The IGBT Driver Board is ready!

The completed PCB arrived from Ragworm on Saturday, and I’m quite pleased with the end result. The board arrived vacuum packed into bubble-wrap, with a couple of free LEDs thrown in (and some orange paper hearts!) Here’s a view of the top surface:

PCB Top Surface

I think this  looks pretty good for a first version. The silk-screen looks very slightly offset (see JP2, JP3) but nothing to worry about. Here’s the underside of the board:

PCB Bottom Surface

Today I made a start on assembling the board. I was relieved to find that all the components fitted fine, with no obvious layout errors. Unfortunately, I was missing one component (5mm pitch screw terminals for the mains input), so it’s time to place another order! Here’s the partially assembled board:

PCB Assembly

2 thoughts on “The IGBT Driver Board is ready!”

    1. Hi Rowan

      You’re right, the isolation wasn’t why I chose this model: the reasons for selecting the VTX-214-003-105 were because it was small, inexpensive, has low ripple and noise on the 5V output voltage and 70% efficiency (i.e. won’t run hot). It’s not just a transformer, it’s a full switch mode regulated 5V power supply with output filtering, in an encapsulated package.

      The 5V output has to be referenced to Neutral in order to be able to switch the IGBT, so that’s unavoidable, but the 5V supply is only used on this PCB, and the PCB is completely isolated from the rest of the system with Opto-Isolators and mounted in a plastic enclosure.

      I did look at a number of other ideas before finding the VTX-214-003-105, but none of them are ideal. For example, instead of this module we could have used a small transformer, bridge rectifier, linear voltage regulator and filter capacitors, or we could use a capacitative dropper circuit with a zener diode and filter capacitors. If this was mass produced commercial product, they would probably use a capacitative dropper to save component cost (not because it’s the best solution, but definitely the cheapest way to do it).

      The most obvious would be to use a transformer, which would require: 1) the transformer, 2) bridge rectifier, 3) 7805 linear regulator, 4+5) electrolytic filter capacitors. So that’s 4/5 components in total and would be physically quite big with a transformer (that alone might be bigger than the VTX-214-003-105). It wouldn’t be much cheaper either.

      If we used a capacitative dropper, it could be made a similar size to the VTX-214-003-105. You would need 1) input resistor, 2) beefy 400V metal film capacitor, 3) discharge resistor across capacitor, 4) bridge rectifier, 5) Zener diode, 6+7) output filter capacitors. So that’s 6/7 components (but smaller ones). If we wanted to make sure it is safe in case of individual component failures, we would need to add some more parts. There would probably be a fair bit of ripple, and it’s not very efficient so the Zener might run warm. However, the component cost would be cheaper, and this is most likely what would be used in a mass produced “real world” product. It could be a good solution if designed carefully.

      There’s a bit more detail on the electronics here:
      http://int03.co.uk/blog/2014/12/11/ulka-ep5-pump-pwm-pressure-modulation/

      Hope this helped with your question. Are you building something similar?

      Best regards
      James

Leave a Reply

Your email address will not be published. Required fields are marked *