1

enter image description here

I'm using a ZVS driver to create a spark for a plasma speaker. The circuit above is the one I'm using for my project however I have run into some issues whihc I do not understand. First of all the circuit works when tested in LTspice but I had to use a startup to get the oscillation or else the MOSFETs did not switch.

However I made a prototype to compare the simulation results. When doing so I got no plasma spark, but when I disconnected the powersupply cables and held it on to the connection it made a spark. This made me think that the powersupply has a slow ramp up maybe. So to counter this I had a switch before the ZVS driver hoping this would solve the issue but it still does not work. All connections seem to be okay since a spark can be created. So I am confused how to solve this issue now. Perhaps there is some advanced circuit which I can add before the ZVS driver.

The components are not final, I was in a rush to have a design for a school project and from different ZVS drivers made the component choices.

1 Answers1

1

It seems a problem of power supply impedance: the real power supply \$Z_{V_1}\$ (using the notation in your LTspice schematics) seems so high that it prevents the oscillation of the prototype circuit to start. In order to lower it, I advise you to mount a bank made of large electrolytic capacitors in parallel with small high frequency decoupling capacitors (in the schematic below I've chosen \$C_3=C_5=1000{\mu\text F}\$ and \$C_4=C_5 = 100\text{nF}\$ but other values of the same order of magnitude will possibly do) as near as possible tu the power supply connections of \$L_1\$, \$R_3\$ and \$R_4\$ and the source terminals of the MOSFETs M1 and M2 in order for the power supply to have an almost ideal output impedance at the frequency of oscillation of the circuit.

schematic

simulate this circuit – Schematic created using CircuitLab

Daniele Tampieri
  • 3,161
  • 4
  • 14
  • 29