You are here

High Voltage Power Supply Overview

Overview: In many experiments it's necessary to utilize high voltage. I've created this high voltage power supply using a flyback transformer I salvaged from a discarded CRT monitor, and I'm driving it with a 555 timer and a power MOSFET (FS10KM). The timing of the 555 chip is controlled by the resistors R1, R2 and the capacitor C1. The MOSFET enables the low power output of the 555 timer to gate a high power throughput for the transformer. The MOSFET is designed to be very fast switching, and because the output of the 555 is a square wave, the magnetic field in the transformer forms and collapses very suddenly, which has the intended effect of inducing the current at very high voltages in the secondary, but it also induces some currents in the primary, and these spikes can be high enough voltages to ruin the 555 chip. So in order to protect the circuit from this, there is a resistor that reduces the voltage from the primary and the 555 supply voltage, and two capacitors of different capacitance (C2 and C3) connected to the 555 supply voltage and ground to act as a filter. Also, I know that the transformer will have a substantial amount of inductance which should serve to limit the current, however since I am using a car battery, and I want to be safe if for example the driver fails and the current is left on through the transformer, so I've added a resistor of low resistance in series with the transformer (R6, ~ 2 ohm). Additionally my circuit has an LED to show when the power is on, and I've also at times connected a speaker to the output of the 555 timer to hear the tone for the frequency produced by the 555.

Parts List:
  • Flyback transformer from a CRT computer monitor
  • FS10KM - Nch Power MOSFET high-speed switching use
  • NTE955 - 555 equivalent timing IC
  • R1 & R2 - 1000 Ohm
  • R3 & R4 - 47 Ohm
  • R5 - 1000 Ohm
  • R6 - 2 Ohm
  • C1 - .01 uF
  • C2 - .01 uF
  • C3 - 220 uF
  • F - 15A

Since making the video and schematic, I have played a bit with the timing resistors and capacitor for the 555. I put a smaller cap (.001 uF), and put a POT between R1 and R2, with the middle connected to pin 7, as illustrated in the following picture. This allows me to tune the duty cycle in a crude way, which also affects the frequency, but allows me to tune the circuit to get the best arc. As a result of making the duty cycle more efficient or generally running at a higher frequencies the arc becomes more like a flame and has more current, whereas low frequencies or a less efficient duty cycle give the same arc length but sharper finer discharges (less current).

And here are some oscilloscope readings from the power supply:


Hello, I have built this circuit but the MOSFETs keep on dieing.

Is there a way to stop this from happening?


Yes heat is an issue for that MOSFET. I actually salvaged that MOSFET straight off of the CRT motherboard I got the flyback from, and it was already thermal pasted and mounted to that giant aluminum block you see in the video.

So my suggestion is try a heatsink.

The heat sink I am using is HUGE yet it still gets hot lol. Do you think lowering the value of the gate resistor would help the MOSFET switch more efficiently and using some sort of TVS to clamp the back EMF spikes would help the MOSFET last.

BTW here is what I managed to achieve using this driver. BTW I changed the two resistors to two 50k pots so I could tune it to the "sweet spot".

I also audio modulated it via pin 5 of the 555 chip and a 100nF capacitor.


You 1 upped me :D

I like the audio modulation, very nice. I tried the pot's for tuning too :) Good call.

As for the heat sink problem, you could try to limit the input amperage, I think my breadboard layout is so ghetto there's a lot of contact resistance that's not really identified in the schematic. Can you show a pic or video of your circuitry as well?


Hi again,

Here is my circuit

Everything is soldered onto a piece of strip-board.

I think the problem is that the two 47 ohm resistors are limiting the amount of current getting to the MOSFET base too much there for the MOSFET stays in the linear region for longer and generates heat. When I have another try at this circuit I think I will use a separate supply for the 555 and ditch the 47ohm resistors all together in order to give the gate more current, perhaps even use a push-pull buffer transistor stage.