The spark gap in Tesla coil operation acts as a Switch. The gap may contain only stationary electrodes separated by a small distance or can contain one or two electrodes and a rotary break wheel. The Project Tesla arrangement contained two spark gaps, a rotary break and a stationary spark gap. The rotary break or break wheel looks somewhat like a saw blade. The break wheel used in this case was approximately 18" in diameter, had 20 teeth, and was made of a high quality grade of aluminum. Horizontal tungsten electrodes were placed opposite each other on either side of the break wheel. The electrodes were separated from the teeth by a distance of about 0.125 inches. The break wheel was driven at 7110 rpm by a 40 hp turning at 3555 rpm. An insulating belt and 2:1 step up pulleys were used to obtain the desired wheel speed. The break wheel and electrode assembly were well shielded by a one-quarter inch steel housing.

The stationary gap may more accurately be referred to as a quench gap. The quench gap is in series with the rotary break and is pressurized with sulfur hexaflouride. The quench gap contains a tungsten electrode separated from a metal base plate by a distance of 0.125 to .500 inch and is housed in a chamber made from a ceramic insulator taken from a large oil filled transformer. The sulfur hexaflouride (SF6) in the gap is at a pressure of between 20-60 psi and may also be mixed with air to provide a range of mixtures. SF6 is a gas that has many advantages as a dielectric medium. Two properties of SF6 taken advantage of in this application are dielectric strength and arc quenching capability (1-4). At atmospheric pressure, SF6 has three times the dielectric strength of air. At 10 psi the dielectric strength is equal to oil. And at 50 psi, the break down voltage between a one-inch tungsten rod separated from a one-inch tungsten sphere by one inch is 650 kv (1-5). Arc quenching in SF6 takes place in 1/100th of the time that it takes in air. The quench gap is used to insure that the discharge process initiated by the rotary break is as short as possible (50 to 100 microseconds).

A Tesla coil is nothing more than an air core resonant transformer with a primary and secondary. The input to the primary is a series of pulses of relatively high voltage which can be from 7 kv to 70 kv. The pulses are derived from the discharge of a capacitor bank that is triggered by a rotary break. Future designs will incorporate thyratrons and remove the inherent mechanical dangers of the rotary break. The amount of power that the coil will use is dependent on the energy stored in the capacitor bank. The capacitor bank consisted of five 1 uf capacitors connected in series. The SI unit of energy is the joule. The amount of energy stored in the capacitor bank during alternating current charging is equal to one-half of the quantity of the total capacitance times the peak voltage across the capacitor bank squared. Energy in joules is equal to (1/2) (CV²)

For the case under discussion, the capacitance was equal to .177 uf and the rms voltage rating of the high voltage step up transformer was 45 kv. The peak voltage was 1.4 times 45 kv or 63 kv. Using the above formula, the maximum energy that could be obtained from the charged capacitor bank was 351 joules.

The total power requirements for Project Tesla can be determined. With the capacitors containing 351 joules and charged to 63 kv and break wheel with 20 teeth spinning at 7110 rpm, the input power could have a maximum value of 831 kw - almost a megawatt! This power consumption was not verified due to lack of instrumentation to measure the output of the generator.

The generator was a Magna One, star wound, synchronous AC generator driven by tandem V-8 Detroit Diesels that were manufactured by Marathon Electric. The output power capability was stated by the manufacturer to be 625 kw at 0.8 power factor (with derating for altitude). The top end of the meters on the output line was 600 volts and 1000 amperes. With the generator up and running, the meters would peg and the diesels would slow down when the Tesla Coil was switched on line. These figures indicate that the system was drawing well over 600 kw!

Now, the curious question was why would a system that required over 600 kw on a 150 kw generator in Utah not run in Colorado. The Utah experiments had been driven off a single leg of a 150 kw delta wound generator. A generator of the same type was obtained for further tests. It was not delta wound but was star wound with the sane rating of 150 kw. The difference between delta and star wound generators is that star windings yield higher voltage and lower amps. The reverse would be true of delta wound generators. The other difference between the Utah experiments and the Colorado experiments were the number and size of the transformers between the Tesla coil and the generators. A number of configurations were tried with the basic obiective - choke the current flow. The Utah experiments had utilized a number of high impedance transformers which were hooked in a step up/step down/step up configuration. The Colorado experiments initially used four 75 kva power transformers (240/480x13,800 VAC) and a 45 kv step up transformer (240x45,000 VAC). Two of the 75 kva transformers were paralleled at the output of the generator and these were connected to two paralleled 75 kva transformers connected to the input of the 45 kva step up transformer. The 240 volt output of the generator was connected to the 240 volt side of the first two 75 kva transformers. The 13,800 volt output of the first two paralleled transformers was input to the 13,800 volt side of the second two paralleled transformers. The 240 volt side of the second set of paralleled transformers were connected to the 240 volt side of the 45 kv transformer. The output of the 45 kv transformer was the input to the capacitor bank, rotary break, and the primary of the Tesla coil. Eventually several of the 75 kva transformers burned out. This can easily be understood if the Tesla coil is using close to a megawatt as shown above.

A number of combinations of transformers were utilized in an effort to recreate the magical setup that worked in Utah which had contained even more step down/step up transformers. Transformers of 667 kva were tried in a number of configurations. One of the 75 kva transformers was disassembled, its secondary stripped off, and used as a variable inductor and current choke in series with the generator. The final circuit consisted of three 333 kva, 240vx480v/50kv transformers connected in a step up/stepdown/step up configuration. In order to get a feel for the size of this arrangement, you must realize that each transformer weighs 5600 pounds, contains 266 gallons of oil, stands 8 feet tall, and with cooling fins is about four feet on a side. The set up apparently gave the best results with the 480 volt input to the final transformer connected to the 240 volt line.