The project require a compact plasma arc generator, which needs a lot of current so I chose two FETs driving a transformer.

The problem is I cannot find commercially available high-voltage transformer with center tap. So I found two methods for driving a transformer without a CT:

enter image description here

The first one is cheap and dirty ZVS, but I haven't built ZVS without CT before. It seems this circuit will short through L1 and L2 when either FET is open. Also, the oscillation can sometimes fail to start in ZVS.

The second one is a half-bridge driven by self-oscillating gate driver (not shown in the diagram). The back-EMF is a major concern and I'm not sure if I placed the snubber diodes correctly.

Are there better ways to drive a pulse transformer without a CT?

  • \$\begingroup\$ In your second circuit, why do you need 2 capacitors? Can't you just use one capacitor and AC couple the coil to ground? \$\endgroup\$ Commented Jun 28, 2022 at 13:01
  • \$\begingroup\$ @BruceAbbott split capacitors approach brings some advantages but it's out of topic for now. \$\endgroup\$ Commented Jun 28, 2022 at 14:32
  • \$\begingroup\$ The first circuit is popular amongst the low-voltage self-oscillating induction heater circuits. The input voltage must be stepped for the oscillation to start. Some intrinsic asymmetry causes a preference for one transistor to switch above the otehr. \$\endgroup\$
    – GLK
    Commented Aug 3, 2023 at 20:43

1 Answer 1


Both circuits are LLC (two L's and one C) tanks:

  • First one is series-parallel LLC tank (*)
  • Second one is series LLC tank

The second one is quite popular amongst the power electronics industry.

Remember that you still need an extra inductor for the second circuit unless you rely on the primary leakage.

(*) I'm not sure about if this one can properly work. L1 and L2 should be quite small compared to the magnetising inductance of the primary maybe.

  • \$\begingroup\$ Can you specify the inductor usage in the second circuit ? \$\endgroup\$
    – 7E10FC9A
    Commented Jun 28, 2022 at 16:17
  • \$\begingroup\$ @7E10FC9A basically, what you are building is a Royer oscillator using a non-centre-tapped transformer (Never seen an example, BTW). Without the inductor the circuit can still work in voltage mode but the series inductor will change the operation to current mode and the stress of the switches will be reduced (the primary magnetising inductance and the capacitor will resonate in either mode). I can't give exact numbers for neither the series inductance nor the primary magnetising inductance but you can study Royer oscillator for more technical details. \$\endgroup\$ Commented Jun 29, 2022 at 6:50
  • \$\begingroup\$ @7E10FC9A One other thing I'd like to mention about is driving the high-side switch: I hope you are aware of the fact that the top switch needs a bootstrap or a similar approach for proper high-side drive. \$\endgroup\$ Commented Jun 29, 2022 at 6:50
  • \$\begingroup\$ there is no oscillator in the second circuit. the circuit is driven by a self-oscillating gate driver IC, which includes the bootstrap. \$\endgroup\$
    – 7E10FC9A
    Commented Jun 29, 2022 at 11:47

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.