I have an air coupled transformer consisting of a primary L C resonant circuit fed with feedback transistors and a secondary L C circuit feeding a bridge recitfier to power a small circuit over an air gap. It is almost identical to this circuit: https://en.wikipedia.org/wiki/Royer_oscillator

I also use a FET shorting the rectifier on the secondary side that is triggered occasionally to pass a signal back to the primary circuit. It looks like this example, the secondary coil and resonance capcitor is on the left, in this case a forked light barrier is used as source of the signal to be transmitted: enter image description here The lowered resistance leads to a small change (about 2%) in frequency which can be sensed on the primary side. That worked well in several places, but if the constant power consumtion is higher, it sometimes fails: While the signal is triggered, it tends to overload the primary circuit, giving an trashed non-sine waveform.

However, by changing the capacitance C4 mounted to the secondary resonator, the frequency can be changed easily over a broad range of about 30% without much additional power consumption.

So I wonder if there any way to change the AC capacitance with the FET or some other simple circuit?

  • 1
    \$\begingroup\$ A varicap maybe? \$\endgroup\$ – Arsenal Nov 16 '15 at 16:11
  • \$\begingroup\$ It seems varicaps are only useful for small AC voltages with almost no power, that are then amplified. I need to tune the high power secondary oscilator (about 30V peak-to-peak). \$\endgroup\$ – dronus Nov 16 '15 at 16:56
  • \$\begingroup\$ What range of capacitance change do you need? Is driving a classic air-spaced tuning capacitor with an RC servo an option? \$\endgroup\$ – Brian Drummond Nov 16 '15 at 17:41
  • \$\begingroup\$ You could put in a bank of 8 capacitors each 2x the previous one, and switch them on and off similar to a DAC. So, 2.2, 4.7, 10, 22, 47, 100, 220, 470. This would give you 256 possible capacitance in more or less uniform steps. This is just a wild idea. I don't really understand your circuit. So I haven't put in any values for the caps. \$\endgroup\$ – mkeith Nov 16 '15 at 18:50
  • \$\begingroup\$ I just need to capacitance levels (binary signal). Using a motor driven caps is not an option, there is not enough space (flat circuit board) and I don't like worrying about mechanical reliability. \$\endgroup\$ – dronus Nov 16 '15 at 22:19

Use mosfet to switch in or out the extra capacitance .Its been done for years on CRT monitors .If you use 2 fets they both can lowside switch making your circuit simple to produce.

  • \$\begingroup\$ How would you do it? The circuit is fed by a rectified and regulated voltage from the secondary coil, so it's GND is "somewhere in the middle" of the secondary coil AC voltage. I think the FET would be reversed in the process and destroyed? \$\endgroup\$ – dronus Nov 22 '15 at 14:33
  • \$\begingroup\$ It can be done because a cap conducts AC and not DC so a mosfet is good .When the Fet is off the cap is connected via the reverse diode thats in powermosfets.This doesnt matter because the cap charges to the peak and stops charging .So you can have simple lowside switches that switch caps that are connected to the AC side of the bridge. \$\endgroup\$ – Autistic Nov 22 '15 at 18:45
  • \$\begingroup\$ Guess I don't get it.. The cap is currently in parallel to the secondary coil. Do you suggest to add another cap via FET there, or should the FET connect another cap to "ground" of the DC circuit behind the rectifier? The resonant cap in parallel to the coil need to be very high quality for a high resonance. I doubt that a FET that drops some of the energy would allow strong resonance. \$\endgroup\$ – dronus Nov 25 '15 at 13:11
  • \$\begingroup\$ I guss the reverse diode would connect any other wave, and in on state, the FET would then connect all waves? Sounds kind of plausible... however, the reverse voltage would be 15-30V maybe and no large losses are allowed as they would brake resonance. \$\endgroup\$ – dronus Nov 25 '15 at 13:14
  • \$\begingroup\$ You are breaking resonance with C3 anyway,I hope that you are doing a series resonant current based scheme which is what I did.AIf you must for other reasons do your paralell scheme then place a choke between the bridge and C3 .The choke should have high impedence at twice the resonant frequency so as a guide try 10 times the inductance of L2 it is not critical. \$\endgroup\$ – Autistic Nov 25 '15 at 19:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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