In my previous series about VFD filament driver, I built a power supply based on a few operational amplifiers and a transistors. I actually implemented this design in a fully fledged PCB:
However, the method used is not perfect. I can still see some biasing issues proper to the VFD technology: the bottom of the tube is slightly less luminous than the top. To avoid this issue, I decided to build a high frequency square wave power supply based on a H-Bridge:
There are many H-Bridge designs out there, but I just needed a very simple design so I stopped on the one above. An ATTINY13A drives both the SQW1 and SQW2 high and low through a bit banged square wave. I won’t get into the details of a H-bridge design but you can’t drive the SQW1 and SQW2 high at the same time or it creates a short circuit. Instead you have to turn them on and off alternatively.
Slow transistor woes
At first I designed the attiny program to oscillate at about 60Khz.
This results in a broken design. The circuit uses 800ma without any load. I confirmed this result on the oscilloscope and came to the conclusion: the transistors I chose are too slow. Because the transistors have a turn on/turn off time, the switch is not immediate and there’s a brief period of time where the H-Bridge generates a short circuit. Slowing doing the design to a few hertz confirms the problem with the switching speed.
So I have learned a valuable lesson here: be very wary of your transistors Ton, Toff, fall time and storage time! I have now ordered faster parts to see if this design can be salvaged. Stay tuned for the next adventures!