Colorado Springs Notes - October 1-31, 1899

Following items partly completed omitted for want of time:

  • Oct. 5. More complete description of photographs taken.

  • Oct 14, 22, 23 (page 5, page 6), 25 and 29 corrected results deduced from experimental data recorded.

  • Patent matters nearly completed:

    • a) Method securing excessive e.m.f. momenta.

    • b) Various ways of avoiding use of elevated terminal in power transmission etc.





Colorado Springs
Oct. 1, 1899

The new secondary was wound with 22 turns in all. The last turn was placed on top on porcelain insulators to prevent injury to the wood and breaking through on the last turn where danger greatest. It was evident that the coil could not stand the strain as there was only 1" between the turns. But the winding was tried for trial. The total length of coil was now disregarding the last turn which was on top of frame 27" approx. Now the length of the old coil was 63" (last turn excepted).

Colorado Springs
Oct. 2, 1899

As expected the coil wound before was unable to stand the strain and a different distribution of turns of the secondary was made. The ten turns nearest to the ground were left as before and the remaining were placed one turn in each second groove making the distance between the upper turns 2 5/8" and lower ones 1". Even with this arrangement, some of the upper and also some of the lower turns would break through.

A change was again made and only 4 of the turns the lowest were left 1" apart and all the others were placed in every second groove. The tests showed that the turns could now withstand the full charge. Perhaps up to six first turns might have been left at a distance of 1" but it was thought that the distribution was good enough as it was.

The last change reduced the number of turns to 18 (the uppermost not counted). Compared with the last form of tapering coil of 25 turns the period of the new was now approximately estimated from rough data below:

or very nearly equal.


Therefore the distribution of the turns will secure nearly the same period of vibration as in the old coil.

Colorado Springs
Oct. 3, 1899

Useful data in estimating possible errors due to the proximity of the ground or conductors or other causes.

Colorado Springs
Oct. 4, 1899

Test to determine more exactly influence of elevation on capacity of an insulated body.

A coil was wound to suit best the special conditions of this test. On a drum of 2 feet and 1 1 /4" diam. were wound 400 turns, wire cord No. 20. The insulation was very thick but of small specific inductive capacity. This wire with widely separated turns was used in order to make the capacity of the coil itself as small as possible compared with capacity of insulated body. The latter was in this case the one sphere of 30" diam. arranged to be elevated at will to a height up to nearly 40 feet from the ground.

The wire had a diam. approx. No. 20 B. & S.=0.032" or 0.8128 mm. The circumference of wire if solid would have been about 2.4 mm. Now one turn of wire was π x 25.25" or about π x 641 mm. The total length π x 641x 400=256,400 x π mm. Hence total surface of wire only π x 256,400x2.4=615,360 π or 6154π Now the diam. of the sphere was 30" or 76.2 cm. The surface π d2=18,231 sq. cm. approx.


The utmost we could take would be 1/2 surface of wire, that is 3000 π and this would be only 1/2 of surface of sphere. But other things considered, it would appear that the error due to electrostatic and distributed capacity of the coil itself would be small. This is however to be further investigated and allowance made for. It would be, of course, desiderable to entirely do away with the capacity of the coil to make the results of the observations rigorously true, but this will be hardly possible.

The number of turns on the coil was selected so as to be suitable to the apparatus used normally. The total length of the coil so far as wound was 57 1/8" or 145.1 cm, diam. of drum 25.25"=64.14 cm. To calculate self-induction we have then the following data:

This is a vibration far too quick, in reality it will be much slower because of capacity in the coil. Diagram shown illustrates arrangement used in experiments: The coil of 400 t. was excited by secondary of oscilator, and capacity and self-induction in primary were varied until resonance of free system comprising coil of 400 t. and ball of 38.1 cm cap, was obtained. This was evident from spark length and other indications, as streamers.

To be followed up.

Colorado Springs
Oct. 5, 1899

Test of secondary last pattern, 17 turns in all.

To ascertain the period a spark gap adjustable was used in the secondary from end to earth, as usual, and the capacity and self-ind. of primary was varied until maximum spark length in secondary and other indications showed maximum resonant rise.


Now a ball 38.1 cm. was added on the end and placed near the earth plate at a distance of about 3 1 /2 feet but very little affecting the vibration. This shows that distributed capacity in secondary is very large as before.

Capacity of one tank being 0.03816 mfd. Taking approx. Lp=56,400 and 13 turns box 46,800.

We have:

Latest secondary in receiving apparatus, No. 30 wire, 90 turns per layer, 35 layers Res. 424 ohms.

Tests showed best result with 5668 primary turns. 62 turns, No. 20 cord, are now used. Res. of cord 0.39 ohm.

Test with special coil for determining more accurately the law of variation of capacity with elevation. The coil with 400 turns No. 20 cord was used as previously and the adjustable ball of 38.1 cm capacity. The ball was normally above the extra coil repeatedly referred to, which on account of its great internal capacity was not used. Although unconnected some error was necessarily caused by the presence of the coil near the ball when the latter was near its lowest position.

A further error was anticipated from the influence of the roof at the points when the ball, which was being gradually lifted during the test, was nearest to it. The smallest distance or nearest point was about 11 feet. Nevertheless some action was bound to occur although the wood is very dry here and it is proposed in later tests to eliminate these errors as much as possible.


The roof was covered with some sort of tar paper the influence of which can only be conjectured at present. The most reliable data will be those obtained with ball at the highest points when it is above all structures.

The connections used are shown in diagram. The ball was lifted by steps of one foot each and the primary self-induction and capacity was adjusted until maximum resonant rise was observed on special coil, and adjustable secondary spark gap serving as analyzer, besides streamers which served even as a better guide. The results of the test are shown on the following table:

Table showing results. The capacities are all reduced to the same inductance so that as the primary capacity changes, the secondary is changed in the same ratio.

Table showing results of observations on influence of height in determining capacity of a sphere connected to the coil before described. The sphere of 38.1 cm. electrostatic capacity was gradually elevated and the period of vibration determined for every position of the sphere, by varying the capacity and self-induction of primary circuit.


As the self-inductions of both circuits remained the same the secondary capacity varied exactly as the primary.

Tuning was very sharp from line 27.66 to end. The arc in the primary was getting stronger and stronger. This probably because of lower frequency. The flaming of primary arc also because of this. For same reasons secondary arc was getting continually weaker as the ball was elevated.

  • m read measured data

  • c read calculated data

  • r read reduced data.

From the sign* two primary turns were used in series as the vibration got too low and not enough primary capacity was available with one turn. This was thought preferable to tuning to harmonic. From the figures on the side of 25.66 line it will be seen that the capacity with two turns was not exactly one quarter of 8, that is 2 tanks, but less.


This was probably due to the fact that the tanks are not exactly equal and more so because when two primaries are used in multiple as one single turn the self-induction is less than 1/4 of that of two turns in series. To reduce to same self-induction as with one turn the capacities obtained with two turns should be multiplied by 6.64/6.64.

Colorado Springs
Oct. 6, 1899

Measurement of coefficient of self-induction and mutual induction.

Secondary last form 17 turns of which total 16 turns disposed on frame described before, one turn in every second groove, one groove near the primary free. Average of readings: p=880 Voltage across secondary E=122.5 V, I=13.8 amp.; Res. R=2.804 ohms. From this

Colorado Springs
Oct. 7, 1899

The secondary was again changed by displacing the 15th and 16th turn, the rest remaining as before: namely, 14 turns, a turn in each second groove and the 15th and 16th turn each in every third groove and the top turn on porcelain insulators as before. This changed very little the constants of the circuit.

The test was now made closed with the following results:

Tuning of extra coil to suit vibration of the secondary latest design, as specified on another sheet today.

An attempt was made to get the vibration of the coil with ball elevated exactly as that of the secondary. To adjust the vibration the ball was elevated to various positions and soon the adjustment was reached.

With an elevation of 2 feet lower than the highest point (35.66 feet from the ground) that is 33.66 feet from the ground the maximum resonant rise on coil, as evidenced by the spark, was obtained with 7 2/3 tanks capacity on each side and 13 turns in the self-induction coil. Lowering the ball just a trifle and making capacity on each side 8 tanks the resonant maximum rise in coil took place with 10 turns in regulating coil. This was almost exactly the vibration of the secondary as previously ascertained.

The two were now connected in series and discharges on a spark gap of something like 12 feet were obtained, although the W.T. was not strained to the utmost. When the spark wire was taken off and ball with a rubber covered wire No. 10 specially prepared left alone, streamers formed on top of ball, but little as the wind was blowing. The ball being disconnected and rubber wire alone left, the streamers were very fierce reacting sometimes 1618 feet. Best results were with 6 turns in coil in the last experiments.

A more careful tuning of extra coil without ball, only rubber covered wire or cord No. 10 (which was referred to before), the tip of cable being brought out about 2 feet and inclined to horizontal about 45. This wire to prevent streamers was specially made and was covered with rubber. Composition Habirshaw, 40% pure Para. The tuning of the coil alone with the self-induction coil specially wound in series, gave:

The coil used in series with the extra coil was one especially adjusted so as to give the same vibration to the system as when the extra coil was used alone and with ball on top at highest point. The coil has 160 turns and is wound on a drum 2 feet in diam. with wire No. 10, same as used in the secondary and extra coil.

Now the secondary with a ball on the end and elevated at a height of 2.66 feet from the ground gave also exactly, with 8 tanks in primary on each side and 6 1/2 turns in regulating coil, maximum effect. When the two were connected in series the display was magnificent, sparks flying to the ground a distance of over 16 feet. Their curved paths stretched out would be certainly 24 feet long. This was not the maximum of the power of the apparatus as the spark in the primary could have been still lengthened without difficulty.

Colorado Springs
Oct. 8, 1899

Experimental data in connection with tuning of extra coil and secondary as recorded yesterday.

Now the length of wire in the secondary was 803 meters, namely 17 turns each of a diameter of 15 meters.

The total length of wire in the extra coil circuit was: the extra coil itself 889 meters, namely 149 turns each of diameter of 1.9 meters plus special coil inserted in series: 307 meters, namely 160 turns each of diam. of 0.61 meter.

This special coil was used when the ball on top was not employed as capacity and the coil was so adjusted that the vibration was the same without the ball as with the ball and without the special coil; it being understood that the ball was at its highest position in such case. The total length of wire was therefore:

with all connections the length was increased to 2030 meters (17 meters rubber wire on top; 13 meters lower connecting wire).

Now capacity in primary was 4 tanks=4x0.03816=0.15264 mfd.


Colorado Springs
Oct. 9, 1899

Today an other effort was made to ascertain more closely rate of increase of capacity of a body with the height above the ground. The same adjustable sphere was used and to eliminate some influences which might cause errors the special coil, which was rewound with same wire as before and had now 404 turns, the wire being wound a trifle higher, was connected to the first turn of secondary of oscillator, that is the turn first from the ground.


This gave a small initial e.m.f. and reduced the ei rcr due to capacity very materially. (Here the distributed capacity is meant). Also, since the pressures developed were much smaller owing to small initial pressure on the special coil the streamers did not appear and did not therefore complicate the observations as in some previous cases.


The plan of connections is clear from the diagram below.

The object was specially to obtain a number of values which were as closely determined as possible and with the ball in positions entirely above the building. Only three values could be obtained owing to darkness setting in, but these seemed fairly close as the tuning was done over and over with same results.


These were as follows:

Now since the capacity in the primary was the same the capacity in the secondary varied directly as the self-induction of the primary. The above figures show that from 33.66 feet to 34.66 feet, that is an elevation of one foot, the increase was 9.1%, while for the next foot higher it was nearly 9.9% on the average, say, 9.5%. At this rate the increase of capacity of the elevated sphere with the rise from the ground would be greater than before found.


The absolute rate of increase can be approximately estimated from the period of vibration. As before found the special coil with 400 turns had a self-induction of 44,772,000 cm. Now, however, with 404 turns this would be increased about 1% so that the self-induction would now be

But to this should still be added the self-induction of one turn of the secondary and wire leading to the ball and also the wire leading from the bottom of the special coil to the first turn of the secondary. The total length of these three wires is 240 feet and this is about 12% of the total length of the wire in the special coil which is 2854 feet.


But inasmuch as the one secondary turn was very close to the primary and inasmuch as the other two wires were not coiled up, the self-induction of these wires was comparatively small, estimated a little over 200,000 cm so that the total self-induction was with fair approximation: 44,500,000 cm, or about 0,0445 henry (calculated). Now, with a ball of 38.1 cm capacity the period of secondary would have to be:

This capacity c is now the actual capacity of the sphere at an elevation above considered, that is 33.66 feet. Namely, the total capacity was that of the sphere at that elevation that is c plus the ideal capacity of the coil derived from the computation of the primary vibration.


This latter capacity has been called C. The inductance of the special coil being as before found, 0.0445 henry, and there being resonance under the conditions of the experiment at the elevation, named Tp', gives the value for the secondary period.

Now at another elevation, say 34.66 feet, we shall have similarly

and from this: 1.092x(403.75+c)-403.75=c', but c being 38.1 we find for c'=78.75 cm, which result shows that by lifting the ball from an elevation of 33.66 feet to 34. 66 feet or one foot higher the capacity has been increased from 38.1 to 78.75 cm, or nearly 106.7%. Similarly we find the increase from 34.66 to 35.66 feet by computing the primary vibrations at these elevations.


By analogy to the previous we have

And now we have:

The value at one foot lower was, as before found 78.75 cm, therefore by lifting the sphere from 34.66 to 35.66 feet, the capacity was further increased by 126.5178.75= =47.76 cm, or about 125%. The value which would correspond to the mean would therefore be about 116% per foot. The method followed contains still some possible errors.


One of them lies in the assumption that the capacity of the sphere was 38.1 cm at the starting point. Also there may be an error in the estimation of self-induction of the turns of the regulating coil.

Colorado Springs
Oct. 10, 1899



Colorado Springs
Oct. 11, 1899

Photographs of streamers were taken late last night again and at an early hour this morning under the following conditions.

First two plates exposed to ten flashes, 1/2 second duration each. These flashes issued from the tip of rubber-covered cable or wire No. 10 which was on top of the extra coil. The tip was inclined about 45 degrees to the vertical and pointing downward. The full front view was taken. A curious observation was made.


One of the large streamers, about 22 feet long, disappeared at that length for a space of about a foot and continued again after that for a distance of about 2 feet, so that the total length of it was about 25 feet with a dark interval of one foot.


Evidently, the current passed for a distance of a foot through air or dust particles which were better conducting and the path was of a greater section in all probability. Perhaps the air on that spot might have been electrified in such a way as to produce the phenomenon.

The next experiment was made with an exposure of two plates to about forty flashes, the view being the same as before.

After this two plates were exposed to but a single short Sash about one second, the view being still the same as before. Now a round sheet zinc disk was fastened to the tip of the wire and two plates were again exposed, there being about twenty flashes. Next, the coil was turned and a side view taken with about forty flashes, two plates being again exposed as in all previous cases, two cameras being used for the sake of safety.


Upon this the zinc disk was taken off and a ball of 4" diam. fastened to the tip of the No. 10 rubber-coveted wire. A long exposure of about 50 flashes was again made. The streamers were as expected a little stronger from the ball then from the point as the breaking out took place at a greater pressure. Two plates were used also in this instance.

The next experiment consisted of taking an impression on two plates of the secondary alone in resonating condition. The phenomenon was beautiful to an extraordinary degree. Not only did the top wire glow but from the under wire (turn next below) a steady sheet of streamers of very fine texture issued of an area which must have been many hundred square feet.


The free end of the secondary had the ball of 38.1 cm connected to it at a distance of 32" from the ground but owing to the large radius of curvature sparks did not leap from the ball to the grounded zinc plate below though the distance was small comparatively. During the experiments a short but thick stream issued from the free terminal of the extra coil which had its other end connected to the ground and was thus excited through the vibration of the secondary, having, as stated on a previous occasion, the same period of vibration.


The color of the light issuing from the secondary wires, particularly in the neighbourhood of the condensers, was remarkably blue, and should affect the plate strongly, though the intensity was not great. The switch was thrown in fifty times, the duration being about 1/3 of one second, possibly 1/2.

The last experiment consisted of establishing the resonance of the extra coil and secondary in series connected and with the ball of 38.1 centimeter on the free end of the extra coil, the ball being at its lowest position, 20.66 feet from the ground.


The other ball of the same size remained as before, connected to the end of the secondary where it was joined to the lower end of the extra coil. As there was great danger of inflaming the roof the power was somewhat reduced, but the display was wonderful in spite of this. This was the most significant experiment showing streamers from the ball of 38.1 cm capacity from which is evident the enormous tension, as well as the inconceivable rate at which the energy was delivered in the vibrating system.


Forty flashes were made and afterward the background was illuminated by the arc of the primary circuit to complete the picture.


Colorado Springs
Oct. 12, 1899

Measurement of inductance of 404 turns coil used in determining influence of elevation.

The measured value should be smaller than the calculated but not so much. The internal capacity may be responsible but very likely the current was not quite exactly measured. Corrections to be taken after calibration.

Colorado Springs
Oct. 13, 1899


Measurement of inductances of extra coil and secondary, latest design:

Colorado Springs
Oct. 14, 1899

Determination of the natural period of the secondary with and without capacity on free terminal, also of the extra coil and coil of 404 turns used in investigations before described.

The tuning was effected in a manner later to be more fully dwelt upon, which secured closer readings than when exciting, as in some experiments before made, by the primary current or from a turn of the secondary of oscillator.

The excitation was effected in these tests by connecting directly one of the terminals (the lower) of the coil to be tested to one of the terminals of the primary condensers (the one connected to the tank of W. Transformer).

The results of the test are given below:

Colorado Springs
Oct. 15, 1899

Continuing the considerations made Oct. 9 on the influence of elevation upon the capacity and taking, instead of the calculated value of the self-induction of the coil used in the experiments, the value ascertained by experiment which, with corrections for one turn secondary and connecting wires, may be put at 0.04 henry, we have, assuming now the coil to have a capacity entirely negligible:

This value is but little smaller than that before found. From this result it would then appear that by lifting the sphere from 33.66 to 34.66 feet the capacity was increased from 38.1 to 75.53 cm, or approximately 98.3%, a trifle less than before found.