of H determined by a magnetometer differs consistently
from that determined by the sine galvanometer or other standard
instrument. This difference cannot usually be ascribed to one specific
error in the determination of the constants of a magnetometer but is
probably a cumulative effect of small uncertainties in measuring
several of the constants. When the correction is small and con-
sistent it may be reduced to zero (approximately) by readjustment of
log C for each deflection distance. Observations for adjustment of
log C consist of not less than four complete sets of the usual observa-
tions for H made at each deflection distance for which log C is to be
determined. The average value of H for each whole set of observa-
tions is derived from the magnetograph by scaling average H ordi-
nates for those time intervals during which the observations w^ere in
progress. The H base-line value is derived from measurements of H
with the standard absolute instruments at the observatory. (At
Cheltenham Magnetic Observatory the sine galvanometer is used as
the standard for horizontal intensity.) Note also that a standard
56
MAGNETIC OBSERVATORY MANUAL
[Ch.
Constants of Magnetometer 31
Effective Oct. 22, 1941
Moment of Inertia, log w^ K at 20° C 3.58 233
Temperature coefficient of magnetic moment, q 0.00 018
Log (1 + ^) 0.00 008
Induction factor, /j. 3.5
Scale value of telescope scale, min. per div. 1.37
Middle division of telescope scale 50.00
Declination:
Reduction to middle =1.37 (50.00 — mean scale reading)
Oscillations:
Correction to log T^ for temperature; log [1 + {t-t')q]= +0.00 008 (^
-n
Correction to log T^ for induction ; log M + nj^ j from table below
Correction to log ir^K for temperature; +0.00 001 (^-20°)
Deflections:
For r = 30 cm; log C=5.86 826+ (0.00 0023) (20°-^
For r = 40 cm; log C=5.49 406+ (0.00 0023)(20°-0
Log ilf2o = log M+0.00 008 (t-20°)
^«^i
log(l+.f)
^«^5
IOg(l+Mf)
^«^i
logM+
4)
5. 79 587
6. 00 848
6. 12 990
0. 00 010
0. 00 016
0.00
021
5. 83 934
6. 03 563
6. 15 059
11
17
22
5. 87 885
6.06 119
6. 17 033
12
18
23
5. 91 506
6.08 532
6. 18 922
13
19
24
5. 94 848
6. 10 818
6. 20 732
14
0. 00 020
0.00
025
5. 97 952
6. 12 990
6. 22 469
0. 00 015
6. 00 848
Notes
Log (1+0.0000116 d)2=0.00001 (rate of chronometer in seconds per day
)
Log y^ =0.00008 (h in minutes)
Figure 35.— Constants of magnetometer.
value of H may also be determined directly by simultaneous obser-
vations with the standard instrument and the instrument being
standardized.
148. We make use of the following relationships:
M-
HM
H
log M=log ^M-log H
log C=\og iJ— log M+log sin u
4.] CONSTANTS AND CORRECTIONS
From equation (448) , appendix I,
A log C
At
= -3(0.434)a.
149. The H value from the magnetogram is combined with HMt
TJ TJ
from observed oscillations to give a standard value of ^- Log y^
is added to log sin u for the corresponding deflection observations
to obtain log Ct. Log C at 20° C is derived from log Ct by applying
proper correction for temperature. (See eq. (446), app. L) Figure
36 is a sample set of computations for adjustment of log C for Mag-
netometer RIC-3626. The computed values are compared with the
values calculated from dimensions; lines 17 and 18, figure 36. Similar
procedures may be followed in deriving certain constants for other
instruments such as the Quartz Horizontal Magnetometer.
MAGNETIC STANDARDS
150. International standards, — A new or reconditioned instru-
m.ent is not ready for use until it has been standardized. Instrum.ents
that are apparently identical yield results that are not identical,
despite the exercise of the utmost care in their construction and
standardization. For this reason, there m.ust be occasional intercom-
parisons to safeguard the accuracy of even the best available instru-
ments. It is no simple matter to make correlated measurements of a
shifting magnetic elem.ent in such a way that the purely instrumental
discrepancies may be isolated from other effects. Programs for such
comparisons are described by Hazard ^^ and others. General specifi-
cations for national and international standards are recommended and
published from time to time in the Bulletins of the International
Association of Terrestrial Magnetism and Electricity.
151. Inter comparison of instruments, — In the United States,
all of the m_agnetic instruments of the U. S. Coast and Geodetic Survey,
as well as those belonging to other agencies both foreign and domestic,
are compared (standardized) directly at Cheltenham Magnetic Observ-
atory or indirectly by comparisons first at Cheltenham and then at
other observatories by use of an intermediate instrument.
152. The Quartz Horizontal Magnetometer (QHM)^^ has proved to be
quite useful for international comparisons in horizontal intensity
measurements and the Magnetometric Zero Balance (BMZ)^^ holds a
similar position for vertical intensity. For magnetic declination com-
parisons the ordinary magnetometer is, of course, the most satisfactory
instrument for use in indirect com^parisons.
153. The Sine Galvanometer,^'^ figure 37, requires special standard-
izations of its standard cells, potentiometer, and standard resistances
at regular intervals of one to two years. There is no indication that
the coil constant of sine galvanom.eter No. 1, now at Cheltenham, has
changed appreciably with time.
" Dir. for Mag. Meas., pp. 37-40, (see item 4 of bibliography).
12 D. la Cour, Le Quartz-Magnetometre QHM (see item 9 of bibliography).
'3 D. la Cour, The Magnetometic Zero Balance, the BMZ (see item 10 of bibliography).
1* S. J. Barnett, A sine galvanometer for determining in absolute measure the horizontal intensity of the
earth's field (see item 1 of bibliography).
210111 — 53 5
58
MAGNETIC OBSERVATORY MANUAL
[Ch.
Computation of Log C (H known)
Place Cheltenham, Pier 6 H. s v: 2.53 Date: Oct. 31-Nov. 1, 1951
Magnetometer: RIC-3626 7/mm Observer: R. L. Viets
Long Magnet: 4.9 cm Induction factor: r25 = 24.996 r3o = 29.988
Short Magnet: 4.0 cm m = 0.43 Deflection bar: Duralumin
Oct. 31
Nov. 1
Oct. 31
Nov. 1
Line
Terms
Step
Begin 14:03
End 15:49
15:09
16:40
14:03
15:49
15:09
16:40
1
2
3
4
5
6
Distance
Set
hmm
Bh
H
4+5
25 cm
1
28.3
72
18287
18359
25 cm
3
36.5
92
18288
19380
30 cm
1
30 cm
3
7
8
9
log HMt
log//
log Mt
7-8
1.86 547
9.26 385
2.60 162
1.86 626
9.26 435
2.60 191
10
11
12
1 H
log sin u
log Ct
8-9
10+11
6.66 223
9.44 490
6.10 713
6.66 244
9.4.4 482
6.10 726
6.66 223
9.20 776
5.86 999
6.66 244
9.20 780
5.87 024
13
14
15
16
t
20-t
Corr'n
log C20
12+15
25.20
-5.20
+ 0.00 015
6.10 728
21.90
-1.90
+ 0.00 006
6.10 732
+ 0.00 015
5.87 014
+ 0.00 006
5.87 030
17
18
19
Mean 6.10 730
log C20 (dimensions) 6.10 733
log C20 (adopted) 6.10 730
5.87 022
5 87 013
5.87 022
Notes
Line 3. Scaled from magnetogram Line 13. Mean temperature of
or taken from standardization cards, deflections (erect and inverted).
Line 4. H ordinate in gammas. Line 15. Correction to reduce
Line 5. H base Une value. log C« to log C2o=- — .0.000029 (20-0
Line 6. H from magnetograph for duralumin; —0.000024 (20—
(same for all distances of same set), for brass.
Line 7. Mean value from oscilla- Line 18. Log C20 from dimensions
tions (erect and inverted). of magnets and deflection distances.
Line 11. Log sin u; mean value
from deflections (erect and inverted) .
Figure 36.— Computation of log C.
154. Summary of comparisons. — The results of comparisons of
certain instruments are usually tabulated as shown in figures 38, 39,
40, and 41. For inclination and declination the final correction is
shown as an additive index correction in minutes of arc. In compar-
isons for horizontal intensity, difference in results obtained with two
magnetometers may be due primarily to errors in the adopted con-
4.]
CONSTANTS AND CORRECTIONS
59
stants, assuming of course that the instruments have no magnetic
parts. All of these quantities enter factorially in the formulas from
which H is derived, consequently the effect on II of an error in any one
Figure 37. — The Sine Galvanometer.
of them can better be expressed by a constant of multiplication rather
than by an additive index correction. The errors in the constants are
essentially independent of H. Hence, the correction is assumed to be
proportional to H and is given as a factor
H
60
MAGNETIC OBSERVATORY MANUAL
[Ch.
Correction to Declination Observed With Magnetometer No. 19
AND Computed With Constants Dated October 10, 1941
Cheltenham Magnetic Observatory; Pier 6
Observer
Sets
1
2
3
4
WEW
Apr. 21
11:37-11:46
WEW
Apr. 21
12:06-12:15
WEW
Apr. 21
13:16-13:25
WEW
Apr. 21
13:32-13:41
Date of observation 1947 ._ __ _. .
Time interval (75 M. T.)
-8.6
7°08.'6
-8.8
7° 10 /I
-8.4
7°12.'9
-8.6
7° 13. '8
Observed declination
Variometer ordinate in mm.*
-13.1
1.00
-13.1
-11.8
-ii.'s
-9.0
'-9."0
-8.1
"-s'i
Scale value in '/mm. b
Variometer ordinate in minutes
7°21.'7
7°22.'2
7°21.'9
7°22/2
7°21.'9
7°22/2
7°21.'9
7°22/2
True base-line value "
Correction to W. declination
Mean correction
-f0.'5
+0.'3
+0.'3
+0/3
+0/4
a Ordinates corrected to 100.0 mm. shrinkage distance. Parallax allowed for.
b Scale value for a shrinkage distance of 100.0 mm.
Includes pier correction O.'O added to regular base-line value to reduce to pier 6 and instrumental
correction 0/0 added to regular base-line value to reduce to International Magnetic Standard.
Total correction O'.O added to regular base-line value.
Figure 38.— Standardization of magnetometer for declination.
Correction to Inclination Observed With Earth Inductor No. 106
Cheltenham Magnetic Observatory; Pier 5
Observer
Date of observation 1946
Time interval (75 M. T.)
Observed value of inclination
Value computed from Mgph. No. 5 ».
Index correction
Mean correction
Sets
1
2
3
4
WEW
June 18
15:38-15:47
WEW
June 18
15:49-15:57
WEW
June 18
16:01-16:10
WEW
June 18
16:18-16:26
71°18.'8
71°18.'6
71°19/2
71°19.'2
71°19.'4
71°19.'5
71°18.'7
71°18.'6
-0/2
O.'O
+0/1
-0/1
O.'O
a Includes a pier correction —0/2 added to dip to reduce to pier 5, and an instrumental correction
O.'O added to dip to reduce to International Magnetic Standard.
Figure 39. — Standardization of earth inductor for inclination.
4.]
CONSTANTS AND CORRECTIONS
61
Correction to Horizontal Intensity Observed With Magnetometer
No 19 AND Computed With Constants Dated October 10, 1941
Cheltenham Magnetic Observatory; Pier 6
Sets
1
2
3
4
WEW
June 30
12:41-13:46
WEW
July 5
10:07-11:21
WEW
July 5
12:02-13:07
WEW
Julys
14:01-15:03
Date of observation 1945
Total time interval (75 M. T.)
M-"("'4).-('°^s).
Observed value of log Af2o
-f 0. 00024
2. 80660
18235
+0. 00012
2. 80671
18195
+0. 00009
2. 80665
18197
+0.00010
2. 80666
18208
Observed value of Hia y
42.8
2.67
114
27.6
'"74
29.1
""78
33.4
""89
Scale value in 7 'mm **
18121
18120
18121
18120
18119
18120
18119
18120
True base-line value***
Correction to H- (AH)
-1
-1
+1
+1
AH
^^j 0.00000
*Ordinates corrected to 100.0 m m . shrinkage distance. Parallax allowed for.
**Scale value for a shrinkage distance of 100.0 mm.
***Includes pier correction +87 added to regular base-line value to reduce to pier 6 and instru-
mental correction —27 added to regular base-line value to reduce to International Magnetic Stand-
ard. Total correction +I7 added to regular base-line value.
Figure 40. — Standardization of magnetometer for horizontal intensity.
Correction to Horizontal Intensity Observed With QHM No. 48
and Computed With Constants Dated April 23, 1942
Cheltenham Magnetic Observatory; Pier 6
Observer
Sets
1
2
3
4
WEW
Sept. 6
15:10-15:19
WEW
Sept. 7
08:52-08:59
WEW
Sept. 24
11:31-11:39
WEW
Sept. 24
11:45-11:51
Time interval (75 M T )
Mean temperature in °C
26.2
18234
24.9
. 18190
23.6
18199
23.5
18199
38.7
2.67
103
21.7
""58
24.9
""66
24.9
""66
Scale value in 7/mm.**
Observed base-line value
18131
18120
18132
18120
18133
18120
18133
18120
Correction to H; (AH) .
-11
-12
-13
-13
Mean correction
AH -12
AH
1 nn(M\(\
H
♦Ordinates corrected to 100.0 mm. shrinkage distance. Parallax allowed for.
**Scale value for a shrinkage distance of 100.0 mm.
***Includes pier correction +37 added to regular base-line value to reduce to pier 6 and instru-
mental correction —27 added to regular base-line value to reduce to International Magnetic Stand-
ard. Total correction +I7 added to regular base-line value.
Figure 41.— Standardization of QHM for horizontal intensity.
CHAPTER 5. OPTICAL SYSTEMS
AND
PHOTOGRAPHIC REGISTRATION
COLLIMATION
155. Both absolute and variation instruments involve applications
of simple geometrical optics which are not adequately explained in
standard books on the subject. The ensuing discussion assumes a
basic knowledge of the elementary principles involved.
156. Simple lens, — In figure 42, let r be the radius of curvature of
the convex side of the planoconvex lens, L. The optical axis is defined
as the line, CO, perpendicular to
both surfaces. Lenses are usually
centered, that is, they are cut and
edge-ground so that the optical axis
passes through the geometric center
of the lens itself.
157. Parallel rays of monochro-
matic light falling upon the plane
side of L, figure 42, at normal inci-
dence will converge at a point, P, called the principal focus. The
distance, OP, is called the focal length of this simple lens. A plane
passing through P and perpendicular to the optical axis, PC, is called
the focal plane of the lens.
158. If a point source of light be placed at the principal focus, P,
figure 42, the rays will be parallel after passing through the lens.
This process of producing parallel rays is called collimation. If an
illuminated scale be placed at P and a telescope, previously focused
on a distant object, be placed at T and directed parallel to OP, the
image of the scale will be in sharp focus in the telescope. Many
Figure 42.— Planoconvex lens.
For a thin lens, the index of refraction
IS l+r/f
Figure 43.— Illustration of both the Gaussian ocular and the modified
Gaussian ocular in an autocoUimator.
magnetometer magnets are made in the form of hollow cylinders
with a scale or index lines ruled on glass fixed in one end and a col-
limating lens fixed in the other end. Such a system is called a coh
Kmator.
159. AutocoUimator; Gaussian ocular. — In figure 43, a tele-
scope, T, is provided with a scale, S, ruled on a glass reticle mounted
in the principal focal plane of the objective lens, L. The eyepiece is
fitted with a thin, transparent glass plate, G, set at 45° to the optical
axis of the telescope. Diffuse light, after entering a small window at
W, in the side of the eyepiece, is partly reflected and illuminates the
scale. The telescope then serves as a collimator and if a plane mirror
is interposed at M, normal to the axis of L, a real image of the scale
will be found superimposed upon the scale itself. The scale and its
62
OPTICS AND REGISTRATION
63
image may be brought into precise coincidence by rotation of the mir-
ror or telescope in incUnation or azimuth or both.
160. Modified Gaussian ocular. — In this type of ocular, only the
upper portion of the central line of the scale is illuminated. This is
accomplished by using, instead of the inclined glass plate, an adjust-
able 90° prism mounted in the ocular close to the vertical line of the
scale as shown at P, figure 43. Hence only the image of a portion of
the central vertical line would be formed in the focal plane of the
telescope. By adjustment of the telescope in azimuth and inclination
the image of this central line may be made to fall upon any part of
the scale and serves as an index of relative angular motion between
the mirror, M, and the axis of the telescope. This type of optical
system is used on some magnetometers.
MAGNETOGRAPH OPTICS
161. Optical scale value, — If an optically plane mirror is inter-
posed at any point, M, figure 44, so that its reflecting surface is normal
to OP, the reflected rays will be parallel and will come to focus pre-
^
5
^
_J(
^P
^=^
I
P
P'
n
\29^
L
i
Figure 44. — Optical lever.
Figure 45. — Planoconvex
mirror.
cisely at the position of the source, P. This will be true regardless of
the distance between the mirror and the lens so long as the lens col-
limates the incident light. If the mirror is rotated through a small
angle, d, the reflected rays will sweep through an angle 2 d, remaining
parallel until they enter the lens. After passing through the lens they
will be brought to focus at some point P' , in the focal plane, and the
angle P'OP will be practically equal to 2B. The distance PP^=n
through which the image moves in the focal plane for a given small
value of 6 is directly proportional to the focal length of the lens. For
all practical purposes this rule holds regardless of the distance between
lens and mirror. If the mirror is translated in any direction without
rotation, the image P' will not move. If 26 is the angle through which
the reflected ray turns when the mirror turns through an angle d, then
tan 2 d-
OP
(81)
and for small angles,
20P
The ratio of the angular motion of the mirror to the linear motion of
the image (in the focal plane of L) is called the optical scale value,
designated by e. That is,
64
MAGNETIC OBSERVATORY MANUAL
[Ch.
Note : See paragraphs 200-202, pages 74-76, for more exact expressions
for the optical scale value and its reciprocal the optical lever.
162. The optical scale value may be halved (optical lever doubled)
by allowing the recording mirror, M, to make an angle of 45° with the
collimated rays whereupon the rays will be reflected at an angle of 90°
to their original direction. These rays are then allowed to fall approx-
imately at normal incidence upon a fixed plane mirror after which they
will be reflected back to the mirror, M, thence through the lens, to
form an image in the focal plane of the lens. Since the light is reflected
twice from the recording mirror, M, the reflected ray will turn through
an angle 4 6 when M is turned through an angle 6. The final result is
to double the sensitivity of the system. The optical scale value is
now equal to ^-tt^ instead of ^ ^^ -
^ 4 OP 2 OP
163. Planoconvex mirror, — Suppose the lens is reversed as in
figure 45 so that the light is incident on the convex surface of the lens.
Some of the light will not pass beyond the lens but will be reflected
from the inner surface of the plane side. If the lens is now inclined
slightly making an angle d with OP, the reflected rays from the plane
surface, now collimated, will pass back through the lens and form a
faint real image at P\ such that the angle POP' is again equal approx-
imately to 2 6. If the plane side is silvered or aluminized, practically
(b)
(c)
Figure 46.— (a) Origin of ghost images from front and back surfaces of plane mirror; (b) appearance o
main image and ghosts on screen; (c) alignment of ghosts with main image.
all of the light will be reflected. Planoconvex mirrors of this type are
frequently used as the moving mirror on some galvanometers, optical
thermographs, and other laboratory instruments. Images formed by
unsilvered plane surfaces of lenses or prisms are sometimes called
ghost images.
164. Multiple reflections. — Suppose collimated light from P, fig-
ure 46 (a), falls upon a back-surfaced mirror (front and back surfaces
not quite parallel) . Most of the light will be reflected at and after
again passing through the lens will form a bright image at Pi . Some
of the light will be reflected at the front surface at A and since the
surfaces of the mirror are not precisely parallel, a second faint image
will be formed at P3. Also some of the light reflected from will
be internally reflected at B back to 0' and again back through the
mirror, emerging at C, and forming another faint image at P2. P2 and
P3 are also called ghost images. When the light source, P, is an illu-
minated slit or a straight-filament incandescent lamp, the three images
will usually appear as one central bright image with faint images on
either side as shown at b, figure 46. Usually they may be brought
into alignment as shown at c, by rotation of the mirror, M, on an axis,
XX, normal to M. Ghost images of this kind cannot be formed when
front-surfaced mirrors are used.^
1 D. G. Knapp, Curiosities of magnetographs, Trans. Amer. Geophys. Union, p. 539, 1944.
5.]
OPTICS AND REGISTRATION
65
165. Totally reflecting prism, — 90° prisms, figure 47, are used
extensively in magnetic instruments for changing the direction of
incident hght approximately 90° and, as in the vertical intensity vari-
ometer, for changing the plane of motion of the reflected light beam
from a vertical to a horizontal plane. Light which falls on the face
AB at or near normal incidence is totally reflected by the surface AC
and emerges approximately normal to the face BC
166. Planoconvex prism, — If a planoconvex lens, L, be cemented
to one of the faces of a 90° prism adjacent to the 90° angle, this com-
Figure 47.-90°
prism.
Fig are 48.- Plano-
convex prism.
bination will serve as a collimator for light from a source P when P is
at the proper distance from the prism-lens combination. When such
a combination is made from one piece of glass, figure 48, it is called a
planoconvex prism. The la Cour D variometer is equipped with such
a prism.
167. Cylindrical lens (planoconvex), — This is equivalent to a
right cylinder cut parallel to the long axis of the cylinder, figure 49.
Figure 49.— Planoconvex
cylindrical lens.
Figure 50.— Action of cylindrical
lens in bringing a plane light
beam to a point focus.
Parallel rays incident on either surface will converge in a line focus or
line image parallel to the axis of the cylinder. If the incident light is
confined to a very narrow plane at right angles to the axis of the
cylindrical lens, the rays will be brought down to a point focus,
figure 50. Cylindrical lenses of short focal length, 2 to 4 cm, are
used on photographic recorders. Long-focus cylindrical lenses and
piano cylindrical mirrors are frequently used on variometers and
galvanometers. In such cases the axis of the cylinder must be
adjusted so that it will be parallel to the filament of the lamp and at
right angles to the axis of the cylindrical lens of the recorder in order
to obtain a sharp image.
168. Three-faced mirrors. — A convenient type of mirror for use
on magnetic variometers to produce regular recording spots and
upper and lower reserve spots is shown in figure 51. The three faces,
a, b, and c are ground and polished optically flat and then aluminized.
In order that the images from all tliree faces shall lie in the same
horizontal plane, the faces must be so ground that they will all be
perpendicular to a common plane. The areas of the faces are equal
66
MAGNETIC OBSERVATORY MANUAL
[Ch.
and the dihedral angles are such that when the regular recording
spot from face b passes off the magnetogram at the top or bottom
edge, the image from a reserve face, a or c, will come on the gram at
the bottom or top. The size of the desired dihedral angle is a func-
tion of the optical lever of the variometer lens and the width of the
magnetogram. For example:
Let y=ihe width of the gram =200 mm;
a;=the distance between reserve spots=180 mm;
20P==the optical lever=3460 mm (for a D variometer);
a=ihQ acute angle between adjacent faces.
Then tan a=iw=^TS=
180
20P 3460
and q:=2° 59'
and the interior dihedral angle is 177° 01'. In the Eschenhagen ar-
rangement of variometers the mirrors have certain specifications for
the angles when the width of the drum face is 20 cm.
169. The following table gives a current set of specifications for
dihedral angles, together with the approximate reserve distances which
are obtained by using those angles:
Variometer
Approximate
recording
distance
OP
Specifications for dihedral angle
Approximate
reserve
distance
X
Interior
180°— a
Exterior
a
Tolerance
H
D
Z
mm
1200
1730
2300
o /
175 45
177 00
177 40
4 15
3 00
2 20
±2
±2
±2
mm
178
181
188
a b c
Note: In some of the older variometers, separate plane mirrors were
mounted on thin aluminum frames. With this arrangement it was
. ^ismm^ necessary to bend the frames relative to each other in
._ order^to^obtain proper adjustment of the spots — a
I tedious and time-consuming process,
i 170. Auxiliary spots by other means, — La Cour
Y devised] a method of producing multiple spots from a
i. single recording mirror, by using small, 90° prisms
spaced along a rack in front of the recording lamp.
Figure 51.— Three- The prisms wcre so adjusted that each one provided a
faced mirror. beam of light f OT ouc variometcT mirror — the assembly
having the effect of several light sources, with the corresponding
number of spots for the recording drum. Schmidt ^ used a system
of inclined mirrors to reduce the sensitivity. Both the normal and
the low-sensitivity traces were recorded on the same magnetogram,
thus eliminating the need for reserve mirrors.
171. Base-line mirror. — A plane mirror attached to an adjustable
bracket within the variometer housing and back of the variometer
lens reflects a beam of light to provide a fixed reference line, or base
line, on the magnetogram. Each variometer usually has its own
base-line mirror.
2 Zs. f. Instrumentenkunde, 27, 137-47, 1907.
5.] OPTICS AND REGISTKATION 67
172. Diaphragms, — The quality of an image may be improved in
most cases by use of horizontal diaphragms (narrow horizontal
windows) mounted directly in front of the variometer lens. The
effect is to reduce the intensity of the light leaving the variometer
and at the same time confine the reflected light or transmitted light
to a small segment of the lens or mirror. Another diaphragm on the
cylindrical lens of the recorder serves also as a screen to protect the
photographic paper from extraneous light.
173. Character of image, — The light source is usually a straight-
filament incandescent electric lamp or an illuminated slit placed at
either end of the recorder at approximately the same height as the
cylindrical lens of the recorder. It is important that the axis of the
filament or the slit be strictly vertical so that images from the variom-
eter mirrors will be vertical as they fall upon the cylindrical lens
of the recorder. The recording distances should be such that these
images will be sharply focused line images on the magnetogram before
the cylindrical lens is adjusted. Great care should be used in making
these adjustments in order to insure satisfactory photographic reg-
istration. If the variometer lens is of the long-focus cylindrical type
its axis must be vertical (parallel to the lamp filament).
PHOTOGRAPHIC REGISTRATION
174. Recorder, — This is usually a clock-driven drum having a
20-cm face width and so geared as to have a peripheral (paper) speed
of 20 mm per hour. To prevent loss of record and to provide space
for a paper-clamp the drum makes one revolution in 25 hours. The
recorder and accessories are housed in a lighttight case, figure 52. The
light spots from the variometers are focused sharply on photographic
paper wrapped tightly around the drum. As the drum rotates, these
spots trace latent records of the base lines and variations of D, H,
Z, and temperature on the photo paper.
175. Parallax, — In figure 53(a) the horizontal line represents a
time line across the photographic paper on the drum as seen from
the variometers. When properly adjusted the D, 77, and Z recording
spots should be bisected on the magnetogram by this time line. In
figure 53(b) the H and D spots are not properly adjusted with respect
to the time line.
176. In photographic recording this maladjustment of a spot
relative to a time line (or time mark) is called parallax. Obviously
parallax of the base-line spots is of no consequence. When the centers
of all variometer lenses, axis of cylindrical lens, and center of time-
flash mirror are all in the same horizontal plane, little or no parallax
between any recording spot and the time line would be expected.
In practice, however, because of lack of perfect centering of lenses and
effects of spherical aberration, the alignment of lens centers is often
not sufficient to eliminate all parallax, and the final adjustments must
be made by the trial-and-error method. Adjustment of an^^ individaul
spot is made by raising or lowering the whole variometer, perhaps by
several millimeters, until the recording spot is on the time line. The
time line itself maybe adjusted by raising, lowering, or tilting the time
mirror or time lamp.
177. Sensitized recording paper, — Photographic paper from
several manufacturers has been used successfully for recording mag-
68
MAGNETIC OBSERVATORY MANUAL
[Ch.
Figure 52. — Magnetograph recorder showing driving clocli, gears,
and cylindrical-lens diaphragm and ligh t shield.
Figure 53. — Parallax between time line and recording spots:
(a) satisfactory adjustment, (b) poor adjustment.
5.] OPTICS AND REGISTRATION 69
netograms. In general the paper should have high contrast qualities,
producing a dense black line on a white background. Its sensitivity
to red light should be as low as possible, to facilitate handling the
paper by red light at the recorder and in the photographic dark room
with a minimum of background fogging. Some paper that has been
used appears to be quite sensitive to fingerprints, and no uniform
specification as to resistance to fingerprints has been found. It has
occasionally been necessary to have the observer wash his hands
thoroughly with soap and water just before making the daily change
of paper on the recording drum, or even to wear thin cotton gloves
while handling the paper, in order to reduce the smudging effect of
finger contacts.
178. Standard brands of high-contrast developing solutions for
photographic prints have been used for satisfactory processing of
magnetograms. It has sometimes been found that more uniform
results and greater contrast can be obtained with developing solutions
weaker (i. e., with more water) than the strengths recommended by
the manufacturer of the chemicals. Developing is generally continued
until the magnetogram has a satisfactory appearance under the red
light, rather than until a specified time has elapsed. The developing,
fixing, and washing baths should be warm enough to produce satisfac-
tory speed of chemical action, but not so warm as to risk damage to
the photographic emulsion by heat; temperatures of 15° to 25° C. are
generally acceptable. Magnetograms will have more sharply defined
traces if the intensity of the recording lamp is kept relatively low and
the developing process is forced somewhat by leaving the gram in the
chemical solution longer, although care must be taken that the gram
is not left in the solution so long that the white background becomes
fogged. Thorough washing in clean running water is, of course, neces-
sary for magnetograms that are to be kept on file permanently.
CHAPTER 6. QUARTZ-FIBER TECHNIQUES
179. Procurement of fibers, — The manufacture of quartz fibers
of suitable dimensions for use in magnetic variometers and astatic
galvanometers requires special equipment not usually available at
magnetic observatories.^ For this reason it is well to procure com-
mercially a supply of fibers ranging in diameter from 10 to 75 microns
(1 micron =0.001 mm) and 30 cm in length. A fiber should be reason-
ably uniform in diameter over its entire length, and circular in cross
section. Quartz suspensions of the la Cour type, figure 54, can be
manufactured with a minimum of equipment.^
180. Uses of fibers, — Fibers ranging from 7 to 12 microns in
diameter are used on astatic galvanometers and require special equip-
ment for mounting. A frame of mounted galvanometer fibers is
shown in figure 55.
181. Those ranging from 15 to 20 microns are used in D variometers
and those ranging from 35 to 75 microns are for H variometers.
^^
2 mm
R=^
r*
lO —
in
« — —
dinder and calculate
its moment of inertia about an axis normal to its long axis (see par. 72,
p. 28). Note: The dimensions of the inertia weight used in this work
by the Coast and Geodetic Survey are as follows: Length, 49.4 mm;
72
MAGNETIC OBSERVATORY MANUAL
[Ch.
diameter, 1.66 mm; mass, 0.90 gram; moment of inertia, 1.83 gram-
cm^.
193. Select a fiber and with fused shellac attach one end to the
inertia weight and the other end to a laboratory stand so that the
length of fiber between support and inertia weight is about 15 cm.
Surround the suspension by a glass cylinder to protect it against air
currents. Using a stop watch or a chronometer, determine the period,
T^, (time of one cycle) of the cylinder while oscillating as a torsion
pendulum through an angle not exceeding 20°, by timing 50 to 100
oscillations. Repeat the operation 4 times and take the mean as T^.
Remount the fiber in a separate wooden frame and label as to meas-
ured nominal diameter and period (say 45 microns; T^=9.8 sec).
Figure 56.— Apparatus for mounting brass stems on quartz fibers.
Repeat this operation with several fibers suitable for H and D vari-
ometers. The classification of fibers by this method of oscillations is
quite satisfactory and is preferred to the method of diameter meas-
urement by microscope. Calculate from equation (85) the torsion
constant, k', for each fiber, and record these values on the labels.
Compare with the values obtained from direct measurement of the
diameter (par. 191).
194. Installation of fibers in H and D variometers, — A simple
apparatus for mounting upper and lower brass stems to a fiber is
shown in figure 56. It consists of a brass bar carrying two clamps,
Ci and C2, for aligning and clamping the brass stems and two small
fiber supports, Ti and T2. The bar is attached to a wooden block by
a single screw, aS. By turning the bar at right angles to the wooden
base, the latter may be used as a holder when the stems are heated
for melting the shellac. The stems are slotted and notched as shown
in the figure.
195. To attach the stems to the fiber, proceed as follows: Usmg an
alcohol flame and hand blowpipe (or an oxyhydrogen torch) , bend the
ends of the fiber at right angles, leaving about 15.5 cm of straight
fiber between the bent ends. Feed the ends into the stems A and B
and adjust the stems in the clamps so that the fiber is drawn taut as
the bent ends engage the notches in the stems. Gently warm the
6.1
QUARTZ FIBERS
73
end of the bar near ^ in a direct flame and apply a small amount of
dry shellac to the notch allowing the shellac to melt and flow around
the bent end of the fiber and fill the slot and notch completely. Con-
tinue heating gently until the shellac is free of moisture and air bub-
bles. Avoid burning the shellac. See that the fiber is well centered
in the stem at the point of emergence from the slot. In like manner
attach end B. Before the shellac is entirely set, hold the bar with
its long axis vertical, end B down, loosen C2 and allow the fiber to take
the load gently. After the bar has cooled and
the shellac is well set, tighten C2, leaving the
fiber slightly slack. Break off the ends of the
fiber which protrude laterally from the stems.
In this condition the fiber may be mounted at
once in a variometer or stored in the holder for
future use. Experience has shown that fibers
mounted in this manner give satisfactory per-
formance and that there is little or no yielding
of the shellac when the fiber is subjected to large
torques over long periods of time.
196. Attach the coupler, figure 57, to the
stem A, figure 56, and screw it on rather firmly.
Hold the apparatus with its long axis vertical,
end A down, loosen Ci and allow the fiber to
take the load gently. Avoid bending the fiber near its point of
attachment to a stem. Remove the fiber and attachments from
the holder and while the system is still hanging vertically, attach
the stem B firmly to the torsion head spindle of the variometer.
Lower the whole suspension into the variometer suspension tube and
adjust the torsion head and the foot screws until the suspension
swings freely and the coupler disk is centered and parallel to the
jaws of the fiber clamp. Then set the clamp and tighten the torsion-
head set screw.
Figure 57.— Coupler for at-
taching quartz fiber to
suspended magnet.
210111—53
CHAPTER 7. THE DECLINATION VARIOMETER
197. Function of a D variometer, — This instrument should indi-
cate visually or by photographic registration the variations in the
direction of the horizontal component of the earth's magnetic field,
that is, the variations in magnetic declination.
198. Basic requirements, — The essentials of & D variometer are:
(a) A permanent magnet suspended in a nonmagnetic housing by
a fine quartz (or equivalent) fiber;
(b) An optical system suitable for visual observation or photo-
graphic registration of the variations that occur in the direction of the
D magnet as the declination changes ;
(c) The recording ray and the optical axis of the D lens to be ap-
proximately normal to the recording drum;
(d) Magnetic damping (recording magnet surrounded by a copper
or silver box) ;
(e) The magnetic axis of recording magnet to be parallel to the
magnetic meridian in the absolute observatory; see paragraph 331,
page 135 for effects of exorientation angles;
(f) A fixed base-line mirror so arranged that a straight reference
line (base line) is recorded on the magnetogram near the D trace.
199. D scale value, — The D scale value or minute scale value,
Sd\ of ei D variometer is defined as the change in magnetic declina-
tion, AD, in minutes of arc, corresponding to a change of ordinate of
one millimeter. An, of the D spot. Then
S.'=^. (90)
The D scale value is made up of three parts: (a) the optical scale value;
(b) the torsion factor ; and (c) the Jield factor. These component parts
will be described in detail in the following paragraphs.
200. Terms in the expression for the D scale value, — The var-
ious symbols used in the derivation of the general equation of the
minute scale value of the D variometer are:
Sd' =the minute scale value; minutes per mm;
Sd'^ =the gamma scale value; gammas (E'-PT field) per mm;
k^ =the torsion constant of the suspension fiber; dyne-cm per
radian twist;
Ms =the magnetic moment of the recording magnet; cgs units;
k = ratio of torsion constant of of the fiber to magnetic mo-
ment of suspended magnet
t .
H = horizontal intensity of the earth's magnetic field;
C =the algebraic sum of the north components of the fields
of all other magnets of the magnetograph (Note: the C
field must be the same in torsion tests as it is in normal
operation of the D variometer) ;
R =the effective distance from the D lens to the recording
drum, in mm ;
74
D VARIOMETER 75
2R =tlie optical lever;
_]_
2R
^-^ =6= the optical scale value in radians per mm;
e' =3438e=the optical scale value in minutes per mm;
Ari =a small increment in the ordinate of the D spot, in mm;
A^ =a small increment in the angular motion of the D recording
magnet (or the D magnet mirror) ; minutes or radians ;
h =the angle through which the D magnet turns when the
upper end of the fiber (torsion head) is turned through an
angle, /;
/ =see h above;
/—/i = total twist in the fiber, in the torsion test;
IX =the rigidity modulus of quartz; 2.83X10^^ dynes per cm^
per radian;
I =the length of the fiber in centimeters;
d =the diameter of the fiber in centimeters;
/ =the moment of inertia of the inertia weight suspended on
the fiber and oscillating as a torsion pendulum;
T^ =the period of the torsion pendulum in seconds.
201. The approximate value of E is the distance from the D lens to
the recording drum. Since a small portion of the path of the refiected
ray is made up of the D lens and the cylindrical lens of the recorder,
certain corrections must be made to the measured distance, Z, to obtain
the effective recording distance, R. Assuming the glass of these lenses
to have a nominal refractive index of 1.5, it can be shown that
R=L-^+bl (91)
o
in which
i?=the derived effective recording distance to be used in scale-
value equations;
Z=the measured distance in mm, from the front of the D lens to
the point on the recording drum at which the D spot normally
falls ;
c=the maximum thickness of the cylindrical lens in mm;
2
b=— for a planoconvex lens, plane side out;
o
6=— for an equi-convex lens;
o
b=0 for a planoconvex lens, plane side in;
/=the maximum thickness of the variometer lens in mm.
202. The optical lever and optical scale value, — When the
recording mirror (and the recording magnet) turns through a small
angle A^, the D spot moves through a small ordinate. An. If R is the
effective recording distance (optical lever =2R), the small angle, A^, in
radians is — ^- Thus
A<'=§ (92)
J A0 1
An 2R
76 MAGNETIC OBSERVATORY MANUAL [Ch.
But the optical scale value, designated by e, is
4 <»«
and from equation (91)
' 2(i-.|+w)
\VUJ
In radians,
Ad=e{An).
(96)
And in minutes,
Ao 3438,^ ,
A^- 2^ (An).
(97)
203. Torsion tests, — If the torsion head is turned through an
angle /, and the D recording magnet turns through an angle A. as a
result of the torsion in the fiber, the system will be in equilibrium when
the opposing couples, k' (f—h) and (H-]-C)Ms sin h, are equal. That
Jc' (f-h) = {H+ C) Ms sin h. (98)
For small values of h, write sin h^h radians. Then
^^=k=iH+C)(j^} (99)
Divide by H,
K'+i)(/4s)
(103)
{H+C)Ms
204. Physical picture of the torsion factor, — In figure 58, let
OiHi denote the original direction oiHand M, where His the horizon-
tal intensity and M is the magnetic moment of the D magnet; no
torsion in the fiber. Now turn the torsion head through an angle /;
whereupon the magnet will turn through an angle h, and the torsion in
the fiber is f—h. Compare figure 59 in which the original line of no
7.]
D VARIOMETER
77
torsion remains along O2H2 but the declination, D, changes through a
small angle AD and H is now along O2H2 . But the magnet will be
restrained by torsion and its axis will lie along O2M2. Since the
direction of the field has changed, O2H2 corresponds to OiHi and O2H2
corresponds to Oi T but the torsion is oppositely directed. The change
in the direction of the field, AD in figure 59, is equivalent to the amount
the torsion head is turned in figure 58, since by keeping H constant in
magnitude and direction and turning the torsion head we get the same
result. Thus, when D changes, the magnet turns through the angle
A^, and
Ad=f-h
(104)
and AD=f.
Dividing equation (105) by equation (104)
AD /
A8 j—h
(105)
(106)
The term
/
J-h
is the torsion factor.
"2 M2H2
Figure 58.— Torsion factor,
measured by turning tor-
sion head.
Figure 59.— Effect of fiber
torsion on D variometer
magnet.
205. If the preliminary optical scale value, e', is near one minute
per millimeter, so that h in millimeters on the drum is practically
f
equal to h in minutes of arc, the factor, 7^77? will not change appreci-
ably for relatively small changes in the recording distance. If the
preliminary (or final) value of e' differs appreciably from unity, then
h must be converted to minutes before calculating the torsion factor
since the torsion-head reading w^ill be in minutes.
206. Effect of the C field, —In figure 60:
0H= the horizontal intensity, H;
OC=HX, the vector sum, C, of all stray fields parallel to the
magnetic meridian OH]
0X= resultant of O^plus 0(7= original resultant.
78
MAGNETIC OBSERVATORY MANUAL
[Ch.
Now suppose declination changes:
AZ>= small change in declination;
HH'=XB=esist field corresponding to AD; (=HaD);
OH' =new horizontal intensity, approximately equal to OH in
magnitude ;
CB=OH' and is parallel to OH']
OB =new resultant = vector sum of OC and OH'
= vector sum of OC and CB;
A^= angle between OX and OB.
For a torsionless fiber, the D magnet will originally be aligned wdth
OX; when D changes by AD, the D magnet will turn through the
angle A^ and into alignment with the new resultant OB. From the
figure,
AD -44- (107)
A^
OX
XB
OX
(108)
Dividing"
AD XX' XB+BX'
Ad~ XB
XB
BX'
XB
= 1 +
BX'
HH''
Figure 60.— Effect of
the C field on D
variometer magnet.
and
In the similar triangles H'BX' and OHH',
BX'_H'B_ H'B
HH' 0H~ H '
Since H'B=HX=OC=C by construction,
BX'^C^
HH'~H
AD C _ H+C
Ad '^H H
(109)
(110)
(111)
207. The D scale-value equation. — Let a small field, AE, act on
a prefectly oriented D recording magnet, Mg, as in D scale-value
deflections, or as in the normal operation of a Z> variometer, producing
a small angular deflection, A^, of the recording magnet. The couples
tending to hold the recording magnet in the magnetic meridian are
(H+OMs sin A^ and k'AO. The deflecting couple is (AE)Ms cos AS,
When equilibrium is established these couples are equal and
{H+C)Ms sin Ad+k'Ad=iAE)Ms cos Ad.
But sin AB^AO (radians) and cos A0 — 1.
Hence, {H+C)MsAe+k'Ad=MsAE
(112)
(113)
7.] D VARIOMETER 79
ae
Therefore '^=H+C + k. (114)
From equation (96), A6=( (An). Then also
A 7^
Transposing, and substituting {H-{-C-\-k) for — r- from equation (115),
Au
^=e{H+C+k) (116)
and
So^ = e(H+C+k) (117)
since -— is, by definition, the gamma scale value, Sd'^, of the D vari-
ometer. By equation (301), page 231,
Sn'^=Sn'Ht2inV = ^^' (118)
Then
_3438^z)>
= (^)€(^ + C + ^).
But « = sp' hence,
(119)
^■=(i|f)(.+l4).
From equation (101),
A:
Substituting this value of yj in equation (119) gives
(120)
80 MAGNETIC OBSERVATORY MANUAL [Ch.
Equation (122) is the usual expression for the minute scale value of
the D variometer.
208. Other forms of the D scale-value equation, — From equa-
tion (103)
J^= ^' . (123)
Substituting this value of 7 — ^ in equation (121) gives the D scale
value in terms of the torsion constant of the fiber, thus
209. From equation (150), chapter 8, the torsion constant, k', of
the fiber is given by,
Substituting this value of A:' in equation (124) gives the D scale value
in terms of the period, T^, of the fiber and an inertia weight (known
value of I) when oscillating as a torsion pendulum. Thus
210. From equation (148), chapter 8, the torsion constant, k\ of
the fiber in terms of its dimensions is,
" ~ 32Z
Substituting this value of k' in equation (124) gives the D scale value
in terms of the dimensions of the fiber. Thus
211. Equations (121), (122), (124), and (125) are all useful forms
for the evaluation of the D scale value when the effective distance,
R, is known, and the C field is negligible or small compared to H (no
control magnet). Equation (126ji has limited usefulness because of
the difficulty in measuring the diameter of the quartz fiber to the
desired accuracy, as illustrated by example (b) given in paragraph
215.
212. Low -sensitivity D variometer {scale value). — Note that
all of the D scale-value equations contain the factor — fj^^' indicating
that the scale value might be increased indefinitely by simply increas-
ing the field factor, C, by use of a control magnet attached to the D
variometer, A^ end to the north, magnetic axis in the magnetic meridian
7.] D VARIOMETER 81
through the D recording magnet, and at such a distance as to give the
required field. Suppose a scale value of 5' per millimeter is desired.
To obtain a scale value of this magnitude by use of a control magnet
attached to a sensitive variometer, for example where aS/)' = 1.00, it
would be necessary to make C=^H. If H is 10,0007, C would have
to be around 40,0007 at the center of the D recording magnet. A
control magnet having a magnetic moment of 100 cgs units properly
placed 8 cm north or south of the D recording magnet would supply
the approximate field required, but due to the great uncertainty of
distribution at such a short distance, the effective C field would be
quite uncertain even with the theoretically correct distribution factor
applied. The method is not satisfactory and the use of a larger fiber
without control magnet is recommended. Note also that a stronger
control magnet at a greater distance, though eliminating some uncer-
tainty in the distribution effect, would be undesirable because of
the large fields it would produce at the other variometers of the
magnetograph.
213. Alignment chart for estimating D scale values; the D
nomogram, — Figure 140, appendix VI, is a graphic solution of equa-
tions (124), (125), and (126) for the special case that 2i?=3438 mm;
1=15 cm; (7=0; 7=1.83 gram-cm^; and /jl is the rigidity modulus of
quartz. For this special case,
1-f
47r'I
T^'HM.
^"^ 32lHMs
= l+[l.85X10^(^3} (129)
214. Note that equivalent values of JL, k\ and d, are read hori-
zontally on the nomogram. In using the k' scale in conjunction with
the Sd^ scale and the (HMs) scale, it is first necessary to find the equiva-
lent diameter (on the d scale) or the equivalent period (on the T^
scale). This is illustrated in example (a) of paragraph 215.
215. Use of the D nomogram. — The following examples serve to
illustrate the use of the D nomogram. (Quartz fibers only.)
(a). Given: HMs= 1.85 dyne cm;
aS£,' = 1.01 minutes per mm;
2i?=3438 mm;
C=0;
Z=15.0 cm;
7=1.83 gram-cml
Required: d, k' and JL.
Solution: The straight line passing through *S'o' = 1.01 and
707s=1.85 intersects the d scale at 17.7 microns (0.00177 cm) and
the J_ scale at 61.4 seconds. A horizontal line through 17.7 on the
82 MAGNETIC OBSERVATORY MANUAL [Ch.
d scale intersects the k' scale at 0.0185 dyne-cm per radian. There-
fore,
d=l7.7 microns; T^=Q1A sec; and ^'=0.0185.
(b). Given: 1=15 cm;
7=1.83 gram-cm^;
d=l7.7 microns (±20%) measured.
Required: k^ and T^.
Solution: 17.7±20% = 17.7±3.5 microns.
This means that d lies somewhere between 14.2 and 21.2 microns
Opposite d=14:.2, read r~=100 seconds and A:' =0.007. Opposite
d^=21.2, read T^=45 seconds and A:' =0.038. Thus if a torsion test
were made with this fiber the value of k' might turn out to be any-
thing between 0.007 and 0.038. If the period were determined with
the inertia weight, 7=1.83, it might have any value from about 45
seconds to 100 seconds. This example illustrates why it is impractical
to rely on the measured diameter of the quartz fiber as a measure of
its torsion constant or its period, T'-, as described above.
(c). Given: 2/?= 3438 mm;
(7=0;
7=1.83 gram-cm^;
1=15 cm;
r_=4.6 seconds;
7Z=9000 gammas (=0.090 cgs) ;
aS',>' = 5.0.
Required: Magnetic moment, Ms, of the recording magnet,
to give a scale value of 5.0 minutes per millimeter.
Solution: The straight line through Sd' = 5.0 and r_=4.6,
intersects the HMs scale at 0.83 dyne-cm. Therefore if 7^"= 0.090,
, , HMs 0.83 n o ^u -J
Ms= j-r = ^ Ann =Q-^ ^S^} ^be required magnetic moment.
M U.U9U
216. Evaluation of the recording distance for unit scale
value. — From equation (122), for *S£,' = one minute per mm, we have
and from equation (91),
L=R+^--bL (131)
o
Then Sd' will be one minute per millimeter when
7.] D VARIOMETER 83
For example: Let C= + 100 gammas;
i?« 20,000 gammas;
/=3 mm;
and c = 9 mm.
Then from equation (132),
Z=1719 (1.006) (1.005) + 3.0-2.0
= 1739 mm.
In the above arrangement, if the L distance (measured distance
from the front of the D lens to the drum) is made equal to 1739 mm,
the D scale value will be one minute per mm.
217. Since it is desirable to operate at a scale value near unity,
the focal length of the D lens should be such that the product of the
three factors on the right hand side of equation (122) shall be unity
when the D spot is in sharp focus on the recording paper and when the
D ordinate is about normal. It is good practice to specify a focal
length of 173 to 174 cm for the focal length of the D lens. If it
happens that the derived value of the distance, L, differs from the
focal length by as much as one centimeter, the variometer or the
recorder (preferably the recorder) may be adjusted to the correct
distance for unit scale value without appreciable impairment of the
character of the D spot, since the depth of focus of a well-diaphragmed
D lens of this focal length may be as much as 3 or 4 cm.
CHAPTER 8. THE HORIZONTAL-INTENSITY VARIOMETER
218. Function of an H variometer, — This instrument should
indicate visually or by photographic registration the variations in
the horizontal component of the intensity of the earth's magnetic
field.
219. Basic requirements, — The essentials of the instrument are:
(a) A permanent magnet supported by a bifilar or relatively coarse
quartz unifilar suspension so that its magnetic axis is horizontal and
is kept approximately in the magnetic prime vertical (see par. 39)
by the twisted suspension;
(b) A nonmagnetic housing with suspension tube and torsion head
for regulating the torsion in the suspension;
(c) An optical system for visual observations or for photographic
registration ;
(d) A copper or silver box surrounding the magnet to provide
necessary damping;
(e) A temperature compensation device.
Note: See pages 116 and 124-125, chapter 10, for temperature
compensation of the H variometer, and page 135, chapter 12, for
the effects of an exorientation angle.
220. Operating principle, — To place the variometer in operation,
the torsion head is turned until the mechanical couple of the fiber
is just sufficient to turn the magnet into a position at right angles
to the magnetic meridian through the variometer. In this position
the mechanical couple of the fiber is equal and opposite to the magnetic
couple or restoring torque which tends to turn the magnet back into
the meridian. Any slight change in H results in unbalanced couples,
and the magnet turns in azimuth until the torque applied through
the fiber again balances the magnetic couple. It is this motion
of the magnet that is observed and recorded.
221. Symbols, — The following notation will be used:
T =the total torsion in the fiber, expressed in radians;
6 = angle between the magnetic axis of the recording magnet
and the magnetic meridian ;
k' = torsion constant of the fiber in dyne-centimeters per radian
twist ;
A:'t= total couple applied to the fiber in dyne-centimeters;
H = horizontal intensity;
Ms = magnetic moment of the recording magnet, cgs units;
e = optical scale value = ^^? in radians per mm;
R = effective distance from variometer lens to drum;
f^ =the magnetic east field at center of H variometer due to all
magnets of the magnetograph, except the H recording
magnet;
84
H VARIOMETER
85
V
component of the earth's field along Ms when the recording
magnet is not precisely in the prime vertical; plus the
sum of the components of the fields of all other magnets,
including a sensitivity-control magnet, parallel to Ms] p is
positive when it is directed from the south to the north
pole of Ms]
angular displacement of the recording magnet for a change
A//;
Ai7= small change of H]
n = ordinate in mm of the H spot measured from a fixed line
(the H base line) on the gram;
An =small change in the ordinate, n;
k = ratio of torsion constant of the fiber to magnetic moment of
Ad
suspended magnet =
M.
Ej: =exorientation angle; the small angle between Ms and the
magnetic prime vertical. For an H recording magnet
with A^ end east, E^ is positive when Ms is turned clock-
wise, looking down, through a small angle from the
prime vertical.
222. The H scale value, — When there is no torsion in the fiber
and the suspended magnet, Ms, is not acted upon by the fields of any
other magnets, Ms will come to rest with
its magnetic axis coincident with the mag-
netic meridian, ON, figure 61. This is
called the initial line of no torsion, and the
torsion-head reading establishes the direc-
tion of this line. Now turn the torsion
head through the angle NOB so that Ms
is precisely at right angles to the magnetic
meridian, that is, so that 6 is 90°. OB is
the new line of no torsion, that is, the
new direction the magnet would take if
all fields were removed. The mechanical
couple in the fiber is now k'r, r being the
difference between the angle NOB through
which the upper end of the fiber turns and
the angle NOE, the angle through which
the lower end turns.
223. The mechanical couple A-'/tends to
turn the magnet clockwise out of the mag-
netic prime vertical. The magnetic couple HMs sin 6 tends to turn
the magnet counterclockwise out of the magnetic prime vertical.
Then since equilibrium has been established the couples are equal and
Figure 61.— Simple // variometer.
and since 0=90°,
HMs sin e=k\
UM,=k'T.
(133)
(134)
\
224. Now let H increase by a small increment AH. This will
cause the recording magnet to turn through a small angle AB, so that,
{H-^AH) M, sin (9O°-A0)=/:' (r + A0).
(135)
86 MAGNETIC OBSERVATORY MANUAL [Ch.
225. If there is a fixed field, /^r, say the field of a sensitivity-control
magnet, directed along the prime vertical, there will be an additional
couple, JeMs sin A^, supplementing the couple of the fiber, and equa-
tion (135) becomes,
(i7+AiJ)MsSin (90°-A^)-/^M,sin A^ + ^'(t + A). (136)
226. When ^^ is small, sin Ad^Ad; cos A(9^1; sin (90° — A^) =
cos A(9^1; and cos (90° — A^) = sin A^~A^; then
HMs+MsAH^fEM.Ad+k'r+k'Ad. (137)
Subtracting (134) from (137)
MsAH^fEMsAd+k'Ad. (138)
Transposing,
k'
227. Let ^=k. (It will be shown in paragraph 232 that k may
be taken as an equivalent field, expressible in gammas.) ^ Substituting
in equation (139) gives
^'^k+U (140)
AH
228. The ratio, — -t-» represents the rate of change of i7 with respect
Au
to unit angular displacement of the recording magnet and is called
the scale value of the variometer. Equation (140) gives the approxi-
mate scale value in cgs units per radian. In practice the scale value
is expressed as the change in H, in gammas, when the H spot moves
one millimeter on the magnetogram. If An is a small change in the
ordinate and R is the effective recording distance in the recording
system,
^=A^=eA7i (141)
and
e=A« (142)
An
where e is the optical scale value in radians per millimeter. Multiply-
ing the left-hand side of equation (140) by ■— and the right-hand side
by e, we have
— -X-r-=€(^+/^), Cgs units per mm (143)
isu i\n
and the H scale value, Sh, in gammas per millimeter, is
SH=^=e{k^h)XW . ^ (144)
in which k and/e are in cgs units.
> Ad. Schmidt, Ergebnisse der magnetischen Beobachtungen in Potsdam und Seddin im Jahro 1908,
Veroff. des Preuss. Met. Inst., Berlin, p. 39.
8.] H VARIOMETER 87
229. A detailed analysis ^ shows that it is more accurate to use,
instead oifs, the field p which is the component of all fields along Mg.
Thus
SH=^e{k-\-p)XlO\ (145)
The chief difference between p and Je is the component, along Mgy
of all north or south magnetic fields. This component arises when
Ms is not exactly in the magnetic prime vertical and gives rise to the
a factor (change of H scale value with change of H ordinate, para-
graphs 254-260), and to the b factor (change of H scale value with
change of declination, paragraphs 261 and 262).
230. Equation (144) is useful in calculating the effect of k and Je
on the H scale value.
231. Significance of k. — From equation (134) we have
H_k^_,
r ~Mr^
i_k'_ dyne-cm/radian _ oersted
Ms dyne-cm/oersted radian
(146)
and since angular values are dimensionless numerics, the dimensions
of k are those of field intensity.
232. Also, let Ms be the recording magnet, precisely oriented in the
magnetic prime vertical, north end east. In this position the mechan-
ical couple is just balanced by the magnetic couple (equation 134).
Suppose that both of these couples are reduced to zero but Ms remains
suspended by a (hypothetical) torsionless fiber and free to turn in any
direction in a horizontal plane. It would then take up the direction
of any applied horizontal field of any magnitude. To maintain Ms in
the prime vertical or to turn it into the prime vertical (if displaced)
would require an east field directed along the prime vertical through
the center of Ms. Suppose the system had a scale value of 3.00 7 per
mm operating at a recording distance of i?=1180 mm (optical scale
value, €=^=0.000424). Then from equation (144), since A:=0,
XIO^ ^^A^Q5
and JE=hm ^-"^^^^2360 mm)/' 0.00001 ^-^^^
\ mm / \ gamma/
/^=0. 07080 oersted =70807.
In other words, if A:' = ik = 0),fE would have to be 7O8O7 to produce
the same scale value as if k were acting alone without /^r. Again, if
Ms is acted upon only by the field, /s, and a field, x, of 37 be applied
normal to Ms, the latter will be deflected (take up the direction of the
resultant field) tln-ough a small angle, A^, such that,
tan A0=^=^=O.OOO424. (147)
2 H. H. Howe, On the theory of the iinifilar variometer, Terr. Mag., 42, 29^2, 1937, (see item 6 of
bibliography) .
88 MAGNETIC OBSERVATORY MANUAL [Ch.
But M=-^yr< and An=2RAd, in which An is the motion of the H spot
on the gram when 2R = 23Q0 mm. Then An = 2360X0. 000424 = 1 mm.
This means that a field of 7O8O7 applied parallel to Mg (no other fields
acting, and a torsionless fiber) produces the same scale value for Ms
k'
as we had in the original system when^^=0 and only y^ (=^) deter-
mined the magnitude of the scale value. We can state then, that k
is equivalent to the field which will produce the same scale value,
and that k has the same dimensions as Je, that is, k may be regarded
as an equivalent field expressible in gammas.
233. Evaluation of k' , k, and other constants, — In the preceding
paragraphs it has been assumed that Ms, k' , and/^ remain constant.
In practice all of these factors change with time and/or temperature.
It would not be feasible to determine the values of all of these factors
under changing conditions every time the i^ scale value is determined.
(Practical methods for determining the H scale value are given in
chapter 11.) However, much guesswork may be eliminated in pre-
paring a variometer for routine operation, if one has a good under-
standing of the meaning of the factors which make up equation (144).
234. Since each of the terms e, k' , Ms, and Je may vary contin-
uously over wide ranges, there will be an infinite number of combina-
tions of these terms which will give the same scale value. However,
various practical considerations make it necessary to keep the ranges
within fairly narrow limits.
235. The value of k' may be estimated from fiber dimensions
(paragraph 238), from oscillations (paragraph 239), from torsion
observations (paragraph 240), or from the H scale value (para-
graph 242).
236. The magnetic moment Ms can be determined directly by
deflections (par. 319, p. 129).
237. The term Je can be estimated if the magnetic moments and
relative positions of all nearby magnets are known. Usually Je is
negligible for first-order effects unless there is a sensitivity-control
magnet on the variometer.
238. The quartz fiber and the torsion constant, k\ — The
torsion constant, k', of the fiber is the torque in dyne-centimeters
required to turn one end of the fiber through an angle of one radian
while the other end remains fixed. It may be calculated from Cou-
lomb's equation,^
in which example
ju = the rigidity modulus of quartz, 2.83X10^^ dyne/cmVradian
d = i\ie diameter of the fiber, 0.00372 cm
/ = the length of the fiber, 15.0 cm
^' = from equation (148), 0.354 dyne-cm/radian.
Since the diameter enters as the fourth power in equation (148), it
should be measured with great precision by a suitable microscope at
3 Max Planck, The Mechanics of Deformable Bodies. London, 1932, p. 77.
8.] H VARIOMETER 89
many equally spaced places along the fiber. The average of the fourth
powers of d rather than the fourth power of the average should be
used.
239. Torsion constant by oscillations, — (Preferred method).
The period, T^, of a torsion pendulum oscillating under the directive
force of the suspension is given by
T^=27r^Jj (149)
and k'=~i, (150)
where example
7= the moment of inertia of a sus- 1.83 gram-cm^
pended brass cylinder,
T^^the period of the system, 14.3 sec.
A:' = torsion constant, by equation (150), 0.35 dyne-cm.
The moment of inertia of the cylinder is calculated from its dimensions
and mass using equation (45), page 28. Length of cylinder: 4.94
cm; diameter: 0.166 cm; mass: 0.90 gram. The period is estimated by
timing 50 to 100 oscillations with a stop watch.
240. Torsion constant by torsion observations. — The value of
k' may also be determined by torsion observations as in a Z) variometer.
From equation (99),
k' = {H+C)Ms(^j^y (151)
The values of h and / are observed by the method described in para-
graph 203, page 76. Then, knowing H-\-C and Mg, k' may be cal-
culated. Table 5 gives some values of k' determined from torsion
observations, from oscillations with known moment of inertia, and
from the dimensions of the fiber, using equations (148) (150), and (151 ).
241. Evaluation ofk from t and M^.— Let A:'=0.35 and M3 = 5.0.
Then from equation (146)
, 0.35 dyne-cm r^ r^^ ^ ^ /^ ^^.n
''•— -^ —0.07 oersted. (152)
5.0 dyne-cm/oersted
242. Evaluation ofk from scale-value observations and optical
lever, — Suppose that the H scale value has been determined in the
usual manner by deflections as explained in chapter 11, and that
^^=2.977/mm=0.0000297 cgs/mm, when e=0.000424 and there is no
control magnet (no /^ field). Then from equation (144)
k=^ (153)
0.0000297
0.000424
0.07 cgs = 70007.
210111—53-
go
MAGNETIC OBSERVATORY MANUAL
[Ch,
243. If Ms is known, then k' may be evaluated from the relation,
k'
k=-irr' Suppose Ms=5.0, then
^'=A:M,=:0.07X5=:0.35 dyne-cm. (154)
TABLE 5. — Torsion constant, k' , of quartz fibers by different methods.
Place: Cheltenham Magnetic Observatory Date: July 31, 1951
Horizontal Intensity: 0.183 Observer: J. B. Townshend
Symbol
Quantity
Fiber
No. 10
No. 36
d
I
I
f
h
f-h
h
f-h
Ms
(H-hC)Ms
Diameter cm_ _
Length cm_ _
Oscillations:
Moment of inertia of weight gm-cm2__
Period with this inertia weight sec__
Torsion observations:
Angular motion of torsion head
0. 0018
15. 4
0. 00655
15.6
1. 83
47. 6
5. 798
5.26
1800'
24.'9
1775. '1
0.014
12.0
2.20
100'
88.'2
11. '8
7.48
5.8
1.06
Resulting angular displacement of magnet- _
Twist in the fiber
Displacement per unit twist
Magnetic moment of suspended magnet
cgs-emu- -
Magnetic couple per unit displacement
dyne-cm/radian __
Torsion constant, computed:
From oscillations dyne-cm/radian _ _
From torsion observations_dyne-cm/radian_..
From dimensions dyne-cm/radian _ _
0. 0319
0. 0308
0. 0190*
9.26
7.93
3.28*
♦Computed k' is sensitive to small error in diameter (d).
Example: If error in d is 10%, error in k' will be approximately 46%; if error in d is 20%, error in k' will be
approximately 107%. For this reason, torsion constants computed from dimensions and assumed rigidity
modulus are not reliable and may be used only for rough estimates in preliminary tests.
244. Scale value from constants, — Let 2/? =2360 mm
(€=0.000424), M,=5.0, ^' = 0.35 (/:=0.07), and /s=0. Then from
equation (144), the H scale value is given by
^ /0.000424\ ,^ ^^ , ,, /
Sh=\^ mm ) ^^-^^ oersted) (^
10^ gammas X
oersted /
(155)
= 2.977/mm.
245. Thus, it is possible by careful selection of quartz fiber and
recording magnet to assemble a system which will have a desired scale
value at a given recording distance.
246. The sensitivity -control magnet {scale-value control). —
In figure 62 let Mg be the recording magnet of the H variometer,
operating in the magnetic prime vertical with its A^ end east. Ma is
a sensitivity-control magnet mounted on the variometer with its
8.1
H VARIOMETER
91
axis in the magnetic prime vertical, A^ end east, and at such a distance
from Ms that its field, /^, at the center of M^ is equal to 5OOO7 (0.05 cgs
units). Then the i/ scale value, from equation (144) will be
5'^=0.000424(0.07 + 0.05)-10^=5.097/mm. (156)
If the control magnet is reversed, the /^ field is negative, and
AS'^=0.000424(0.07-0.05)-10^=0.857/mm. (157)
It is evident from the above that the H scale value may be adjusted
over a wide range by simply adjusting the control magnet along its
bar. The approximate field of a bar magnet, along its magnetic axis
extended, at a distance r from its center is
/.-^-^"X10^7
(158)
in which /^^ is the field, and Ma the magnetic
moment, of the control magnet. If Ma= 100
cgs and r=10.0 cm, then /^= 20,0007 at 10 ^
cm. The change in Je as r is changed is
given by
dr
r
and
^ r
For Ar=:— 0.1 mm (=—0.01 cm)
-3X20,000X-0.01
A/g=
10
h607.
(161)
Figure 62.— // variometer with
sensitivity-control magnet.
Disregarding distribution effects, this means that by reducing r by
0.1 mm the effective field at the center of the recording magnet (due
to Ma) has been increased by 6O7. The nomogram of figure 138
gives the same result.
247. Variation of the scale value with k and /^. — From equa-
tion (144)
SH = ke + eU (162)
The relation between changes of Sh and/^, when k is constant, is
dJE ''
and when J E is constant
(164)
(163)
dk ^'
and when both/^ and k change,
dSH=edfE-\-^dk.
(165)
92 MAGNETIC OBSERVATORY MANUAL [Ch.
Equations (160) and (163) will be found quite useful for evaluating
the distance at which the sensitivity magnet must be placed in order
to change the scale value by a desired amount. For example: Let
^^=5.00; €=0.000424; M«=100; and r=10 cm. How far should Ma
be moved in order to change Sh from 5.00 to 5.10? In this case,
AASV=0.107/mm, A/:=0, and from equation (163)
AS'
A/^=^^^= AAS'HX2i? = 0.1 X2360 = 2367. (166)
An additional east field, A/g, of 2367 will increase the scale value from
5.00 to 5.107/mm. To produce this change in the Je field at the
center of Ms, find Ar by applying equation (160),
^r--r(^\=- ^^^^^^ (167)
\3/J 3X20,000 ^^^^^
= — 0.039 cm = — 0.4 mm, approximately.
248. Again, what /^ field would be required to change Sh by just
one gamma per millimeter, say from 5.49 to 4.49 by means of a con-
trol magnet, and at what distance should the control magnet, Afa=
100, be placed to accomplish this change? From equation (163)
dy,='^=-^^^l^= -0.02360 cgs=-2360T.
That is, a field of — 236O7 applied at the center of Ms and directed
to magnetic west would reduce Sh from 5.49 to 4.49. From equation
(158), r''=^^-=^r-p-r— -=8474 cm^ and r=20.39 cm, the distance at
which Ma should be placed to produce the change of one gamma per
millimeter in the scale value. If Ma is reversed it will increase the
scale value from 5.49 to 6.49.
249. Use of two control magnets, — The distance between the
control magnet and the recording magnet may change slightly as a
result of seasonal tilting of the pier (causing the suspended magnet
to swing toward or away from the control magnet) . This would cause
a seasonal change in the scale value. The effect may be compensated
in part by use of two control magnets symmetrically placed in the
prime vertical, east and west of the variometer. (A second control
magnet on the H variometer in the Eschenhagen type of magnetograph
might have an undesirable effect on the D recording magnet. Its
net effect would depend upon the direction and intensity of the other
east or west fields at the center of the D magnet.)
250. Table 6 gives various values of Sh, determined experimentally
for various positions of a control magnet having a magnetic moment
of 50.2. The results are shown graphically in figure 63.
H VARIOMETER
93
TABLE 6. — Variation of H scale value with distance between control
magnet and recording magnet.
[2i2=2360mm; e=0.000424; Afo=50.2; N-end of recording magnet east.]
Observed scale
value Sh
Distance, r,
between
Ma and M.
•
Control magnet,
N-end
k
fE
k+fs
7/mm
cm
7
T
y
2. 29
11
w
12950
-7540
5410
3.03
12
w
12950
-5810
7140
3.55
13
w
12950
-4570
8380
3.94
14
w
12950
-3660
9290
5.49
Away
Away
12950
12950
7.04
14
E
12950
+ 3660
16610
7.43
13
E
12950
+ 4570
17520
7. 95
12
E
12950
+ 5810
18760
8. 69
11
E
12950
+ 7540
20490
10
Scale-Value=5.49 Vmni with No Control Magnet
Control Magnet Distance in cm.
\ I
10 11 12 13 14 15
Figure 63.— Variation of scale value with control magnet distance.
16
251. Alignment chart for estimating H scale values; the H
nomogram, — Figure 141, appendix VI may assist in attaining a par-
ticular scale value. It is based on the equation
Sh —
(w+^^)
(168)
with k' represented by its equivalent in terms of d (eq. 148) or of
TL (eq. 150). When a straightedge is laid across this nomogram, the
left-hand scale should be taken to represent k alone. A reading so
made on the adjacent Sh scale is valid only if /^.^O. However, when
Je^O the Sff scale is useful in another way. Any displacement along
this scale corresponds rigorously with the identical displacement along
the (k-^-fs) scale, permitting additive correction for /^ when necessary.
Special conditions assumed in the construction of this nomogram are:
2R=23Q0 mm (c=0.000424); /=:15 cm; 7=1.83 gram-cm^; and
94 MAGNETIC OBSERVATORY MANUAL [Ch.
M=2.83 X 10^^ (rigidity modulus of quartz) . For this special case, from
equation (144),
= (^+A) (0.000424) X 10^
= 42.4(A:+/^).
When k-{-fE=^l cgs unit, AS'//=42.47/mm. The S„ scale is placed so
that 42.4 on the Sh scale is opposite 1.0 on the {k-\-fE) scale. For
any other value of e, the pDsition of the S^ scale should be changed
accordingly.
252. Note that equivalent values of T^, k' , and d are read hori-
zontally. In using the k' scale in conjunction with the Ms and Sh
scales, it is necessary first to find the equivalent diameter, d, (on the
d scale) or the equivalent period, T^ (on the T^ scale).
253. Examples of the use of the H nomogram,— The following
examples serve to illustrate the use of the H nomogram :
(a) Given: Ms =5.0 cgs units;
2i?=2360 mm;
€-:^ = 0.000424;
/=15 cm;
^^=3.07/mm;
Required: d, diameter of the quartz fiber;
T~, period when 7=1.83 gram-cm^;
k\ torsion constant of fiber.
Solution: The straight line from 5.0 on the Ms scale to 3.0 on the
Sff scale intersects the d scale at 37.1 microns and the period scale at
14.3 seconds. That is, a recording magnet having a magnetic moment
of 5.0 when suspended on a 37.1 micron fiber of uniform diameter
throughout its entire 15 cm length and operating precisely in the
magnetic prime vertical would have a scale value of 3.0 7/mm. The
horizontal line through d = 37.1 microns (7"^= 14.3 seconds) intersects
the k' scale at 0.35 dyne-cm per radian. This is the torsion constant
of the fiber.
(b) Given: r^= 10.1 sec for 7=1.83 gram-cm^;
^=15 cm;
2i? = 2360 mm;
€=^=0.000424;
8.]
H VARIOMETER
95
Required: Value of magnetic moment, Ms, of recording magnet
to give a scale value, *S'h=3.0 7/mm.
Solution: The straight line from Sh = ''^-0 on the Sh scale, through
10.1 on the T^ scale {(1 = 49 microns) intersects the Ms scale at 10.0.
That is, if M«= 10.0 the scale value will be 3.0 for the given conditions.
(c) Given: Sh=(^-0 j /mm.]
2R=2360 mm:
__1_
^'~2R
Je=o.
0.000424;
Required: Sh=^.0 7/mm;
/^= sensitivity -control -magnet field (E-W field at
center of Ms) to reduce Sh from 6.0 to 3.0.
Solution: On the {k-\-fE) scale opposite Sh = Q.O find k+fE=0.14:l
(14,1007), but/£=0, so ^ = 0.141; and on the same scale find k-\-fE=
0.071 (71OO7) opposite &=3.0 7/mm. ThenfE={k+fE) -^ = 0.071 -
0.141= -0.070= -70007, the required Je
field. The negative sign means that the
field must be directed toward magnetic
west. This required field may be supplied
by a small control magnet. Ma, having a
magnetic moment of, say, 80 cgs, fixed on
the variometer deflection bar at a distance
r=13.2 cm (center of Ma to center of
Ms) in the mxagnetic prime vertical, east or
west of Ms, with A^ end west. In this case,
from equation (144)
SH=42A(k^fE)
42.4X0.071
:3.007/mm.
Figure 64.— The a factor in the //scale
value.
254. Variation of H scale value with ordinate; the a factor. —
In figure 64, AOB represents the recording magnet of the H variometer.
It is held in the magnetic prime vertical by the couple of the suspension ,
A^ end to the east. Ma is a control magnet, A^ end east, so that its
field is directed along magnetic east. A magnetic field in the magnetic
meridian is produced by other magnets of the magnetograph and is
denoted by C. The C field is primarily a temperature-compensating
field directed to magnetic south and approximately equal to half of H.
255. When H changes, the recording magnet turns through angle
Ex. The field, /e, of the control ma-^net along Ms is now fs cos E^;.
But there is also a component of {H-\-C) parallel to the recording
magnet when the magnet turns out of the prime vertical. This component
is (H-\- C) sin Ej^. Then the total field parallel to the recordmg magnet
is
p =Je cos £-,+ (i7+ C) sin E, (169)
so that equation (145) becomes
*S'^==€ [k-\-jE COS E,^ (H+C) sin E,]X10'
(170)
96 MAGNETIC OBSERVATORY MANUAL [Ch.
For angles of E^ up to about 3°, cos Ej, may be taken as unity without
introducing an appreciable error in Sh, but {H-\-C) sin E^ is appreciable.
For example: Let i7=27,0007; €=0.000424 (2i?=2360 mm); k=
12,950;/^ = 6,8207; C= -13,500t. When E^=0,
^^=0.000424 (12950 + 6820) =8.383 7/mm.
Now let H increase so that E:,=2°, figure 64. The term (H+C)
sin Ex must be added to the scale value. The scale value is now
&=0.000424 (12950 + 6820 cos 2° + 13500 sin 2°)
= 0.000424 (12950 + 6816+471)
= 8.58l7/mm.
The difference in the scale value is ASh = S.5S1 -8.383 = + 0.1987/mm.
But 2° angular motion of the recording magnet =4° motion of the
spot, and the change in ordinate is
Ari=1180 tan 4°=82.5 mm
and the change in H scale value per milhmeter change in ordinate is
ASh , 0.198
An ' 82.5
0.002407/mm/mm.
This factor, 0.00240, is called the a factor of the scale value. It
means that the H scale value increases with ordinate; the a factor is
primarily a function of (H+C) and is proportional to (H-{-C). When
{H-\-C) is reduced, from H io H-\-C {^y2H), as in magnetic tempera-
ture compensation, the a factor is reduced in the same proportion.
256. Thus it is seen that as the H ordinate, n, changes, the cor-
responding rotation of the recording magnet and change of Ex change
the H scale value through the {H-\-C) sin E^ term of equation (170).
The rate of change, -— ^j is the a factor, so that if *S'o is the base-line
scale value (value of Sh when n is zero) then
S„ = S^^an. (171)
257. It has been shown "^ that the a factor is given approximately
by
a^{H-^C)e\ (172)
In magnetic temperature compensation (see par. 297, p. 115), C is
approximately —]{ H, H-{-C^}2H, and
a^'AHe'. (173)
258. Table 7 gives the a factors computed from equation (173)
(using C= -V^H), and the observed a factors, for the H variometers
at four Coast and Geodetic Survey magnetic observatories.
< H. H. Howe, op. cit., p. 36 (see item 6 of bibliography).
H VARIOMETER
97
TABLE 7. — Computed and observed a factors at U. S. Coast and Geodetic
Survey observatories
Recording distance, R
1
^ = 2^
H
H
Observatory
Honolulu
San Juan
Sitka
Tucson
1126 mm
0. 000444
28620
14310
0. 0028
0. 0028
1144 mm
0. 000437
27470
13735
0. 0026
0. 0020
1127 mm
0. 000444
15500
7750
0. 0015
0. 0020
1215 mm
0. 000412
26020
13010
0. 0022
0. 0024
2
a=HHe2
a (observed).
259. Use of the a factor, — The base-line scale value, aS'o, and the
point scale value, Sh, at ordinate n, are connected by the relation
SH=So-\-an.
(174)
To obtain the value of the ordinate in gammas from the ordinate, n,
in millimeters, multiply the ordinate in mm by the average of the
scale values at zero ordinate and at the ordinate n. This average
scale value is given by
S^=So+y2 an (175)
in which
>Sff=the scale value at the ordinate -J
aS'o = the base-line scale value.
260. To find the scale value for zero ordinate or for a particular
ordinate when it is known at another ordinate, use equation (174).
Note that a scale value to be used for conversion of ordinates is based
on equation (175) but that the scale value applicable to small changes
at a particular ordinate is based on equation (174).
261. Correction to H scale value for change in declination, —
For greater precision in calculating the H scale value, a correction for
change in declination is sometimes applied. By taking the differ-
ential of equation (145), with k constant,
ASH=eAp.
(176)
Suppose ^=90° and D changes but H does not change, figure 65.
There will now be a component of H parallel to Ms which did not
exist when Ms was normal to H. This parallel component is, from
figure 65,
HsmAD=Ap. (177)
Substituting this value of Ap in equation (176)
ASff^eHsin AD.
(178)
98
MAGNETIC OBSERVATORY MANUAL
Let AD=4:' at time of H scale-value observations; iJ= 18316; and
e = 0.000424.
Then
A&=:0.000424X 18316 Xsin 4'
= 0.009 7/mm.
262. In paragraph 261, AD is equal to the difference between the
mean value of D, say for 1 year, and the D ordinate at the time of the
N Ap
Figure 65.— Effect of change of
declination on // scale value.
Figure 66.— Value of declination
affecting H scale value.
scale value observations (See fig. 66.). Note: Equation (178) may
be written (see eq. 381, p. 184),
ASh=(AD) eHsin V
(179)
when AD is a small angle. The value of e ZZ sin T may be taken as a
constant for 1 year. In this case it is equal to 0.0023. And
A&=4'X0.0023 = 0.0097/mm.
To illustrate how large this correction may become even for moderate
changes in declination, suppose ;S'^=3.00; and AD=13'; then
A^H=13'X0.0023 = 0.037/mm
or about 1 percent of the uncorrected H scale value.
CHAPTER 9. THE VERTICAL-INTENSITY VARIOMETER
263. Function of a Z variometer. — This instrument should indi-
cate visually or by photographic registration the variations in the
vertical component of the intensity of the earth's magnetic field.
264. Basic requirements. — The essentials of the instrument are:
(a) A permanent magnet (or pair of magnets) equipped with a
quartz (or equivalent) knife edge near its center of gravity, a balanc-
ing poise for adjustment of the magnetic axis to the horizontal, and
a sensitivity poise for adjustment of the scale value;
(b) A nonmagnetic housing with suitable agate or quartz supports
so arranged that the magnet may oscillate in a vertical plane;
(c) An optical system suitable for visual observations or photo-
graphic registration;
(d) Copper boxes or chambers surrounding the ends of the magnet
system to provide magnetic damping; and
(e) A temperature compensation device.
Note: See chapter 10 for temperature compensation of the Z
variometer and paragraph 331, [page^ 135, for the effects of an
exlevel angle.
265. Operating principle. — To place the instrument in operation,
the magnetic system is balanced so that its magnetic axis is horizontal
and (usually) in the magnetic meridian. When the vertical com-
ponent, Z, of the earth's field increases, the magnetic couple increases
and the south (south seeking) end rises. When Z decreases the
north (north seeking) end rises. Thus any slight change in Z causes
the system to become slightly unbalanced, causing the magnet to
turn slightly in a vertical plane. It is this motion that is observed
or recorded as a change in Z.
266. The equation of equilibrium. — In figure 67, NS represents
the magnetic axis oi sl Z recording magnet of moment Mg, in a bal-
anced horizontal position, N end north.
i7= horizontal intensity of the earth's magnetic field;
Z= vertical intensity of the earth's magnetic field;
iV= north (north seeking) pole of recording magnet Ms]
*S= south (south seeking) pole of recording magnet Ms',
Opposition of the knife edges;
C=the center of gravity of the complete system supported on
the knife edges at 0;
" mi = mass of the supported system = mass of recording magnet
system, hereafter called the mass of the magnet system;
a=jB(7=horizontal distance to C from the vertical, OAB;
positive toward the S end of the recording magnet Ms]
A = OjB= vertical distance to C from the horizontal, OS; positive
downward.
267. When the magnet comes to rest in the horizontal position the
clockwise mechanical couple, rriiga, just balances the counterclockwise
magnetic couple, M^Z sin 90° =M,Z. That is,
MsZ=miga. (180)
100
MAGNETIC OBSERVATORY MANUAL
[Ch.
268. Now let Z increase so that the Z magnet turns counterclock-
wise through a small angle 6. The new positions are shown in figure
67, N having moved to A^', S to S', B to B' , and C to C\ The hori-
zontal distance of the center of gravity from the vertical line OB is
now AD, the new mechanical lever arm. With the recording magnet,
Ms, in the position N'S^, the magnetic couples acting on the magnet
system are:
MsZ sin (90° — 0)=M,.Z cos d, counterclockwise.
MgH sin d, clockwise.
The mechanical couple is rriig AD, clockwise, where g is the accelera-
tion due to gravity. The clockwise and counterclockwise couples must
Figure 67.— Diagram for Z variometer.
be balanced for the system to be in rotational equilibrium in the new
position. Therefore
MsZ cos d=MsH sin d+m,g AD. (181)
From figure 67,
AD=AB'+B'D',
AB' = OB' sin d=OB sin d=h sin d;
B'D=B'C' cos e=BC cos e=a cos 6)
and AD=h sin d+a cos 6.
Substituting into equation (181)
MsZ cos d=MsH sin d+niig (h sin d+a cos 6). (182)
9.] Z VARIOMETER 101
269. Equation (182) must be modified if extraneous fields exist and
if Ms is not in the magnetic meridian. Let
^ = Magnetic azimuth of the recording
magnet = angle between magnetic north
and the direction of M^, as shown in
figure 68 ;
/p=horizontal component in azimuth A of
all extraneous magnetic fields; the/p field
will be small unless the Z variometer
has sensitivity-control magnets;
C= component of all extraneous magnetic
fields parallel to Z,positive if C is down-
-^OT-/^ Figure 68.— View. looking
^* down, of Z variometer mag-
When A, /», and C are taken into account, net oriented at an angle A
,• /ioo\ X. with the magnetic meridian.
equation (182) becomes
Ms (Z+C) cos d=Ms (HcosA+Jj,) sin d+niig (h sin d+a cos (9). (183)
This is the more general equation of equilibrium for the Z variometer.
Solving equation (183) for 6,
tan^= ,,,y'(^ + q,-"'f'^ ■ ■ (184)
Ms(H COS A-\-fp) + m]gh
Sol^'ing equation (183) for Z,
Z=[(;/cosA+/,) + ^]tan9+"^-C. (185)
In normal operation 6 is small and tan 6^6. Then
Z=e [iH cosA+,f,)+"^]+'^£-C. (186)
270. The scale-value equation. — In the operation of Z vari-
ometers at magnetic observatories we are concerned with the Z scale
value, that is, the ratio of the change in Z to the change in the ordinate,
n, of the Z spot. The rate of change of Z with respect to 6 is given
by differentiating equation (186) with respect to 6,
dZ=^{H cos A+f,) + '^'jdd. (187)
This is the Z scale-value equation in which all quantities are in cgs
units and 6 is in radians. In paragraph 202, page 75, it has been
shown that
ci«=g (188)
where c^n= change of ordinate of spot, in millimeters;
and i?= effective distance, Z variometer to drum, in millimeters.
Hence from equations (187) and (188)
dZ
d
l=^[(HcosA-,U) + ^] (189)
Ai/t
102 MAGNETIC OBSERVATORY MANUAL [Ch.
In gammas per millimeter:
Sz=^[(ffcos4+/,) + ^]xiO^ (190)
Equation (190) is the general equation for the scale value of the Z
variometer in gammas per millimeter.
Example: R =2244 mm;
H =0.1875;
A =0;
/. =0;
mi =20 gm;
g =980 cm/sec/sec;
h =0.00322 cm=32.2 microns;
M,=314 cgs;
by equation (190):
aSz = 8.667 per mm.
As explained later (par. 276), /i. may be either positive or negative.
271. Center of gravity and poise displacement. — It can be seen
from equation (190) that the scale value can be increased by increas-
ing h. Before taking up this point (paragraph
274) it will be convenient to see how the posi-
I tion of the center of gravity is affected by dis-
^ placement of each of the poises. SomeZvariom-
1^2 ^ eters have three poises, as shown in figure 69,
Gr^ where
m3=mass of balancing (latitude poise) on
a horizontal invar spindle;
m,=mass of temperature poise on a horizon-
Figure 69.— The adjusting ^ i i • • ii
poises of the z variometer tal alummum spmdlc ;
™*^"®*- and m2 = mass of sensitivity poise on a vertical
spindle.
It may be shown that when the sensitivity poise, 1712, is displaced
upward on its spindle a distance Ay from A to B, figure 69, the center
of gravity will be displaced upward from C to C , a distance — ^ Ay
where m^ is the mass of the magnet system. Since h is measured
positive downward and Ay is taken positive upward, then
dh^-'^dy. (191)
TYli
272. Similarly when nit is displaced outward a distance dLi, then
the mechanical lever arm (a in figure 67) of the center of gravity, C,
will be changed in amount da, given by
da=-^dLt (192)
Ml
and when m^ is displaced outward a distance dL^
da= + —'dLs. (193)
nt]
N-g-
mt
4 1 "3
9.
Z VARIOMETER
103
When there are no temperature changes and the poises lUt and m^
are adjusted,
da-
mi rrii
(194)
273. When a temperature change dt occurs, it will cause displace-
ments dLt of the temperature poise, m,, and dL^ of the latitude
poise, m^. Moreover the centers of mass of the aluminum and invar
spindles, and perhaps of M,., itself, will also be displaced; if these are
neglected, then
dt nil dt nil dt
274. Effect of h on *Sz.— From equation (190)
l5
dSy
niigXlO^
(196)
The change in h may be effected by changing the position of the
sensitivity poise as shown schematically in figure 69. Substitute
for dh the right-hand member of equation (191), obtaining
(197)
icy ms^XlO^ ,
Example: m2 =0.0955 gram;
g =980 cm/sec/sec;
R =2244 mm;
Ms=314 cgs;
dy =0.1370 cm = 5 turns;
by equation (197),
o?*^3 = 0.9l7/mm;
also, if mi = 20 grams, then by equation (191),
rf^=— 0.00065 cm=— 6.5 microns for 5 turns of the poise.
Table 8 shows the variation of scale value with h for two typical cases,
TABLE 8. — Variation of Z scale value with adjustment of sensitivity poise
^ = 0°
^=180
°
TIVITY
Ah
POISE
SCALE
VALUE
HCOS^^j^
^'^'^ X105
'^^'■'xio'
SCALE
VALUE
turns
microns
7/mm
7/mm
7/mm
7/mm
7/mm
8.66
+4.18
+4.48
-4.18
+0.30
5
-6.6
7.87
+4.18
+3.69
-4.18
Unstable
10
-13.1
6.97
+4.18
+2.79
-4.18
Unstable
15
-19.7
5.89
+4.18
+1.71
-4.18
Unstable
20
-26.2
5.12
+4.18
+0.94
-4. 18
Unstable
25
-32.8
4.08
+4.18
-0.10
-4.18
Unstable
30
-39.3
3.19
+4.18
-0.99
-4.18
Unstable
35
-45.9
2.23
+4.18
-1.95
-4.18
Unstable
40
-52.4
1.33
+4.18
-2.85
-4.18
Unstable
45
-59.
0.64
+4.18
-3.54
-4.18
Unstable
47.6
-62.4
0.00
+4.18
-4.18
-4.18
Unstable
50
-66.5
Unstable
+4.18
Unstable
-4.18
Unstable
104 MAGNETIC OBSERVATORY MANUAL [Ch.
275. Equation (190) may be solved for h\
A = [2i?^,X10-^-(iJcos^ + /p)]-^. (198)
rriig
Example: R =2244 mm;
*S'z=8.667 per mm;
H -0.1875 cgs;
A =0;
/. =0;
M,=314 cgs;
mi =20 gm;
g =980 cm/sec/sec;
by equation (198),
A=0.00322 cm=32.2 microns.
Example: Same as above, except that Sz is now instead of 8.66.
By equation (198),
A=— 0.0030 cm=— 30 microns, that is, the center
of gravity will be above the knife edge.
276. The ordinary analytical balance (nonmagnetic), is so con-
structed that the center of gravity is very close to the central knife
edge, but always below it. Such a balance may be adjusted by means
of a counterpoise so that its center of gravity will be at the same
height as the knife edge or above it. In either case the balance beam
would be unstable. This would be true also of the magnetic balance
(Z variometer) if it were operated with no sensitivity-control magnets
and with its magnetic axis strictly in the magnetic prime vertical,
N end east or A^ end west; that is, if no component of H or fp acts
on the magnet. Suppose now the recording magnet is operated with
its axis in the magnetic meridian with its A^ end to the north. As
stated in paragraph 268, there will be a clockwise couple acting on
the magnet and this couple is equal to MsH sin 6. Thus it is possible
to operate a magnetic balance with its center of gravity above the
line of support {h negative) and maintain stability for long periods
of time. It should be noted in passing that the distance h for a Z
variometer as it is normally operated, is extremely small, usually of
the order of 20 to 50 microns.
277. Effect of fp on Sz* — If /p is positive, that is, if it is directed
the same as the Z recording magnet, Sz will be increased, or what is
the same thing, the magnet will become less sensitive to changes in Z.
If /p is negative (directed opposite to Mg) it becomes more sensitive
to changes in Z, and Sz will be decreased. Thus the scale value of
a Z variometer may be adjusted over a wide range by use of control
magnets properly placed and oriented with respect to the recording
magnet. Control magnets have been used on la Cour Z variometers
when it was not possible to obtain the desired sensitivity by other
means.
278. Differentiating Sz with respect to/p, in equation (190),
dS,=^XlO'; (199)
9.] Z VARIOMETER 105
and if /p is in gammas
dSz='^^- (200)
Example:
i?=2244 mm;
/S^^ 8.667 per millimeter;
4488
change /p to +44887; then by equation (200) , dSz= r, w no a a ^ ~^ ^ 7/mm
and /S'^ becomes 9.667/mm; or change fj, to —44887, dSz is now
— 17/mm and Sz becomes 7.667/mm.
279. Effect of H on Sz* — Differentiating equation (190) with
respect to H,
dSz=^^X10'dH (201)
^^^"^dHy (202)
2R
It is seen that the scale value is not affected by H when M^ is directed
to magnetic east (^ = 90°) or to magnetic west (^=270°).
Example:
i?=2244 mm;
C^iJ^ = 1007.
By equation (202),
^&=^=0.0227/mm.
It is evident from the above that the change in Sz due to a small
change in H is negligible for practical purposes,
280. Effect of A on &.— From equation (190)
dSz=-^^^^XlO'dA, (203)
When dA is in degrees (c^^° = 57.3 dA) and 77 is in gammas (i7^=^X 10^),
^o _ Hy sin A dA""
^^^ 114:0 ^^^^)
Example: ^-y= 187507=0.1875 cgs;
A° = 90°;
i?=-2244 mm;
change A to 95°; then c?^° = 5° =0.0873 radians; and by equation
(203) or (204),
dSz = 0.36y/mm.
Table 9 shows the variation of scale value, Sz, with azimuth A of
recording magnet for a typical case.
210111—53— — 8
106 MAGNETIC OBSERVATORY MANUAL [Ch.
281. Scale values of a typical Z variometer, — In order to give
some idea of the magnitude of the various quantities entering into
the Z scale value, results of various tests with a Schulze variometer
are given in table 8. In these tests scale-value observations were
made first with the sensitivity poise in its lowest position. The poise
was then raised by steps of 5 turns (5 turns=0.1370 cm), up to the
point of instability (47.6 turns =1.304 cm) keeping A constant at 0°,
that is, N end north. The first column of table 8 gives the 5-turn
steps. Column 2 is derived from equation (191) for m2=0.0955 gm
and mi =20 gm. Five turns upward adjustment of the sensitivity
poise decreases h and raises the center of gravity by 6.5 microns.
Column 3 is the observed scale value; column 4 is the value of the
constant term, — —^ — XIO^, of equation (190); and column 5 is the
difference between columns 3 and 4, that is, the variable term,
^1^^X10^ By equation (197),
For mi = 20 gm, ^=980 cm/sec/sec, i?=2244 mm, and Ms=314 cgs,
by equation (205),
—7-^=— 6.64 7 per mm, per cm of upward displacement of the poise.
Since 5 turns=0.1370 cm, 1 turn=0.0274 cm; and for one upward
turn of the poise,
dSz= -6.64X0.0274= -0.1827/mm.
282. Suppose now that the variometer described above is operated
with the A^ end of the recording magnet to the south and with the
magnetic axis in the magnetic meridian. For this condition, ^=180°,
cos ^= — 1, and the term
— ^P — XIO^ becomes — 4.187/mm (column 6 of table 8).
Then for the first position of the poise (0 turns), the scale value
would be
&=-4.18 + 4.48 = 0.37/mm,
as shown in column 7. This means that the magnet would be ex-
tremely sensitive, almost to the point of instability. In fact it would
require an increase of only 0.45 mm (1.64 turns) in Ay to make it
unstable.
283. Table 9 gives the computed values of Sz for all azimuths in
steps of 30° from 0° to 360°.
i
9.] Z VARIOMETER 107
TABLE 9. — Variation of Z scale value with azimuth of recording magnet
A
Cos A
2R ^^^
Z Scale
value
o
yimm
y/vam
7/mm
+ 1.00
+ 4. 18
4.48
8. 66
30
+ 0. 866
+ 3. 62
4.48
8. 10
60
+ 0. 500
+ 2.09
4.48
6. 57
90
4. 48
4. 48
120
-0. 500
-2.09
4.48
2.39
150
-0. 866
-3. 62
4. 48
0.86
180
-1.00
-4. 18
4. 48
0.30
210
-0 866
-3.62
4.48
0.86
240
-0. 500
-2.09
4.48
2.39
270
4.48
4. 48
300
+ 0. 500
+ 2.09
4.48
6. 57
330
+ 0. 866
+ 3. 62
4. 48
8. 10
360
+ 1.00
+ 4. 18
4. 48
8.66
284. Adjustment for latitude, — The vertical intensity varies
from zero at the magnetic equator to approximately ±70,0007 at the
magnetic poles. If the variometer is moved to a site where Z is con-
siderably different, it will be necessary to rebalance the magnet by
adjustment of the latitude poise, ma, figure 69. If nii is the mass of
the magnet system, m^, the mass of the latitude poise, and dL^ the
distance it must be moved outward along the invar spindle in order
to keep the magnetic axis horizontal, we have by equation (193)
mi
By equation (186) when only Z and a change,
dZ= qr-^ da
Ms
and, combining equations (206) and (207),
Example:
^ By equation (208)
dZ^^'^dU
m3=2.444 grams;
^=980 cm per sec^;
M,= 1225 cgs;
dLs= one turn =0.04 cm.
dZ=0.0782 cgs=78207.
(206)
(207)
(208)
I That is, one turn of the poise on the latitude spindle is equivalent to
78207 change in Z.
108 MAGNETIC OBSERVATORY MANUAL
285. Adjustment of both latitude and temperature poises, —
For very large changes in Z, it would be necessary to adjust the tem-
perature poise, nit (see par. 292, ch. 10) and also to adjust the lati-
tude poise to keep the magnet balanced in the horizontal plane.
Combining equations (194) and (207),
dZ=^-'gdLt+^fjL, (209)
and dL,^^ dZ+^' dLt. (210)
m^g ma
The term — - dZ is the adjustment of the latitude poise needed to bal-
ance the change in Z and the term — ' dLi is the additional adjustment
TYl^
of the latitude poise needed to balance out the effect of the shift of
the temperature poise.
Example: Z changes from 0.538 to 0.200 cgs; pitch of temperature
thread 0.06 cm per turn; pitch of latitude thread 0.04 cm per turn;
dZ=-0.33S;
mi=2.030 grams;
c?Zj= — 1.074 cm to maintain temperature compensation
= — 17.9 turns;
m3=2.444 grams;
Ms=1225 cgs;
^=980 cm per sec^.
By equation (210)
dL^=— 0.1729 cm=— 4.3 turns to balance dZ;
and c?Z3=— 0.8921 cm =—22.3 turns to balance dL(.
Therefore the total adjustment of the latitude poise is
c^Z3= -0.1729 cm-0.8921 cm=- 1.0650 cm
= —4.3 turns — 22.3 turns=— 26.6 turns.
286. Further details on the adjustment of the temperature poise
are given in chapter 10, pages 111-112.
CHAPTER 10. TEMPERATURE COEFFICIENTS
OF VARIOMETERS
287. Meaning of the variometer temperature coefficient, — As
stated in paragrapli 267, page 99, wlien ^=0, the magnetic axis
of the recording magnet is horizontal, and
MsZ=miga. (211)
Suppose that Z remains constant and the temperature of the recording
magnet increases, resulting in a decrease in Ms- Then the product
MgZ will be less than ruiga (a is assumed to be constant in this example)
and the A^ end of the magnet will rise, indicating on the magnetogram
an apparent decrease in Z. Now let Z increase by an amount sufficient
to balance the effect due to the change in Ms- Then we may say
that the change in Z is related to the change in Ms as follows:
dZ m,ga ^^12)
dM, M
and since miga=MsZ
dZ M.Z Z (2^3^
dMs Ms' Ms
But, by definition, — ^r— per degree centigrade change in temperature
is the temperature coefficient of the magnetic moment. That is,
i,fj = — gi. Transposing equation (213) and dividing by dt,
dZ dMs , /r.-. 4\
Zdt—Mjt=+^^ (21^)
and '^=+g,Z=Q,. (215)
The term q^Z is the temperature coefficient, hereafter called Qz, of
this Z variometer. In general the temperature coefficient will contain
other terms in addition to qiZ.
288. Qz may be defined as the temperatu? e coefficient oj the Z spot,
always expressed in gammas per degree C. It is equal to the negative
of the apparent change in Z, as recorded on the magnetogram, caused
by a change in temperature of 1° C. It is also equal to the real
gamma change in Z that would be required to return the Z spot
to its original position after a temperature change of one degree
centigrade causes the spot to be deflected. The temperature coeffi-
cient, Oh, of the H variometer is similarly defined. The temperature
coefficient, Qd, is defined as the declination change, in minutes of arc,
per degree centigrade. It may be noted that this definition of Q
leads to the correction term Q (t—to) in the derivation of values from
the magnetogram (par. 424, p. 165).
109
no
MAGNETIC OBSERVATORY MANUAL
[Ch.
289. Mechanical compensation of the Z variometer, — Figure
70 is a plan view and side elevation of a Z recording magnet equipped
with a device for mechanical compensation.
Let
mi =the mass of the recording magnet system in grams;
m2 =the mass of the sensitivity poise;
Ms =the mass of the balancing (latitude) poise;
rUt =the mass of the temperature poise;
Lt = effective length of the temperature compensating arm
(aluminum) ;
the latitude adjustment arm (invar);
a =the lateral distance of the center
of gravity of the magnet system
from the center of support when
^ = 0;
6 = angle between the magnetic axis
of Ms and the horizontal;
Ms = magnetic moment of the record-
ing magnet;
Z = vertical intensity ;
i = temperature of the magnet sys-
tem;
qi = temperature coefficient of the
magnetic moment Ms ;
temperature coefficient of the variometer in gammas per
degree C;
g = acceleration due to gravity =980 cm/sec/sec;
a = coefficient of linear thermal expansion of aluminum =
0.000023 per degree C;
p = temperature coefficient of the distance a.
The bronze mass rrit is threaded on the aluminum spindle of effective
length Lt', the bronze mass, rris, is threaded on the invar spindle of
effective length L3. The two masses m, and rris, are approximately
equal. The coefficient of linear thermal expansion of aluminum is
taken as 0.000023 per degree C and that of invar as practically zero.
290. If the temperature of the aluminum spindle increases by an
amount A^, the mass rut moves outward through a distance aLtAt, and
the increase in the mechanical couples on this side of the magnet sys-
tem is rrit gaLtAt. The product of rrit and its horizontal displacement
due to a moderate temperature change is approximately equal to the
product of the total mass of the magnet system rrii and the horizontal
displacement of the center of gravity of the system due to the same
temperature change. That is
Figure 70.— Vertical-intensity recording
magnet, Schmidt type.
ft
mtOiLi
rriiap
(216)
or
^ nil
(217)
in which the minus sign is required because a and Lt are taken positive
in opposite directions. The temperature coefficient ^ is a complex
function of the masses, lever arms, and temperature coefficients of the
various parts of the recording system.^ If both the change of M,
J C. A. Heiland and W. E. Pugh, Am. Inst. Min. and Met. Eng., Tech. Pub. 483, p. 13, 1932 (see item
5 of bibliography).
10.] TEMPERATURE COEFFICIENTS 111
and that of a with t are taken mto account, equation (215) becomes
<2z=2,Z+g'a?'. (218)
Substituting for ap from equation (217),
Q,^q,Z-'^iaL, (219)
and ^^?=2,. (220)
It should be noted in passing that the coefficient of linear thermal
expansion of invar is only about Ke that of aluminum and even though
the change in the mechanical moment on the invar side of the system
(due to a temperature change) is extremely small, it is appreciable.
This effect is almost exactly balanced by the change in the mechanical
moment of the aluminum spindle itself for the same temperature
change.^
291. Compensation for moderate temperature changes will be
achieved when Qz=^, that is,
Cz=giZ-^g^'=0 (221)
Z.=^^^. (222)
amtg
Example (a) :
gi = 0.000 138 per degree C;
Z=:0.538 cgs;
then by equation (221),
Lt=0 (uncompensated variometer);
^2= 0.000 074 cgs per degree C;
= 7.47 per degree C.
Example (b) :
Z=0.538 cgs;
M,= 1225 cgs;
2i = 0.000 138 per degree C;
a=0.000 023 per degree C;
m«=2.030 grams;
^=980 cm/sec/sec;
then by equation (222),
Z, = 1.99 cm.
If the center of m, is set precisely at 1.99 cm along the aluminum
spindle and the latitude poise, rriz, is adjusted along the invar spindle
so that the magnet is just balanced in the horizontal plane, the vari-
ometer will be compensated for temperature at a site where Z is 0.538
cgs (53,8007). It is apparent from equation (222) that Lt must be
2 J. W. Joyce, Manual on Geophysical Prospecting with the Magnetometer, U. S. Dept. Int., Bur. of
Mines, 1937, p. 42 (see item 7 of bibliography).
112 MAGNETIC OBSERVATORY MANUAL [Ch.
increased or decreased in direct proportion to Z in order to preserve
perfect temperature compensation. That is,
L/ 7J
(223)
292. If Q is of the order of d=l7 per degree, it will be satisfactory
for all practical purposes. What change in Z would be required to
change Qz by I7 per degree centigrade if gi =0.000 138 per degree
centigrade? By equation (220),
^Qz=q^^Z (224)
and AZ=^. (225)
Then ^Z= ^
0.000138
= 72007 approximately.
That is, Z may change by approximately 72OO7 before Qz will change
by more than I7 per degree for this variometer. This variometer,
compensated for temperature where Z= 53,8007, would be considered
practically compensated for values of Z ranging from 46,0007 to
61,0007, a range of 14,4007. Beyond this range it would be necessary
to adjust the temperature poise in accordance with equation (223)
in order to preserve compensation within one gamma per degree C.
For very large changes in Z, for example when a Z variometer is
moved from a place where Z is 0.538 cgs to a place where Z is 0.200
cgs, both the temperature and latitude poises must be adjusted, as
explained in paragraph 285. It is important always to make latitude
adjustments with the latitude poise only and not to adjust the temper-
ature poise unless one wishes to change the temperature coefficient of
the instrument.
293. In south magnetic latitudes Z is negative, and the A^ end of
the recording magnet will move downward when there is a small
increase in temperature and there is no compensation. In order to
achieve compensation when Z is negative, the aluminum and invar
spindles should be interchanged, or what amounts to the same thing,
the magnets should be removed from the central block of the system
and reversed, so that the aluminum spindle is on the same end as the
S end of the system. Under these conditions giZ is negative. For
an uncompensated variometer, Qz is always positive in north mag-
netic latitudes and negative in south magnetic latitudes. Qz then
will have the same sign as Z; and Qz(t—to), the correction for temper-
ature, is applied algebraically to observed Z in order to obtain the
true Z.
294. Temperature coefficient and scale value, — Heiland^ has
shown that the temperature coefficient of a variometer compensated
as described above is practically independent of the scale value of the
instrument. In other words, Qz is independent of Sz when Qz is ex-
3 Heiland and Pugh, op. cit., p. 21 (see i tern 5 of bibliography).
10.1
TEMPERATURE COEFFICIENTS
113
pressed in gammas and not in arc or millimeters of ordinate. Heiland
has shown also that the temperature coefficient of the scale value itself
is negligible.'' These conclusions are based on the assumption that
the axis of translation of the sensitivity poise is precisely perpendicular
to the magnetic axis of the recording magnet. (It will be shown
later, paragraph 310, that for optical compensation the temperature
coefficient is dependent on the scale value.)
295. Magnetic compensation, — Temperature compensation may
be accomplished by use of an auxiliary magnet (or magnets) to reduce
the component of the vertical field acting
upon the recording magnet. An increase
in temperature reduces the magnetic mo-
ment of both magnets. Reducing the
magnetic moment of the recording magnet
causes an apparent decrease in Z. At the
same time the field of the compensating
magnet is reduced numerically, causing
an increase in the resultant field acting
on the recording magnet, an apparent
increase in Z. When the proper relation
exists between the temperature coefficients
of the two magnets and the amount of
reduction in the field acting on the record-
ing magnet, first-order temperature com-
pensation is effective. In practice the
temperature magnet is fixed relative to the recording magnet as
shown in figure 71. The position of Ma is reversed in south mag-
netic latitudes, that is, the A^ end of Ma is down. Equation (182)
may be written, for 6=0°,
Mr
Figure 71. — Magnetic temperature
compensation of Z variometer.
or
Ms{Z+C)=m,9a
MsZ+MsC=m,ga
(226)
(227)
in which C=the amount by which the vertical field acting on the
recording magnet is altered;
Ms==the magnetic moment of the recording magnet;
and mi^a=the mechanical couple balancing the magnetic couple
Ms{Z+C).
(It is to be noted that Z and (7 have opposite signs. In north magnetic
latitudes, Z is positive and C is negative.) Let the temperature of the
whole system rise by an infinitesimal amount, dt. This will cause a
change in C, Ms, and a, and the result will be an apparent change in
Z (increase or decrease) . C decreases as a result of a decrease in the
magnetic moment of the temperature magnet and because of the
increase in the distance r (fig. 71), due to the expansion of the material
separating the magnets (usually this material is brass).
296. Now let Z increase by an infinitesimal amount, just sufficient
to keep the magnetic axis horizontal. Then from equation (227) we
have
(M,+AM,)(Z+AZ) + (M,+AM,)((7+A(7+A'(?)=mi^a + m,^Aa. (228)
Heiland and Pugh, op. dt., p. 11.
114 MAGNETIC OBSERVATORY MANUAL [Ch.
In this equation ACis the change in Cdue to a change in the magnetic
moment of the temperature magnet and A'(7 is the change in C due to
the change in the distance between magnets. Expanding equation
(228) and neglecting second order terms,
MsZ+MsAZ+ZAM.+MsC+MsAC+MsA'Ci-CAMs
=miga-\-migAa. (229)
By equation (227),
MsZi-MsC=miga.
Subtracting equation (227) from equation (229),
MsAZ+ZAMs+MsAC+MsA'C+CAMs=migAa. (230)
Dividing by A^,
MsAZ , ZAMs , MsAC , M.A'C , CAMs _ m,gAa
At "^"aT"^ A^ "^ A^ + A^ - A^ ^^'^^^
A '7
Let — 7 =()2= temperature coefficient of the variometer in
gammas per degree C ;
— 71-/- Af ~Q.^ = temperature coefficient of the magnetic moment
of the recording magnet ;
AC
—-pXi ~9^2 = temperature coefficient of the compensating field
at constant r;
A'C
—yyir, —^3 = temperature coefficient of the compensating field
at constant Ma]
— — =gi = temperature coefficient of mechanical couple =
p (par. 289).
Substituting these values of the g's in equation (231), having regard
to signs, and noting that MsZ-\-MsC=miga,
MsQz-ZMsgi-MsCg2-MsCq,-MsCg,=m,gaq,=q,{MsZ+MsC) (232)
from which
Qz=Z(q,+q,) + C(g, + q2+q.z+g[4)- (233)
When compensation is effective, Qz=0, and
Z(q^ + g,) = -C{q, + g2 + q^ + qi) (234)
or
§= J't^\ • (235)
Z qi+q2+q3 + qi
Let S^=9:i + 9'4
10.] TEMPERATURE COEFFICIENTS 115
and S'?=£> = — 0.078 minutes of arc of east declination
per degree C, which means that a temperature increase of one degree
G will produce an apparent increase of east declination equal to
0'.078. This example shows that Qd will usually be negligible.
304. Optical compensation of the Z variometer, — In the optical
system of temperature compensation, described by la Cour and Peter-
sen,^ the motion of the Z spot caused by a temperature change of the
Z recording magnet is compensated by passing the incident and
reflected light through a 90° prism attached to a nonmagnetic,
bimetallic strip of silver and platinum. The strip is assumed to be
at the same temperature as the Z magnet at all times. Let the prism
be clamped and let Z remain constant while the temperature rises a
definite amount. The magnetic moment, Ms, of the recording magnet
will decrease. The Z spot will be displaced, indicating an apparent
decrease in Z. Now let Z continue to remain constant but with the
Z magnet clamped and the prism undamped and let the system be
subjected to the same temperature change as above. If the strip is
properly faced and is adjusted to the correct sensitivity, the displace-
ment of the Z spot caused by the bending of the strip will be just equal
and opposite to that caused by the change in Ms described above.
The resultant motion of the Z spot, when both magnet and prism are
undamped, will be zero. Under these conditions the variometer is
optically compensated for temperature.
305. In the la Cour variometers, a portion of the light which passes
through the reflecting prism is intercepted by a fixed mirror and reflected
back through the prism to the drum, forming the temperature spot.
The bimetallic strip then serves two purposes, to provide optical
compensation for the variometer and to act as an optical thermograph
for recording changes in the temperature of the instrument.
306. The constant, Ct, of the bimetallic strip. — In order to
adjust properly the length of the strip for a particular instrument, one
should know the sensitivity of the strip to temperature changes.
The strip constant, Ct, is defined as the motion of the T spot in mm on
the drum, per mm length of strip, per degree change in temperature,
at a recording distance of 1 cm. The constant may be furnished by
the manufacturer, or determined experimentally by subjecting a known
length of the strip to a known temperature change and noting the
motion of the T spot at a known recording distance. If the length /
mm of the strip is subjected to a temperature change tz—ti when the
effective recording distance is R and the ordinate of the T spot
changes from Ui to 712 millimeters, then the strip constant, Ct, is
given by
Also, n2-ni = CTRl(t2-ti). (266)
7 D. la Cour. La Balance de Godhavn, p. 17 (see item 8 of bibliography).
10.] TEMPERATURE COEFFICIENTS 121
307. The sensitivity and scale value of the bimetallic strip, —
The sensitivity, Tg, of a strip is defined as the change in the miUimeter
ordinate of the T spot caused by a change in temperature of one degree
C. Using the same notation that was used in equation (265),
rp_n^ni pg^j
and by equation (265)
T^^'^^—^ = CtRI (268)
The reciprocal of the sensitivity, that is the change in temperature
in degrees C represented by one mm change in ordinate of the T spot,
is the scale value of the T spot, and is designated St. Thus if St is
known, variations in temperature may be scaled directly from the
temperature curve. By definition,
(269)
^»
and by equation (268),
/ /, 1
(270)
Example: A platmum-silver strip was tested at Cheltenham
Observatory. It was found that the angular displacement of the
end of the strip was proportional to the temperature change over a
range of 15° C. In these tests,
St-
1
iS'.,-
t2-
-ti
1
Or
712-
-Ui
CtRI
I
R=231 cm;
1 = 22 mm;
n2—ni = 20 mm;
^2-^1=8° a
equation (265),
Cr^ 0.00049 per cm per degree.
equation (267),
r,=2.5 mm/°C.
equation (270),
^r=0.4°C/mm.
308. The temperature coefficient with optical compensa-
tion, — For effective temperature compensation the ordinate of the
Z spot should not change when Z is constant and temperature changes.
Let
AZi= apparent change in Z due to A^ change in temperature with
recording magnet free, strip clamped, and Z constant;
AZ2= apparent change in Z due to A^ change in temperature,
with recording magnet clamped, strip free, and Z constant;
q_i=--irf —17^= temperature coefficient of magnetic moment of
Z magnet;
210111—53 y
122 MAGNETIC OBSERVATORY MANUAL [Ch.
$^= temperature coefficient of the variometer, 7/° C;
aS'z== scale value of the Z variometer, 7/mm;
i?= distance from Z lens to drum;
Cr= strip constant, or motion of T spot in mm, per mm length
of strip, per degree C, for R=\ cm (furnished by maker
or determined from equation (265));
/= length of strip in mm.
Now let Z remain constant while the temperature of the Z recording
magnet changes by A^. This will cause the Z spot to move as a result
of the change of magnetic moment of the recording magnet, and the
apparent change in Z is, by equation (214) (p. 109),
AZi=-(7iZA^. (271)
Again, let the temperature of the strip change by A^. With the strip
facing properly, the Z spot will move in the opposite direction due to
the bending of the strip and the motion of the Z spot, An, will be, by
equation (266),
Au^RICtM (272)
and
AZ2=+SzAn=RlCTMSz- (273)
The net apparent change in Z will be
AZa = AZi + AZ2
= -qiZAt + SzRlCTM; (274)
and by definition of Qz (par. 288),
Also
and
Q^=-^=q,Z-RlCTSz. (275)
l=^W^ (276)
^^=-^%- ^277)
Also, from equation (275),
^_ ^z I x..^.^z . (278)
Qz ~h RICtSz
When the two apparent changes AZi and AZ2 just balance each other
and there is no motion of the Z spot resulting from a temperature
change, the variometer is compensated and ^^=0. When Qz=^,
q,Z-RlCTSz = (279)
and
10.] TEMPERATURE COEFFICIENTS 123
309. Estimation of strip length for optical compensation, —
The following examples illustrate the application of the equations of
paragraph 308:
Example (a): Let qi =0.00030 per degree C;
z =232007;
R =225 cm;
Cr= 0.00049 per cm per degree;
Sz =3.007/mm.
Z=21 mm.
By equation (280):
This variometer, having a scale value of 3.007/mm and operating
where Z= 23,2007, will be optically compensated for temperature
when the length of the strip is 21 mm. It is seen from equation (275)
that the temperature coefficient of the variometer will change if the
Z scale value changes. Equation (280) shows that for Qz=0, I will
vary directly as Z and inversely as R and Sz- For a constant recording
distance and scale value, I must be made longer as Z increases and
shorter as R and Sz increase. This condition is expressed in the
following equation:
1^ = 1 ~y o/e /' (281)
Example (b): Let Z =53,8007;
R =231 cm;
^, =3.607/mm;
Cr =0.00049 per cm per degree;
I =22 mm.
By changing the temperature artificially, or by comparing with another
variometer, it is found that
&=7.37/°C.
What is the strip length for Qz=07 Let A0z=-7:SyrC.
By equation (277) A/=18 mm. Then, for Oz=0,
1=22 mm + A/=40 mm.
Also, what is the value of qi in this case? By equation (278), ^1 =
0.00030 per degree C.
Example (c): Suppose it is desired to move a Z variometer from
Cheltenham where it is properly compensated for temperature to
San Juan. What should be the new length of the bimetalhc compen-
sating strip? At Cheltenham /=40 mm, Z=53,8007, i?=231 cm,
124 MAGNETIC OBSERVATORY MANUAL [Ch.
and S^=3.Qy/mm. At San Juan Z will be 35,5007, R will be 240 cm,
and the required scale value S^, will be 2.557/mm. By equation (281)
/'=40x|Sx^4ix'-'
53,800'"240""2.55
=36 mm.
310. Optical compensation of the H variometer. — This may
be accomplished in the same manner as that used for the Z instrument
except that we must now take into consideration the changes in the
torsion constant of the quartz fiber with changes in temperature. Let
qi ==— TT --7T^= temperature coefficient of the H recording
magnet ;
1 dk^
gs =-p -7^ = temperature coefficient of the torsion constant of
the quartz fiber=0. 00016 per degree C;
Sh =H scale value, in gammas per mm;
Ct = bimetallic-strip constant;
R = recording distance from drum to H lens;
I = length of the strip in nam;
Qjf = temperature coefficient of the variometer, in gammas per
degree C;
A^i= apparent change in jFZdue to change in temperature. At, with
recording magnet free, strip clamped, and H constant;
AZZ2=apparent change in H due to change in temperature, At,
with magnet clamped, strip free, and H constant.
With the prism clamped and the magnet free, and assuming that the
extraneous fields at the center of H are negligible.
AH,= -mq^ + q,)At, (282)
and with the magnet clamped and strip free,
AH2=lCTRSHAt. (283)
The net apparent change when both the magnet and strip are free is
AHa=AH, + AH2
= -H(qi + qs)At+lCTRSHAt, (284)
10.] TEMPERATURE COEFFICIENTS 125
and by definition of Qh, paragraph 288,
QH=-^=H(q, + q,)-mCrS^. (285)
For temperature compensation, Qh=0 and
It is seen that in this system of temperature compensation the value
of Qh will depend upon the // scale value. The formulas developed
for the optical compensation of the Z variometer, paragraph 308,
will also apply to the H variometer if qi is changed to ^i + Ssj ^ to H,
and *S'^ to Sh-
CHAPTER 11. DETERMINATION OF SCALE VALUES
USE OF DEFLECTOR MAGNET
311. The H scale value, — For small angles, the angular displace-
ment of the recording magnet is approximately proportional to the
field applied in the magnetic meridian through Ms. Then the H
scale value is
St=^ (287)
in which aS^ = observed H scale value in gammas per mm;
/n= applied field in gammas;
7/^^= deflection of H spot in mm on the magnetogram.
312. The field, /„, is supplied by a small magnet mounted on the
deflection bar of the variometer, or by a large deflector of high magnetic
moment placed 2 or 3 meters from the variometer (in the magnetic
meridian through the center of Ms and at the same elevation as M^ ,
or by a Helmholtz-Gaugain coil. When a small deflector is used at
short distances, a correction must be applied for distribution by means
of one of the formulas in table 2. When a large deflector is used at
distances which are large compared with the dimensions of the
magnets, corrections for distribution may be entirely negligible.
313. In practice, three deflections are made with the deflector in
the ^position, figure 118 (p. 204), top, first with the N end to the north
and then with the A^ end to the south, the reversals being made in the
manner shown in the sample set of observations, figure 72. For a
double deflection, 2 Uh, equation (287), may be written
S*=|£. (288)
314. In gammas, the field, /„, of the deflector is — ^XIO^, Ma
being the magnetic moment of the deflector and r the deflection dis-
tance in centimeters (center of Ma to center of Ms) . Substituting this
value of /„ in equation (288), and including a distribution factor, a^r,
calculated by means of the top fine of table 2 (p. 13), we have
^^ maOH , (289)
This is the observed scale value at the away ordinate, that is, at the
ordinate, hmm, of the undeflected H spot.
126
SCALE VALUES
127
374d
OePAm-MtNT OF
COMMCNCK
COACT AND
OEOOmC SURVCY
DIV: OftS
Observatory TUCSON
Date Mi4. 1951
S^^ y^UES OF INTENSITY VARIOMETERS
Deflector ^-^1 cm Deflection distance 3^6 cm Observer J B Campbell
(Die l»*»>
H VARIOMETER
(Deflector face up)
D VARIOMETER
(Deflector face up)
Z VARIOMETER
(Deflector face East)
Nend 2u,
N end 2Ub
Nend 2u,
I
i
I
Mean(
! > "7.5
E > ?«-5
Mean(E-W)-2uo 26.7
u -■"
> .«9.o
i > ^7v.^
N-S)=2ue 67.4
u ^ ^9.0
Mean(U-D) = 2u, 29.0
Away
onl.
Before 16.4
Away
Old.'
Before 4« 1
Away Before -6,1
After 14.7
[After 5.4
ord. After _ ^, fi
Mean
^ 15, 6 Aim
Meand
L. 4. 8 mm
Mean z„ - d.o"™"™
S, «-97 r/mm
S; 0. 50 /mm
105 M.Time
K 46 r
d 2'
Began 18 '22
Prel. Hblv. ^593^ 7'
Prel. D blv. 13' 21
Ended i5 53
H^ 25975 r
D 13' 23
Mgph. Temp. 20. 4*C
D 13 24'"
Remarks
(D-D) _ 1'
S;-S^H^tanl'
S: = 4M(10)yH/'2uH
S, = 2M(10)»ot,/r'2u,
log ton r 6.4637
log 4(10)' 5.6021
log 2(10)' 5.3010
logSp' 9.6990
log eX-a 0.0006
lo«o<-r 9-9 9 91
logH^ 4, 4146
cologr* 2, 54^9
colog r* 2. 5429
logSS 0. 5773
log 4(10)»oC^/r' 5. 1456
log2(10)'oL,/r'7. 843^
^ 3. 75
logM 4. 1603
logM 4. 1603
M = r'2u„SS/2c=Co(10)'
colog 2Uh 8. lyi 3
colog2u^ 8. 5376
logr/2(10)» 2. 1561
log S^ 0. 4112
logs, 0. 5415
cologoi„ 0. 0004
Sh' 3. 00 1
s. 3.45
log2u„ 1.4365
-0^0390-5) ■'■.003
Sh=S;- 0» 0030 (D-D)
S. = Se- 0. 0024 h_
logS^ 0.5113
s, 3-004
logM 4. 1603
-0.0024^^ -.031
Sh=S, + 0. 001 a h_
h,= Seh„
M 14460
s. a. 9 7
Scaled
to lpp,p™n
Scalings
by ^M eheck^ by M-?.
Computed
. by
RLR
itioa (or 195-'- ^^ " ^3 ^4 (to nmrml minut
Computations Abstracted
checked by Mp by Mp
Figure 72.— Scale-value observations and computations.
128 MAGNETIC OBSERVATORY MANUAL [Ch.
315. The H scale value of a unifilar variometer depends on the
ordinate of the H spot at the time the deflections are made and on the
magnetic declination, in accordance with the equation,
Sl=S^+ah+h{P-B) (290)
where a =the a factor described in paragraph 255 (p. 95) ;
h =the h factor
= eHsuiV (see par. 262) ;
h =the away ordinate of the H spot, denoted kmm in figure 72;
D == standard declination value, such as the mean for one year;
I) =the declination at the time of the ZZ" scale-value deflections;
5^0= the H scale value that would be observed when /i=0 and
Sh = the observed H scale value when the H ordinate is h and
the declination is D.
Also, if Sh is the H scale value at ordinate h and declination D, then
SH=So+ah (291)
and S*=SH+b{D-D),
In figure 72, Sh and So are calculated from the observed i? scale value,
Sh, as follows:
SH = S*-b{D-D) (292)
So=SH-ah. (293)
If the terms b{D—D) and ah are negligible, then Sh and So may be
taken as equal to Sh- This is usually the case for low-sensitivity
H variometers.
316. The Z scale value, — For the Z variometer, the deflector is
used in the B position, figure 118, center, that is, the center of the
deflector, Ma, is placed in the magnetic meridian through the center
of the Z recording magnet, Ms, axis of Ma vertical and with its mag-
netic center at the same elevation as the center of Ms. In this
position, the field of Ma at the center of Ms is
Substituting this value of /„ in equation (288), changing H to Z,
and including a distribution factor az, we have
&=^^X10'. (294)
11.] SCALE VALUES 129
in which 2 Uz is the double deflection of the Z spot in mm on the
magnetogram, the deflector being reversed in making the deflection
observations in accordance with the example in figure 72.
317. Equation (294) gives the Z scale value at the observed ordinate,
Zmm, which is the away ordinate. The Z scale value varies slightly
with ordinate, but the variation is usually negligible ajid will not be
considered here.
318. The D scale value {gamma scale value), — The deflector is
used in the B position, figure 118 bottom — that is, it is placed with
its center in the magnetic meridian thi-ough the center of the D
recording magnet, at a distance, r, and with its magnetic axis in the
same horizontal plane as the D recording magnet, at right angles
to the magnetic meridian through Mg. The field, /„, producing the
deflection is — 3^X10^, and for a double deflection it is — 3-^x10^.
Substituting this value of /„ in equation (288) , changing H to D,
and including a distribution factor ao, we have
^-^XlO^- (295)
This is the intensity scale value (gamma scale value) of the D variom-
eter. The relation between the gamma scale value and the minute
scale value is described in paragraphs 321-23.
319. Com,putation of scale values from, deflections. — In the
case of a sensitive magnetograph, the D scale value is usually known
quite accurately, and is quite constant since jt is small compared
with unity (see eq. 119). It is then advantageous to derive Ma from
the following equation, based on equation (295) :
, , _ 2unr^S l
^«-2a^>
" >
U
2uz
4. 8 mm.
N end
' >
W
>
E
2u
8.5 mm
N end
N
>
^ >
N
2uh
14. 1 mm
4.8
8.6
14;1
Mean (U-0) ^. 8
Mean (E-W) g. 6
Mean (N-S) j^. ,
Away
Ordinate
Before 14> ^
After 14*
Away Before 13. 6
Ordinote After I3. 2
Away Before $0. 3
Ordinote | After 30.4
All scolings corrected
for shrinkage to..??.?.^.mm.
H from sensitive
magnetograph...l55?P...T
Mognetogroph
Temperature X^ .'c
Sj- 2Mltfotj/r*2u2
Sj « 2MIOi'e»o/r'2u
So • S; cot I'/h
S„- 4Mltfoc„/r»2uH
log 2 10^
log M
log o-j
colog r'
5.3010
4. 0849
0.0000
2.795S
log 2 10'
log M
log oto
colog r'
log Co
log 2 u
log So
log cot 1
colog H
log So
5.3010
4.0849
9-9993
2.7958
log 4 10'
log M
log >
B
1
T
/« +/?/"^^^^^:::^;^^
y
1
1
fo
H-fa
^i^^
^a
H
Ms
r
w
Figure 77.— Test of orientation of Z)-variometer magnet;
deflector placed A^ end south
The equivalent west field that would produce this deflection if /„ were
zero is Ay 2. Then,
Ay2=Slu2
and
By similar triangles
and
For small angles
and
tan (a— ,82) =
y
H+fa
y _A2/2.
tan (a— /32)
H^-fa fa
A2/2=/a tan {a—^2) = SlU2.
tan (a+iSi)«tan a+tan ^i
tan (a— i32)~tan a— tan ^2-
r
— 1 ^
^^
i
5:
~~~^~--^_^
?7?^^^-^
-^^
' /T"
y
Ma
, '«
H ^
^ ^
H+fa
^
r
w
Figure 78.— Test of orientation of Z>-Tariometer magnet;
deflector placed A^ end north.
Then from equation (315)
fa ■
and from equation (319)
210111—53 10
fa ■
tan a+tan ^i
tan a— tan ^2-
(316)
(317)
(318)
(319)
(320)
(321)
(322)
C323)
138 MAGNETIC OBSERVATOKY MANUAL [Ch.
Adding equations (322) and (323),
la
2 tan a+tan /3i — tan ^2 (324)
and tana=^^^%'t^-i(tani8i-tan|82). (325)
Let Mi+U2=2u; 2ja= ^ — — ; tan i8i = i8i; and tan i32 = i82; then
tana=^|g^,-J(ft-&). (326)
337. In normal operation oi a, D variometer, it is evident from an
examination of figures 77 and 78 that larger values of /3i and ^2 will
be obtained when /« increases with respect to H. An extreme case
might be when ZZ"=80007 and /„=1000t. From equations (313)
and (317),
tan {a-{-M=jjziJ^ (327)
and tan {a—^2)=i-rr^' (328)
^-rja
Applying equations (320) and (321) and assuming that tan a=a;
tan i8i = i3i; and tan i82=i32,
and q:-/32=y7t-7- (330)
J^ -Tja
Adding equations (329) and (330),
y , y
2a+i8,-^2=
H-Ja ■ H-^Ja
Since tan «=-^'
Q _R- y I y ?i
Divide both sides by 2a,
y I y 2^
2a 2y 2y
H H
(331) i
Hh^.^h^j)-' ^'''^ ^
12.] ORIENTATION 139
which simplifies to
i \ /2 1
Thus if /a= 1000 and //=8000,'^=- and the ratio 7^2 ^2 =^ — ^ =
-^ ' // 8 H^—ja 64-1
0.016. This shows that in equation (326) the term |(/3i— i^o) may
be safely neglected in computing tan a and tan E^. Then
tan a=tan E,^^^, (334)
in which Ex=a and 2i^ is the double deflection of the D spot in mm.
Note: The scale value of the D variometer will be changed as a result
of the large component of the deflector field parallel to Ms. This
component enters as the term C in the general equation of the gamma
scale value. That is,
In orientation tests the C field, practically equal to fa, may be of the
TJ_\_ n
order of 5OO7. Suppose //=20,0007. Then the factor, -^j- =
1+Yt==1+ 7^777^77?,= 1.025 for the A^ end of the deflector north and
rl zU,UUU
C
1 —yj= 0.975 for the A^ end south. As a result of this change in scale
value, the deflections will be unequal by a small amount; in this case,
by 5 percent. Through the use of the mean gamma scale value in
the computation of Ex by equation (334), these effects will cancel each
other for all practical purposes. The foregoing proof disregards torsion
in the fiber, but even for a stiff fiber the important results (eqs. 326
and 334) still hold.
338. Z variometer. — In making orientation tests of the Z recording
magnet the deflector is placed in the same relative position as for the
D variometer. Figures 76, 77, and 78 apply except that they now
represent side elevations as seen from the west side of the Z variom-
eter, and the line marked magnetic meridian is now the horizontal
plane through the center of the deflector and the center of the Z
magnet. The development of the orientation equation follows the
same steps as for D, and
in which Sz is the mean Z scale value and 2u is the double deflection
of the Z spot in mm.
339. H variometer. — In figure 79, let a: be a small field or its
equivalent that is deflecting the recording magnet, Ms, out of the
magnetic prime vertical by a small angle a. Then since k (par. 232,
p. 87) may be treated as an equivalent field,
tan a~ (337)
140
MAGNETIC OBSERVATORY MANUAL
[Ch.
and if there is a sensitivity magnet having a field, p, at Mg, parallel
to and directed the same as Ms, then
tan a=
k-\-V
(338)
Figure 79.— H-variometer magnet not in magnetic prime vertical.
340. With the deflector in position for orientation tests, M, will
be deflected through a small angle jSi, figure 80, such that
(339)
Figure 80.— Test of orientation of /7-variometer magnet;
' deflector placed iST end west.
in which ja is the field of Ma at the center of M^, and ^x^ is the com-
ponent of ja practically normal to Ms and causing the small deflection
/3i. When the deflector is reversed, figure 81,
tan (a +182) =
Aa:o
Jc + p—fa fa
(340)
H^
^p^~""
m
r
Ma
-^
^X2
\
X
^^-^^^^ST
V^
■^ k + p-fa
fa
Ms
k^p
Figure 81.— Test of orientation of H-variometer magnet;
deflector placed N end east.
12.] ORIENTATION 141
For small angles,
tan (a— i3i)«tan a— tan /3i,
and tan (a+/32)~tan a + tan ^2-
Then tana-tan^i = ^=%^ (341)
and tan a+tan ^,=^=^ (342)
J a J a
in which Sh is the H scale value, and Ui and U2 are the deflections in
mm of the H spot corresponding to ^i and ^2- Putting Ui-{-U2=2u
and adding equations (341) and (342),
2 tan a=?^+ (tan ^i-tan ^2). (343)
Ja
Neglecting the term (tan /3i — tan ^2), which will always be small
compared to a,
2uSh
tan a-
2/«
tan a=tan E,-^^^,^ (344)
since 2/a= — s-^XlO*^ (deflector in B position).
DIRECTIONS FOR ORIENTATION TESTS
341. Critical adjustment, — Read carefully appendix III, giving
special attention to those parts dealing with critical adjustments of
the deflector. Table 18 of appendix III gives a summary of critical
adjustments.
342. Orientation bench, — If space in the variation room permits,
construct a permanent bench or table similar to that shown in figure
82. Use nonmagnetic materials throughout. Adjust the bench
plates (wood, brass, or aluminum plates upon which the deflector
rests) in elevation so that when the deflector is in proper position for
deflections, its geometric axis will be horizontal and at the same
elevation as the recording magnet to be tested.
343. Adjustment of the deflector to correct elevation, — Place
a short piece of glass tubing in each end of a long piece of flexible
rubber tubing and fill the tube with water, free of air bubbles. Hold
one tube vertical near the recording magnet to be tested and the other
end near the bench plate opposite the variometer. When the water
surface, bottom of the meniscus, is at the precise elevation of the
center of the recording magnet and the water is at rest, mark on the
bench-plate support the height of the water meniscus in the adjacent
glass tube. Adjust the bench-plate supports in thickness until the
deflector is at the correct height when resting in proper position on
the plate. Check carefully. (Note: The recording magnets may not
142 MAGNETIC OBSERVATORY MANUAL [Ch.
all be at the same elevation; the three bench plates must be adjusted
separately.)
344. If it is not feasible to construct a permanent bench as de-
scribed in paragraph 342, use a small nonmagnetic plane table for
supporting the deflector.
345. Establishing magnetic meridians through the variom-
eters, — If magnetic meridians have been established previously,
proceed as outlined in paragraph 348. Otherwise establish these
magnetic meridians as outlined below.
346. Using a transit or theodolite, run a traverse from the declina-
tion pier in the absolute observatory to the variation room. This
line, preferably about parallel to the magnetic meridian, should be
Figure 82.— Orientation bench.
marked with permanent vertical lines inscribed on brass plates on
the north and south (inside) walls of the variation room. The plates
should bear the inscription: True Bearing, (For example:
N 14° E.) A simple computation will then, at any time, furnish
data for establishment of the magnetic meridian by offsetting one
end of the line established by the two permanent marks. An illustra-
tion of this is given in figure 83. The line AB, established by traverse,
has a true bearing of N 14° E. If the mean magnetic declination
in the absolute observatory is 13°48' East, then the desired magnetic
meridian in the variation room will be AB\ making an angle of 12'
with AB. If the length of the line ^5=300 cm, the offset distance,
BB' will be 3000 tan 12' or 10.5 mm to the west of the point B.
347. Make a sketch of the traverse, showing all turning points,
courses, and angles on the sketch. Furnish also a record of all circle
readings. Check carefully.
348. At approximately the same elevation as the marks described
in paragraph 346, mount two wooden strips horizontally on the north
and south walls of the variation room. A convenient size for the
strips is 2 by 10 cm in section by 3 meters in length, and they should
be held away from the walls by spacing blocks 2 cm thick.
12.1
ORIENTATION
143
349. Stretch a white linen or cotton thread between the wooden
strips and attach a small weight to each end of the thread to keep
it taut as it passes over the top edges of the strips. Attach a light-
weight plumb bob to the thread in such a way that the point of the
bob can be precisely centered over the top of the D variometer. By
moving the thread along the wall strips so the two ends of the thread
are the same distance (within \ mm) from the points A and B' ,
figure 83, the thread can be made to lie in the magnetic meridian
passing through the center of the D variometer. Now file small
grooves in the wooden strips directly beneath the ends of the threads
and mark these grooves with a label; such as, D, Jan. 1, 1951.
B B
North Wall
D =13°48'E
"\\4-
12'
A
0)
2
Variation Room
\
H \
'z
D
\
Magnetograph Pier
\
South Wall
Figure 83.— Establishing the magnetic meridians through the variometers.
i4i?= established true bearing;
^iV= true north;
yiB'= magnetic meridian (for magnetic declination of 13°48' East).
350. In like manner establish magnetic meridians through the H
and Z variometers. (Note: Center of variometer means center of
recording magnet.) Label and date each mark.
351. Suspend a light-weight plumb bob (perhaps a suitably sharp-
ened copper nail or brass screw so tied to a piece of thread that it
will hang straight with the vertical) from the meridian thread of the
D variometer and carefully mark a point on the bench plate. Move
the bob along the meridian thread and mark a second point, making
the two marked points as far apart as possible. These two points
then mark the centerline of the orientation deflector. Prepare a
wooden strip and fasten it to the bench plate with brass screws
(preferably in a way that will permit final adjustment of the guide
strip after the screws are started) making due allowances for the
size of the deflector magnet or its case so that the deflector will be
centered on the meridian and parallel to the meridian. Fasten a
144 MAGNETIC OBSERVATORY MANUAL [Ch.
second wooden strip to the bench plate to form a stop that will
determine the distance between the center of the deflector magnet
and the center of the D recording magent. Both wooden strips
(guides) should be well made, with straight sides, so that the deflector
will fit firmly and positively in one position only when placed against
them on the bench plate. All markings of points and lines should
be done as precisely as the materials at hand will permit.
352. Mark a meridian line on the bench plate that will be used
for the H variometer in the same manner as that described for the
D plate. With a draftsman's triangle having a true 90° angle, mark
a line on the bench plate perpendicular to the meridian line at the
desired distance from the center of the H recording magnet. This
prime vertical line will be the guide line for determining the position
of the guide strip that should now be fastened to the bench plate as
was done for the D position. Fix an end position stop, also, to the
bench plate so that the center of the deflector magnet will be pre-
cisely in the magnetic meridian through the H variometer. Check
the work carefully. It is important that the work be done with
great precision.
353. Establish the meridian line on the Z bench plate in the manner
described for H and D. Make a second check on the Z bench plate
to see that the center of the deflector magnet will be at the same
elevation as the Z recording magnet, and to see that the deflector
will be level. (Test the level of the deflector with the sensitive stride
level of the earth inductor.) Fasten wooden guide strips to the
bench plates to serve as guides or stops against which the deflector
may be placed in making orientation tests.
354. The deflection distances, r, need not be the same for all
variometers. Measure and record these distances to the nearest
millimeter.
355. It cannot be emphasized too strongly that every effort should
be made to obtain precisely fixed positions for the deflector magnet for
orientation tests, particularly to insure that the deflector shall be
accurately parallel to the established magnetic meridian for Z>,
accurately parallel to the established prime vertical for H, and accu-
rately level for Z.
356. The procedures described above have been outlined in detail.
There are, of course, other ways of accomplishing the results with
equal if not greater precision, and the observer may adopt whatever
methods seem to be most desirable. Experience has shown the author,
however, that in any event a complete, detailed record of the methods
and procedures followed will inevitably be highly useful at some later
time.
357. Orientation deflections, — On a magnetically quiet day,
make orientation deflections precisely as outlined in figm-e 84, making
certain that the deflector is always against the guides. Begin with
the D variometer and see that the first deflection is made with the
deflector /ac6 up, N end to the north.
358. Complete all of the deflections indicated on the form, in pre-
cisely the order indicated and with the deflector oriented properly
for the indicated deflections.
359. Measure and record, as in figure 84, the deflection distances
for each variometer. Record also the magnetic moment of the
deflector (in terms of colog M, line 19), and the gamma scale value
of all variometers (line 15).
12.1
ORIENTATION
145
360. From the magnetogram, scale and record all deflections and
the away ordinates, and complete the computations for the numerical
values of the orientation angles.
Orientatiok Tests
(Deflector North or South of Variometer)
Magnetograph:{f «gf«;^«f^« ^ ^ ^^-
Magnetic Observatory: Cheltenham
Observer: S. O. Townshend
1. Variometer
2. Date
3. Approximate time
4.
5. Away
6. Face up; N end of deflector
7. Face up; N end of deflector
8. Face down; N end of deflector
9. Face down; N end of deflector
10. Away
11. Mean, N end (A)
12. Mean, N end (B)
13. 2u (A) -(B)
14. r
15. sy
16. Log 2u
17. 3 1ogr
18. LogSy
19. Colog Af
20. Colog 2X105
21. Colog 4X105
22. Log tan Ex
23. Ex (to nearest minute of arc)
24. A'' end of recording magnet
25. Formula tan Ez =
D
June 2k 19S7
15 20
Ord. mm
-S.O
: +2.8
-9.5
-9.5
I +2.8
-3.0
r +2.8
-9.5
+12. S
129 cm
5. SO y/mm
1.090
6.SSS
0.72j^
6.129
I'm
8. 674
2° 42'
EofN
2uT^ Spy
4MX105
H
June 24 1937
15 40
Ord. mm
4.6
W 6.1
; 4-3
: 8.7
V 7.0
10. o
V +6.6
; +6.5
+0.1
129 cm
3. 30 y/mm
9.000
6. 3SS
0.519
6.129
4.699
6'.680
0° 02'
SofE
2ur^ Sny
2MX105
Z
June 24 1937
1610
Ord. mm
34.8
36.6
32.6
32.9
36.7
34.4
36.6
32.8
+3.8
134 cm
5. 54 y/mm
0.580
6.381
0.549
6.129
1398
8.037
0° 37'
Up
2ur3 Szy
4MX105
Record all ordinates to nearest 0.1 mm with same sign as in scaling an ordinate for
Colog M derived from: Regular scale-value obser
base line value.
Figure 84. — Observations for orientation tests.
I
361. Prepare a tabulation similar to that shown in table 12.
362. By means of table 13, identify the quadrant in which the
N end of a recording magnet lies for the away position of the deflector.
Record the results on line 24, figure 84.
363. Adjustment of recording magnets, — If a calculated E^
angle for any recording magnet is in excess of 1°, the recording magnet
should be adjusted for correct orientation and the orientation tests
repeated. Adjusting and testing should be continued until satis-
factory orientation is indicated by the recorded deflections.
TABLE \2.^Miscellaneous data
Observatory: Cheltenham Date: Dec. 31, 1951
D variometer
(a) Declination is West.
(bj D spot moves up on the magnetogram for numerical
increase in declination.
B variometer
(c) N end of H recording magnet is toward the East.
(d) H spot moves up on the magnetogram for numerical
increase in horizontal intensitv.
Z variometer
(e) N end of Z magnet is toward the North.
(f) Absolute value of vertical intensity is positive ( + ).
(g; Z spot moves up on tne magnetogram for numerical
increase in vertical intensity.
146
MAGNETIC OBSERVATORY MANUAL [Ch.
TABLE 13. — Orientation of variometer magnets
LINE
D Variometer
West Declination
East Declination
1
2
3
4
When N end of deflector is toward.. .
The field of the deflector at the D
variometer is directed
If the D spot moves in the direction
of
The N end of the recording magnet
lies in
Magnetic north
Magnetic north
Magnetic north
Magnetic north
(a)
Increasing
West!)
(b)
Decreasing
West 2)
(c)
Increasing
East D
(d)
Decreasing
EastD
NE quad-
rant
NW quad-
rant
NW quad-
rant
NE quad-
rant
5
H Variometer
N end of recording magnet
east
N end of recording magnet
west
6
7
8
9
When N end of deflector is toward.. .
The field of the deflector at the H
variometer is directed
Magnetic west
Magnetic west
Magnetic east
Magnetic east
If the i^spot moves in the direction
of
Increasing H
Decreasing H
Increasing H
Decreasing H
The N end of the recording magnet
SE quad-
rant
NE quad-
rant
NW quad-
rant
SW quad-
rant
10
Z Variometer
N end of recording magnet
north
N end of recording magnet
south
11
When N end of deflector is toward.. .
Magnetic north
Magnetic north
12
The field of the deflector at the Z
variometer is directed ... . -
Magnetic north
Magnetic north
13
If the Z spot moves in the direction
of numerically
Increasing Z
Decreasing Z
Increasing Z
Decreasing Z
14
In North magnetic latitudes, the N
end of the recording magnet is
Too high
Too low
Too low
Too high
15
In South magnetic latitudes, the N
end of the recording magnet is
Too low
Too high
Too high
Too low
364. The D recording magnet can be brought into correct orienta-
tion by sHght adjustment of its corrector magnet (see par. 381, p. 152).
If a corrector magnet is used the E^ angle should not exceed 20';
if a corrector magnet is not used, an E^ angle up to 1° may be per-
mitted. In any event, when acceptable orientation of the D magnet
is achieved, if the D ordinate on the magnetogram is not correct (too
large, too small, or even negative in sign), it will be necessary to
adjust the D magnet relative to the mirror frame, in the manner
described in paragraph 377, page 150. Recording of orientation
deflections must, of course, be repeated after any mechanical adjust-
ments are made.
365. In like manner the H and Z recording magnets may be
adjusted to correct orientation by slight adjustments of the tem-
perature-compensating magnets. In the case of H, for example:
Let £'a:=l°, L (recording distance) = 1200 mm, and A7i=change in
ordinate corresponding to 1° change in orientation. Then
and
tan E^=2Z
A7i=0.0175X2400=42 mm.
12.] ORIENTATION 147
If the H scale value is 5.07/mm, this would be equivalent to a change
of 21O7 in the field of the temperature-compensating magnet, an
amount which would not appreciably affect the temperature coefficient
of the variometer. In like manner the change of ordinate of the Z
trace corresponding to an angular motion of the Z magnet may be
computed. After proper orientation of the recording magnets is
obtained, if the H or Z ordinates on the magnetogram are too large,
or are negative, it will be necessary to adjust the recording mirror
without changing the orientation of the magnet — for H, by turning
the magnet relative to the mirror; for Z, by rotating the whole variom-
eter and then readjusting the base line mirror, or by adjusting the
prisms if the variometer provides for such adjustment.
366. In ordinary routine operation, readjustments of a recording
magnet should not be made without the prior approval of the adminis-
trative authority responsible for the processing of the results.
CHAPTER 13. DIRECTIONS FOR INSTALLING
A MAGNETOGRAPH
PRELIMINARY STEPS
367. The variation room, — These directions apply to the instal-
lation of a complete magnetograph, in which the instruments are
arranged as shown in figure 13 and schematically in figure 85. It is
assumed that the absolute values of D, H, and Z on the observing
piers in the absolute observatory are practically identical with the
respective values on the variation-room pier; that the magnetograph
pier and scale-value and orientation shelves and guides have been con-
structed in accordance with the general specifications in chapter 3;
and that a line with known bearing has been established and perma-
nently marked on the walls of the variation room.
368. Comparison observations. — If there is any doubt about
magnetic materials having been introduced into the piers or the
structural parts of the buildings, compare the values of D and H
on the variometer pier and the piers in the absolute observatory by
making one or two sets of horizontal intensity with a magnetometer
or a Quartz Horizontal Magnetometer and at least two sets of decli-
nation with a magnetometer or a compass declinometer at both sites.
Differences of more than a few minutes in declination or 5O7 in
horizontal intensity should not be tolerated.
369. Constants of component par ^s.— Prepare or have on hand
several D and H quartz fibers, calibrated and mounted as explained in
chapter 6. Determine the dimensions and magnetic moments of all
of the magnets to be used with the variometers. The magnetic
moments may be determined by deflections, using a magnetometer,
at a place where H is known. Label these magnets by number and
by magnetic moments for future use. Determine the temperature
coefficients of magnetic moment of the H and Z recording magnets
and of the temperature compensating magnets for the H and Z
variometers for use in the computations which follow, or use the
temperature coefficients furnished with the magnets.
370. Record all available data on the appropriate lines in tables
14, 15, 16, and 17, as the work progresses.
371. Calculate the approximate distances at which the temperature
compensating magnets for the H and Z variometers should be placed
on these variometers for effective temperature compensation, and
mount the magnets at those distances on their respective variometers.
The H temperature magnet should be mounted with its N end to the
south and the Z temperature magnet with its N end up in north mag-
netic latitudes and with its A^ end down in south magnetic latitudes.
372. Test of the optical system, — Set up the three variometers,
the recorder, time flasher, and thermograph at the approximate posi-
tions they will occupy in routine operation of the magnetograph. See
that the recording magnets are in the instruments and that the A^'end
of the Z recording magnet is to the north; also that the temperature
magnets and sensitivity magnets are attached to their deflection bars
148
INSTALLING A MAGNETOGRAPH
149
or holders in approximately the same positions and orientations in
which they will be operated later. Details regarding the methods of
leriving this information may be foimd in the chapter on temperature
'^efficients and in paragraphs 387, 390, 393, 394, and 401 of this
chapter. Adjust the D recording distance to 174 cm (D lens to record-
ing drum) and adjust the D base-line spot to the desired ordinate on
5 Pq S
3 1
g
CONCRETE WALL
-=f
m
}
« I
CONCRETE BLOCKS
4='
3=J
I I
I i
4
t-^^^'i
C
CONCRETE BLOCKS
the drum. In like manner adjust the H and Z recording distances so
that the fl' base-line spot and the Z base-line spot are in good focus on
the drum, axes of all variometer lenses approximately perpendicular to
the face of the recorder. See that the Z base line is not eclipsed by the
D or H v^ariometers when the Z variometer is rotated on its vertical
axis for the full aperture of the recorder window. In like manner see
that the D base line is not eclipsed by the H variometer. Using marble
150 MAGNETIC OBSERVATORY MANUAL [Ch.
blocks or nonmagnetic metal spacers, adjust the heights of all of the
variometers so that the centers of all lenses are in the same horizontal
plane as the axis of the recorder when all instruments are resting on
their footplates and when each foot screw is approximately at the
middle of its run. Also adjust the axis of the cylindrical lens to this
same elevation. Mark on the pier top the positions of the three vari-
ometers and then take the H and Z instruments several meters away
from the pier.
INSTALLING THE D VARIOMETER
373. Removal of torsion. — Consult the nomogram, figure 140,
appendix VI, and select a D fiber and D magnet combination that
will give a D scale value close to one minute per millimeter at a re-
cording distance of approximately 174 cm. Make preliminary vis-
ual tests for the torsion factor. Level the instrument and install the
D fiber and the mirror frame without the D magnet but with a small
torsion weight attached. Adjust the torsion head in azimuth and the
mirror in inclination until the regular D spot (from the central mirror
if a triple-faced mirror is used) falls at the desired ordinate when the
suspended system is centered in the magnet chamber and is at rest.
Clamp the torsion head in this position and set the fiber clamp.
Remove the mirror frame from the hooks, then replace it, release the
fiber clamp, and allow the system to come to rest. If the D spot
comes to rest at approximately the same ordinate as before, the mirror
frame is in good adjustment and probably hanging evenly on both
hooks. Always set the fiber clamp before removing or replacing the
mirror frame.
374. Record the torsion-head reading, Rq, when the torsion- weight
axis is approximately in the magnetic meridian and the regular D spot
is at the desired ordinate. Record also the position of the D spot on
the recording-box scale, figure 86. Call this original scale reading Uq.
Clamp the fiber.
375. Replace the torsion weight with the D magnet, N end to the
north. Release the fiber clamp; carefully raise the damping chamber
around the magnet. See that the magnet is weU centered within the
damping chamber by slight adjustment of the level of the instrument.
Test the damping by deflecting the magnet and allowing it to come to
rest. The system should be so adjusted that it will be somewhat
underdamped. If the regular D spot comes to rest within 10 mm of
the original scale reading, Uq, the D magnet is practically in the mag-
netic meridian and there is little or no torsion in the fiber.
376. If the D spot does not come to rest within 10 mm of the
original scale reading, no, the magnet is not properly oriented with
respect to the mirror, the mirror frame is not properly suspended
on the hooks, magnetic conditions are not quiet, or there may be
some magnetic material near the instrument. Observe the spot for
several minutes. If magnetic conditions are not quiet, postpone
operations. If magnetic conditions are quiet, remove the mirror
frame and readjust the magnet in azimuth relative to the frame and
test again. Repeat this operation until satisfactory^ adjustment has
been obtained. If available, use the mirror-adjusting apparatus
described below for adjusting the magnet relative to the frame.
377. Mirror-adjusting apparatus, — This apparatus, figure 87,
is used for turning the mirror frame and the attached mirror through
13.
INSTALLING A MAGNETOGRAPH
151
a small angle relative to the magnetic axis of the recording magnet.
The lenses have focal lengths equal, respectively, to those on the H
and D variometers. The whole apparatus is mounted on a non-
magnetic laboratory stand. To turn the mirror through a small angle
relative to the magnet, proceed as follows: Place the stand on the
variometer pier so that the longer focus lens is parallel to and at the
same height as the D lens, also at the same distance from the recorder.
Clamp the magnet so that the frame stands vertically and the triple-
Figure 86. — Magnetograph recorder showing scale on front of cylindrical-lens window.
face mirror faces the lens. Find the image of the lamp filament as
reflected from the regular mirror and adjust the stand in azimuth
and the apparatus in elevation until the image of the regular D spot
falls on the recorder scale. Then by means of a long adjusting pin,
turn the mirror frame in azimuth until the image has moved through
the desired change of ordinate.
378. Adjust the D base-line mirror to correct position in azimuth
and elevation so that the D spot has an ordinate of about 20 mm and
so that numerically increasing ordinate corresponds to increasing
magnetic declination. Note that in some cases it may be necessary
to have the D base line above the D trace in order to accomplish this.
379. Establish the magnetic meridian through the D variometer,
adjust the guides of the orientation bench for D, and make orientation
tests as described in chapter 12. Record the observations as in
figure 84.
380. Measurement of stray east fields. — When magnetic condi-
tions are reasonably quiet, record D and the D base line photograph-
ically for one hour. Close the recorder window and quickl}^ replace
the H and Z variometers in their correct positions on the pier with
temperature magnets, sensitivity^ magnets, and recording magnets
all in place, as described in paragraph 371 . Cover the H and Z lenses
152
MAGNETIC OBSEKVATORY MANUAL
[Ch.
with paper screens. The D spot may be deflected away from its
original position, Uq, because of the resultant east (or west) com-
ponent, Je, of the fields of the magnets of the H and Z variometers.
Record D and the D base line photographically for 30 minutes. Close
the recorder window. Develop the magnetogram and scale the change
in ordinate, A?i, of the D spot due to the resultant field, /^, at D.
Then
fE={^n)Sl. (345)
Compute Je, noting the sign. It may be as large as ±1507.
381. The D corrector. — If the deflection, An, is large, attach a
short deflection bar to the D variometer on the side nearer the Z
variometer, axis of the bar in the magnetic prime vertical through D.
Attach the D corrector to this bar and adjust the deflection distance
until the D spot returns to its original position, Uq, on the scale of
the recorder; that is, make An=0. AD corrector having a magnetic
moment of 5 cgs units at a distance of 20 to 25 cm would furnish the
required compensating field at D. If the field, /e, deflects the D spot
only a few mm, the D corrector is unnecessary, since an exmeridian
angle this small would introduce no appreciable error in the recorded
changes in declination.
382. Determine the torsion factor, tt^; by turning the torsion
head 30° clockwise, then 60° counterclockwise (that is, 30° counter-
clockwise from its original position), then 30° clockwise. Record
the torsion-head readings and the corresponding ordinates of the D
spot. Record photographically or observe the ordinates of the D
spot on a paper millimeter scale attached to the drum.
TABLE 14. — D variometer; miscellaneous data
Observatoey
Variometer:
Cheltenham, Md. Date: December 12, 1948
Schulze No. XI Observer: J. Q. P.
Line
Symbol
Items
Examples
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
c
b
I
f
c
H+C
H
L
f
R
e
M.
Ma
T
SECTION I, the optical SCALE VALUE
9 mm
2
3
3 mm
1.006
183007
+1007
1.005
1739 mm
1.011
1748 mm
1747 mm
0.000 2862
8. 8 cgs
5. cgs
West
Factor in equation (91); plane side of J) lens out
Thickness of D lens at center.
Estimated torsion factor (or derived from preliminary visual tests) . . .
Horizontal intensity
Estimated north field or caculated (see appendix V)
Estimated distance, D lens to drum, for imit scale value, equation
(132).
Torsion factor, mea''ured; see Section IV below
Calculated from equation (132) using new torsion factor
Optical scale value, — , in radians per mm ...
2R
SECTION n. variometer MAGNETS
Magnetic moment of D recording magnet (Alnico II; cylindrical; 2 by
10 mm).
Magnetic moment of D corrector (Alnico II; 2 by 10 mm)
13.] INSTALLING A MAGNETOGRAPH 153
TABLE 14. — D variometer; miscellaneous data — Continued
Line
Symbol
Items
Examples
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
I
d
I-
k'
k'
f
n
h
h
f-h
f
f-h
C
H
H+C
H
no
ni
An
'A-
So'
Soy
D
H
7
DR-D
D-Dr
e
Li
«i
no
n
n— no
50
s
s—so
h
f
f-h
f
f-h
h
f-h
SECTION III. QUARTZ FIBER
Effective length of fiber
15 cm
0. 0017 cm
63 sec
0. 0155
0. 0182
0.0181
-I-IOO7
183007
1.005
47.5
60.0
12.5
667
7O7
1.000
5.32
0.1 mm
0.0
0.0
187. 1 mm
186. 2 mm
0. 000 2862
1000 mm
0. 000 500
Diameter of fiber; several measurements uniformly spaced ; calculated
diameter as described in paragraph 188.
Period of system with torsion weight, 7=1.83, attached _
Torsion constant from dimensions; from equation (148) . .
Torsion constant from period; from equation (150)
Torsion constant from torsion tests; from equation (151)
SECTION IV. TORSION OBSERVATIONS; SENSITIVE VARIOMETER
Torsion head readings 30° right 30° left
Angular motion of torsion head 1800' 3600' 1800'
Ordinate of the D spot 42. 2 62. 1 22. 1 42. 2
Displacement of the Z> spot 19.9' 40.0' 20.1'
Mean h for/=1800' 20.0'
Twist in the fiber 1780.0'
Torsion factor _ _ . . , _ .1.011
SECTION V. FIELD FACTOR
Resultant N or S field at D magnet due to stray fields
Mean value of H at the variometer pier
Field factor; equation (111).. .....
SECTION VI. EAST FIELD AT D VARIOMETER
D ordinate on magnetogram, H and Z variometers in place
Change in Z) ordinate due to resultant east field, /b .
East field from equation (345)
East field, calculated from equations in appendix V (Item 42 is op-
tional) .
SECTION VII. COMPUTATION OF D SCALE VALUE
From equation (122); minute scale value. . ....
From equation (301); gamma scale value
SECTION Vm. PARALLAX TESTS
Correction in mm between time line and Z> spot...
Correction in mm between time line and H spot
Correction in mm between time line and Zspot
SECTION IX. RESERVE DISTANCES
Distance in mm on magnetogram, regular spot to upper reserve
Distance in mm on magnetogram, regular spot to lower reserve
SECTION X. LOW-SENSITIVITY VARIOMETER; TORSION TESTS
Optical scale value, D variometer (from line 13).
Recording distance, auxiliary scale
D ordinate, no torsion in fiber . 46.0 46.0
D ordinate, magnet deflected by torsion 116.6 188.2
Change of D ordinate 70. 6 142. 2
Auxiliary scale reading, no torsion 1.5 1.5
Auxiliary scale reading, magnet deflected by 52. 1 103. 3
torsion.
Change in scale reading 50.6 101.8
Angular motion in radians of the D magnet = 0.0202 0.0407
«(n-no).
Angular motion in radians of torsion head= 0.0253 0.0509
ei (s-so).
Torsion in the fiber 0. 0051 0. 0102
Torsion factor . ... ... 4.96 4.99
Mean torsion factor.. 4. 98
3.96 3.99
383. Adjust the time flasher apparatus in elevation until there
is no parallax between the D spot and the time line. The parallax
may be examined visually while the D spot and the time flasher lamp
are activated simultaneously.
210111—53-
-11
154 MAGNETIC OBSERVATORY MANUAL [Ch.
384. Computation of the effective D scale value, — Compute
the D scale value (minute scale value) from equation (122) having
regard for the direction (sign) and magnitude of the (7 field (see app. V)
H-\-C
in the factor — yy— • If (7 is directed to magnetic north it is positive
(+) ; if directed to magnetic south it is negative (— ) . If the calculated
D scale value is (1.000 ±0.002) minutes per mm, the installation of the
D variometer may be considered satisfactory in this respect. If the
scale value differs from unity by more than 2 parts in 1000, adjust the
D recording distance by moving the recorder directly away from, or
toward, the D variometer by the required amount. For example:
Suppose the effective recording distance is 1720 mm, and the D scale
value, calculated from equation (122), is 1.005. Then the distance
should be increased approximately 0.005 X 1720 = 9 mm. In that case
move the recorder away from the D variometer 9 mm. Check this
distance by direct measurement.
385. If the recorder has been moved as above, repeat torsion obser-
vations, recording several tests photographically, at the new recording
distance; then make a final computation of the D scale value using the
new torsion factor, the new effective recording distance, and the field
factor. Calculate also the gamma scale value from equation (301).
INSTALLING THE H VARIOMETER
386. Torsion and compensation, — These directions apply to H
variometers designed for magnetic temperature compensation. Ob-
servations for the torsion constant, k', of the H fiber are highly de-
sirable though not strictly necessary.
387. Examine the H nomogram, figure 141, appendix VI. Select
a combination of size of fiber, magnetic moment, Ms, of the recording
magnet, and a suitable sensitivity control magnet. Determine and
record all of the constants and miscellaneous data for the H variometer
as outlined in table 15 as the work progresses.
388. Kemove all magnets from the variometer, install the H fiber,
and remove torsion as for the D variometer. Adjust the damping
chamber for proper damping as described in paragraph 375 for a D
variometer, and then make tests for residual torsion as described in
the same place. Adjust the mirror face (central mirror if a triple-
face mirror is used) approximately parallel to the axis of the torsion
weight as for a D variometer, and then adjust the torsion head until
the H spot falls near the center of the recording drum when the system
is at rest and the axis of the torsion weight is in the magnetic meridian
through H, as judged by eye. Note the scale reading on the recorder
scale, or on a paper millimeter scale attached to the drum. Replace
the torsion weight with the selected H recording magnet and allow
the system to come to rest. Note the position of the H spot. If it
is close to Tio the H magnet is in the magnetic meridian with practically
no torsion in the fiber. If not in the magnetic meridian within 2°,
readjust the magnet relative to the frame by use of the mirror adjust-
ing apparatus, figure 87. When this apparatus is used with the H
variometer the shorter focus lens should be installed. Continue
adjustments until the H spot falls near the original ordinate, Uq, with
no torsion in the fiber, magnetic conditions quiet, and no appreciable
stray fields. Read the torsion head. This is Rq, table 16. Deter-
13.
INSTALLING A MAGNETOGRAPH
TABLE 15. — H Variometer; miscellaneous data
155
Observatory
Variometer:
: Cheltenham, Md. Date: December 14, 1948
Toepfer No. XII Observer: J. H. D.
LINE
SYMBOL
items
EXAMPLES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
L
t
M,
Ma
Ma
'A
Sh*
qi
Qi
Qi
Qh
^q =
S'(/ =
H
C/H~
C
T\
ro
I
d
I-
k'
k'
k'
k'
n
2uh
Sh*
An
ASh*
An
L
f
U
«i
no
n
n—m
so
s
s-so
ft
,u
f
f-h
h
f-h
HR-H
H-Hh
section I. OPTICAL SCALE VALUE
1180 mm
0.000 424
5.0 cgs
78.5 cgs
50.0 cgs
-0.0677
1.52
2.66
0.000 50
0.000 48
0.000 06
0.000 16
0.000 66
0.001 20
0.183 cgs
-0.55
-0.101
11.58 cm
11.63 cm
15.0 cm
0.0040
10.5
0.474
0.655
0.641
0.652
0.655
1180
0.000 424
1000
0.000 500
15.0
112.2
97.2
0.2
200.2
200.0
0.0412
0.1000
0.0588
1.70
1.70
0.70
182.0
181.5
Approximate optical scale value — radians per mm
SECTION II. VARIOMETER MAGNETS
Magnetic moment of H recording magnet (Alnico II; cylindrical;
2 X 10 mm).
Magnetic moment of H temperature magnet (Alnico II; cylindrical;
5x 25 mm).
Magnetic moment of sensitivity magnet (Alnico II; cylindrical;
5 X 25 mm).
Field of sensitivity magnet at 11.4 cm; A^ end west -.
Distribution coefficient (temperature and recording magnets)
SECTION III. TEMPERATURE COEFFICIENTS
Temperature coefficient of H recording magnet
Temperature coefficient of temperature-compensation magnet
3 times the coefficient of thermal (linear) expansion of the bar
Temperature coefficient of the torsion constant of the quartz fiber
qi-\-qi (see equation 251)
9l+?2+?3+95 - ------ - -
Temperature coefficient of the i/ variometer (from equation 249)
Temperature coefficient of the H variometer (by test)
SECTION IV. TEMPERATURE COMPENSATION
Ratio of compensation field to H (from equation 253)
Estimated correct distance for temperature magnet (equation (346))--
Calculated distance for temperature magnet (from equation (347))
SECTION V. QUARTZ FIBER
Effective length of the fiber
Diameter of fiber; give several measurements at uniformly spaced
intervals. Calculate mean as explained in par. 188.
Torsion constant (approximate) from dimensions (equation (148))..--
Torsion constant from torsion observations (equation (151))
Torsion constant from scale value observations (equations (153) and
(154)).
SECTION VI. a FACTOR DETERMINATIONS
Ordinate of undeflected spot in mm —37. 8 +15. 75. 6
Double deflection of H spot in mm 135. 5 130. 4 126. 1
Observed ii" scale value in 7/mm 2. 56 2. 66 2. 75
Increment in /i scale value 0.10 0.09
a factor in 7/nim/mm 0.0019 0.0015
Mean 0.0017
SECTION VII. TORSION OBSERVATIONS (OPTIONAL)
Recording distance, H variometer, in mm
Recording distance, auxiliary scale, in mm
Optical scale value, auxiliary scale
//ordinate, magnet deflected by torsion
Scale reading, no torsion in fiber
Change in scale reading
Angular motion, in radians, of torsion head, «i(5— so)
Torsion in the fiber
Mean torsion factor
SECTION Vm. RESERVE DISTANCES
Distance in mm on magnetogram, regular spot to upper reserve
Distance in mm on magnetogram, regular spot to lower reserve
156
MAGNETIC OBSERVATORY MANUAL
[Ch.
mine the torsion factor by the method described for low-sensitivity
variometers, paragraph 417. Adjust the H base-Une spot so that
increasing H ordinate corresponds to increasing horizontal intensity
and so that there may be few negative ordinates on quiet days. Test
the H spot for parallax. If the amount is appreciable, reduce it by
raising or lowering vertically the entire H variometer (see par. 176).
If this adjustment results in reflecting the recording spot too high or
too low in relation to the cylindrical lens of the recorder, the variom-
eter mirror must be tilted slightly by adjusting the mirror screws, or
by bending the mirror frame in older instruments.
Figure 87.— Mirror-adjusting apparatus, for adjusting mirror
relative to recording magnet.
TABLE 16.— i^ variometer, torsion in the fiber
Figure
Estimated
torsion (t)
C field
Magnet (6)
m
Torsion-head circle
R=Ro+e+r
Remarks
about M.
88(a)
88(b)
88(c)
o
ro =
ri=145
r2 = 65
7
10100
o
90
90
o
75
75
75
o
7?o = 75
7?i = 310
/?2 = 230
N end N.
A^ end E.
A^ end E.
389. Calculate the torsion constant, k\ of the H fiber from its
dimensions (eq. 148) ; from the period observations (eq. 150) ; from
torsion observations (eq. 151); and from scale-value observations
(eq. 153 and 154). Compare the results. The torsion constant
determined from oscillations is quite likely the most reliable.
390. Calculate the factor, -^; from equation (250) or (253), or esti-
mate its value from table 10 (p. 117), and then compute C, the value of
the compensating field necessary to make the temperature coefficient,
13.] INSTALLING A MAGNETOGRAPH 157
Qh, of the variometer equal to zero. From the nomogram, figure
137, appendix VI, estimate the magnetic moment of a temperature
compensation magnet that will provide the required C field at a
distance, ri = 12 cm, along its magnetic axis produced, and select a
magnet of this approximate moment. Assuming P^ = 0, calculate
the approximate value, ri, for the temperature-compensation magnet
from the relation,
=[!^»]'- (346)
Calculate the distribution coefficient. Pa, from the dimensions of the
magnets or estimate its value from table 21a, and then the distribution
P
factor, \-\ — Y' Finally, calculate the more precise value, ro, from
the relation
Tq
[l+^]*- (347)
391. Uncompensated H variometer, — Clamp the fiber, remove
the mirror frame, and turn the magnet until its long axis is approxi-
mately perpendicular to the regular mirror face, N end to the rear of
the mirror face. Suspend the mirror frame and turn the torsion head
clockwise,- looking down, until the magnet is in the magnetic prime
vertical as estimated by eye. If the H spot does not fall at the de-
sired ordinate repeat the adjustment of the magnet relative to the
frame until satisfactory. Use the mirror adjusting apparatus for this
operation. Preliminary orientation tests may be made at this point
if desired. If these tests indicate that the H magnet is closely in the
prime vertical (say within one degree), and the H spot is at the desired
ordinate, the variometer is adjusted for routine operation but it is
uncompensated for temperature. The torsion-head reading is E^,
table 16.
392. Compensated H variometer, — Attach a variometer deflec-
tion bar to the north or south side of the variometer parallel to the
magnetic meridian and set the temperature-compensation magnet
(center of magnet) at the calculated correct distance, Vq, to give the
required C field to make Qh=^. A more uniform C field may be
obtained by using two temperature-compensation magnets as shown
in figure 88 (c). In this case the deflection distances should be equal
so that each magnet provides half of the C fleld. The A^ end of the
compensation magnet should be to the south to make C negative.
Reduce the torsion in the fiber by turning the torsion head counter-
clockwise, looking down, until the H spot comes to rest at the desired
ordinate, tiq, making allowance for possible change of H during the
interval. Observe the torsion head reading, i?2- The variometer is
now compensated for temperature. The example given in table 16
is shown graphically in figure 88. Make orientation observations as
described in chapter 12.
393. Adjustment of the sensitivity- control magnet, — If the
scale value is too high or too low, attach a variometer deflection bar
to the H variometer on the side near the recorder. Note the scale
reading of the H spot. Mount a sensitivity magnet on the bar with
its A^ end toward the variometer, say in this case A^ end west, and
158
MAGNETIC OBSERVATORY MANUAL
[Ch.
then move it slowly along the bar toward Ms until r is about 10 cm.
If Ms shows no appreciable deflection during this operation, the bar
is closely parallel to Ms and the sensitivity magnet remains practically
parallel to Ms throughout the range of r. If Mg is deflected through
an appreciable angle for small changes of r, it signifies maladjustment
of Ms, or of the bar, or that the axis of Ma is not parallel to the bar or
to Ms.
394. Determine the H scale value by deflections with no sensitivity
magnet attached to the variometer bar and for several equally spaced
positions along the bar, first with the A'' end of Ma to the west and
then with the AT^ end to the east, noting the away positions of the H
310
255"
(a)
0)
Figure 88. — Torsion in H variometer fiber: (a) 75°, line of no torsion; (b) uncompensated
H variometer; (c) variometer with two temperature compensation magnets.
spot before and after each setting of Ma. When Ma is directed the
same as Ms the scale value is increased (sensitivity diminished) and
when it is directed opposite to Ms the scale value is decreased (sensi-
tivity increased). Plot a curve showing variation of scale value with
r. (See ^g. 63.) From this graph estimate the correct distance at
which Ma should be placed to give the desired scale value. If the H
spot is not at the desired ordinate (within a few mm) adjust the tem-
perature magnet by a very small distance, just sufficient to bring the
H spot to the desired ordinate. (See par. 301 regarding the effect of
this adjustment on the temperature coefficient of the variometer.)
Some sensitivity-magnet holders are equipped with a slow-motion de-
vice for making fine adjustments of the sensitivity magnet. De-
termine the scale value again and if satisfactory, clamp the control
magnets in place. Note: For most precise work the observations with
the sensitivity magnet in different positions should be made with the
away position of iheH spot the same for all cases, since the scale value
varies with ordinate. However, since these observations are made for
the purpose of simplifying the process of adjustment of the sensitivity
magnet, the change of scale value with ordinate may be ignored for
that particular operation.
395. Computation of the Je field (optional), — Compute the /^
field from equation (11), omitting a distribution factor, and then
compute the scale value from equation (144) and compare the result
with the observed value.
13.] INSTALLING A MAGNETOGRAPH 159
396. Experimental determination of the a factor. — Change
the H ordinate by 50 to 75 mm by deflecting the //recording magnet by
means of a large (auxihary) deflector placed in the magnetic meridian
on the opposite side of the variometer from the scale value shelf. The
deflection distance should be quite large, say 2 or 3 meters. Make
scale-value observations at this ordinate. Reverse the auxiliary
deflector and repeat the scale-value deflections at low ordinate. The
difference in the scale values divided by the difference in away ordi-
nate, in mm, is the a factor. Some experimental determinations of
the a factor are given in section VI of table 15.
397. Damping, — Adjust the damping chamber in elevation relative
to the H magnet so that the system is rather highly damped but not
critically so.
INSTALLING THE Z VARIOMETER
398. Preliminary work, — The directions in this section apply to
vertical-intensity variometers equipped with assembled recording-
magnet systems (Schmidt type) as distinguished from those made
from one piece of steel (la Cour type).
399. Instrumental data, — From table 11, page 134, select a
scale value at which the variometer should be operated. Determine
and record as in table 17 all of the constants and pertinent instru-
mental data, as the work progresses.
C
400. From table 10 estimate the value of -^; or estimate the value
of this ratio from the g coefficients, and then calculate C, the required
vertical field to be applied opposite to Z, for effective temperature
compensation.
401. From the nomogram, figure 137, appendix VI, estimate the
magnetic moment of a temperature-compensation magnet that will
provide the required C field at a distance, fi, along its axis for r^ equal
to approximately 13 cm. Select a magnet that has approximately^
this moment.
402. Determine the magnetic moment, il/j, of the recording magne-t
by deflections on a magnetometer at a place where H is known.
403. Unless such data are furnished with the instrument, determine
the temperature coefficients of the magnetic moments of Ma and M^
C
and calculate more precise values of -y and C.
404. From the equation ri^= — -^^ calculate the approximate value
P Q
of Ti and then allowing for a distribution factor, l-\ — f +^j calculate
l-\ — ^+-^ )^; with
due regard for the signs 'of Pa and Qa. This value, ro, is the dis-
tance at which the temperature magnet, Ala, must be set vertically
above or below the center of Alg, N end of Ma up, to provide the nec-
essary C field for effective compensation. For south magnetic lati-
tudes the A^ end of Ma should be down.
160
MAGNETIC OBSERVATORY MANUAL
TABLE 17. Z variometer: miscellaneous data
[Ch.
Observatory:
Variometer
Cheltenham, Md.
Toepfer No. XIII
Date: December 15, 1948
Observer R. X. R.
Ms
Ma
Pa
Qa
(li
Qi
Oz
Qz
Sz
ZR-Z
Z-Zr
39
mi
40
rm
41
vn
42
mt
43
TOi
44
45
46
section I. OPTICAL SCALE VALUE
Center of front of Zlens to Zspot on the magnetogram
Optical scale value, — radians per mm
SECTION II. VARIOMETER MAGNETS
Magnetic moment of recording magnet (cobalt steel)
Magnetic moment of temperature magnet (Alnico II)
Magnetic moment of A?" sensitivity magnet (if used)
Magnetic moment of S sensitivity magnet (if used)
Length of recording magnet
Length of temperature magnet
Length of A'" sensitivity magnet (if used)
Length of S sensitivity magnet (if used)
Distribution coefficient, recording magnet and temperature magnet
(from table 21a).
Distribution coefficient, recording magnet and temperature magnet
(from table 21a).
SECTION HI. TEMPERATURE COEFFICIENTS
Temperature coefficient of magnetic moment of recording magnet..
Temperature coefficient of magnetic moment of temperature magnet
3 times the coefficient of thermal (linear) expansion of the bar
Temperature coefficient of mechanical couple (lever arm a)
qi+qi
qi+qi+Qi+qi
Temperature coefficient of the variometer (calculated) (Eq. 233) . . .
Temperature coefficient of the variometer from tests
SECTION IV. TEMPERATURE COMPENSATION
Vertical intensity at the Z variometer
Ratio of required compensating field to vertical intensity (Eq. 238)
Required compensating field to make Qz=0; C=— 0.507 Z
Calculated approximate distance (see par. 404) assuming Pa=0 and
Oa=0.
Calculated distance (see par. 404) when Pa = -28 and Qa = +S20--
SECTION V. SCALE VALUE
Z scale value from deflections, at average ordinate; 7/mm
SECTION VI. RESERVE DISTANCES
Distance in mm on magnetogram, regular spot to upper reserve spot
Distance in mm on magnetogram, regular spot to lower reserve spot
SECTION VII. SHRINKAGE
Shrinkage gauge ; distance between points
SECTION VIII. MISCELLANEOUS DATA (OPTIONAL)
Mass of Z recording magnet including sensitivity poise but without
balancing poises.
Mass of sensitivity poise (without set screw)
Mass of south latitude poise
Mass of north latitude poise
Mass of temperature poise
Mass of assembled magnet system
Pitch of sensitivity thread, ^^o-inch (nominal)
Pitch of invar spindle thread, ^V inch (nominal)
Pitch of temperature spindle thread
2322 mm
0.000 215
877
404
100
100
8.1 cm
3.5 cm
4.0 cm
4.0 cm
-28
+320
0.000 34
0.000 28
0.000 06
0.000 01
0.000 35
0.000 69
0.538 cgs
-0.507
-0.273
14.36
13.72
4.52
175.6
175.2
100.0 mm
37.55 grams
1.397 grams
2.396 grams
2.396 grams
2.921 grams
42.34
0.0638 cm
0.0397 cm
0.0397 cm
405. Installation and major adjustments. — Set up the Z vari-
ometer at the Z position on the variometer pier and mount the tempera-
ture magnet in its holder at the calculated distance Tq. This is the
distance in cm from the center of Ma to the center of Ms, when Mj is in
operating position. The center of Ms may be taken as the knife-edge
13.] INSTALLING A MAGNETOGRAPH 161
(or pivot) support. Level the instrument and adjust it in azimuth so
that the axis of the Z lens is approximately normal to the drum. With
a camel's-hair brush remove any dust particles from the recording mag-
net, knife edges, bearings, and the damping chamber. Clean the opti-
cal parts and the magnet with lens cloth. Set the sensitive poise quite
low (large scale value), and the counterpoises at their midpositions.
Place the magnet in its cradle and lower it carefully by means of the
cradle mechanism so that the knife edge (or pivots) will take the load
gently. Quite likely the magnet will not be balanced. Lift the mag-
net by the cradle mechanism, adjust one or both counterpoises until
the recording magnet remains horizontal (as estimated by eye) when it
is resting on its bearings, N end to the north. Turn the whole variom-
eter, slowly, 180° in azimuth so that the A^ end of Ms is to the south.
At this azimuth, Ms is more sensitive. Rebalance if necessary, and
continue these tests until the magnet remains practically horizontal (as
estimated by eye) for any azimuth, N, E, S, or W. Finish the test
with the A^ end to the north and with the Z spot at the desired ordinate.
Readjust the Z base-line spot so that increasing Z ordinate corresponds
to numerical increase of Z.
406. Examine the Z spot and time line, and if there is appreciable
parallax between them, raise or lower the Z variometer until the
parallax is negligible.
407. Determine the scale value by deflections. Raise the sensitivity
poise several turns, rebalance if necessary by adjustment of a latitude
poise so that the Z spot is at approximately the same ordinate as in the
first test. Make scale-value deflections for this position of the sensi-
tivity poise. By interpolation or extrapolation, estimate the number
of turns of the sensitivity poise that will be necessary to produce the
scale value desired. Make this adjustment, repeat scale-value deflec-
tions, and continue this process until the deflections show that the
proper sensitivity has been attained, say within 5 percent. (Note: By
observing the period of the system for each position of the counter-
poise and then plotting a curve of period or turns of counterpoise vs.
double deflection or scale value, the value of a period which will cor-
respond to a desired scale value or value of 2 Uz may be easily
determined.)^
408. Make orientation tests, and calculate the exlevel angle. If
necessary, readjust the latitude poises to correct for a large exlevel angle.
Small adjustments in level may be made by slight adjustments to the
temperature-magnet distance. Continue orientation tests and adjust-
ments until the exlevel angle is 1° or less.
409. Unless the variometer is provided with extra prisms or mirrors
for independent adjustment of the ordinate of the Z spot, turn the vari-
ometer in azimuth until the Z spot is at the desired ordinate for routine
operation. Again adjust the Z base-line mirror so that increasing Z
ordinate corresponds to numerically increasing vertical intensity and
so there may be no negative ordinates on quiet days.
410. Repeat orientation tests and scale-value observations, record-
ing final values as in figures 72 and 84. The variometer is now ad-
justed for routine operation.
411. Final check, — Remove all tools and superfluous magnetic
materials from the variation room. Examine all of the recording spots
and the time line for parallax. Also, see that all of the spots are record-
> H. E. McComb, The sensitivity of magnetic variometers, Terr. Mag., 33, 65, 1928.
162 MAGNETIC OBSERVATORY MANUAL [Ch.
ing at the desired ordinates and positions on the magnetogram, when
magnetic conditions are quiet. If the instruments seem to be operat-
ing satisfactorily, attach the foot plates to the marble blocks or metal
spacers with plaster of paris and attach the spacers or blocks to the
pier in the same manner, exercising great care not to move the foot
plates or blocks from their proper places.
412. Make scale-value deflections and orientation tests for all the
variometers, recording all deflections photographically.
413. Report on the installation. — Furnish a complete, detailed
report of the project including sketches that show all pertinent
dimensions (see fig. 85 and fig. 131, page 211) and, if possible, a
photograph of the complete magnetograph made ready for operation.
LOW-SENSITIVITY MAGNETOGRAPHS
414. Object. — In high geomagnetic latitudes, where daily variations
are large and the amplitudes of the recording spots may be very high
during magnetic storms, variometers must be operated at low sensitivi-
ties. Even in low latitudes, many observatories operate both high
and low-sensitivity instruments to guard against loss of record during
magnetic storms.
415. Procedure. — From table 11, determine the approximate scale
values at which the variometers should be operated. Using the D
nomogram, figure 140, appendix VI, select a combination of size of
quartz fiber and magnetic moment of recording magnet that will give
the desired D scale value. (The use of sensitivity-control magnets
on a Z) variometer is not recommended.) Using the H nomogram,
figure 141, appendix VI, select a combination of size of fiber and
magnetic moment that will give the desired H scale value. A sensi-
tivity-control magnet may be used if necessary for final adjustment
to the desired value. The sensitivity of the Z recording magnet may
be adjusted to any desired low value by means of the sensitivity
poise.
416. In estimating the C fields for effective temperature compensa-
tion of the H and Z variometers, and in measuring the magnetic
moments of the temperature magnets and other magnets of the
magnetograph, follow the same procedures used for sensitive vario-
meters.
417. Auxiliary optical lever. — The torsion factors of large fibers
may be determined as follows: With a piece of laboratory wax, mount
a planoconvex mirror (fig. 45) centrally on the torsion head of the
D variometer so that the reflecting surface is vertical. Set up a
straight-filament incandescent lamp and a millimeter scale at the
same elevation as the mirror and at the proper distance for sharp
focus of the image. Using this optical lever as a means of estimating
accurately the angular motion of the torsion head (in radians) , make
torsion observations as described in paragraph 382, page 152.
Record aU observations and data, as shown in table 15, section VII,
for the sensitive H variometer. Calculate the torsion factor, . , ?
H-\-C
and the field factor, — yj^' ^^^ ^^^^ ^h® ^ scale value from equation
(122). It is desirable, though not strictly necessary, to determine
likewise the torsion factor for the H variometer.
13.] INSTALLING A MAGNETOGRAPH 163
418. Make orientation tests, scale-value observations, parallax
tests, and all other adjustments in the manner prescribed for sensitive
variometers except that the deflectors for scale-value and orientation
deflections should have larger magnetic moments or should be used
at shorter deflection distances.
419. Record all pertinent data as in tables 14, 15, 16, and 17.
Furnish a complete report of the installation with sketches showing
all pertinent dimensions as in figures 85 and 131, appendix V, and if
possible, a photograph of the complete magnetograph as it will be
operated.
THE LA COUR MAGNETOGRAPH
420. Description of the instruments, — Complete descriptions of
the instruments and directions for their installation are contained in
publications of the Danish Meteorological Institute.^ Only a brief
Figure 89. — La Cour vertical-intensity variometer equipped with sensitivity-control magnets, as
operated at Honolulu Magnetic Observatory.
summary will be given here. The general principles involved in the
selection of quartz-fiber suspensions and magnets are similar to those
outlined elsewhere in this manual. The use of adjustable prisms on
the la Cour D and H variometers simplifies many of the problems of
installation, such as adjustment of the azimuth and inclination of the
rays reflected from the miiTors of the recording magnets. La Cour's
method of attachment of fibers to the torsion-head stem and to the
magnet mirror frame is indicated briefly in paragraph 179. The Z
recording magnet, magnet mirror, and knife edges are fabricated from
2 La Cour, Danske Met. Inst. Pubs, (see items 8 and 11 of bibliography).
164 MAGNETIC OBSERVATORY MANUAL
one piece of magnet steel. The magnet is ground, polished, and
balanced by the manufacturer for a particular magnetic latitude and
sensitivity. Final balancing and/or adjustment of the sensitivity may
be accomplished at the observatory where it is to be operated, by the
simple process of careful honing of the magnet with a fine-grained
carborundum hone. To increase the sensitivity it is honed below the
longitudinal axis of symmetry that is normal to the knife edges, and
for balancing it is honed on one end. The knife edges are ground and
polished to a definite small radius, and at the same time the two edges
are made collinear by a simple and ingenious process. The polished
steel mirror is on the upper side of the magnet, and an adjustable 90°
prism mounted above the mirror affords a means of controlling the
azimuth and inclination of the incident and reflected light. The
magnet is balanced on a pair of agate cylinders and is in a sealed
chamber which is partially evacuated. At the Honolulu Magnetic
Observatory the la Cour Z variometer is equipped with a pair of
sensitivity-control magnets (see fig. 89.) Its performance for a number
of years has been quite satisfactory.
421. Optical compensation, — The 90° prisms on the H and Z
variometers are mounted on bimetallic (silver-platinum) strips. As
the temperature changes the strip bends, causing angular motion of
the prism. By proper facing of the strip and adjustment of its length,
the linear motion of the recording spot due to a change in temperature
of the recording magnet (and the resulting change in its magnetic
moment) may be just compensated by the angular motion of the 90°
prism through which the incident and reflected rays pass.
422. Reserve spots, — By the use of a series of 90° prisms mounted
in front of a straight-filament recorder lamp, two or more reserve images
are produced for each variometer, as mentioned in paragraph 170,
page 66.
CHAPTER 14. PROCESSING OF DATA AT THE OBSERVATORY
PROGRAM OF OBSERVATORY WORK
423. Routine duties. — After the observatory has been estabUshed
and the magnetograph and other instruments are functioning satis-
factorily, most of the regular magnetic work is of a routine nature.
However, to carry on this routine and to achieve continuously the
objectives for which the observatory has been established requires
great care on the part of the observer in operating delicate instru-
ments, considerable skill in the observing programs, and perseverance in
keeping the routine work on a current basis. Some of the principal
items are: (a) absolute observations for magnetic declination, usually
with a magnetometer, (see fig. 90) ; (b) absolute observations for
horizontal intensity with a magnetometer, sine galvanometer (fig. 37) ,
Quartz Horizontal Magnetometer (fig. 91), or other equally precise
instrument; (c) absolute observations for inclination (dip) with an
earth inductor (fig. 92), or (d) observations for vertical intensity with
a standard magnetic field balance or Magnetometric Zero Balance
(BMZ),^ or some equally precise instrument; (e) scale-value observa-
tions; (f) time observations for comparison of chronometers and
pendulum clocks; (g) changing the traces daily on the magnetograph
recorder and developing the magnetograms ; (h) applying legends on
the magnetograms (see fig. 93) ; (i) scaling of hourly ordinates of H,
D, and Z, from the magnetograms (see fig. 100); (j) computations of
absolute values of D, H, and / from absolute observations; (k) com-
putations of D, H, and Z base lines; (1) estimation of character
figures; (m) scaling of K indices; (n) tabulation of times of sudden
commencements of magnetic storms, times of natural disturbances of
the variometers (seismic or other) ; (o) solar flare studies; (p) observa-
tions for orientation of variometer recording magnets; (q) operation
of auxiliary magnetographs, such as low-sensitivity variometers,
photoelectric magnetographs,^ etc.; (r) abstracting of pertinent
records and monthly transmittal of data and magnetograms to the
home office.
424. Conversion to absolute values, — In order to determine the
absolute value of D, H, or Z for any moment from the magnetogram
it is necessary to know: (a) the base-line value, that is, the absolute
value of the element when its trace and base line coincide; (b) the
scale value ; and (c) in the case of H and Z, the temperature coefficient
of the variometer. Let d, h, and z denote the ordinates in mm at the
temperature t, corrected for shrinkage of the magnetogram (increasing
ordinates corresponding to increasing D, H, and Z). Let Sd\ Sh, and
' La Coiir, Danske Met. Inst. Pub. 19 (see item 10 of bibliography).
2 R. E. Gebhardt, T. J. Hickley, and T. L. Skillman, A photoelectric magnetograph, Trans. Amer. Geoph.
Union, 32, 322, 1951 (abstract only). This arrangement is illustrated in figure 94.
165
166
MAGNETIC OBSERVATORY MANUAL
[Ch.
Figure 90.- Observatory magnetometer (Ruskatype) with telescope having modified Gaussian eyepiece.
Sz represent the scale values of the D, H, and Z traces respectivelv,
and let i^ J;
*So=the base-line scale value for H;
a=the a factor for H;
SH=So-\-i a h (see par. 259);
^o=the standard temperature, usually 20° C.
14.1
PROCESSING DATA
167
a
Figure 91.— Quartz Horizontal Magnetometer (QHM) mounted on a special base
with divided circle and verniers.
Finally, Qh and Qz are the temperature coefficients of the H and Z
variometers (par. 288) ; and Bd, Bm, and Bzo are the base-line values
for D, H, and Z (in the case of H and Z, reduced to the standard
temperature ^o).
Then
D=Bj,+Sn'd
=B^o + {So + iah)h+QH(t--to)
Z=:Bzo+SzZ + Qz(t-to).
(348)
(349)
(350)
(351)
168
MAGNETIC OBSERVATOEY MANUAL
[Ch.
14.]
PROCESSING DATA
169
^
^-^
2
N N
510H1— 53-
12
170
MAGNETIC OBSERVATORY MANUAL
[Ch.
Figure 94.— Special light source and photoelectric ce.. ^^d „.»h » h^
visual-recording magnetograph.
ll'^^J.ftOll'J
=r for
425. Base-line values, — For the determination of base-line values,
absolute observations are made at least once a week. From an
inspection of the above formulas it will be seen that if the ordinates
d, h, and z be read for the times at which absolute observations have
been made, the base-line values may be computed, provided the scale
values and the temperature coefficients are known. Generally the
absolute value of the vertical intensity must be computed, however,
from values of H and /. It is, in general, not feasible to make simul-
taneous absolute observations of H and /, but the value of H at the
time of the absolute observations of dip may be determined from the
magnetogram after a preliminary H base-line value has been com-
i
A
14.]
PROCESSING DATA
171
DEPARTMENT OF COMMERCE
' U. S. COAST AND CEODCTIC SURVa
Form No. 3«7
Ed. July 1048
Magnetic Observatory ..S9M.?.91t.JM?JA.
JAIf.r..J.SliP:.
(Month) (Ymt)
DECLINATION BASE-LINE VALUE
Maineto^raph MCMNMGM Magnetometer Xo ?3..
Date
.A5g. M.
Mian
SCAUNO •
Computed
OBDlNATKt
OsaCBTED
D
Basb-une
Valob
Observer
Remabes
Jan 3
-f.
mm.
11,6
11,4
39 U.5
2g 03.:
VAT
3
3
3
"-4
o8.8
o8.8
12. 1
_o8,6
o8,6
i6.o
12.1
11.8
03.9
03:5
03.2
..JAL.
_JLC_
MLC
8
"it
11,3
11, 1
14.9
03.8
MLC
8
-?|
13-5
13' s
17.4
04.2
MLC
8
>3?l
og.g
og,n
13.6
03.9
7AT
8
'4^
og.8
og, 6
13.1
04.1
7AT
~
Mean
^.9.03:1
♦Corrected for shrinkage to ...1?.?.. :?.... mm.
t Assigned scale value = 0.«..9.9- '/mm. at zero shrinkage.
Scaled bv Y.J...Thomas Checked by .LL.Q.h.V.^J}....
Computed by f.A.T. Checked by ...K..k..Q. Abstracted by ...H...k.^..
u. s. eovMNHCNT miNTiNS orricc 1»— 50307-1
Figure 95.— Declination base-line computations.
puted from equation (355). Z is computed from this value of H and
the observed / by the relation
Z=H tan /.
(352)
172
Department of commerce
u. s. coast and geodetic survey
Form No. 368
Ed. Oct. 1042
MAGNETIC OBSERVATORY MANUAL [Ch.
Magnetic Observatorij .TUCSM,...ARIZOIfA...
Sept 1951
(Month) (Year)
HORIZONTAL INTENSITY BASE-LINE VALUE
Madnetodraph }.9. 9.9.V:V. Magnetometer ^o 3^. ...
-2
V
CO
H
O 0)
=5S
Date
4 1
4 i
11
i 05. M. TIME
Mean
Time
^rKAN
Scaling
Time
Mean
Begin
End
SCAUNG
mm.
Becin
A. m.^
..13.38
End
h. 171.
i.a.45..
Bigiu
End
SCAI.INO
Oscill. E
ft. m.
.09.30..
ft. m.
.09.57..
mm,
21..$..
A. m.
.09.37..
A. m.
.09.. 44..
&
Defl. E
.^.0.03.
,10 13
.^3.'.4
..^3.5.4.
M.O.8.,
94.'.^..
.P9.53.
10 07
?5r.4
Dcfl. I
_ipj6
_ip,26
.13!.9
.1.4..IJ?.
14..^...
95.'.o.
.1P..1.2_
.10.. 26.
.^f.O.
Oscill. I. ..
10 33
10 40
14.1
H 35
14 42
24.1
10 38
10 4*5
1Q.4
Mean Ord
i<^8t
^r7A
24.00
100,6 -.06
.100,6
100. 2 -. Of^
Corr. Ord.* .
13^8
2^.6
24.0
*A (7/mm.).* 2!
h (7)
^32
2,98
2.q8
.4^...
70...
.79...
H (Observed)...
sf)98s
a6ooq
26001
H Base Line.. ..
.^9.^...
.259.39...
.^59.^9..
Observer
...l.B.Camt^.en
L R Southwick
G^EM.k^r.
Dati
1
Time
Mean
SCALI.\G
Time
Mean
SCAUNG
Time
Mean
Begin
End
Begin
End
Begin
End
Set UNO
h. m.
ft. 771.
mm.
A. m.
A. J7I.
mm.
A. m.
A. m.
77J7:.
Oscill E.
Defl E
Defl. I
1
Oscill. I.
Mean Ord
\
1
Corr. Ord.*
1
h(-Y)
H (Ob.served)..
H Base Line
Observer...
^ ?^ • Ordinate corrected to ...?P?.».P. mm. shrinkage distance.
? Scaled by L..R..§. E..li..R. Checked by 9...?...A.
Computed by ..9.^.4.*....?.^.^... Checked by ...l?.^.'....^!!^... Abstracted by
GEA
u. s.
NTINS orrict 10—50430-1
Figure 96.— Computation of horizontal-intensity base-line value.
We have for the computation of the base-Hne values from the observed
absolute values, Da, Ha, and I a,
Bo=DA-S'^d
BHo=HA—SHh—QH (i—to)
=HA-{S,^hh) h-Qn {t-k)
(353)
(354)
(355)
14.1
PROCESSING DATA
Magnetic Observatory ...MfiQMMj...L'J:
J.^.h...J95j
(Month) (Yet)
VERTICAL INTENSITY BASE-LINE VALUE
Ma$neto$raph 4 Earth Inductor ^o 4..
173
Department or Commerce
U. S. COAST AND CCOOCTK SURVIY
Fy>nn No. 359
(Rev. Oct. 1042)
Date
T
i
*!
1
10
..1<55.M.
Tim
Mian
scalinos
Time
Mean
scalings
Time
Mean
scalinos
H
z
H
z
H
Z
lat half
h. m.
10 Jg
mm.
17. £
mm.
3P,o
'3^"
mm.
12,7
mm.
32.7
ft. ffl.
"48
mm.
27.8
25.0
OH hftlf
^'t^
17,0
29.9
«^3
12,7
32-9
"S
28.1
25,0
Mean Ord
17-10
29.95
12,70
32B0
27-95
25X>0
99-5
,01
.1'
..9M
. Ol ' .16
..99.S
V06
x>5
O .: fJorr Ord *
17,^
30,1
12.8
33.0
~^o
25,0
° «2 «fc and €,
3..?<5
.2-.i?9
2,76
^.?9.
2.7^
3-29
•S^"** At and Ah
i; o a
? **. ! H base line
47
35
78
28504
i 28504
28504
5 ac^i
° S d! ~ H
! 28^51
! ^539
28582
8:1
Icrj ! I ^
51,81
....53:
4.4
2n
48.38
1di|
4., 4
jii^^
1544...
10663
4.456og
g. g
10625
1 9.?
9. 90552
inates separately
aled interval
time flashes
3 CSJ N
a.
...4:.3^.i8l..
22008
..4.'.^.20X..
22018
...4-..3^i§.i..
.^.^9.9.4..
a . a ^t and Az
la 1
99
log
82
1 « -
1 ? Z base line
.?2.9.9.9.
! 2i?9?.9.L...
229 12_
rt " Observer
T H Pearce R F White
D C McGowan
' Ordinate corrected to '.... mm. shrinkage distance.
Scaled by
.P.S...^. Checked
by
D C M
Computed by Checked by
Abstracted by
Z revised 194
R F V
using differential formula AZ- AH+ AL
«. eOVHMMCNT PRIMTINfi OmCI 10 — 60438-1
Figure 97.— Computation of vertical-intensity base-line value.
and
Bzo=H tan h-Sz z-Qz {t-to) (356)
= [BHo+{So-^icih)h+QH{t-to)]tein I^-Sz z-Qz (t-to), (357)
Figures 95, 96, and 97 will illustrate how D, H, and Z base-line
values are computed for the cases where the temperature coefficients
of the H and Z variometers are zero. Note: In figures 96 and 97
the symbol e is used in place of S for the scale value.
174 MAGNETIC OBSERVATORY MANUAL [Ch.
IMPORTANCE OF ADEQUATE CONTROL
426. Control observations for deriving absolute values. — Di-
rections for making absolute observations with a magnetometer and
earth inductor are given by Hazard.^ Directions for operating the
magnetic field balance are well covered by Joyce* and by Heiland
and Pugh.^ A complete description of the sine galvanometer is
given in the Researches of the Department of Terrestrial Magnetism.^
The publications of the Danish Meteorological Institute ^ give com-
plete details on the operation of the la Cour instruments.
427. In making absolute observations at a magnetic observatory
greater care must be exercised in the operations and a greater degree
of accuracy is required than is the case for work in the field. How-
ever, it IS not possible, as a rule, to achieve in the individual observations
an accuracy on a par with the short-term stability of the variom-
eters. Usually the base-line values resulting from a series of observa-
tions, when plotted against time, will show more or less spread due to
accidental error in the absolute observations, or to the application of
temperature or other corrections that are made with faulty coeffi-
cients. They may also show a progressive change or drift with time.
This drift may result from loss of magnetism of the suspended mag-
nets, from the seasonal effect of error in the adopted temperature
coefficient of the variometer, or (for ET) from changes in the elastic
properties of the quartz-fiber suspension. Hence, in determining
what base-line values to adopt, it is necessary to adjust the observed
values with due regard for these progressive changes. For any par-
ticular set of instruments it can be determined only by experience
how closely the adopted values should correspond to those resulting
from actual observations.
428. Time observations, — Chronometers, pendulum clocks, or
other timepieces used in any part of the observing program should be
compared daily with radio time signals broadcast by the U. S. Naval
Observatory, the National Bureau of Standards, or other reliable
source. Where a time-flashing mechanism is used to place time marks
on the magnetogram, it is convenient to keep the time correction of
the time-marking clock small so that it will not be necessary to allow
any time correction in scaling values from a magnetogram. The
method of keeping a record of the performance of a chronometer or
clock is shown in figure 98.
DIRECTIONS FOR PROCESSING RECORDS
429. Producing the magnetograms. — Change the paper on the
magnetograph drum daily at the same hour, on the hour, noting on
the magnetograph record (fig. 99) any pertinent facts, however trivial,
regarding any adjustments of the instruments, natural or artificial
mechanical disturbances, or other unusual events related to the
operation of the magnetograph and of the absolute instruments.
Develop the traces as soon as possible after exposure, and examine
them carefully for quality and for possible malperformance of any of
3 D. L. Hazard, Dir. for Mag. Meas. (see item 4 of bibliography).
* J. W. Joyce, Manual on Geoph. Pros, with the Mgr., U. S. Dept. Int., Bur. of Mines, 1937 (see item 7
of bibliography).
« C. A. Heiland and W. E. Pugh. Am. Inst. Min. and Met. Eng.. Tech. Pub. 483, 1932 (see item 5 of
bibliography).
« Barnett, DTM, CIW, Pub. No. 175 (see item 1 of bibliography).
7 La Cour, Danske Met. Inst. Pubs, (see items 8-11 of bibliography).
14.
PROCESSING DATA
175
DEPARTMENT OF COMMERCE
U. S. COAST AND GEODETIC SURVEY
Form 698— Ed. Jan. 1950
18— 29fl5.V.>
TUCSON, ARIZONA Chron. 1.^95
(Cbron. or Clock No.)
Jan,
^95a
TIME
RECORD
Date
BnoAD-
CA8T
Station
-_/*°-i_ Mer. Time
OF Signal
^°5 Mer. Time
BY Chron. or Clock
1
WW7
h. m. s.
09 00 00
h.» in s.
09 0^ 49. 2
2
II
52.8
4
II
f^^.8
4
II
68, g
5
II
04 02, 5
6
II
06. ^
7
II
OQ, 4
8
II
1^.2
9
n
16.0
n
^9' 3
,
Figure 98.— Time comparisons between chronometer and continuous time signals from Station
WWV. U. S. National Bureau of Standards.
176 MAGNETIC OBSERVATORY MANUAL
DEPARTMENT OP COMMERCE
US. COAST AND GEODCTIC SURVEV
Form 247
Re.. Feb 1938 MAGNETOGRAPH RECORD
[Ch.
Oct
19_5Li
TUQSOlf Magnetic Observatory.
Magnetograph No-^omt
S5
TiMK BY Chron.' No. 1"*^ ch
Tkmpkb-
ATUBI
RXIIABXS
02
Stop
Begin
Othert
Chion.'
Clw5k
H
z
A. TO.
ft. m.
ft. TO.
10 04
10 00
m. t.
m. 1.
Parallax Test
10 08
10 49
Scale Value
IS
05 00
26,^
08 02
13
08 00
26.6
08 02
14
08 00
26. i
08 09
^5 57
Check Operations
ifi
26,Ji
16 02
16 10
Visitors
16
oR 00
3iJ3
08 02
17
08 00
a^/i
08 02
iH
f^S 00
^ ^
nR n^
10
2'y ^
Cleaned jJoPer clamp bar
08 05
20
08 00
^f'4
08 02
t Thto coloaa la to b« DMd fei fiagb Hm* anblM, uraollx whm Ih* photograpUe trao* to not iBtoiraplad.
w. •. ceviiNMNT raiNTiNC omci 10— 41W(hl
Figure 99.— Magnetograph Record, Tucson Magnetic Observatory.
the variometers or of the time-marking system. Estabhsh a systema-
tic and fixed routine for this part of the work. Carefully apply all
necessary legends on the magnetograms as shown in figure 93. This
work should be done neatly and accurately, because the magneto-
grams are subsequently reproduced for publication.^
« U. S. Coast and Geodetic Survey, Magnetograms and Hourly Values MHV-Ch50 (in press, 1952).
14.]
PROCESSING DATA
177
SCALING OF ORDINATES
430. Types of scaling, — Three varieties of scaling are customarily
done on the magnetograms at the observatory, namely, those con-
cerned with: (a) hourly values, or mean ordinate over each of the
twenty-four hourly intervals during the day; (b) base-line scalings, or
mean ordinate for the time interval during which an observation has
been made for absolute value of a magnetic element; and (c) scale-
value deflections.
431. Hourly values, — For measuring the average ordinate for an
hour a special scaling glass is used, accurately engraved with rulings
sc w » » w
16-
."S
Figure 100.— Magnetogram reading scale superimposed on magnetogram
for scaling ordinates. Cut-out portion, upper left, shows method of
averaging hourly ordinates. Parallax test and scale-value deflection on
right.
and graduations as shown in figure 100. Vertical or lengthwise lines
on the scaling glass are 20 mm apart, corresponding to a time interval
of one hour on the magnetogram. Horizontal or transverse lines are
1 cm apart, with finer divisions of 1 mm. The scaling glass is laid
on the magnetogram with the ruled surface next to the paper, and with
the vertical lines coinciding with the hour marks on the gram. The
space divided into millimeters on the glass should be placed across
the base line. The scale is then moved up or down until one of the
traverse lines is set for the average ordinate of an hour-long interval.
With a little practice this can be done rapidly and accurately by
making equal the areas between the trace being scaled and the trans-
178 MAGNETIC OBSERVATORY MANUAL [Ch.
verse line, as illustrated by the shaded areas in the figure. The
number of whole centimeters is read at the end of the transverse line,
and the fraction of a centimeter is read to tenths of millimeters at
the base line. The tabulations are ordinarily made directly in tenths
of millimeters for the whole ordinate. Thus, for the example illus-
trated the tabular entry would be 257.
432. Occasionally during disturbed magnetic conditions it is difficult
to judge accurately the proper setting of the scaling glass for the
mean ordinate of the whole hour. It is then easier to use half-hour
or quarter-hour intervals, calculating the mean of the two or four
values thus scaled for use as the hourly value. One section of the
scaling glass is divided into half-centimeter (quarter-hour) intervals.
Under extremely disturbed conditions it may be found desirable to
divide the hourly interval into even smaller time intervals by drawing
auxiliary time lines directly on the magnetogram, but it is seldom that
the increase in accuracy thus obtained for the hourly value can justify
the extra time required by this procedure.
433. Tabulations of the hourly values may be made in any con-
venient form. Figure 101 illustrates the form now being used by the
Coast and Geodetic Survey. This style of scaling sheet expedites the
step of transferring the values to punch cards for further processing of
the data at the Washington office, where the publications are compiled.
434. Base-line settlings. —The same procedure described for hourly
value scalings is used for base-line scalings, except that the interval
of time is determined by the duration of the absolute observations
with a magnetometer or earth inductor. Extra time lines are placed
on the magnetogram, through a push-button circuit from the absolute
building, when an observation is begun and again when it is finished.
For instance, observations for declination usually require 9 minutes.
The extra time lines would thus be 9 minutes, or 3 millimeters, apart
on the magnetogram, and the mean D ordinate for that 9-minute
interval would constitute the base-line scaling for that declination
observation. Base-line scalings are customarily tabulated in milli-
meters, to the nearest tenth-millimeter.
435. Scttle-Vttlue deflections. — Figure 100 also illustrates a typical
set of scale-value deflections for all three elements, D, H, and Z. In
the illustration, deflections of the Z trace are made first (two deflec-
tions with the deflector magnet placed N end up, separated by a
deflection made with the A^ end down) , followed by deflections of the
D trace, and finally deflections of the H trace. All three variometers
are disturbed, of course, when the deflector magnet is in place for
any deflection, so that care must be used in selecting the proper
deflection spots on the magnetogram.
436. Scalings of the deflected spots are made, not to the base line
but from one spot to the immediately succeeding spot — that is, from
an up spot to a down spot and then from a down spot to the next up
spot. (All measurements should be made in a direction perpendicular
to the base line^ — ^it is the difference between ordinates of deflected
spots that is being measured.) This is a measurement of 2Uy the
effect caused by a complete reversal of the deflecting magnet, and the
quantity used in computing the scale values of the several element
traces on the magnetogram. The sample scale-value computation
shown in figure 72 is derived from the deflections of figure 100.
14.1
PROCESSING DATA
179
DEPARTMENT OF COMMERCK
U. S. COABT AND OlODBTIC SUIVIT
Cto & Stis
gOJJi^GE^ ALASKA 1951
(Obiirytiori', (i'mt)
OCT,
"■"(MoiHh)
HOURLY MEAN ORDINATES FROM MACWETOGRAMS
Average values for successive periods of one hour beginning at midnigh t 150 M.T.
Tenths
Signs
of mi
are as
llimeters corrected to shrinkage distance
follows unless otherwise noted: D +
Signs reviewed by
of
; H_
+
•>
100.0 mm
Z +
Char.
1
*Shr.
99.0
99.0
99.0
Day-
25
26
27
Hr.
D
H
Z
Hr.
D
H
Z
Hr.
D
H
Z
01
101
375
71
01
02
lip
197
26..
199
8.2.
14
01
02
53
94
.447
475
98...
95
02
97
372
72
...P.?
04
.....1.9.?...
103
364...
366
.71...
69
03
106
446
51
.....03
04
121
528
82
04
109
383
75
130
241
52
05
110
.....3.67.
69
05
146
362
73
05
110
292
5
06
113
368
70
06
153
352
65
06
135
567
■ 40
...07
08
...1.2.5...
130
.....3.65...
368
.7.0..
70
...0.7
08
.1.4.4.
163
.3.61.
363
.6.2...
61
07
132
362
61.
08
125
358
64
..0?
10
...1^1..
112
.....3.6.8....
362
.6.7....
68
....0.9
10
.1.5.8..
146
.57.2...
356
.5.8...
62
.....09
10
.1.11.
98
.5.5.2.
345
67.,
68
...i.l
1?
.....102...
89
.....3.63...
366
.6.7...
65
....1.1
12
99
48
332.
311
59
77
11
92
538
70
12
83
345
72
...1.2
14
82
....368...
67
13
-115
278
63
.....13
14
.7.2.
71
.552..
551
7.5.
71
90
363
70
14
-62
341
68
..15.
1*1
81
365
70
....1.5
16
10
578
89
....IS
16
.71.
70
.5.55.
560
...7.2..
70
82
363
70
31
391
104
...17.
IR
SI..
84
.....3.7.Q...
373
.6.9....
69
...17
IR
50.
-15
4.Q2..
441
92...
95
....1.7
Ifl
.77.,
3:7.0.
71.
74
577
72
..19.
?0
86..
75
37.4
376
69...
69
...19
20
12.
49
543..
510
126..
131
...1.9
?0
71
53
3,81
510
.7.6.
122
..21
??
42
78
450
521
90
102
...21
22
40.
. 61
473
396
127
109
....21
22
.162
98
.458
422
145.
124
.22.
.6.5..
....16.7....
452
94...
65
...25
?4
67.
68
3.8.7..
422
9.2...
91
...25
24
.6.7.
68
.4.1.6.
587
Ill
100
Sum
2182
9246
1733
1735
8825
1926
2256
8985
1859
Mean
91
385
72
72
368
80
95
574
77
« Seated 6y
W.H.S,
K.H..S..,
W.H.S,
»CK,ek»d 6v
1
V.A.T.
V.A.T.
V.A.T 2
tThis entr
y refe
rs to
magnet
Figu
ogram w
re 101 —
hich
Tabi
1 begir
ilation (
IS on t
>fscale
log (r)2=log T2-log (l+2aA^).
By equation (367),
log iTy=los T'-0MSa{t-20) ^40(3)
and, for a=0.000 Oil,
log (r)^=log r2-|_o.OOO Ol (20-^f). (407)
The factor 0.000 01 (20 — ^) is the correction that must be apphed to
log T^ to give the value of log {Ty at 20° C. In this case a= 0.000 Oil
the coefficient of thermal expansion (linear) of steel. The coefficient
is slightly larger for aluminum -nickel-steel alloys.
450. Correction for change in (K^Ki) due to temperature
change (loaded system), — It may be shown that the treatment is
the same as for changes in the dimensions of the magnet (or the
inertia weight) due to temperature change, except that in this case
the effective coefficient of expansion is taken as approximately the
mean of the coefficients for steel and brass. Taking ai = 0.000 014 as
this mean coefficient, and letting AK{=Kt—K2Q) represent the total
increase in the moment of inertia of the magnet, stirrup, and inertia
weight combined (due to an increase in temperature A^(=^ — 20),
with all other coefficients equal to zero), and substituting in equation
(406),
log (ri)2=log rf-0.868X0.000 014 (^-20) (408)
=log Tf +0.000 012 (20-0, (409)
190 MAGNETIC OBSERVATORY MANUAL [App.
in which (Ti) is the time of one oscillation at 20° C (loaded system),
and Ti is the observed time of one oscillation at t° C (loaded system) .
The term, 0.000 012 {20— t), is the correction that must be applied to
log Tf to give the value of log (TiY at the standard temperature,
20° C, under the imposed conditions.
451. Correction to T^ for torsion factor, — Let
/ =angular motion of torsion head in minutes, in torsion test;
h = angular deflection of magnet in minutes, for (f—h) minutes
of twist ;
(T) =true time of one oscillation (torsionless fiber; H, Ms, K,
chronometer rate, etc. constant) ;
T = observed time of one oscillation, torsion effective ; (ii/, A/^,
K, etc., constant).
For a small amplitude and torsionless fiber,
HMs'
(T)'=^y (410)
and with torsion effective,^
HMs+t
T'=^ rTL ir (411)
By equation (151), when (7=0
j-h
k'=HMsTK (412)
and T'= —
Divide equation (410) by equation (413):
T' ^j-h f-h
HMs-\-HM, J^ (413)
(414)
Letting /== 5400', the usual angular motion of the torsion head in
magnetometer work,
/ _ 5400' _ h , .
/-/i"'5400'-/i,"" ^5400'-/i ^ ^
Since h is usually small compared to 5400', equation (414) may be
written:
^«l+j (416)
> Dir. for Mag. Meas., pp. 16, 17 (see item 4 of bibliography).
I.] APPROXIMATIONS 191
By equation (362),
log {Ty^log r-hOAU ^, (417)
'--+'(°Jr)
«log T^+OmOOSh. (418)
Note: Even when h is as large as 60', which is unusual, equation (418)
holds with sufficient precision in this work.
452. Correction for change in H (all other factors constant) . —
Let (T)^ be the time of one oscillation, reduced to a standard H {=Hq) ;
Then (^)^=|§; (^19)
Now let H change by AH, such that AH is small compared to Hq;
H is now Hq+AH, and
Divide equation (419) by equation (420):
(D^=r(l+^ (422)
log {T)'=\og 7^+log (l +^- (423)
From equation (362),
aTT
log (T)'^\og 7^4-0.434 ^. (424)
When the observations are made at a magnetic observatory it is proper
and convenient to choose the H base line value (Bh) for Hq. Then AH
becomes the mean ordinate of the H trace during the time of a set
of oscillations. Thus
H=Ho+AH (425)
becomes H=BH+hy (426)
=B„+hrr,^S„ (427)
and ^HKnA (428)
192 MAGNETIC OBSERVATORY MANUAL [App.
Then, from equation (424),
log {Ty=\og T2+0.434 ^ hmm- (429)
Example: Let 5^=18,200
&=2.60.
Then 0.434 ^=0.000 062.
Bff
It may be shown that any value of Bh within about one-half percent
of the value of H is sufficiently accurate to reduce T^ to the value (TY
that would be observed if the H ordinate were zero.
453. Correction for induction in oscillations. — Let
AM=the change in the magnetic moment due to induction;
M=the induction factor (see par. 110, chap. 4).
r^, AM .AM H
Then M^-^, and ^j^=M^-
In oscillations,
iTy=TfTr (everything constant, fjL=0). (430)
When induction is effective and everything else is constant,
^'^HiM-hAM)' ^^^^^
Divide equation (430) by equation (431),
(Xy M+AM AM
As explained in paragraph 110, AM may be replaced by yiH, and
equation (432) becomes
{Tf , , H.
then
log (T)^=log T^+log (l+M^) (433)
From equation (362),
log {TY=\og r^+0.434M ~ (434)
I.] APPROXIMATIONS 193
In practice it is convenient to use a table of log ( I+m t>) against
TT / TT\
log Yf ^or a given value of /x, to get log ( 1 H-/z ^r? )• In the lower part
of figure 35, page 56, such a table is illustrated, showing critical
TT / TT\
values of log ^ for values of log f 1 +/x ^ b for ju=3.5.
454. Variation of log C with temperature in magnetometer
deflections, — In the deflection equation
^ ^ (435)
M sin u
ture because
r varies
with
temperature.
That is,
^=1
"20
~
dC 3(7
dr r2o
and
dC= ^^dr.
^20
2 .
the constant C is made up primarily of -3; which varies with tempera-
(436)
(437)
(438)
But dr=r2oadt (439)
in which a is the coefficient of thermal expansion (linear) of the
deflection bar, and dt is a small increment in the temperature. When
everything is constant at 20° C,
C^=^- (440)
'20
When the temperature changes by A^ and everything else remains
constant except r,
Divide equation (440) by equation (441),
C20 (r2o + Ar)
(■+47
(442)
log C2o=log C,+3 log (1 + 7)- (443)
194 MAGNETIC OBSERVATORY MANUAL
From equation (362),
hence
and
log (720 «log 0,+ 1.30 —
(444)
(439), j=otM',
log (72o«log Ct^-l.^OaM
(445)
«logO,+ 1.30a(^-20)
(446)
log Ct—\og ^20= A log C
«-1.30a(^-20)
«1.30q!(20-0
(447)
A-/
(448)
Examples:
For brass, «=0.000 018;
log C2o=log C'.+O.OOO 023(^-20). (449)
For duralumin, a=0.000 022;
log C2o=log C. + O.OOO 029(^-20). (450)
APPENDIX II. TEMPERATURE COEFFICIENT OF MAGNETIC MOMENT
455. In Chapter 4, the temperature coefficient of magnetic moment
was taken as a constant over a temperature range of 50° C. However,
the relation between temperature and magnetic moment is not strictly
linear. A more precise expression is given by
Mt=Moil-qt-qr) (451)
in which Mt=the magnetic moment at a temperature t° C,
ilio=the magnetic moment at ^o° C,
and q and q' are coefficients which must be evaluated from experi-
mental data for each magnet.
Differentiating equation (451) with respect to t,
^^=-Mo(q + 2q't) (452)
AM= -Mo{q+2q't)At. (453)
Equation (452) gives the rate of change of the magnetic moment at the
temperature t° C. In general we are concerned with the change in
magnetic moment for a particular change in temperature, hence the
average rate, -— -; between the reference temperatures must be used in
calculating AM from equation (453).
456. It may be shown that the average rate of change over the
range, ^o to ti, is
AM
At '~
Mo(q + 2q^ ^^) (454)
(455)
and AM= -MJq-{-2q' ^-^)^^'
457. Example: Let
A/o=1000 cgs at fo;
^0=0° C;
^1=50° C;
2=0.000329;
and g'= 0.000 000 91.
195
196 MAGNETIC OBSERVATORY MANUAL
Then from equation (454),
^=_1000r0.000 329+2(0.000 000 91)(' ^^°^^° ')1
= —0.3745 cgs units per degree C, for Mo=1000.
When Mo=l, ^=-0.000 3745 = 2-
Then q is the mean temperature coefficient between 0° and 50° C.
From equation (455)
AM= — 1000X0.000 3745X50 =-18.7 cgs.
Equation (451) will give the same result.
458. Summary, — This analysis may be summarized as follows:
(a) Equation (451) gives the magnetic moment at a temperature
t° C when Mq and the q coefficients are known;
(b) Equation (452) gives the temperature-rate of change of magnetic
moment at the temperature t;
(c) Equation (454) gives, for the mean temperature between ^o
and ti, the temperature rate of change of the moment of the magnet
whose moment is Mq at fo or 0° C.
(d) Equation (455) gives the approximate change of the magnetic
moment, AM, when the temperature of the magnet changes by A^;
(e) The process is reversible ; that is, to go from a higher temperature
to a lower temperature, change the sign of A^.
APPENDIX III. ORIENTATION ERRORS
459. Errors in the computed value of Ex. — The formulas for
computing E^ for D, H, and Z variometer recording magnets assume
that the deflector is placed properly and that the deflection, 2u, if it
exists at all, is due entirely to the misorientation of the recording
magnet. However, if the deflector is improperly placed, there will
be a deflection of the recording spot despite correct orientation
of the variometer magnet. If the usual orientation formula is applied,
using the 2u recorded on the magnetogram, the apparent (computed)
value of Ex will be made up of two parts: first, that due to the real
misorientation of the recording magnet; and second, that due to
improper placing of the deflector. There is no way to separate these
two parts. It can only be assumed that the second part can be kept
small by using great care in properly installing the stops or guides
that hold the deflector in place while the orientation deflections are
being made. The effects of improper placing of the deflector are
summarized in table 18, page 203.
460. Discussion of errors, — In this discussion it is assumed that:
(a) The deflector, or its holder, is a rectangular parallelepiped;
(b) the deflector is symmetrically magnetized with its magnetic axis
approximately coincident with its geometric axis or with the geometric
axis of the holder;
(c) the test deflections are made by placing the deflector (or holder)
against the fixed stops or guides which definitely i^x the direction
of the geometric axis of the deflector (or holder) and the geometric
center of the deflector or holder with respect to the stops or guides;
(d) a sufficient number of deflections are made with the deflector in
the four positions shown on figure 84, lines 6-9 (see p. 145) to eliminate
the residuary effect of nonco incidence of magnetic and geometric axes;
(e) errors in fixing the guides or stops are small, say of the order
of one degree or less in azimuth and level ; and in any direction normal
to the magnetic meridian the departure of the geometric center of the
deflector does not exceed H of one percent of the value of the deflection
distance, r (5 parts in 1000); and
(f) the recording magnet is in perfect orientation.
Note: It can be shown that the error equations derived below also
hold approximately for the condition that the recording magnet
under test is improperly oriented by a small angle of the same order
of magnitude as the error angles introduced in setting the deflector.
461. A perfectly adjusted deflector will not deflect a perfectly
oriented recording magnet. Imperfect adjustment of the deflector,
Ma, will cause a deflection of the recording magnet, Ms, when Ma
produces an effective field perpendicular to Ms, in the plane of rotation
of M,.
Let
/p= component of the test field parallel to the recording
magnet;
197
198 MAGNETIC OBSEEVATORY MANUAL [App.
/„= component of the test field normal to the recording magnet
(in the horizontal plane for D and H) in the vertical
plane for Z) ;
'Ui = the real ''up" deflection of the spot (taken as +);
^^2=the real ''down" deflection of the spot (taken as -f");
2u=Ui-\-U2',
iS^the normal scale value (gamma scale value);
(£'a:)c==the calculated exorientation angle of M,;
and Sj^the real exorientation angle of M,=0.
Then from equation (334), page 139,
(456)
tan
But by definition of S
Jn=u, S;
and upon reversal of Ma
fn = U2S.
Hence
2uS=2fn
and
tan
jp
(457)
Therefore, part of the analysis reduces to a comparison of /« with/,.
462. By introducing small known errors of placement of the
deflector, the effect of such errors on the calculated exorientation
angles may be computed. In the following paragraphs the errors
(Ex)c in Ex, for critical and noncritical errors of placement, are
derived for 5 possible cases for each variometer. The results are
summarized in table 18, page 203.
463. D variometer, — In all the cases, Ms is assumed in perfect
adjustment.
Case (a), figure 103. — Ma is in perfect orientation except that it
makes a small horizontal angle, p, with the magnetic meridian through
Ms.
fp=fr=~^ COS p (458)
Jn=Je=^ sin p; (459)
Ma .
tan {Ex),=^-f=^^ =i tan p (460)
^ :r COS p
r
{Ex)c^hp (for small values of p). (461)
III.]
ORIENTATION ERRORS
199
Case (b), figure 104.^A/a is in perfect orientation except that the
geometric center of Ma is displaced a small distance, y, east or west
of the magnetic meridian through Mg. Angle A0B=8.
Jv=h
Ma
(3 cos2 5-1) =
2M„
since 8 is small;
fn = h =
SMa
sin 8 cos 8-
SMg
SMa
sin 8, since 8 is small;
tan {Ej:)c-
sin 8
(EA
8^
2Ma
2 r
sm 8
(462)
(463)
(464)
(465)
Figure 103.— Looking down.
Figure 104.— Looking down.
Case (c), figure 105. — Ma is in perfect adjustment except that it is
above or below the horizontal plane through M^ by a small distance,
h. There is no component normal to Ms in a horizontal plane, hence
no deflection of M., and (Ex)c=0.
Figure 105.— Looking east.
Figure 106.— Looking east.
Case (d), figure 106. — Ma is in perfect adjustment except that it is
tilted in the magnetic meridian plane through a small vertical angle,
p'. There is no component normal to M^ in a horizontal plane,
hence no deflection of Af,, and {Ej:)c = 0.
f-
w
Figure 107.— Looking down.
Case (e), figure 107.^ — Ma is in perfect adjustment except that the
deflection distance, r, is uncertain by a small distance, Ar. Ms will
not be deflected and (£'x)c = 0.
200
MAGNETIC OBSERVATORY MANUAL
[App
464. H variometer, — In all the cases, Ms is assumed in perfect
adjustment.
Case (a), figure 108. — Ma is in perfect adjustment except that it
makes a small horizontal angle, p, with the magnetic prime vertical
through the center of Ma.
r r 2Ma
2Ma
•>3
sm p;
(466)
/p=/e=-3- sm e
M„
^ cos p;
(467)
2Ma
sm p
tan {E^)c
cos p
(468)
tan {Ex)c=2 tan p
(Ex)c^2p, since p is small.
E
S
M,
M.
W
Figure 108.— Looking down.
B
(469)
o s
M.
W
Figure 109.— Looking down.
Case (b), figure 109. — Ma is in perfect adjustment except that its
center lies to the east or west of the magnetic meridian through Mg,
by a small distance, b, such that the angle A0B=8.
, , ZMa . . .
j^=j^=-—- COS 6 sin 5
3ilf„
6, since 5 is small.
(470)
/.-/ii=^M3cos^^-l)
M ■'
=-^«(3sm2 5-l)
III.]
ORIENTATION ERRORS
—Ma . ^ . ,1
« — 3— J Since 8 is small.
^=tan iE,\
Jv
(^x)<
35
3M«
201
(471)
(472)
(473)
Case (c), figure 110. — Ma is in perfect adjustment except that it is
above or below the horizontal plane through Ms by a small distance, h.
There will be no component normal to Ms in a horizontal plane, hence
no deflection, and {Ex)c=^>
Up
I^s
w
A/.
Figure 110.— Looking east.
Down
Figure 111.— Looking south.
Case (d), figure 111. — Ma is in perfect adjustment except that it is
not level — that is, it makes a small vertical angle, p', with the hori-
zontal plane through Ms (in the prime vertical). There will be no
component normal to Ms in a horizontal plane, hence no deflection,
and {E^)c=^.
Ma
Ar
W
Figure 112.— Looking down.
Case (e), figure 112. — Ma is in perfect adjustment except that the
deflection distance, r is uncertain by a small distance, Ar. There
will be no deflection and (£'a;)c=0.
465. Z variometer, — In all the cases, Ms is assumed in perfect
adjustment.
Case (a), figure 113. — Ma is in perfect adjustment except that it
dips through a small angle, p, in the magnetic meridian plane through
Ms', that is, Ma is not level.
r r Ma .
Jn=j6=-:^ sm p
Jv=Jr-
2Ma
COS p
/»_
= tan
(E,)c-
Ma
sin p
/,
2Ma
,»,3
cos p
J tan p
210111—53-
14
(£'^)c~ Jp, since p is small.
(474)
(475)
(476)
(477)
202
MAGNETIC OBSERVATORY MANUAL
[App.
Case (b), figure 114.— il/^ is in perfect adjustment except that it is
above or below the horizontal plane through Mg by a small distance, h,
such that the angle AOB=b.
/.=/!! =
= -^ (3 cos2 5-
1)
^ — ^> smce 5 is small.
Tn=J± = ^ Sm 5 COS 8
« — 3- 8, smce 8 is small.
iE,).4j-l7
M
u
p
B
u
P
— 1^
. N_—
^s
h
^
-n
1
^5
^""^S
"^•^--J^ ^
1
wn r"
MR
r
"" t
P
r
Figure 113.— Looking east.
(478)
(479)
(480)
(481)
Figure 114.— Looking east.
Case (c), figure 115. — M^ is in perfect adjustment except that it is
displaced in the horizontal plane through Mg by a small distance, y,
such that the angle A0B=8\ There will be no component normal
to Mg in the vertical plane, hence no deflection, and {Ex)c=0.
Figure 115.— Looking down.
[Figure 116.— Looking down.
Case (d), figure 116. — Ma is in perfect adjustment except that it
makes a small horizontal angle, p', with the magnetic meridian through
Mg. There will be no component normal to Mg in a vertical plane,
hence no deflection, and (E3:)c=0.
Case (e), figure 117. — Ma is in perfect adjustment except that the
deflection distance, r, is uncertain by a small distance, Ar. There is
no component of Ma perpendicular to Mg, hence no deflection, and
{E,)c=0.
I
Ill]
ORIENTATION ERRORS
203
466. Practical applications, — It is estimated that the deflector
stops or guides can be set to an accuracy of 1 mm in 200 mm, so that
p=0.005 radians=17'. Also, it is estimated that the guides can be
placed so that the center of the deflector wiU be on the magnetic
meridian or at the proper elevation within 2 mm. At r=100 cm. 5
Up
Down
Figure HT.— Looking east.
should not exceed 0.002 radians or approximately 7\ Then referring
to table 1§ for these critical errors, assuming the recording magnet
is perfectly oriented, the apparent exorientation angles would be:
YorD: E, (apparent)=|a+ip=| X 7'+ix 17' = 19'
For^: Ex (apparent)=:36+2p-=3X7' + 2X17' = 55'
For Z: Same as D.
It is obvious that the utmost care must be taken in fixing the posi-
tions of the stops or guides for the critical adjustments, otherwise the
calculated ~QXOv\QnLi2iiion errors may be quite large even though the
recording magnets may be in excellent adjustment.
TABLE 18. — Critical adjustments of the deflector in orientation tests.
APPENDIX IV. VARIOMETER SCALE-VALUE ERRORS
467. Errors in scale values arising from maladjustment of the
deflector, — In order to simplify the analyses and to concentrate on
the physical picture, it will be assumed that the undeflected variometer
magnets are in perfect adjustment, that is, that the D magnet is in
the magnetic meridian, the H magnet is in the magnetic prime vertical,
and the Z magnet is level; that there are no appreciable errors in the
magnetic moment of the deflector and in the values of 2u (these errors
are not relevant here); that operational procedures in scale- value
observations are performed in the manner described in chapter 1 1 ;
that the deflector is an ideal magnet; and
that all other conditions fare ideal, aside
from maladjustment of the deflector.
From the definition of scale value, the
deflecting fields should be directed as fol-
lows (see fig. 118):
For H: Magnetic north or magnetic
south.
For Z: Vertically upward or downward.
For D: Magnetic east or magnetic west.
468. Maladjustment of the deflector will,
in general, change the magnitude of the
component of the field in which we are
interested. At the same time this mal-
adjustment will introduce a small field
parallel to the recording magnet which will
affect the scale value ; however, this latter
effect will cancel out on reversal of the
deflector.
469. Therefore, under the assumed con-
ditions, the maladjustment of the deflector
will affect the calculated field, fc, insofar
as this field differs from the real deflector
field, / (/ being the effective field — the field which actually produces
the deflection), as described in the last part of paragraph 467. In
computing scale values we assume ideal conditions and compute a
deflector field, fc, as in the examples which follow.
470. In this discussion, the deflection, u, is the deflection of the
recording spot on the gram caused by the deflector in one position
only, and the double deflection, 2u, is the sum of the deflections
caused by the deflector direct and reversed. 2u is always considered
positive (+).
471. Magnetic meridian and prime vertical, — The magnetic
meridian plane at a point is the vertical plane containing a magnet
freely suspended at the point and acted on only by the earth's magnetic
field. In this discussion the magnetic meridian will be taken as thei
horizontal line in the magnetic meridian plane and passing through]
the point.
204
Figure 118.— Positions of deflector rela-
tive to recording magnet in observa-
tions for scale values by magnetic
method. H deflections at top, Z in
center, D at bottom.
SCALE-VALUE ERRORS
205
472. The magnetic prime vertical plane at a point is the vertical
plane perpendicular to the magnetic meridian plane, and containing
the point. In this discussion the magnetic prime vertical will be taken
as the horizontal line perpendicular to the magnetic meridian at the
point.
473. The magnetic meridians in the variation building and absolute
building are assumed to be parallel.
474. Calculated scale value,—
-The calculated scale value is
2/c
2u
The real (true) scale value is
e_2/
The difference is
^c-^^
2u
_2_
2u
_2/
2u
(482)
(483)
(484)
The fractional error in the scale value will be
2
Sr-S
2u^' ^^ _jc-j.
and the percent error will be
Sc-S
2/c
2u
100
fc-f
Jc
xioo.
(485)
(486)
475. D scale value. — (a) Consider first the deflector correctly
oriented (fig. 119); deflector level; center on the magnetic meridian
Ma
Ms
w
Figure 119.— Looking down.
Ma
Ms
W
Figure 120.— Looking down.
through the recording magnet, M^; axis normal to the magnetic
meridian through Af,; same elevation as Mg. In figure 119, the effec-
tive field is fE=Je = fc] 61 = 90°; sin ^=1.
, Ma . ^ Ma
ic=^ sm e=^
(487)
206
MAGNETIC OBSERVATORY MANUAL
[App.
This is the value of /c in the following cases.
(b) Deflector correctly oriented except that it makes a small angle,
p, with the prime vertical. In figure 120, 6 is positive. d=90°—p.
Effective field: fE=fe.
/,=M?sin (90-p)
Ma ,
= — ^ COS p^c COS p
Jc—Je=fc(l — COS p).
The fractional error is
-^-l-cosp.
Jc
(488)
(489)
(490)
(c) Ma is correctly oriented except that its center is horizontally
displaced from the magnetic meridian through Ms by a small amount,
y (fig. 121). Effective field: /^=-/|,. Angle A0B=8; 6=90-5.
cos^ (90-5) = sin2 5.
/ii = +^"(3cos2^-l)
= +^«(3sin^5-l)
= +/c(3sin2 5-l)
Jc -f=fc 4-/,, =/c (1 + 3 sin^ 8-1)
=M3sm'8),
^''+•^'1 = 3 sin^ 6.
The fractional error is
(491)
(492)
(493)
Figure 121.— Looking down.
Down
Figure 122.— Looking south.
(d) Ma is correctly oriented except that it makes a small angle, p',
with the horizontal but remains in the magnetic prime vertical through
its center (fig. 122). Effective field: Horizontal component of the fe
field; call itfe.
fE=fe=je COS p'==—^ COS p'=/c COS p' (494)
fc — fe=fc{l—COS p').
The fractional error is
j-f.
1 — cos p' ,
(495)
(496)
IV.1
SCALE-VALUE ERRORS
207
(e) Ma is correctly oriented but its distance, r, from M, is uncertain
by a small amount, Ar (fig. 123). Effective field: /b^/o.
Af=-^ Ar by eq (160).
(497)
But
and
h
f
[Ma
Ar=r — rc, hence Aj=Je—jc
fe-fc=-%'Ar
fc-Je
Jc
3 ^
— Ar-
r
Tf- = measured distance
(498)
(499)
Up
Ms
W
Figure 123.— Looking down.
f
tA/o
A
N
"Mi ■
r^. = measured distance
Down
Figure 124.— Looking east.
(f) Ma is correctly oriented except that it is above or below the
horizontal plane through Ms by a small amount, h (fig. 124). The
effect is merely to increase the true deflection distance, r, by a small
amount, Ar. In figure 124, OA is the measured distance, Vc', OB is
the true distance, r; AB—h, the displacement above the horizontal
plane. Effective field: Je=J%'
Ar
r—r cos 6' = r(l— cos b').
As in case (e) above
Jc-Jb 3 . 3
/c
- Ar=- [r(l— cos 5')]
r r
= 3(1— cos b').
(500)
If the measured distance, r^, is OB rather than OA, then jc=ji and
Ar=:0.
476. H scale value; maladjustment of the deflector, — (a) Ma
is in perfect orientation with deflector axis in the magnetic meridian
through the center of Ms (fig. 125). Effective component producing
deflection: jN=Jr=Jc'
/c=^- (501)
This is the value of /c in the following cases.
(b) Ma is in perfect orientation except that it makes a small
horizontal angle, p, with the magnetic meridian through M, (fig. 126).
208
MAGNETIC OBSERVATORY MANUAL
[App.
Effective component producing deflection is the radial component:
Jn=Jt-
r 2Ma .
Jr=—^ COS p=jc COS p
fc—jr=jc{^—GO^p).
The fractional error in the scale value is
je
(502)
(503)
(504)
Ma
Ms
W
Figure 125.— Looking down.
Figure 126.— Looking down.
(c) Ma is in perfect orientation except that it is east or west of the
magnetic meridian through M, by a small horizontal distance, y, such
that the angle A0N=8 (fig. 127). Effective component producing
the deflection is the parallel component: fN =J\\-
Ji\=^ (3 cos^ 5-1)=^ (3 cos2 5_i)
fc-fn
U2-3 cos2 g^i)_y^(3_3 ^os^ 8)
(505)
^ 3/c(l-cos^ 8) ^ Sjc sin^ d
2 2
The fractional error is
fc—J\\_^ „• 2
sin=^ 5.
(506)
(507)
Figure 127.— Looking down.
Figure 128.— Looking east.
(d) Ma is in perfect orientation except that the deflector is above
or below the horizontal plane through Ms by a small distance, h,
such that the angle A0B = 6' (fig. 128). Effective component
producing the deflection: Jn=J\\- The result is the same as in
case (c) . The fractional error in the scale value will be
•^V^=| sin2 8\
(508)
IV.
SCALE-VALUE ERRORS
209
(e) . Ma is in perfect orientation except that it dips through a small
angle p', while its magnetic axis remains in the magnetic meridian
plane through M, (fig. 129). Effective component producing the
deflection is the radial component: Jn=Jt as in case (b). The frac-
tional error in the scale value is
/c-/.
/«
1— cos p'.
(509)
Ma
— Ar —
Tf. s measured distance
Ms
Down
Figure 129.— Looking east.
W
Figure 130. — Looking down.
(f). Ma is in perfect orientation except that the deflection distance,
r, is uncertain by a small distance H-Ar (fig. 130). Effective compo-
nent producing deflection: Jn=Jt- The fractional error in the scale
value is
J-Jr_ , 3
Ar.
(510)
477. Z scale value; maladjustment of the deflector, — In Z
scale-value deflections the deflector is used in the B position, just as in
the D deflections. In deducing scale-value errors for the Z deflections,
figures 119, 120, 121, and 123 may be taken as side elevations of the
Z recording magnet and its deflector, looking east. Figure 122, if
rotated 90° so that E means uj^, represents the condition that the
deflector makes a small angle p' with the vertical in the magnetic
prime vertical plane. Figure 124, if taken as a view looking down,
represents the Z recording magnet and its deflector when the latter
is displaced horizontally to the east by a small amount but its mag-
netic axis remaining vertical. In all cases the pertinent functions and
estimated fractional errors are the same for both I) and Z variometers.
478. Sum,m,ary of errors.— Tables 19 and 20 summarize the errors
in the scale value deflections for all variometers due to small maladjust-
ment errors of the deflector. It is assumed that with reasonable
care the deflector holder can be adjusted to the accuracy indicated
in each case. For example: It should be possible to adjust Ma
(length 20 cm) parallel to a magnetic meridian or to make it level
to an accuracy of 1° or better. For a 20-cm deflector this would be
equivalent to 3.5 mm in 200 mm, a maladjustment easily detectable
by eye. Also, it should be possible to make lateral or vertical adjust-
ments to an accuracy of 2 or 3 mm by the methods described in
paragraphs 343 to 349. As indicated in the tables, an error of 1 cm
inhory will not cause appreciable error in the deflections.
479. The controlling error is in the deflection distance, r, assuming
that the other adjustments are made to the precision indicated in the
tables. In the tables, r is taken as 100 cm. Ma as 10,000 cgs. Usually
r is much greater for scale value deflections. The scale- value error
210
MAGNETIC OBSERVATORY MANUAL
TABLE 19. — Summary of possible errors in D and Z scale values due to
maladjustment of the deflector.
DEFLECTOR
ORIENTATION
INCREMENT OF VARIABLE
ERROR FACTOR
ERRORS
Position
Figure
fc-f
% Error,
•^j^lOO
a
b
c
. d
e
f
119
120
121
122
123
124
y
"o.il"
0.3
0.15
3.0
0.15
p = l°
y—1 cm; 6=0.01 rad...
1-COSp
3 sin2 S- -
0.015
0.03
0.015
0.3
0.045
p— 1° ..._-.
1-COSp'
+:-- -
+3 (1-cosy)
r=100cm;Ar=0.1cm
h=lcm;5'=0.01rad
TABLE 20.
Summary of possible errors in H scale value due to mal-
adjustment of the deflector
DEFLECTOR
ORIENTATION
INCREMENT OF VARIABLE
ERROR
FACTOR
ERRORS
Position
Figure
fc-J
% Error,
a
b
c
d
e
f
125
126
127
128
129
130
Perfect orientation
y
'"6.3"
0.3
0.3
0.3
fi.O
p=l°
1-COSp
-|sin25
^sinU'
1-COSp'
0.015
0.015
0.015
0.015
0.3
y=l cm; 5=0.01 rad
/i=l cm;5'=0.01 rad
p'=l''
r=100 cm; Ar=0.1 cm
due to Ar may be positive or negative; all the others are always posi-
tive. Assuming the most unfavorable condition where all have the
same sign, the maximum possible error would be the sum of all the
individual errors, and in the examples given should not exceed 0.4 per-
cent. Assuming that the angular and linear displacement errors will
be the same for values of r up to 300 cm, the percent error in the deflec-
tions due to error in r would decrease as r increases.
480. Secular change in D and its effect on maladjustment of
the deflector. — After a number of years it may be necessary to
readjust the stops laterally and in azimuth because of possible large
secular change of declination. For example: Suppose r=200 cm and
the declination changes 30' over a period of years. Then the east-
west displacement of the magnetic meridian through the center of Ms,
at a distance of 2000 mm, would be ]/=2000 taa 30'=2000X0.0087 =
17.4 mm. Unless the stops were readjusted to the new magnetic
meridian, a small systematic error would be introduced into the calcu-
lated scale values as a result of this large secular change. It is obvious
that a similar error would be introduced if deflections were made at a
time when the declination differs by 30' from the mean for the day,
assuming that the deflector stops had just been adjusted to the mean
value of D for that particular day. This question of readjustment of
the stops is discussed further in paragraph 346.
APPENDIX V. CHANGE OF ORIENTATION AND SCALE VALUE OF D
VARIOMETER CAUSED BY OTHER MAGNETS OF THE MAGNETO-
GRAPH
481. Assuming that the D variometer is constructed of nonmagnetic
materials, and that there is no gross torsion in the D fiber, an isolated D
variometer will always be properly oriented. The H and Z variome-
ters, however, contain magnets whose fields can modify not only the
orientation but also the scale value of the D variometer magnet. It
is possible to compute these effects, if the magnetic moments of the
several magnets are known, with an accuracy probably greater in all
cases than that obtainable by direct measurement.
I „ Magnetic
I. ^ 1 — ^f-.^.^
h 65.4 -j- 53.6 Ul2.(>-j
Pier Top
PLAN
Horizontal Center Line ,
_D i through D magnet Ji \r ^ i I b^ ^
All Dimensions in Centimeters
Figure 131.— Plan and elevation showing relative locations of all the magnets in the regular
magnetograph at College, Alaska.
482. The example shown below applies to the sensitive magneto-
graph at College, Alaska, installed in February 1949. Figure 131
shows the relative positions in plan and elevation of all the magnets
of the magnetograph. Dimensions are in centimeters. The following
symbols are used, and the magnetic moments are as specified:
H is H variometer magnet; moment =8 cgs units;
Ht is H temperature-compensating magnet; moment =85 cgs
units;
211
212
MAGNETIC OBSERVATORY MANUAL
[App.
Hs is H sensitivity magnet; moment =83 cgs units;
Z \s> Z variometer magnet; moment=215 cgs units;
Zt is Z temperature-compensating magnet; moment =450 cgs
units ;
/„=that component of the field of a magnet that is parallel to the
axis of the magnet;
/x=that component of the field of a magnet that is perpendicular
to the axis of the magnet;
/b= eastward horizontal component of the field of a magnet;
/ivr= north ward horizontal component of the field of a magnet.
483. Values of /^ and Jn at the D recording magnet will be computed
separately for each of the five magnets; the resultant field will be the
algebraic sum of the individual components. Magnitudes of /g and
Jn will be computed by means of equations (27) and (28) ; signs may be
Figure 132.— Position of the H-Tariometer recording magnet relative to the D magnet.
determined by examining the relative physical positions of the D
magnet and of the magnet whose field is being computed, as illustrated
in figures 132, 133, 134, 135, and 136. The figures are not drawn to
scale. Slide-rule computations are used. Distribution coefficients
(see par. 30) are neglected because the other magnets are far enough
away from the D magnet that the effects of distribution will be small
enough to ignore.
484. H magnet, — From figures 131 and 132, it is evident that the
H magnet is 5.8 cm south and 53.6 cm east of the D magnet. Then,
from equation (27),
/^=/ii=^(3cos2^-l)
(511)
_ 8X10^ r./ -53.6 Y -.1
(53.8)3 L \ 53.8 ) J
v.]
INTERACTION OF VARIOMETERS
213
and from equation (28),
-5.14(2.978-1)
= 107 (eastward);
M
Jn=J'x=-3- (3 sin d COS 0)
= —27 (27 southward).
(512)
Figure 133. — Position of the H tempsrature-compensating magnet relative to the D magnet.
485. H temperature-compensating magnet, — From figures 131
and 133, the H temperature-compensating magnet, Ht, is 7.3 cm
north and 53.6 cm east of D. Then from equation (28),
/b=/i.=^- (3 sin 6 cos 6)
85X10^^^3^^53.6^, 7.3
(513)
and from equation (27),
(54.1)3 54.r^54.1
= 53.7X0.401
= 227 (westward);
Jn=A=^ (3 cos' d-1)
(514)
==53.7X (-0.945)
= 5l7 (northward).
214
MAGNETIC OBSERVATORY MANUAL
[App.
486. H sensitivity magnet, — From figures 131 and 134, the H
sensitivity magnet, Hg, is 5.8 cm south and 65.6 cm east of D. Then
from equation (27),
^^=A=^ (3COS20-1)
(515)
_ 83X10^ r / -65.6 Y 1
(65.8)3 L V 65.8 / J
= 29.1X1.982
= 587 (eastward);
and from equation (28) ,
/iv=/jL=-^ (3 sin d COS d)
= 29.1X3Xg^X-g^-g-
= — 87 (87 southward).
(516)
Figure 134.— Position of the H sensitivity magnet relative to the D magnet.
487. Z magnet, — From figures 131 and 135, the Z variometer
magnet, Z, is seen to be 11.5 cm north, 65.4 cm west, and 6.0 cm higher
than D. The east component, which will affect D, is the horizontal
projection of /x-
:fE=J± cos a
=—r (3 sin 6 cos 0) cos a
(517)
v.]
INTERACTION OF VARIOMETERS
215
215X10^
65.7_-11.5_65.4
X3X.^„X
X
(66.7)^ ^'^^66. 7^' 66.7 '65.7
= — 377 (377 westward);
and from equation (27),
M
/a^=/ii=7^(3 cos^^-l)
^ 215X10^ r. / -11.5 Y -]
(66.7)3 \jy ggy ) A J
= — 667 (667 southward).
(518)
Figure 135.— Position of the Z- variometer recording magnet
relative to the D magnet.
488. Z temperature- compensating magnet, — From figures 131
and 136, the Z temperature-compensating magnet, Z,, is 11.5 cm north,
65.4 cm west, and 7.7 cm lower than the D magnet. Again, as in the
case of the Z magnet, the elevations of Zt and D are different. Je is
the eastward component of/j., and is therefore equal to/j. cos /3 (fig.
136); similarly /^^ is the northward component of/j., and is equal to/x
sin jS.
Je— — 3- sin 6 cos 6 cos /3 (519)
3X450X10^ 66.4 7.7 65.4
66.8 66.8 66.4
(66.8)=
= 52X0.985
= 517 (eastward);
216
and
MAGNETIC OBSERVATORY MANUAL
Jn= — 3- sm B cos B sm /3
(520)
= 52X
11.5
66.4
= 97 (southward).
Figure 136.— Position of the Z temperature-compensating magnet
relative to the D magnet.
489. The following table, then, summarizes the resultant field at
the D variometer of all the other magnets in the magnetograph, west-
ward and southward components being given a negative sign:
Field component
H
H.
H.
z
Zt
Total
In
+ 10
-2
-22
+ 51
+ 58
-8
-37
-66
+ 51
-9
+ 60
-34
490. When Je is positive, therefore directed eastward, it causes
the N end of the D magnet to lie in the northeast quadrant; the ex-
orientation angle (par. 338) is arc tan =16 minutes of arc. (Hori-
zontal component of the earth's field at College is I26OO7.)
491. When /at is negative, therefore directed southward, it causes
a change of scale value of the D variometer according to equation
(122), where the (7 of the equation is equal XoJn derived above.
APPENDIX VI. NOMOGRAMS
492. The nomograms given in tliis appendix will be found useful
in the preliminary studies associated with the establishment of a
magnetograph. Beneath each illustration is a brief note explaining
how it may be used and referring to the pertinent text paragraphs.
Figure 137. — Field of a bar magnet at a point on[its magnetic axis produced.
The above nomogram represents graphic solutions of the equation f=— — ; — -, in which 3/ is the mag-
netic moment and/the field in gammas at a distance r along the magnetic axis produced. Example: Given
M=1000 and r=lOO cm. Then a straight line from iV/=1000 on the M scale through r = 100 on the r scale
intersects /= 200 on the /scale. The field is 20O7, the required field. Conversely, if it is desired to know
the magnetic moment required to produce a field of 2OO7 at a distance of 100 cm along the magnetic axis,
simply reverse the process (see par. 16).
210111—53-
217
218
MAGNETIC OBSEEVATORY MANUAL
[App.
figure 138.
-Change in the field of a bar magnet, along its axis produced, with change of distance; magnetic
moment known.^
The above nomogram represents graphic solutions of the equation — A/=
2 MX 105
3^,=6MX10^^^_
t3 r ri
Examples: Given JVf=140 cgs, r= 11.62 cm, and Ar=l mm. A straight line from Af=140 on the M scale
through ?■= 11.62 on the r scale intersects the —A/ scale at 46O7. That is, when r is increased from 11.62 cm
to 11.72 cm the field will be smaller at r=10.1 by 46O7. Likewise if M= 10,000 cgs, r=200 cm, and Ar = l mm,
then A/= — 0.887, approximately (see par. 16). In this nomogram Ar is fixed at 0.1 cm.
VI,
NOMOGRAMS
219
Figure"139.— Change in the field of a
ir magnet with change in distance along its magnetic axis produced;
field at a distance r, known.
I The above nomogram represents graphic solutions of the equation
in which / is the field at a distance r from the center of the magnet, Ar is the increment in r, and a; is mo
corresponding increment in /. Example: Given: /=10007 at a distance r = 10 cm. Required A/ when
Ar = l mm. A straight line from /= 1000 on the /scale through r = 10 on the r scale intersects the —A/ scale
at 30. This means that if a field is IOOO7 at a distance of 10 cm from the magnet and along its magnetic
axis, the field will decrease approximately 30 gammas between the points r = 10.0 cm and r = 10.1 cm. (See
par. 16). In this nomogram Ar is fixed at 0.1 cm.
220
MAGNETIC OBSERVATORY MANUAL
[App.
Figure 140.— The D Nomogram.
An alignment chart showing the relation between the product of the horizontal intenrfity and the mag-
netic moment of the recording magnet, the dimensions of the quartz fiber suspension, and the min-ate scale
value; based on equations 124-128, with optical lever 2R taken as 3438 mm. See paragraph 215 for complete
explanation and examples.
Figure 141.— The H Nomogram.
An alignment chart showing the relation between the magnetic moment of the recording magnet, the
dimensions of the quartz fiber suspension, and the H scale value. See paragraphs 251, 252, and 253, for addi-
tional details and examples.
222
MAGNETIC OBSERVATORY MANUAL
[App.
Figure 142. — Nomogram for conversion of D scale values.
This chart represents graphic solutions of equations (301) and (302). For example: Given a D scale value
approximately equal to 1.00 minute per mm at a place where H= 18300t. Required the gamma scale value,
Sdt, of the D variometer. A straight line from So' = 1.00 on the So' scale to H= 18300^ on the H scale
intersects the Sd^ scale at 5.3 vv^hich is the required gamma scale value, approximately. Likewise the
minute scale value may be evaluated if the gamma soale value is given (see par. 323).
VI.]
NOMOr.RAMS
223
Figure 143.
-Nomogram showing spurious effects on recorded magnetic declination due to changes
the horizontal intensity (see par. 331).
This chart represents graphic solutions of the equation, ADs-
AH
H
tan Ex, in which H is the horizontal
intensity in gammas; AH is the change in Hin gammas, and Ex is the exmeridian angle of the D recording
magnet. Example: Given, H=183007, AH=1007; and £'x = 18.6 minutes. Then from the above equation,
AD, (in radians) =--,-—X0.00532= 0.00029 radians=0.1'. The straight line from 0.00546 on the -^^ scale
through 18.6' on the Ex scale intersects the AD scale at 0.1', the spurious change in recorded declination
due to A//.
224
MAGNETIC OBSERVATORY MANUAL
[App.
H^D
Figure 144. — Nomogram showing'spurious effects on recorded horizontal intensity due to changes
in magnetic declination (see par. 331).
This nomogram represents graphic solutions of the equation
AH. = HAD tan Ex
in which AHs is the recorded change in H, in gammas, due to a small declination change, AD, in radians,
when the ex-prime vertical angle is Ex. Example: Given, H=10fi00y; Ai)=10' =0.00291 radians; and Ex=
10°=600'. Then AHs = 10,000X0.00291 tan 10°=5.1'y, the spurious effect. The straight line from the HAD=
29.1 on the appropriate HAD scale through ^i = 10° = 600' on the Ex scale intersects the Alls scale at 5.I7.
VI.]
NOMOGRAMS
225
150
a; =1 30
< m
6
= 1 3
■i2^r
Figure 145. — Nomogram showing spurious effects on recorded vertical intensity due to changes in the
horizontal intensity (see par. 331).
This nomogram represents graphic solutions of the equation
AZs = AHtanEx
in which AH is a small increment in H: Ex is the ex-level angle of the Z recording magnet. The equation
applies ;to a Z recording magnet operating in the magnetic meridian. Example; Given A/7=1007 and
Ex=&0'. Then from the above equation, AZs = lMy, the spurious effect. The straight line from IOO7 on
the A// scale through 60' on the Ex scale intpi^iects the AZ. scale at I.847.
226
MAGNETIC OBSERVATORY MANUAL
[App.
--.000000 5
Figure 146.— Nomogram for distribution factors ( l-\ — ^-f— ^ ] •
P
This nomogram represents graphic solutions of the terms -^ and — for different values of the deflection
distance r, for the conditions that P and Q are known. The coefficients P and may be calculated from
the equations in table 2 or interpolated from the values in tables 21a, 21b, and 21c. Example: Given, for
P
position A, P=10, 0=5000, and r=40 cm. Required: the factor 1+ —„ + -.• The straight line from 10 on
the P scale through 40 on the r scale intersects the - scale at 0.0064. The straight line from 5000 on the
scale through 40 on the r scale intersects the -^ scale at 0.002. Then the distribution factor is 1.0000+0.0064
+0.0020 = 1.0084.
VI.
NOMOGRAMS
227
TABLE 21a.
Pa
Approximate distribution coefficients for position A.
Px = 2ia2-3i.2=0.32 LJ-QA8 L.2
/o=0.4 La Z.=0.4 L,
\ La
L. \
1
2
4
'
8
10
12
14
16
81.9
18
104
20
128
0.00
+0.32
1.28
5.12
11.5
20.5
32.0
46.1
62.7
1
-0.48
-0.16
+0.80
4.64
11.0
20.0
31.5
45.6
62.2
81.4
103
128
2
-1.92
-1.60
-0.64
+3.20
9.60
18.6
30.1
44.2
60.8
80.0
102
126
4
-7.68
-7.36
-6.40
-2.f6
+3.84
12.8
24.3
38.4
55.0
74.2
96.0
120
6
-17.3
-17.0
-16.0
-12.2
-5.76
+3.20
14.7
28.8
45.4
64.6
86.4
HI
8
-30.7
-30.4
-29.4
-25.6
-19.2
-10.2
+1.28
+15.4
32.0
51.2
73.0
97.3
10
-48.0
-47.7
-46.7
-42.9
-36.5
-27.5
-16.0
-1.92
+14.7
33.9
55.7
80.0
12
-69.1
-68.8
-67.8
-64.0
-57.6
-48.6
-37.1
-23.0
-6.40
12.8
34.6
58.9
QA=3h*-
Qa
la = OALa
-0.384 La^ Z.2+0.144 L,*
l,=OA Ls
V-
1
2
4
6
8
10
12
14
16
18
20
'■\
19.7
O.CO
+0.08
+1.23
99.5
315
768
1.590
2.950
5,030
8,060
12, 300
1
0.14
-0.16
-0.16
+13.7
85.8
290
730
1,540
2.880
4.940
7,940
12, 100
2
2.3C
+0.84
-2.61
-2.61
+46.5
+219
617
1,370
2.650
4,640
7,570
11, 700
4
36.9
30.8
+13.5
-41.8
-84.8
-41.8
+190
+745
1,780
3,500
6,110
9.870
6
187
173
133
-14.9
-212
-384
-428
-212
+427
+1,680
3,770
6,950
8
~ 590
565
493
+216
-195
-668
-1,100
-1,360
-1.280
-668
+689
+3, 050
10
1.440
1.400
1,290
845
+157
-703
-1,630
-2.500
-3, 140
-3, 360
-2,940
-1,630
12
2,990
2,930
2,770
2,120
1,090
-238
-1, 780
-3.380
-4, 900
-6, 140
-6.870
-6,840
TABLE 21b. — Approximate distribution coefficients for position B,
Pb
Pb=-^ ?a2+6i2=-0.24 i„2_(_o.96 L,2
Za = 0.4 La
ls=OA L,
\ La
\
1
2
4
6
8
IC
12
14
16
18
20
L. \
-3.84
-24.0
-34.6
0.00
-0.24
-0.96
-8.64
-15.4
-47.0
-61.4
-77.8
-96.0
1
0.96
+0.72
+0.00
-2.88
-7.68
-14.4
-23. C
-33.6
-46.1
-60.5
-76.8
-95.0
2
3.84
3.60
2.88
0.00
-4.80
-11.5
-20.2
-30.7
-43.2
-57.6
-73.9
-92.2
4
15.4
15.1
14.4
+11.5
+6.72
0.00
-8. 64
-19.2
-31.7
-46.1
-62.4
-80.6
6
34.6
34.3
33.6
30.7
25.9
+19.2
+10.6
0.00
-12.5
-26.9
-43.2
-61.4
8
61.4
61.2
60.5
57.6
52.8
46.1
37.4
+26.9
+14.4
0.00
-16.3
-34.6
10
96.0
95.8
95. C
92.2
87.4
80.6
72.0
61.4
49.0
+34.6
+18.2
0.00
12
138
138
137
134
130
123
114
104
91.2
76.8
60.5
+42.2
Qb
pB=yZJ-^Z„2Z.2+15Z.<
0.048 La^
0.4 La h-
-0.576 L<,2 L,2_|.o.384 L,*
0.4 L»
\^i>a
1
2
4
6
8
10
12
14
16
18
20
+0.77
0.00
+0. 05
12.3
62.2
197
480
995
1.840
3.150
5,040
7.680
1
0.38
-0.14
-1.15
+3.46
+41.9
160
423
913
1, 730
3, 000
4,850
7,450
2
6.14
+3.89
-2.30
-18.4
-14.6
+55.3
+256
+670
1,400
2. 560
4.300
6,760
4
98.3
89.1
+62.2
-36.9
-171
-295
-343
-233 +136
+885
+2, 150
+4. 090
6
498
477
415
+178
-187
-633
-1,100
-1.490-1,720-1.670
-1,180
-117
8
1.570
1.540
1,430
995
+308
-590
-1,630
-2.740,-3.810-4.720
-5. 330
-5.490
10
3.840
3.780
3,610
2,930
1,830
+350
-1,440 -3,460[-5,610 -7, 760
-9.780-11,500
12
7,960
7,880
7,630
6,650
5,040
2,850
+148-2,990-6.450-10,100
-13,900-17,500
1
228
MAGNETIC OBSERVATORY MANUAL
TABLE 21c. — Approximate distribution coefficients for position C.
Pc
Pc=-\ {laHU) = -QM {LJ+L,i)
Zo = 0.4 La
ls=OA Ls
\
\ i^a
L, \
1
2
4
6
8
10
12
14
16
18
20
0.00
-0.24
-0.96
-3.84
-8.64
-15.4
-24.0
-34.6
-47.0
-61.4
-77.8
-96.0
1
-0.24
-0.48
-1.20
-4.08
-8.88
-15.6
-24.2
-34.8
-47.3
-61.7
-78.0
-96.2
2
-0.96
-1.20
-1.92
-4.80
-9.60
-16.3
-25.0
-35.5
-48.0
-62.4
-78.7
-97.0
4
-3.84
-4.08
-4.80
-7.68
-12.5
-19.2
-27.8
-38.4
-50.9
-65.3
-81.6
-99.8
6
-8.64
-8.88
-9.60
-12.5
-17.3
-24.0
-32.6
-43.2
-55.7
-70.1
-86.4
-105
8
-15.4
-15.6
-16.3
-19.2
-24.0
-30.7
-39.4
-49.9
-62.4
-76.8
-93.1
-111
10
-24.0
-24.2
-25.0
-27.8
-32.6
-39.4
-48.0
-58.6
-71.0
-85.4
-102
-120
12
1
-34.6
-34.8
-35.5
-38.4
-43.2
-49.9
-58.6
-69.1
-81.6
-96.0
-112
-131
Qc
Qc=Y (^a2+Z.2)2=0.048 (LJ+Ls'^)i
la = OA La ls = OA Ls
\l.
\
1
2
4
6
8
10
12
14
16
18
20
Ls \
0.00
0.05
0.77
12.3
62.2
197
480
995
1,840
3,150
5,040
7,680
1
0.05
0.19
1.20
13.9
65.7
203
490
1,010
1,860
3,170
5,070
7,720
2
0.77
1.20
3.07
19.2
76.8
222
519
1,050
1,920
3,240
5,160
7,830
4
12.3
13.9
19.2
49.2
130
307
646
1,230
2,160
3,550
5,550
8,S10
6
62.2
65.7
76.8
130
249
480
888
1,560
2,580
4,090
6,220
9.120
8
197
203
222
307
480
786
1,290
2,080
3,240
4.920
7,230
10, 300
10
480
490
519
646
888
1,290
1,920
2,860
4,210
6,080
8,630
12, 000
12
995
1,010
1,050
1.230
1,560
2,080
2,860
3,980
5,550
7,680
10, 500
14, 200
BIBLIOGRAPHY
1. S. J. Barnett, A sine galvanometer for determining in absolute measure the
horizontal intensity of the earth's field, Department of Terrestrial Magnetism,
Carnegie Institution of Washington, Publication No. 175, vol. IV, pp. 373-394,
Baltimore, 1921.
2. S. Chapman and J. Bartels, Geomagnetism, vols. I and II, Oxford, 1951.
3. G. Hartnell, Distribution Coefficients of Magnets, Special Publication 157,
pp. 9-10, U. S. Coast and Geodetic Survey, Washington, 1930.
4. D. L. Hazard, Directions for Magnetic Measurements, Serial 166, 3rd ed.,
corrected 1947, U. S. Coast and Geodetic Survey, Washington, 1947.
5. C. A. Heiland and W. E. Pugh, The American Institute of Mining and Metal-
lurgical Engineers, Technical Publication No. 483, New York, 1932.
6. H. H. Howe, On the theory of the uniular variometer, Terr. Mag., 42,
29-42, Baltimore, 1937.
7. J. W. Joyce, Manual on Geophysical Prospecting with the Magnetometer,
U. S. Department of Interior, Bureau of Mines, Houston, 1937.
8. D. la Cour, La Balance de Godhavn, Danske Meteorologiske Institut,
Communications Magnetiques, etc.. Publication No. 8, Copenhagen, 1930.
9. Le Quartz- Magnetometre QHM (Quartz Horizontal Force Mag-
netometer), Danske Met. Inst., Comm. Mag., etc., Pub. No. 15, Copenhagen,
1936.
10. The Magnetometric Zero Balance, the BMZ, Danske Met. Inst.,
Comm. Mag., etc.. Pub. No. 19, Copenhagen, 1942.
11. D. la Cour and V. Laursen, Le Variometre de Copenhague, Danske Met.
Inst., Comm. Mag., etc., Pub. No. 11, Copenhagen, 1930.
12. J. Lamont, Handbuch des Erdmagnetismus, Berlin, 1849.
13. A. K. Ludy and H. H. Howe, Magnetism of the Earth, Serial 663, U. S.
Coast and Geodetic Survey, Washington, 1945.
14. Smithsonian Physical Tables, Geomagnetism, Tables 495-511, 9th ed. (in
press, 1952).
INDEX
a factor, 87, 95, 97, 128, 166
— significance of, 96
— determination of, 159
— at C. & G. S, observatories, 97
Aberration, spherical, 67
Absolute control, importance of, 174
Absolute observations, 174
Absolute observatory, 21
Activity, magnetic, 181, 182
Alignment charts (see Nomograms')
Anomaly; tolerances in selecting site, 21
Approximations, 183
Aurora borealis, 20
Autocollimator (Gaussian ocular), 62
Auxiliary spots (la Cour), 66
Axis, magnetic (bar magnet), 1
— optical, 62
— of centered dipole, 16
Azimuth of Z magnet; scale value, 105
— true. 21
b factor; if scale value, 87
Balance, analytical, 104
Bar magnet; components of field, 8
Base-line mirror, 66
Base-line mirror, D, 74
Base line, scale value at, 165
Base-line value, H, 30
— drift of, 174
— scaling for, 178
Bench, orientation, 21, 141
Bimetallic strip, 120, 121, 123, 164
— on la Cour variometer, 164
Bifilar suspension, H variometer, 84
BMZ. 57
C^ FIET D*
D variometer, 74, 77, 80
H variometer, 95
elTective compensation, H variometer, 116
computation, H variometer, 156
Z variometer, 101
evaluation of; Z variometer, 159
estimation of; for if and Z variometers, 117
C figures, 181 ^ ^^
C (instrumental constant of magnetometer), 27
— adjustment of, 55
— computation of, 58
Centered dipole, 16
Charts, isomagnetic, 16
Chronometer, 21
— rate of, 27
Clock, time marking, 21 , . ^.
Coefficient of thermal expansion, magnet and mertia
weight, 29
Coil factor, induction tests, 50
Collimation, 62
Comparison observations, 148
Comparison of fields, bar magnet and earth, 18
Comparisons, summary of, 58, 60, 61
Compensated variometer, H, 167
Compensating field, variometers, 114, 116
Compensator distance, evaluation, 118
Components of field:
of a bar magnet, 8, 17
for small angles, 10
Constants:
ofmagnetograph, 148
of magnetometer; summary, 55, 56
Control, absolute; importance of, 174
Control magnet:
D variometer, 81
H variometer, 86, 90
Z variometer, 104
Conversion factors, 30
Conversion to absolute units, 165
Corrections:
index, 25
instrumental, 25
Corrections to //observations, 59
Corrector, D, 152
Coulomb's equation for k', 88
Couples:
in H variometer, 84, 85
in D scale value equation, 78
Coupler, in variometers, 73
Cylindrical lens, thickness,. 75
D nomogram, use of, 81
D scale-value, 74, 82, 83
D scale value equation, 78
D spot, 74
D variometer:
C field of, 74, 77, 80
control magnet, 80
depth of focus of, 83
east field at, 77
effect of cylindrical lens, 75
effective recording distance, 75
evaluation of recording distance, 82
field factor, 74
gamma-scale value, 74
la Cour, 65
lens, thickness of, 75
lens, specifications, 83
line of no torsion, 76
loM^-sensitivity, 80
magnetic moment of recording magnet, 82
measured distance, 75, 83
nomogram, 81
optical axis of D lens, 74
optical lever, 75
optical scale value, 75, 77
orientation formulas, 136
recording magnet, 74
stray fields, 77
torsion constant, 74, 81
torsion factor, 76
torsion in fiber, 75
torsion tests, 76
Damping, 29, 74
— i/ recording magnet, 84, 159
— test, D, 150
Darkroom, 21
Declination, magnetic, 21
— standard value, 128
Deflection angle, u, 27
Deflection distances, 26, 27
— for orientation tests, 144
Deflection, double, in scale value, 128
— exmeridian, 135
— orientation, 144
Deflections, positions of magnets in, 5-8
— for //scale value, 126, 158
— for Z scale value, 161
Deflector orientation, 197
Deflector positions, effect on distributioncoelTicients, 12
Depth of focus, D variometer, 83
Developing processes, photographic, 69
Diaphragms, 67
Dip, magnetic (inclination), 16
Dipole, centered: representing earth's field, 16
Directions for performing operatiors {see supple-
ment, page 232).
Distribution coefficients, 12, 13, 15, 26, 27, 227, 228
— formulas, 13
— special cases, 14
— corrections for,* 5
Distribution effects, // scale value, 91
Distribution factors, 12
Disturbance, irregular, 20
— local, 21
Diurnal variation, 19
D>Tie, 1
Earth inductor, 21
Earth's dipole field, 16
Earth's field:
changes in, 19
east component, 16
at various latitudes, 17
229
230
INDEX
Earth's magnetic moment, 17
East field; at D variometer, 77, 151, 152
— at iJ variometer, 84, 86, 87, 88
Elements, magnetic, 16
Equivalent field; H variometer, 86, 87
Errors in orientation tests, 197-202
Errors in scale- value observations, 204
Exmeridian deflections, 135
Exorientation angles, evaluation, 84, 85, 145
Fiber, torsionless, 77
Field, magnetic:
at a point; distribution considered, 14
around a bar magnet, 2, 17, 18
C, evaluation of, 159
dipole, 16
equivalent, 86
estimation from nomogr'am, 5
factor {D variometer), 74
in terms of magnetic moment, 5
of magnet, direction of, 11
p, at // variometer, 87, 158
resultant, 5, 8
strength, 1
uniform, 3
Focal length of lens, 62
Force, magnetizing, 1
Gamma scale value, D, 74, 79
Gauss, 1
Gaussian ocular, 62, 63
Geomagnetic latitude, recommended scale^values for,
134
Geomagnetic poles, 17
Ghost images, 64
H magnet:
magnetic moment, 84, 87
effect on D variometer, 212
H nomogram ; for H scale value, 93
i7 sensitivity control magnet; effect on D variometer,
214
— reversal of, 92
H spot, adjustment of, by temperature magnet, 158
H temperature compensating magnet; effect on D
variometer, 213
H variometer, 84
— orientation formulas, 85, 139
— scale value, 85
Half period, in oscillations, 27, 29
Heat treatment of magnets, 33, 38
Helmholtz coil; external field, 134
Helmholtz-Gaugain coil, 132
Horizontal intensitv, de'inition of, 16
Hourly values, publication of, 176
— scaling, 177, 178
Housing, nonmagnetic, 84
Image, character of, 67
Impurities, magnetic, in instruments, 25
Inclination (magnetic dip), definition of, 16
Index corrections, 25
Induction coefficient, 42
— Lamont's method, 44
— computations, 47
— Nelson's method, 48, 52, 54
Induction factor, 26, 27, 42
Inductor, earth, 21
Inertia cylinder, 71
Inertia weight, 29
Installation of magnetograoh, report on, 162, 163
Instrumental corrections, 25
Instruments, quality of, 25
Intensity, horizontal, definition of, 16
Intensity, magnetic, 1
Intensity, vertical, definition of, 16
Intensity, total, definition of, 16
Intercomparison observations, 57
International standards, 57
Invar; coefficient of expansion, 110
Inverse square law, 1
Isomagnetic charts, 16
k, evaluation and dimensions of, 87-89
K indices, 181
Knife edge; Z recording magnet, 99
— care of, 161
La C our fibers, 70, 163
— magnetograph, 70, 163
— variometers, optical compensation, 120
Latitude adjustment; Z variometer, 107
Latitude, geomagnetic and magnetic, 18
Lens, cylindrical, 65, 75
Lens, D variometer; specifications, 75, 83
— focal length, 62
Lenses, centering of, 62, 67
Lever, optical, 63, 64
Line of force, equation, 1, 18
Local disturbance, 21
Low-sensitivity variometers, 66, 80, 162
— orientation tests, 163
— scale values, 129
Magnet, length of, 4
— poles of, 1
— axis of, 1, 74, 85
Magnetic activity data, 181
Magnetic compensation, 116
Magnetic elements, 16
Magnetic field strength, 1
Magnetic impurities, 25
Magnetic meridian, 16, 74, 204, 205
— D variometer, 151
— in variation room; establishing, 142
Magnetic moment, 3, 27
— D magnet, 82
— equivalent for a coil, 50
—H" magnet, 84, 88
— of the earth, 17
— temperature coefficient of. 195
Magnetic observatory, buildings and equipment of, 21
Magnetic prime vertical, 85, 87, 204, 205
— at /-/ variometer, 87, 90, 95
Magnetic poles of the earth, 17
Magnetic standards, 57
Magnetic storms, 20
Magnetic temperature compensation, 96, 113, 117
Magnetism tester, 25
Magnetizing field, 42
Magnetizing force, 1
Magnetograms, 67, 69, 176
Magnetograph, 21, 23
— constants, 148
— installation, 148
— la Cour, 5
— optics, 63
— visual, 165
Magnetometer, 21
— equations, 26
— Quartz Horizontal (QHM), 21
— constants, 55, 56
Magnetometric Zero Balance (BMZ), 57
Magnets, fields of, 8-15
— bar; forces between, 1
Maladjustment:
of deflector in orientation tests, 198-202
of deflector in scale-value observations, 205-210
of deflector due to secular change in D, 210
of recording magnets, 135
Meridian, magnetic, 16
— true, 21
Milligauss, 2
Minute scale value, 74
Mirror-adjusting apparatus, 150, 154
Mirrors:
base line, 66
front surfaced, 64
planoconvex, 64
time-flashing, 67
triple-face (three-face\65, 150
Moment of inertia, 26, 27, 28, 30, 31
— directions for observing, 31
— corrections for change of temperature, 187
— of cylinder, 28, 89
— of weight, 29, 75
Moment, magnetic {see magnetic moment),
Monochromatic light, 62
Multiple reflections, 64
Nomograms:
I) variometer, 81
H" variometer, 93, 94
for estimating field of a magnet, 5
Observations, absolute, 174
Observatory, absolute, 21
Observatory, magnetic, 21
Observatory site, selection of, 21
Oersted, 1, 87
INDEX
2:u
Optical axis, 62, 74
Optical compensation, 120, 164
— H variometer, 124
— la Cour variometer, 120, 123
Optical lever, 63, 64, 75
— auxiliary, 162
— double, 134
Optical scale value, 63, 75
— D variometer, 74, 75,77
— 27 variometer, 84, 87
Optical system; 7/ variometer, 84
— magnctograph, 63, 148
Orientation, 135
Orientation bench, 21, 141
Orientation deflections, 144
Orientation errors, 197
— critical adjustment of deflector, 203
Orientation formulas, 136, 139
Orientation of variometer magnets, 135, 146
Orientation tests, 141, 145
— effects of maladjustment of deflector, 141, 144, 161,
198-202
— evaluation of Ex angles, 145
— low-sensitivity variometers, 163
Oscillations, 28, 29, 33
Oscillations, corrections to:
for all effects, 184
for change in H, 191
for change in moment of inertia, 188, 189
for change in magnetic moment, 186
for torsion factor, 190
for induction, 192
for rate of chronometer, 185
p field at H variometer, 85, 87
— computation of, 158
p field at Z variometer, 104
Paper, recording, 67
Parallax 67
Parallax'te^ts, 156, 161, 163, 177, 181
Parallel component; field of bar magnet, 9
Parallel rays, 62
Period, torsion pendulum, 75
Perpendicular component; field of a bar magnet, 10
Photographic registration, 23, 62
Piers, 21
Pivots and knife edges, care of, 161
Poises, Z recording magnet, 161
Pole, unit, 1
Poles, earth's magnetic and geomagnetic, 17
Poles of a magnet, 1
Positions of magnets, 5-8. 12
Potentiometer, precision, 57
Prime vertical, magnetic, 85, 87, 204, 205
Prisms, 65, 163
Processing of records, 165, 174
q coefficients, 115, 116
QHM (Quartz Horizontal Magnetometer), 21, 57
Quality of instruments, 25
Quartz fibers: 70
classification, 71, 72
dimensions, 70. 75, 88
galvanometer, 70
installation, 72, 73
la Cour type, 70
measurement, 71
modulus of rigidity, 70
period of oscillation, 72, 75
procurement, 70
selection, 72, 154, 162
torsion constant, 70, 72
Quartz Horizontal Magnetometer (QHM), 21, 57
Radial component; field of a bar magnet, 8, 10
Radian scale value; //variometer, 86
Rate of chronometer, 27, 185
Recorder, photographic, 23, 67
Recording distance, 82, 84
Recording paper, 67
Recording magnet, D, 74, 82
Recording magnet, //, 87, 88, 90, 95
— damping, 159
Recording magnet, Z:
azimuth and scale value, 100, 101
balancing poise— center of gravity, 99, 100
knife edge, 99
la Cour, 163
magnetic axis, 99, 104
scale value, 101
sensitivity poise, 99, 104
supports, 99
Recording magnets, adjustment, 135, 145
Registration, i)hotographic, 62, 67
Report of installation of magnctograph, Ifi", I 13
Reserve distances, 155
Reserve spots, 65, 164
Resistance, standard, 57
Resolution of fields, resultant field, 5
Resultant field, D, 77
Resultant field, //, 87
Rigidity modulus of quartz, 75
Routine duties, 105
Scale value:
a factor, significance of, 96
a factor tests, 128, 159
adjustment by control magnet, 90
bimetallic strip, 121
calculated, 205
computations, 129
correlated with geomagnetic latitude, 134
double deflections, 128
electromagnetic methods, 132
errors due to maladjusted deflector, 204
Helmholtz-Gaugain coil, 132
low-sensitivity variometers, 129, 162, 163
optical, 63, 75
recommended, 134
scale in magnet, 26
scale in telescope, 26
scahng of deflections, 177
sensitivity-control magnet, 90
shelves, 21
temperature coefficient, 112
Scale value, D variometer:
adjustment, 83
computation, 154
equations, 78, 82
gamma, 74, 79
gamma and minute, 129
minute, 74, 231
optical scale value, 75, 77
Scale value, H variometer:
a factor, 87, 95
adjustment by control magnet, 91, 92
average, 97
b factor, 87
base-line, 97, 128
cgs tmits per radian, 85, 86
control magnet, 157
correction for change in D, 97, 128
deflections, determination, 126, 158
dimensions of k , 87
distribution effects, 91
equation, 71, 128
from constants, 90
gammas per mm, 86
graph, 158
graphic solutions, 93
optical, 86, 87
reversal of control magnet, 92
selection of, 90
use of nomogram, 94
variation with/a, the east field, 91
variation with k, 91
variation with distance of sensitivity magnet, 93
variation with ordinate, 97, 128
with control magnet, 93
with two control magnets, 92
Scale value, Z variometer:
azimuth of recording magnet, 105, 106, 107
center of gravity of recording magnet, 103
deflections, 101
equation, 101, 128
example, 106
effect of //, 105
latitude and temperature compensation. 108
A'' end south, 106
parallel fields, 101
p field, 104
sensitivity poise. 99, 103, 106
Scaling, base line, 177, 178
— during a storm, 178
— hourly values, 177, 178
Scaling base line,
— low sensitivity magnetograms, 180
— ordinates, 33, 177
— scale value deflections, 178
Secular change. 20
Sensitivity, bimetallic strip, 121
— control magnet. 90, 91
232
INDEX
Sensitivity— Continued
— double; D variometer, 162
— Z recording magnet, 104
Sensitivity magnet, adjustment of, H, 157
Sensitivity magnets, two, 92
Sensitivity magnets, la Cour Z variometer, 164
Shelves, scale value, 21
Shrinkage, 180
Sine galvanometer, 57
Site, observatory, selection of, 21
Solar flare effects, 181
Solenoids, axial field, 48
South magnetic latitude; temperature compensation, 112
Spherical aberration, 67
Spots, reserve, 65, 66
Spurious effects, 98, 135
Stabilization of magnets (see heat treatment)
Standard cells, 57
Standard resistance, 57
Standards, International Magnetic, 57
Storms, magnetic, 20
Stray fields at D variometer, 77, 211, 212, 216
Stray fields at Z variometer, 101
Sudden commencements, 181
Tangential component; field of a bar magnet, 8
Temperature bath, 34
Temperature coefficient and scale value, 112
Temperature coefficient, mechanical, 110
Temperature coefficient:
of D variometer, 119
of H variometer, 167
of magnetic moment, 26, 27, 29, 33, 195
of magnetic moment, alternate method, 39
of magnetic moment; directions, 34, 35
of magnetic moment; errors of scale reductions, 42
of magnetic moment; graph, 38
of magnetic moment; sensitivity, 41
of variometers, 109
of variometers; errors, 118
of variometers; meaning, 109
of variometers; optical compensation, 121
of Zspot, 109, 111
Temperature compensating field; C field, i7 variom-
eter, 96, 156
Temperature compensating magnet; adjustment of
Hspot by means of, 158
Temperature compensation:
device for H variometer, 84
estimation of compensation distance, 118
H variometer, 154
H variometer by two magnets, 157
magnetic, 96, 113, 116-118
magnetic; south latitudes, 112, 113
magnetic, Z variometer; evaluation of q coeffi-
cients, 113, 115
mechanical; examples. 111, 112
mechanical; south magnetic latitudes, 112
optical, 120
Z variometer; evaluation of C field, 159
Temperature, effect of, on log C, 193
Temperature magnet, adjustment of, 119, 148
Temperature, standard, 166
Thermal expansion, coefficient of (magnet and iner-
tia weight), 29
Time comparison, 174
Tune flasher, 23, 32, 67, 153
Time signals, 174
Tolerances in local anomalies at observatory sites, 21
Torque, applied; //variometer, 84
Torsion constant; Coulomb's equation, 88
Torsion constant:
by oscillations, 89
by torsion observations, 89
calculations, 71, 72
D fiber, 74
equation, 71
evaluation from k' and Ms, 89
from nomogram, 94
from scale- value observations, 89, 90
H fiber, 84, 94
k', for D variometer, 81
definition, 88
estimation of; H variometer, 88
per unit magnetic moment; H variometer, 85
Quartz fiber, 70, 90
Torsion factor:
D variometer, 74, 76, 152
//variometer, 154, 156
low-sensitivity variometer, 162
Torsion head, 150
Torsion in H variometer fiber, 84, 156, 157
Torsion, line of no; in variometer, 76, 85
Torsion in fiber, magnetometer, 27
Torsion observations, 30-32
Torsion pendulum, 71, 75
Torsion, removal of, 150, 154
Torsion tests, D, 76
Torsion, total; in D variometer fiber, 75
Torsion weight, variometer, 150
Torsionless fiber, 78 , 87
Total field of bar magnet, 9
Total intensity, definition of, 16
True azimuth; true meridian, 21
Unifilar suspension, 84
Uniform field, 3
Unit field, 1
Unit pole, 1
Variation building, 21, 22
Variation, daily, 19
Variometers, 23
Variometer data, miscellaneous, 145
Variometer, D, 74
— installation, 150
— miscellaneous data, 152
Variometer, H, 84
— installation, 154
Variometer, low-sensitivity, Schmidt, 66
Variometer, positions on pier, 149
Variometer, Z, 99
— installation, 159
Vertical intensity, definition, 16
Visual magnetograph, 165
Z magnet, la Cour, 164
Z recording magnet, effect on D variometer, 214
Z scale value, effect of H on, 105
Z spot, adjustment of, 161
Z temperature-compensating magnet; effect on D
variometer, 215
Z variometer equation, 99, 101
DIRECTIONS FOR PERFORMING OPERATIONS
a factor, H scale value; determination of, 159
Damping, H variometer; adjusting, 159
Disturbance, local, testing for, 148
Impurities, magnetic, testing for, 25
Induction coefficient:
Lamont's method, 45
Nelson's method, 51
Magnetic meridian marking in variation room, 142
Magnetograph, installation of, 148
Mirrors, variometer; adjusting, 150
Optical lever, auxiliary, use of, 162
Optical system of magnetograph, testing of, 148
Orientation tests, 141
— critical adjustment of deflector, 141
Parallax, removal of, 67
Parallax test, recording of, 181
Processing records, 174
Quartz fibers, 71, 72
—•installation of, 72
Recording distance, D variometer, evaluation of, 82
Scale values, coil method, 133
Scale value:
D, 129, 154
H, 126, 157
7, 128, 161
low-sensitivity variometers, 129
scale in telescope, 26
scale in magnet, 26
Scaling ordinates, 177
Sensitivity-control magnet, adjustment of, 157
Shrinkage, determination of, 180
Stray east field, D variometer, determination of, 151
Temperature coefficient of magnetic moment, 35, 84
— alternate method of determination, 39
Temperature compensation, optical method, 164
Time flasher, adjustment of, 153
Torsion constant, determination of, 71
Torsion, D variometer; removal of, 150
Torsion, H variometer; removal of, 154
Torsion factor, D variometer, determination of, 152
Torsion factor, H variometer, determination of, 156
Variometer installation:
D, 150
//, 154
Z, 159
low-sensitivity, 162
U. S. GOVERNMENT PRINTING OFFICE: 1953