This is a follow-up project to my first induction heater, which I built for my junior year AP Physics E&M midyear project.
I wanted to have only one circuit board in this project, so I integrated a TL494 VFO, ‘HC14-based PWM generator (for average power control), switching synchronization flop, and IGBT brick driver in one design:
This is the board file for the first prototype control board, it contains several errors that I plan to fix in the next board I etch. For example, I accidentally connected the ‘HC109 flip flop’s supply to +15V instead of +5V in my original schematic, so I was forced to rip up the trace once I’d soldered it and manually wire it. Also, I wired the brick drive MOSFETs to the +15V rail, instead of a separate +24V rail, so the output to too low to drive the brick effectively and there is significant heating of the IGBT half bridge as a result.
The reason I used a giant string of 1uF blocking capacitors instead of one large capacitor is because I just happen to have a bag of small 1uF 50V ceramic capacitors that work well for this application. The space between the control board logic and brick drive is for an intermediate GDT – something I hope to eliminate in my next design by using a totem pole instead of an intermediate bridge:
Construction and Testing
The rev. 1 control board is connected to a half bridge brick, which has an 8uF 800V snubber, 2200uF 250V filter, and a pair of 20uF 600V half bridge capacitors. I decided not to use a voltage doubler to reduce parts count and allow the power supply to fit directly on the brick:
“Flying” gate resistor construction:
Preparing the primary coil water block that will go on the primary capacitor with flux for torch soldering:
The other capacitor water block, soldered:
The water blocks were cleaned with acetone and polished with a belt sander, then bolted to the primary capacitor:
The new primary system vs. the old one from Induction Heater 1.0:
The work coil installed in the water loop with compression fittings:
The liquid cooling pump connected to the primary loop – instead of using a large radiator/fan to dissipate the heat, I just rely on the giant thermal mass of several gallons of water in a large bucket, I’ve had no problems with overheating so far:
A low power (only ~1kW) test heating a large steel bolt:
In-tune primary current waveform:
So far, low power tests have been a complete success (the work coil stays at room temperature with liquid cooling!) – In order to run higher powers, I either need to find a bigger variac or implement my PWM power controller.