How to test this board

[nuno]

Hello All,

I have a board from the external unit of an air conditioned for repair.

These boards include a low power switching PSU (where the problem usually is) to power the logic parts and a PFC+Motor-driver (the compressor's motor) sub-circuits (and other stuff too, of course ). The PSU is feed rectified mains coming from the output of the PFC stage.

I usually get the PFC stage inductor with the board, but it is not the case. Since I need to have the inductor in place for the rectified mains to reach the PSU, how will I get around to test this without the inductor? I'm not sure the PFC stage is ON unless the motor is turned ON (which will never happen in my tests). Do you think I can use an incandescent light bulb in place of the inductor without risking damage to the PFC stage, even if it is turned ON?

Many thanks!

[fzabkar]

Is there any schematic you could show us, or a model number?

[nuno]

I don't have a schematics for this board, but it's similar to the attached one from another brand.

[Armadillo]

No, you cannot just connect any "inductor" there, it may explode when the transistor switch to GND.
You will be dealing with very high DC voltage, so any flying temporary cables with contact exposure dangling will soon be forgotten posing very high risk.
This is the portion you need to be more careful.
I am glad that you ask.

[fzabkar]

ISTM that you could bridge the inductor with a wire link and disconnect the collector of the IGBT. Your PFC supply should then be equal to the rectified mains.

[nuno]

Ok, back to this. Thanks for the comments.

What ended up happening was that I tested components on the low power PSU, without power, and I was not seeing anything wrong. I kept testing until I found out that the PFC's IGBT was totally shorted, the 3 terminals shorted. I was not expecting this. That's why nothing on the board was coming up, the shorted IGBT prevented any power from reaching anything "useful".

I don't know how did the board survive, because there's nothing with burn marks and the fuse is intact. Maybe the control system detected the short before its own PSU went down (capacitors and all), shut all down and then couldn't come up again. A control logic controlled relay prevents any big power from going through (see the left edge of the example schematics in my post above, there's a series relay with a parallel pre-charge resistor).

Since the IGBT was shorted, I obviously removed it. Then I remember of a safer way to check if the PSU was ruined. I remembered many mains switched PSU controller chips can actually work starting at voltages as low as some 30V, so I shorted the pins from the missing inductor and fed the circuit 60V from a current controlled bench PSU. It worked fine and the control logic all came up, the low power switch PSU is working fine.

I then tested components around the IGBT. It being shorted to GND may hint that perhaps the driving circuit wasn't too affected - but still I found a G-E protection bidirectional TVS working only in one direction. I reverse engineered the driver part and it's roughly as attached. TVS ZD01, model ZD022. Now, what puzzles me is, the IGBT max abs ratings for V_GE is -/+20V and this TVS is a 22V model, with a breakdown voltage of 18.7V to 25.3V (see attachment). I understand that the gate voltage is probably 15V (not sure now) and the 18V model (the "previous" one) just wouldn't do it because of the 13.7V stand-off voltage. I know that with component tolerances and all it may still provide protection on most cases, or maybe in all cases if one knows exactly the coefficients or maybe Sanyo just hand picked the TVSs.... what do you think about this Sanyo "design"?

[perieanuo]

Don't have lot of igbt experience but in mosfets gate protection the pro boards use 15V protection zeners.they fu..ed up the BOM.replace with 15 V zener 1W and yhe igbt, you will be covered.


Envoyé de mon iPad en utilisant Tapatalk

[nuno]

Well, Zeners and TVSs are not exactly the same... this TVS withstands 600W of transient power and has a very high breakdown voltage. This TVS is also bidirectional, although a pair of back-to-back zeners is also bidirectional.