Peek The Sphere Gap As A Means of Measuring High Voltage

Presented at the 31st Annual Convention of
the American Institute of Electrical Engineers,

Detroit, Mich. June 24, 1914, under the aus-
pices of the Electrophysics Committee.
Copyright 1914. By A.I.E.E.
THE SPHERE GAP AS A MEANS OF MEASURING HIGH

VOLTAGE
BY F. W. PEEK, JR.
ABSTRACT OF PAPER
A gap method of measuring high voltage is desirable in
certain tests because a gap measures the maximum point of the
voltage wave.
The needle gap is unreliable, and is subject to large cor-
rections due to humidity, etc. The sphere gap is free from
complicated corrections and the curve is subject to calculation.
Wherever voltage is measured by a gap, correction must be
made for air density-that is, altitude or barometric pressure,
and temperature. The laws for this correction have been
deduced for the sphere gap, from tests, over an air density range
equivalent to that from sea level to 18,000 ft. elevation. An
equation is given for calculating sphere spark-over curves for
various spacings, radii, air density, etc.
Standard measured curves for convenient sizes of spheres at
sea level are given for practical use, as well as a table for applying
these curves to any altitude. The altitude correction may be
made very accurately to the standard curve, either by applying
the calculated correction constant or by applying the correction
factor given in the tables.
The complete standard curves should not be calculated when
reliable measured curves are available.
The effects of high frequency and impulse voltages are dis-
cussed. In making measurements certain precautions are neces-
sary for accuracy, as well as for protection of apparatus. These
precautions are discussed and test results are given.
GAP method of measuring high voltages is often desirable

in certain commercial and experimental tests. As a gap
measures the maximum point of the voltage wave it is therefore
used in many insulation tests where breakdown also depends upon
the maximum voltage point. In most commercial tests an ac-
curacy of one or two per cent is sufficient. A gteater accuracy
can be obtained with the sphere gap for special work where
proper precautions are taken.
THE NEEDLE GAP
T'he needle gap is unreliable at high voltages because, due to the
brush discharge and broken-down air that precedesthespark-over,
variations are caused by humidity, oscillations, and frequency.'
The needle gap is also inconvenient because needles must be
1. Discussion, F. W. Peek, Jr., TRANS. A. I. E. E., Vol. XXXII,
1913, p. 812. G. E. Review, May, 1913.
923
zed licensed use limited to: BFH Bern. Downloaded on February 02,2022 at 14:38:01 UTC from IEEE Xplore. Restriction
924 PEEK: THE SPHERE GAP [June 24
replaced after each discharge; the gap becomes very long at high
voltages, and the calibration varies somewhat with the sharpness
of the needle.
The effect of humidity is shown in Fig. 1, where it can be seen
that a higher voltage is required to spark over a needle gap when
the humidity is high than when it is low. (Curves of Fig. 1
are intended only to illustrate this effect and not for use in
making accurate measurements). It is probable that the
corona streamers in humid air cause a "fog ", and then agglom-
erate the water particles,
which, in effect, increase the t1o,PERCENT
size of the electrodes. 200
All spark curves of what- 8 _
ever form of gap must be cor- _ - _
rected for air density; that is, t60 _ i wio i
altitude and temperature. - - - _ 1
For low voltages the spark- 140 7
over voltage of the needle gap cI i / -
decreases approximately as _ _1 T
the air density. At higher I o - 1 / _ -
voltages the effect becomes I / _ _ __
erratic, probably due to t
humidity.2 -
THE SPHERE ~GAP --'- --
The voltage required to 4_
spark over a given gap be- 20-
tween spheres increases with -
the diameter of the sphere. o__
0 10 II20 30 40 50 to
Corona cannot form on SPAC1iG,Cm.
spheres, or rather, the spark- FIG. 1-APPROXIMATE NEEDLE GAP
over point and corona point CURVES FOR DIFFERENT RELATIVE
* if
are coincident * the .
spacing ~~~H UMIDITI ES
is not greater than the diameter of the sphere. Practically
a spacing as great as three times the radius may be used
without appreciable corona. The voltage limit of a given
sphere in high-voltage measurements is thus reached when a
gap setting greater than three times the radius is required. A
larger sphere should then be used.* With this space limit, the
first evidence of stress is complete spark-over; corona can never
form, and all of the undesirable effects and variables due to brush
*More accurate and consistent results are obtained, however, for
spacings less than 2 R.
1914] PEEK: THE SPHERE GA P 925
discharge and broken-down air are eliminated. Humidity has

no appreciable effect.
The space factor is relatively small. Several thousand
measurements may be made without repolishing. The curve
may be calculated. The only correction is the air density cor-
rection. This has been investigated and the results are given
below. The correction is quite simple. Fig. 2 gives typieal
sphere-gap curves for both spheres insulated and for one sphere
grounded. Tables I to V give spark-over curves for 6.25-, 12.5-,
25-, 50- and 100-cm. spheres at sea level (25 deg. cent., 76 cm.
barometer). a = 1. (See Note 2).
MOO -- - .
'a I
4 a IZ 16 -~~~~~~~I
201001 24 28 52 3
0 40 lz 16 20 24 28 3'2 58
SPACING. Cm.
FIG. 2-SPARK-OVER CURVES
25 cm. diam. spheres
Calcu/lation of Curves: The gradient or stress on the air at the

sphere surface, where it is greatest, is, found mathematically:
e
g -f x
kv. per cm.
where e is the applied voltage in kilovolts, x is the spacing in
centimeters, f is a function of R-, and R is the radius of the sphere
in centimeters.
2. The Calibration of the Sphere-Gap V'oltmeter, Chubb and Fortescue,
TRANS. A.I.E.E., Vol. XXXII, 1913, p. 739. Discussion by F.W. Peek,
Jr., p. 812. " The Sphere Gap as a MTeans of Measturing High Volt-
age, F. W. Peek, Jr., G. E. Review, May, 1913.
926 PEEK: THE SPHERE GAP (June 24
TABLE I
SPHERE-GAP SPARK-OVER VOLTAGES
6.25-cm. DIAMETER SPHERES.
Spacing Kilovolts effective
cm. in. Non-grounded Grounded
0.5 0.197 12 12
1. 0.394 22.5 22.5
1.5 0.591 31.5 31.5
2 0.787 41. 41.

3 1.181 57.5 56.
4 1.575 70.5 66.
5 1.969 81. 73.

B6 2.362 89. 79.
7 2.756 96. 83.
8 3.150 102. 88.

9 3.543 107. 90.5
10 3.937 110 93.
Each point is the average of five readings. The average variation between maximum
and minimum for a given setting is less than 0.5 per cent.
TABLE II
SPHERE-GAP SPARK-OVER VOLTAGES.
12.5-cm. DIAMETER SPHERES.

_____________ l______ ___. _
0.25 0.098 6.5 6.5

0.50
1
0.197
0.394
12.
22
12.
22.
1.5 0.591 31.5 31.5
2 0.787 41 41.
3 1.181 59 59.
I 4 1.575 76 75
5 1.969 91 89
6 2.362 105. 102
7 2.756 118. 112

8 3.150 130 120
9 3.543 141. 128
10 3.937 151. 135

12 4.72 167. 147
15 5.91 188 160
17.5 6.88 201 168.

20. 7.87 213. 174.
1914] PEEK: THE SPHERE GAP 927
Then e=
9$-x f kv. per cm.
where eS is the spark-over voltage and g, is the apparent strength

of air.
f is found mathematically and tabulated in Table VJ3 for the
non-grounded and grounded cases. For the non-grounded* case it
TABLE III
25-cm. DIAMETER SPHERES
0.5 0.197 11 11
1. 0.394 22 22
1.5 0.591 32 32
2. 0.787 42 42
2.5 0.983 52 52
3 1.181 61 61
4. 1.575 78 78
5 1.919 96 94
6 2.362 112 110
7.5 2.953 135 132
10. 3.937 171 166
12.5 4.92 203 196
15. 5.91 230 220
17.5 6.88 255 238
20. 7.87 278 254
22.5 8.85 297 268

25. 9.83 314 280
30. 11.81 339 300
40. 15.75 385 325
was found experimnentally that gs, the apparent surface gradient

at spark-over, increased with decreasing radius of sphere, as g,
for corona on wires increases for decreasing radius of wire. For
a given size of sphere, gs is practically constant, independent of
spacing, between the limits of x = 0.54 V\R and x = 2 R.
3. See Law of Corona III, F. W. Peek, Jr., TRANS. A. I. E. E., Vol.
XXXII, 1913, p. 1767. G. R. Dean, G. E. Review, March, 1913;
Physical Review, Dec. 1912 and April 1913. Russel-Philosophical,
May 1906.
*Non-grounded here means with neutral at the mid point. It is thus,
in this case, preferable to ground the transformer neutral.
The average gradient between these limits of separation iS4
=3 27.2 (1 + i) kv. per cm. max. (la)
g, 19.3 (1 + 0.54 )kv. per cm. eff. sine wave. (lb)
TABLE IV
50-cm. DIAMETER SrHERES

l ~~~~~~*
2 0.787 40 40
4 1.575 76.5 76
6 2.362 115.5 112
8 3.150 149 145

10 3.937 189 185
12 4.72 224.3 220
14 5.51 255.5 250

16 6.30 285 275
20 7.87 335 320
25 9.83 393 377t

30 11.81 445 420
35 13.80 493 456
40 15.75 537 489

45 17.72 573 516
50 19.19 605 541
55 21.65 633 561

60 23.62 660 579
65 25.60 684 594
70 27.56 705 608

75 29.55 725 619
*These values are calculated.
tSpacings above 22 cm. are calculated.
The maximum variation from the average between the limits

may be 2 per cent. When x is less than 0.54 v"R, gs increases
very rapidly because the spacing is then less than the " rupturing
energy distance."5 Above x = 3 R, gv apparently gradually
increases5. This increase seems only apparent and due to the
4. See Law of Corona III, F. W. Peek, Jr., TRANS. A. I. B. B., Vol.
XXXII, 1913, p. 1767.
5. Ibid.
shanlks, surrounding objects, etc., better distributing the flux

or lessening the flux density. When both spheres are insulated
and of practical size, the change is not great within the prescribed
limits. When one sphere is grounded, however, this apparent
increase of gradient is very great if the mathematical f, which
does not take account of the effect of surrounding objects, is
used. For this reason this fo was determined experimentally,
for the grounded case, by assuming gs constant within the limits,
TABLE V
100-cm. DIAMETER SPHERE.
cm. in. INon-grounded Grounded
1.0 0.394 20 20
3.0 1.181 60 60
5.0 1.969 100 100
10.0 3.937 195 195

15.0 5.91 283 280
20.0 7.87 364 360
30.0 11.81 520 505
40.0 15.75 650 615
50.0 19.69 770 730
60.0 23.62 870 810

70.0 27.56 956 895
80.0 31.50 1044 956
90.0 35.43 1107 1010

100.0 39.37 1182 1057
110.0 43.35 1238 1090
120.0 47.20 1290 1133

130.0 51.20 1335 1160
140.0 55.70 1378 1189
150.0 59.10 1412 1212

These values are calculated.
as it is in the non-grounded case, and finding values of fO corres-
ponding to the different values of - These measurements
were made with the grounded sphere 4 to 5 diameters above
x
ground. Any given value of the ratio k should require a con-
stant fo to keep gs constant independent of R. This was found
to check.
The curves may be quite closely calculated thus:
eS = gs .f (non-grounded) effective sine wave. (2)
e g x (grounded) effective sine wave (3)

where g, is calculated from equation (ib), and fo or f is found
from the table for the given R Cutrves need be cal-
TABLE VI
SPHERE-GAP GRADIENT FACTOR.
Non-grounded. One sphere One sphere grounded.

Mathematical grounded. Mathematical coeffi-
X and actual Actual cient not correct
R practical practical on account of effect
coefficient* coefficient of shanks, etc.
f f_ f,
0.1 1.034 1.034 1.050

0.3 1.103 1.105 1.162
0.5 1.175 1.180 1.280
1.0 1.366 1.410 1.620
1.5 1.568 1.675 2.000
2.0 1.781 1.965 2.415
2.5 2.002 2.27 2.85
3.0 2.225 2.59 3.30
3.5 2.455 2.90 3.76
4.0 2.686 3.20 4.23
*Non-grounded: f 3 +1 + + 1 + 8)
I fo experimental values.
One sphere grounded: = X + + 4)
fl should not be used-see note above. fo should be used for the grounded case.
The experitnental values were determined with the grounded sphere four to five
diameters above ground.
culated only when standard measuired curves cannot be ob-

tained. Measured curves are given here. The average error,
however, for curves calculated from the above equations, for
2-cm. diameter spheres and over, should not be greater than 2
or 3 per cent. The accuracy of calculations is not as great as
in the case of the starting point of corona on wires.
EFFECT OF AIR DENSITY OR ALTITUDE AND TEMPERATURE

We have found that the average gradient for various air
densities may be expressed
gs -27.2 a (1 + /aR) kvT. per cm. max.
/ 0.54\
gs = 19.3 (1 + )kv. per cm. effective
where a is the relative air density
3.92 b
5a = 273+t
=
b = barometric pressure in cm.

t -temperature in deg. cent.
The standard sea-level curves correspond to 76 cm. barometric

pressure and 25 deg. cent. or where a = 1.
Correction Factor. The standard curve may be made to apply
to any given altitude by multiplying the voltage at different
spacings by the correction factor, thus:
/ 0.54 0.5
(4)
=ea, =ek4
0
Practical Application. In order to avoid the trouble of cal-

culating in practise, the factor is tabulated in Tables VII and
VIII. This correction is very accurate. Table VII gives the
correction factor for different sizes of spheres at different baro-
metric pressures and at constant temperature. When volt-
TABLE VII
Approximate Barometer Values of k at 25 deg. cent. for

corresponding standard spheres of the following diameters.
altitude cm. in. Centimeters.
ft. Hg Hg
6.25 12.5 25 37.5 50 75 100
0 76.00 29.92 1.000 1.000 1.000 1.000 1.000 1.000 1. 000

29.36 0.981 0.980 0.980 0.979 0.979 0.979 0.979 b
500 74.58
1,000 73.14 28.79 0.964 0.963 0.962 0.961 0.960 0.960 0.960
1,500 71.77 28.25 0.948 0.946. 0.945 0.944 0.9493 0.942 0.942
2,000 70.42 27.72 0.932 0.929 0.927 0.926 0.925 0.924 0.924
2,500 69.09 27.20 0.916 0.913 0.911 0.909 0.908 0.907 0.907
3,000 67.74 26.67 0.902 0.899 0.897 0.895 0.893 0.892 0.891 j
3,500 66.51 26.18 0.887 0.881 0.882 0.880 0.878 0.876 0.875 U
4,000 65.25 25.69 0.873 0.870 0.867 0.865 0.863 0.861 0.860
4,500 64.01 25.23 0.859 0.855 0.852 0.850 ,0.848 0.846 0.845 ;
5,000 62.79 24.72 0.845 0.841 0.838 0.835 0.833 0.831 0.830
6,000 60.45 23.80 0.817 0.812 0.808 0.805 0.803 0.801 0.800
7,000 58.22 22.93 0.791 0.786 0.782 0.779 0.776 0.774 0.772
8,000 56.03 22.05 0.765 0.759 0.754 0.750 0.748 0.746 0.744
9,000 53.84 21.20 0.739 0.733 0.728 0.724 0.721 0.719 0.717
10,000 51.85 20.41 0.716 0.709 0.703 0.698 0.694 0.692 0.690 _
12,000 48.09 18.93 0.669 0.661 0.656 0.651 0.647 0.644 0.642 =
15,000 42.88 16.88 0.606 0.596 0.589 0.585 0.580 0.577 0.575 r
Authorized licensed use limited to: BFH Bern. Downloaded on February 02,2022 at 14:38:01 UTC from IEEE Xplore. Restrictions apply.
1914] PE-EK: THE SPHERE GA P 933
age strikes across a given gap, its value, el, may be found by tak-
ing the voltage e, corresponding to the gap, from the standard
curve and multiplying by the correction factor k, or a curve may
be plotted corresponding to a given barometric pressure, thus
e = ek.
TABLE VIII
Calculated values of k for different values of a
+0.541
k= a5 _ 8R !
+
0.54
I%/R j
a R + 0.54
o/R+ 0 54
.~~~~~~~~~~~~~~
Relative Values of k.
air
density Diameter of standard spheres in cm.
6.25 12.5 25. 37.5 50. 75. 100.
0.50 0.547 0.535 0.527 0.522 0.519 0.519 0.516

0.55 0.594 0.583 0.575 0.570 0.567 0.565 0.564
0.60 0.640 0.630 0.623 0.618 0.615 0.613 0.612
0.65 0.686 0.677 0.670 0.665 0.663 0.661 0.660
0.70 0.732 0.724 0.718 0.714 0.711 0.709 0.708
0.75 0.777 0.771 0.766 0.762 0.759 0.757 0.756
0.80 0.821 0.816 0.812 0.809 0.807 0.805 0.804
0.85 0.866 0.862 0.859 0.857 0.855 0.854 0.853
0.90 0.910 0.908 0.906 0.905 0.904 0.903 0.902
0.95 0.956 0.955 0.954 0.953 0.952 0.951 0.951
1.00 1.000 1.000 1.000 1.000 1.000 1.000 1.000
1.05 1.044 1.045 1.046 1.047 1.048 1.049 1.049
1.10 1.090 1.092 1.094 1.095 1.096 1.097 1.098
Table VIII gives the correction factor for various values of &
The value of 6 may be calculated for the given temperature and
barometric pressure, and a correction factor then found from the
table. Fig. 3 gives the standard curve for the 12.5-cm. sphere
(non-grounded, 25 deg. cent., 76 cm. bar. pressure) and curves
calculated therefrom for 25 deg. cent. and various barometric
pressures.
EXPERIMENTAL DETERMINATION OF AIR DENSITY CORRECTION

Tests. The equation (4) for the air density correction fac-
tor was determined by an extensive investigation of the spark-
over of spheres in a large wooden cask arranged for exhaustion
of air. This cask was built of paraffined wood and was 2.1
meters high by 1.8 meters inside diameter.
Tests were made by setting a given size of sphere at a given
spacing, gradually exhausting the cask, and reading spark-over
voltage at intervals as the air pressure was changed. (Tempera-
ture was always read, but varied only between 16 deg. and 21
deg. cent). This was repeated for various spacings on spheres
ranging in diameter from 2.0 cm. to 25 cm. At the start, the
G G - __ I_ a 45.76C om=.tOG
SPAC-ING,
S8CM.E
- - V ___664Cm. A8
~~ I5C - - -
=m09
~~~~~~ 53.2 Cl dQ. 7
-
80Cm.d-Q05
0 200 15 2
PAIGCm.
FIG. 3-SPHERE GAP SPARK-OVER VOLTAGE, AT VARIOUS
AIR PRESSURES
Temperature 25 deg.cent. 12.5-cm. spheres. Non-grounded. Drawn curves calculated.
76 cm. is standard curve. Points are from curves through measured values.
possible effect of spark-overs on the succeeding ones in the cask

was investigated and found to be nil or negligible. A resistance
of one to four ohms per volt was used in series with the spheres.
Wave shape was measured and corrected for. Voltage was read
on a voltmeter coil, by step-down transformer, and by ratio.
Precautions were taken as nloted in the discussion referred to
below-6
In order to illustrate the method of recording data, etc., a
small part of the data for various spheres and spacings is given
in Tables IX to XVIII. Numerous data are plotted in curves,
6. The Calibration of the Sphere-Gap Voltmeter, TRANS. A.I.E.E., Vol.
XXXII, 1913, p. 739. Discussion by F. W. Peek, Jr., p. 812.
TABLE IX
SPHERE-GAP SPARK-OVER VOLTAGES AND GRADIENTS
2.54-cm. DIAMETER SPHERES-NON-GROUNDED.
Spacing Temp Press. Relative kv. gs

cm. in. deg. in. of Hg air eff. max. eff. max.
cent. density measured
0.635 0.25 15 29.78 1.028 15.7 22.2 29.2 41.3

28.63 0.99 15.3 21.65 28.5 40.3
27.73 0.957 14.9 21.1 27.7 39.2
26 93 0.93 14.4 20.4 26.8 23.0

25.93 0.895 14.0 19.8 26.0 36.8
24.68 0.852 13.6 19.21 25.3 35.7
23.63 0.818 13.2 18.7 24.6 34.8

22.38 0.776 12.5 17.2 23.25 32.0
21.o8 0.745 12.1 17.1 22.5 31.8
20.78 0.717 11.8 16.7 22.0 31.1

19.88 0.686 11.3 16.0 21.0 29.8
18.78 0.649 10.75 15.2 20.0 28.3
17.88 0.618 10.4 14.7 19.35 27.4
16 .68 0.576 9.8 13.85 18.25 25.8
15.73 0.544 9.26 13.1 17.2 24.4
14.48 0.500 8.66 12.25 16.1 22.8
_l_ I-_
Barometer 29.78 in.
TABLE X
SPHERE-GAP SPAR K-OVER VOLTAGES AND GRADIENTS
Relative g
Spacing air kv. eff. max.
cdo. in. density eff. mrax. measured
3.81 1.3 1.01 52.4 74.2 29.2 41.4

It
. 0.962 50.5 71.5 28.1 39.8
ss
It
0.923 49.1 69.5 27.4 38.7
ss
if
0.875 45.7 64.6 23.5 36.0
0.834 45.2 64.0 25.2 35.7
0.798 42.8 60.6 23.8 33.8
0.757 41.3 38.5 23.0 32.6

0.734 40.4 57.2 22.5 31.9
0.722 39.0 55.1 21.7 30.7
0.669 38.0 53.8 21.2 30.0

0.629 35.6 50.4 19.8 28.1
0.590 34.1 48.3 19.0 26.9
0.551 32.7 46.3 18.2 25.8
0.527 31.2 44.3 17.4 24.7
£ a ()0.487 30 5 43.2 17.0 24.1
936 PEEK: T/HE SPIIERE GA P [June 24
TABLE XI
5.08-cm. DIAMETER SPHERES-NON-GROUNDED
Spacing Relative kv. Bs

cm. in. air eff. max. eff. max.
density measured
5.08 2 1.018 78.4 111.0 27.4 38.9

0.98 75.6 107. 26.5 37.6
N i
a 0.944 73.1 103.5 25.6 36.2
N uN 0.903 7.07 100. 24.8 35.0
N u 0.872 68.7 97.2 24.0 34.0
0.836 64. 90.5 22.4 31.7
0.798 63.6 90. 22.3 31.5

|Na0.764 61.1 86.4 21.4 30.2
NN 0.726 58.9 83.3 20.6 29.2
N N 0.682 56.1 80.5 19.6 28.2
N a
0.654
I
54.0 76.4 18:9 26.8
^ 0.618 51.o 73.0 18.0 25.6
£ a 0.578 49.0 69.3 17.2 24.3
I a 0.544 45.8 64.7 16.0 22.6
N I
a 0.510 43.3 61.2 15.2 21.4
.,
~~I I_ _I__
TABLE XII
12.5-cm. DIAMETER SPHERES- NON-GROUNDED.
Relative gs
cm. in. density eff. max. measured.
2.54 1 1.016 52.4 74.1 23.6 33.4

N Na 0.966 49.9 67.9 22.4 30.6
N N 0 928 47.4 67.0 21.3 30.2
S N 0.88 45.4 64.2 20.4 28.9
. a 0.83 42.5 60.1 19.2 27.1
£ Na 0.778 40.0 56.6 18.0 25.5
£ a 0.738 38.4 54.3 17.3 24.4
0.705 36.0 52.2 16.6 23.5
4 N 0.644 33.4 47.2 15.0 21.2
N z 0.607 31.4 44.5 14.1 20.0
N a 0.558 29.4 41.6 13.2 18.7
N . Z 0.538 29.6 41.8 13.3 18.8
S N 0.493 27.4 38.8 1.2.3 17.5
19143 PEEK: THE SPHERE GA P 937
TABLE XIII
SPHER E-GAP SPARK-OVER VOLTAGES AND GRADIENTS.
Relative gs
cm. in. density eff. max. measured
5.0S 2 1.02 91. 129. 23. 33.65

0.962 87.4 123.5 22.1 31.25
0.925 84.6 120. 21.4 30.40
0.89 81.9 115.8 20.7 29.25

0.846 78.1 110.5 18.73 27.95
0.81 76.3 107.9 19.26 27.25
0.754 71.6 100. 18.1 25.30
0.707 66. 93.5 16.7 23.65
0.661 64.1 90.6 16.2 22.90
0.603 59.3 84. 15.0 21.25
0.559 53.9 76.1 13.64 19.25
0.532 52.0 73.5 13.15 18.60
0.506 50.1 71. 12.65 17.95
TABLE XIV
SPHERE-GAP SPARK-OVER VOLTAGES AND GRADIENTS.
Relative gs
10.16 4 1.004 145.0 205.0 23.0 32.55

0.965 144. 203.7 22.85 32.30
Is Is
0.93 139 196.6 22.10 31.20
s is 0.90 137. 194. 21.78 30.80
0.865 131.8 186.5 20.95 29.65
0.859 131.5 186.2 20.85 29.60
0.793 123.7 175.0 19.60 27.80

0.762 120.8 171.0 19.20 27.20
0.739 115.7 163.6 18.36 26.00
0.685 110.2 156.0 17.50 24.80
0.67 105.9 149.7 16.80 23.75
0.623 100. 141.4 15.90 22.45
0.595 96.6 136.6 15.34 21.70
0.534 88.3 125.0 14.05 19.85
0.503 82.2 116.4 13.05 18.50
TABLE XV
12.5-cm. DIAMETER SPHERE-NON-GROUNDED.
Relative gs
12.7 5 0.982 163.0 230.0 23.1 32.7

0.951 156. 221.0 22.2 31.4
0.917 150 212.5 21.3 30.2
0.88 147 208 20.9 29.5

0.846 143.5 203 20.4 29.8
0.807 139.5 197.5 19.8 28.0
0.78 134.5 190 19.1 27.0

0.736 131 185.5 18.6 26.3
0.699 125 177 17.7 25.1
0.666 120.5 170 17.1 24.2
a 6 0.637 115.5 163 16.4 23.1
0.598 109.15 155 15.5 22.0
0.561 104.5 147.5 14.8 21.0

0.541 101 142.5 14.3 20.2
TABLE XVI
12.5-cm. DIA-METER SPHERES-NON-GROUNDED.
Relative gs
20.32 8 0.938 198.5 281 23.0 32.6
0.912 197 279 22.9 32.4
0.874 189.5 268 22.0 31.1
0.842 183 259 21.2 28.9

0.804 176.5 250 20.5 29.0
0.767 169 239 19.6 27.8
a^ 0.737 163.5 231 19.0 26.8
£^ 0.704 157.5 223 18.3 24.7
0.674 151 235 17.5 24.8
0.615 140.0 198.0 16.2 23.0
£ a 0.568 130.5 184.5 15.1 21.4
0.528 120.5 171 14.0 19.8
0.488 110 157 12.8 18.2
0.464 105.5 149 12.2 17.3
'I'ABLE XVII
SPHERE-GAP SPARK-OVER VOLTAGLES AND GRADIENTS
12.5-cm. DIAMETER SPHERE-GROUNDED.
Relative gs
6.35 2.5 0.908 95.5 135 21.2 30.0

0.869 92.5 130.5 20.5 29.0
0.828 88.6 125 19.7 27.8
0.796 85.8 121 19.05 26.9

0.758 81.1 114.5 18.0 25.4
0.723 78.2 110.5 17.35 24.5
0.69 73.2 103.5 16.25 23.0

0.653 71.5 101 15.9 22.4
0.62 68.2 96.3 15.15 21.4
0.582 64.3 90.9 14.25 20.2

0.539 60.7 85.8 13.5 19.05
0.439 55.6 78.5 12.35 17.4
TABLE XVIII
25-cm. DIAMETER SPHERE-NON-GROUNDED.
Relative gs
crn. in. density eff. max. measured
7.6'2 3 1.018 139 196.5 22.2 31.5

0.978 133.5 189 21.4 30.3
0.942 129.5 183 20.8 29.3
0.906 126 178 20.2 28.5

0.888 121.5 172 19.5 27.6
0.839 115 163 18.4 26.1
0.796 111 157 17.8 25.2

0.752 105.5 149 16.9 23.9
0.718 101.5 142 16.3 22.7
0.685 96.2 136 15.4 21.8

0.646 91.5 129.5 14.6 20.7
0.608 87.0 123 13.9 19.7
0.570 81.3 115 13.0 18.4

0.527- 74.7 105.5 12.0 16.9
0.491 70.8 100 11.3 16 0
Figs. 4 to 8. The points are measured values. The drawn lines

are calculated by multiplying the voltage values from the stand-
ard curves at 5 = 1, by the correction factor k.
(+ %.4)
V A/6 RJ
k b-lk
= I
The calculated values check the measured values very closely.
40
o. o
30~~~~~~~~
0 0.5 1. 0 0.5 1.0

RELATIVE AIR DENSITY 6 RELATIVE AIR DENSITY d
FIG. 4-SPARK-OVER VOLTAGE FIG. 5-SPARK-OVTER VOLTAGE
WITH VARYING AIR DENSITIES WITH VARYING AIR DENSITIES
A )4cmn (1-in.) spheres. Nion-grounded. 5.08-c:m. (2-in.) spheres. Non-grounded.
Spacing 381 cm. (1.5 in.), 0.635 cm. (0.25 in.) Spacing 5.08 cm. (2 in.).
Tphc
equation for the correction factor was deduced fromn
measulred values as follows:
From a former investigation7 it was found that at 6 - 1 the
average gradient
g5 = go(i ± _-) (1)
Fromn the i)resenlt investigationl it wvas found thlat the average

gradient at various values of 6 is
gS= galv R
7. Law of Corona IlTl. F. W. Pek, Jr., TRANS. A.I.E.E., VOL. XXXII,

1913, p. 1767. Discussion, TRANS. A.I.E E., VO. XXXII, 1913, P. 812.
The average measured gradients for various values of 6 are given

in Table XIX; values calculated from the equation are also
given. The check is quite close. It should be remembered, how-
ever, that these are average values and that there is a small varia-
4--
17.78
cm.1( Ccm.
7.62
cm. (3 in.) Cin
0 0.5
RTLATIVE AR DENSITYD
1VRN-SPARK-OVER
IG
5C
VOLIAGE WITH VARYING AIR DENSITIES
2032cm (mi.
15/16
: 2.5-cm. (4 6 in.) spheresr Non-grounded
254 cm. (1n.)
5.08 cm. (2 in.)
7.62 cm. (3 in.)
Spacing 1o0.16 cm. (4 in.)
12.7 cm. (5 in.)
17.78 cm. (7 in.)
(20.32 cm. (8 in.)
50 O~~50
o o. - .o o - .
-.
~~~~~~~~~~~I 5C
f1 -
12jCms
7.62 Ce
50
1.0 RELATIV AIRDN5TY

0
rRELATIVE0.5AIR DENSITYCY 1.0
FiG. 7-SPARK-OVER VOLTAGE FiG. 8-SPARK-OVER VOLTAGE
WITHi VARYING AIR DENSITIES WITH VARYING AIR DENSITIES
.12.5-cm. (4 15/16-in.) spheres grounded. 25-cm. (9 7/8-in.) spheres, non-grounded.
Spacing 6.35 cm. (2.5 in.). Spacing 12.7 cm. (5 in.).
7.62 cm. (3 in.).
tion at different spacings over this range, as already explained.

(See Figs. 9 and 10).
FREQUENCY
Over the commercial range there is no variation due to
frequency. Fig. 11 shows spark-over curves up to 25 kv.
TABLE XIX
AVERAGE EF;FECTIVE RUPTURING-GRADIENTS FOR SPHERES OF SEVERAL DIAMETERS
AND VARYING AIR DENSITIES.
Diameter of Spheres, cm.
2.54 5.08 12.5 25
6
Surface gradients
Meas. Calc. Meas. Calc. Meas. Calc. Meas. Calc.
1.00 28.7 28.5 25.8 25.8 23.6 23.4 22.0 22.2

0.90 26.2 26.1 23.7 23.3 21.8 21.2 19.9 20.1
0.80 24.0 23.7 21.3 21.3 19.6 19.1 17.9 18.0
0.70 21.3 21.2 19.0 19.0 17.4 16.9 15.7 15.9
0.60 18.7 18.7 16.7 16.6 15.2 14.7 13.6 13.8
0.05 16.1 16.1 14.6 14.3 13.0 12.5 11.6 11.7
0.54
Columns marked Calc. " are from gs - 19.3 6 (1 +
30 -
1~~~~~~~~2.54
Cm. Dio
_50 i _-
z ;508 Cm. Dia:

112.5 Cm.Dia.
0252ZiimDm
US
0 0.5 1.0
RELATIVE AIR OEMSITY&
FIG. 9-SURFACE GRADIENT AT VARYING AIR DENSITIES FOR SEVERAL
SIZES OF SPHERES
Drawn curves calculated. Points measured values:
F-~~~ ~ ~ ~~~~~~0
Si 20 09z
d'= 0. 5
0 5 lO .15 20 25
DIAMETER OF SPHERES Cm.
FIG. 10-SURFACE GRADIENT FOR VARYING SIZES OF SPHERES AT SEV-
ERAL AIR DENSITIES.
0.54X
Drawn lines calculated from formula gs = 19.36 1 + R)
Points measured values.
with 1000-cycle sine-wave current from an alternator. The

voltage was measured by a static voltmeter calibrated at
60 cycles. The drawn curve is the 60-cycle curve and the
points are measured values. Fig. 12 gives a 60-cycle curve,
and also a 40,000-cycle curve from a sine-wave alternator.
25 25
20 t0208
Ui Z0
0~~~~~~~~~~~~~1
Q00. 0.4 0.6 0.8

0.0 01 0 0 0.4 6G
S PACI NG, Cm . SPACINO, Cm.
FIG. 11A-SPHERE GAP SPARK- FIG. 11B-SPHERE GAP SPARK-
OVER VOLTAGE AT 60 CYCLES AND OVER VOLTAGE AT 60 CYCLES AND
1000 CYCLES 1000 CYCLES
5.08-cm. diam. spheres. Drawn curve 12.5-cm. diam. spheres. Drawn curve
60 cycles. Points taken at 1000 cycles. 60 cycles-points taken at 1000 cycles.
02 0. 0j6 - -
S
04 N6 C.( L
FIG. 12 SPHERE GAP SPARK-OVER VOLTAGE AT 60 CYCLES AND 40,000

CYCLES
Points measured at 40,000 cycles. 5.08-cm. (2-in.) spheres, not polished.
The voltage in this case was mneasured by a static voltmeter.

No special care was taken to polish the sphere surfaces. At
low frequencies, at rough places on the electrode surface there
is local over-stress, but even if the air is broke-n down, the loss
at these places is very srnall and the streamers inappreciable.
944 PEEK: TIHE SPHERE GAP [June 24
At continuous high frequency, a local breakdown at a

rough point probably takes place at the same gradient as
at 60 cycles, but the energy loss after the breakdown at this
point occurs may be 1000 times as great. This forms a needle-
like streamer which increases the stress and local loss. Spark-
over then takes place from this " electric needle " at a lower
voltage than the true sphere gap voltage; thus it seems that the
air at high frequency of the above order is only apparently of
less strength. These " electric needles " when once formed
may be blown to different parts of the sphere surface. The
corona starting point appears to take place at a lower voltage
at high frequency, because the local loss at rough points, which
occurs before the true critical voltage is reached, is very high
at high frequency and distorts the field and masks the true
starting voltage. The loss at rough points starts at the same
voltage at low frequency, but is inappreciable and cannot change
conditions. If the sphere surfaces are very highly polished it
seems that the high-frequency spark-over voltage should check
more closely with the 60-cycle voltage for frequencies of the
above order. It also seems that the curves should check very
closely for oscillating voltages of short duration and voltages of
steep wave front, even under ordinarv conditions of surface.
With needle gaps the results are quite different at high and
low frequency and steep wave front. More complete investi-
gations are being made.
Care should be taken in attempting to check voltages across
one gap of any sort by another gap in parallel. For instance,
if a needle gap is set so as to just spark over when a steep wave-
front voltage of a given constant value is applied, and a sphere
gap is similarly set, and these two gaps are then placed in par-
allel and the same impulse voltage applied, apparent discrepancy
results. Spark-over will take place across one gap and not the
other, even when the spacing on the non-sparking gap is de-
creased. This will be noticed in all cases where electrodes of
different shapes are employed. The reason, apparently, is
that energy is necessary to start rupture in the dielectric, the
amount of energy varying with the shape of the electrode.
This introduces a time element which differs for different gaps.
Thus, in this case the " fastest " gap will spark over and relieve
the voltage before the other has time to spark, even if the other
gap is decreased.
It is now often said that " high frequency acts differently,
on insulation, from low frequency ". When the time of applica-
tion is sufficiently short, say a few cycles, so that the effect of

heating may be neglected, and the circuit is such that local
concentration of potential is not possible, it is probable that a
given high-frequency potential is less destructive than the same
low-frequency potential. For steep wave fronts a higher vol-
tage is required to spark over a given gap than for low fre-
quency. Destruction of insulation by high frequency is
generally due to local over-voltage, and not to " frequency '.
Fbr instance, high-frequency low voltage may be applied to a
piece of apparatus containing induction and capacity, such as
a transformer. On account of the capacity- and inductance,
very high local over-voltage may be built up, and breakdown
results, due to over-voltage. The petticoats of an insulator
may be broken down by an " electric needle " forming as de-
scribed above, and bringing the total stress on the thin petticoat.
Spectacular tests are sometimes made by placing an insulator
in multiple with a gap over which an oscillatory discharge is
taking place, 'such as a Tesla coil gap. This gap is set at a
spacing that would limit the voltage to a safe voltage at 60
cycles, yet the insulator is punctured on the oscillatory dis-
charge. This is not remarkable. By the circuit arrangement
many times the insulator puncture voltage is applied to the gap.
This voltage is across the insulator for short intervals just be-
fore the gap discharges take place. It is simply a case of over-
potential.
Such tests are useful in many cases, but are dangerous to
apparatus unless thoroughly understood and necessary pre-
caution taken.
PRECAUTIONS AGAINST OSCILLATIONS IN TESTING
A non-inductive resistance of one to four ohms per volt should
always be placed directly in series with the gap. For the non-
grounded gap, one-half should be placed on each side. When
one gap is grounded, all of the resistance should be placed on the
insulated side. One object of the resistance is to prevent
oscillations from the test piece, as a partial arc-over on a line
insulator, reaching the gap. Accuracy cannot be secured
without this resistance. Another object is to limit the
current discharge. This resistance is of special importance
when tests are being made on apparatus containing inductance
and capacity. If there is no resistance, when the gap sparks
over, oscillations will be produced whiclh will cause very high
local voltage rise over parts of the winding. If sufficient re-

sistance is used, these oscillations will be damped out. This
is illustrated in Fig. 13, which shows results of a test on a high-
voltage transformer on which a 19-cm. gap test is required.
Referring to Fig. 13, the high-voltage winding of the trans-
former under test is short-circuited and connected to one ter-
minal of the testing transformer. The other terminal of the
testing transformer is grounded. The low-voltage winding
of the transformer under test is short-circuited, connected to
the case and ground. Voltage is gradually applied to the trans-
Gen.
Tesf Trans former
F19
CGs.
feasuring "ap g pr o
sraindi wen the misnc

, K et-i'Sml
12 Cm~Im-j
{/ , # 15 ~Case
r-1C. /Cm.-_;
9Q2oAsprV/
02OM
k---105 Cm_-
ll
fote
pont *rs132
Cmo8n
ih-enin
.5 m- ->j "-.025-
te fre b
inig OH/MS fher Vo/O troansfre ne
1 50'Cm.--- - '(,C- Verl/ Smnall
n te s ar di.s a OHf per ovolt
FIG. 1 3
former under test until the " measuringF gap sparks over.
4, are brougit out at equidistant
Insulated taps, 1, 2, 3, r5,
points from the high-tension winding 'of the transformer under
test. Auxiliary needle gaps are placed between I anld 2, 2 and
3, and 1 and 3, to measure the voltage which appears across these
sections of the winding when the main measuring gap discharges.
The numbers on the figure between 1 -2, 2 -3 and 1 -3, rep-
resent the sparking distances of the local voltages caused
by a discharge of the measuring gap. Four cases are given
with different values of resistance w in the main gap. When
w equals one ohm per volt the local voltages are completely
damped out.
With small resistance in the main gap, a 19-cm. spark-over
causes a voltage to build up between coils one and three which
sparks over a 150-cm. gap, although the total applied voltage
across the transformer is only equivalent to a 19-cm. gap. The
apparatus may thus be subjected to strains far beyond reason,
and either broken down or very much weakened. Water-tube
resistance is the most reliable. A metallic resistance, if non-
inductive and of small capacity, may be used. Carbon or
graphite rods should be avoided, because, although they may
measure up to a very high resistance at low voltage, the re-
sistance may become very low at high voltage by " coherer "
action. Where possible the gap should be used only to measure
the equivalent transformer ratio and not during test. When the
tested apparatus is of such nature that there is considerable
incipient arcing before spark-over, it is better to use the spheres
to determine the " equivalent ratio " of the transformer below
the voltage at which this arcing occurs. It must be remembered
that resistances do not dampen out low-frequency surges resultinrg
from a short circuit, etc.
TABLE XX.
POSSIBLE PER CENT ERROR DUE TO PLATES ON
6.25-cM. SPHERE SHANKS.
Non-grounded Grounded
Spacing 5 cm. diameter plate 5 cm. diameter plate
cm. 6.25 cm. back of 6.25 cm. back of
both one insulated grounded both
spheres sphere sphere sphere spheres
1.5 0 -0.2 +0.7 -0.7 0

3.0 +1.0 -1.0 +1.5 -1.5 +2.0
6.0 +2.0 -2.0 +3.0 -2.0 +2.0
This table is not meant for making corrections. The arrangement should be such that
no corrections are necessary.
APPROXIMATE EFFECT OF DISTANCE ABOVE GROUND WHEN ONE SPHERE IS GROUNDED

Percentage variations from standard curves
Diameters of grounded for different spacings.
sphere above ground. X = 2R; X = R; X = R/2
0 -10. 5.5 0.0

1 - 4.5 -3.0 0.0
2 - 2. -1.0 0.0
3 - 1. -0.5 0.0
4 0. 0.0 0.0
5 +.0.5 +0.3 0.0
6 + 1.0 +0.5 0.0
10 + 2.5 +1.0 0.0
20 + 2.5 +1.0 0.0
MISCELLANEOUS PRECAUTIONS
In making tests it is desirable to observe the following pre-
cautions:
The shanks should not be greater in diameter than 1/5 the
sphere diameter. Metal collars, etc., through which shanks
extend, shouldbe as small as practicable, and should not come
closer to the sphere than the gap length. For the non-grounded
case the effect of such objects is a minimum when symmetrical on
either side. For the grounded case the effect is considerable.
The effect of a large plate or plates on the shanks is given in
Table XX. This should show the maximum error in all cases,
as the maximum gap setting, and the extreme diameter of
collar that is likely to be used in practise, is taken. The
sphere diameter should not vary more than 0.1 per cenit, and
the curvature, measured by a spherometer, should not vary
more than I per cent from that of a true sphere of the re-
quired diameter. The spheres should be twice the gap setting
from surroundings.
In Table XX the spacing Xis given in terms of radius R, to make
the correction applicable to any size of sphere. The distance of the
grounded sphere above ground is, for the same reason, given in
terms of the sphere diameter. The (+) sign means that a higher
voltage is required to arc over the gap than that giveln by the
standard curve. The (-) sign indicates that the spheres read
too low. The standard curv,es were made with the grounded
spheres from four to five diameters above ground. In practise
it is desirable to work between four and teln diameters; never
ulnder three. Above ten the variation in per cent error remains
about the same.
When both spheres are insulated, with the transformer neutral
at the mid point, there is practically no variation in voltage for
different distances above ground.
Great precautions are necessary at very high voltages to pre-
vent leakage over stands, supports, etc., and to prevent corona
and brush discharges. Care must also be taken to so place the
spheres that external fields are not superposed above the sphere
gap. This is likely to result, especially in the non-grounded
case, from a large mass of resistance units or connecting leads,
etc., in back of and in electrical connection with either sphere.
The 'error may be either plus or minus, as indicated for the small
1914] PEEK: THE SPHERE GAP 949
plates in Table XX. With the water-tube resistance this condi-

tion is not likely to obtain, as the tube may be brought directly
to the sphere as an extension of the shank.
The author wishes to acknowledge the assistan.ce of Mr. H.
K. Humphrey in laboratory work and in making calculations.

Peek The Sphere Gap As A Means of Measuring High Voltage

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Peek The Sphere Gap As A Means of Measuring High Voltage

Uploaded by

Copyright:

Available Formats

Presented at the 31st Annual Convention of

the American Institute of Electrical Engineers,

THE SPHERE GAP AS A MEANS OF MEASURING HIGH

GAP method of measuring high voltages is often desirable

discharge and broken-down air are eliminated. Humidity has

Calcu/lation of Curves: The gradient or stress on the air at the

cm. in. Non-grounded Grounded

2 0.787 41. 41.

5 1.969 81. 73.

8 3.150 102. 88.

cm. in. Non-grounded Grounded

0.25 0.098 6.5 6.5

7 2.756 118. 112

10 3.937 151. 135

17.5 6.88 201 168.

where eS is the spark-over voltage and g, is the apparent strength

22.5 8.85 297 268

40. 15.75 385 325

was found experimnentally that gs, the apparent surface gradient

The average gradient between these limits of separation iS4

=3 27.2 (1 + i) kv. per cm. max. (la)

g, 19.3 (1 + 0.54 )kv. per cm. eff. sine wave. (lb)

cm. in. Non-grounded Grounded

8 3.150 149 145

14 5.51 255.5 250

25 9.83 393 377t

40 15.75 537 489

55 21.65 633 561

70 27.56 705 608

The maximum variation from the average between the limits

shanlks, surrounding objects, etc., better distributing the flux

10.0 3.937 195 195

60.0 23.62 870 810

90.0 35.43 1107 1010

120.0 47.20 1290 1133

150.0 59.10 1412 1212

e g x (grounded) effective sine wave (3)

Non-grounded. One sphere One sphere grounded.

0.1 1.034 1.034 1.050

4.0 2.686 3.20 4.23

culated only when standard measuired curves cannot be ob-

EFFECT OF AIR DENSITY OR ALTITUDE AND TEMPERATURE

gs -27.2 a (1 + /aR) kvT. per cm. max.

where a is the relative air density

b = barometric pressure in cm.

The standard sea-level curves correspond to 76 cm. barometric

Practical Application. In order to avoid the trouble of cal-

Approximate Barometer Values of k at 25 deg. cent. for

6.25 12.5 25 37.5 50 75 100

0 76.00 29.92 1.000 1.000 1.000 1.000 1.000 1.000 1. 000

6.25 12.5 25. 37.5 50. 75. 100.

0.50 0.547 0.535 0.527 0.522 0.519 0.519 0.516

EXPERIMENTAL DETERMINATION OF AIR DENSITY CORRECTION

possible effect of spark-overs on the succeeding ones in the cask

Spacing Temp Press. Relative kv. gs

0.635 0.25 15 29.78 1.028 15.7 22.2 29.2 41.3

26 93 0.93 14.4 20.4 26.8 23.0

23.63 0.818 13.2 18.7 24.6 34.8

20.78 0.717 11.8 16.7 22.0 31.1

3.81 1.3 1.01 52.4 74.2 29.2 41.4

0.757 41.3 38.5 23.0 32.6

0.669 38.0 53.8 21.2 30.0

Spacing Relative kv. Bs