0
\$\begingroup\$

I'm curious if it is possible to build a spark-gap Tesla coil without using capacitors at all.

Basically, treating the primary and secondary as TX/RX antennas and cutting them to proper length so that the primary is producing a certain frequency and the secondary is tuned to it by using a wire length that is integer multiple of the length used in primary. Then using a HV source and spark gap to excite the primary at its natural frequency (half or quarter-wave, whatever it happens to be.)

Do you think it is possible to create spark gap Tesla coil this way so that it is into tens of MHz range? Do you see any pitfalls here?

Here's a crude drawing of this.

some crude drawing of it

Let's assume L1 and L2 are quarter-wave helical antennas, L2 being a smaller diameter one surrounded by L1. In order to tune the L2 to the frequency that L1 produces, you would normally need them to be the same length, but in this case we can't create a voltage step up configuration (we would need more turns in the secondary). But since L2 is also resonant at odd multiples of the quarter wavelength, like 3/4, 5/4, 7/4 etc, by having the wire length that is N times larger than that of a primary (where N is an odd number), the secondary should also be resonant on the same frequency while still providing means for stepping the voltage up.

Just to give you an idea, here's a video of my B&W 3034 antenna coil in spark-gap Tesla coil configuration (quarter-wave) using resonant caps. What I'm trying to do is remove these caps and tune this as if the two coils were antennas. The length of this B&W coil is around 19 meters, so if I manage to tune it to 5/4 wavelength, the primary will have to be 3.8 meters long which seems to be doable.

\$\endgroup\$
1
  • \$\begingroup\$ Essentially the same question was asked on ham.stackexchange.com here. \$\endgroup\$
    – rclocher3
    Feb 10, 2022 at 19:44

1 Answer 1

1
\$\begingroup\$

I'm curious if it is possible to build a spark-gap Tesla coil without using capacitors at all

Inevitably you will be using capacitance no matter how hard you try not to.

It's impossible not to use capacitance given that free space has a capacitive dielectric called the permittivity of free space (measured in farads per metre) and, that value influences the speed of light and, in turn, the speed of light influences how long an antenna is to become a tuned dipole or monopole.

\$\endgroup\$
8
  • \$\begingroup\$ I understand that self-capacitance will always be involved. I'm curious whether what I described is possible and we can use a bare piece of wire with no externals capacitors attached to do this tuning. \$\endgroup\$
    – elgroovy
    Feb 10, 2022 at 10:13
  • \$\begingroup\$ @elgr not using external capacitors and relying on parasitic capacitors produces the same result. Most tesla coils have no physical tuning capacitance on the secondary thus relying on self-capacitance of the coil i.e. it relies on parasitic capacitance. \$\endgroup\$
    – Andy aka
    Feb 10, 2022 at 10:36
  • \$\begingroup\$ Thanks for your reply. I should have been more clear. I'm referring to the capacitors on a primary side (MMC banks used in Tesla coils). So as I understand, there is nothing that prevents us to use a bare wire on the primary side too. \$\endgroup\$
    – elgroovy
    Feb 10, 2022 at 10:54
  • \$\begingroup\$ @elgr a tesla coil has a massive step up transformer at its heart and, I think it's highly unlikely that this can be achieved with tuned monopoles because they would have a similar length (due to the common operating frequency) and therefore there can be no step-up facility that is inevitably produced with magnetic coupling using two coils of vastly different turns. \$\endgroup\$
    – Andy aka
    Feb 10, 2022 at 11:48
  • \$\begingroup\$ what I envisioned to take advantage of is, since the primary is say a quarter-wave monopole and the secondary is also resonant at odd multiples of the wavelength like 3/4, 5/4, 7/4 etc, by having the wire length that is N times larger than that of a primary (where N is an odd number), the secondary should also be resonant on the same frequency while still providing means for stepping up the voltage. \$\endgroup\$
    – elgroovy
    Feb 10, 2022 at 12:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.