Need assistance to find a short in power supply on BD1000a1

[86boy]

It seems half of the power supply is working correctly, but the other half has a possible short somewhere. With a 5 amp fuse inline with +B and only 1 FET on the Q5-Q8 (Q9-Q12 empty) side it will pop the fuse instantly. If I have 1 FET installed on the Q9-Q12 (Q5-Q8 empty) side it will power up green light and all. I started playing around trail and error style, and removed all the transformer leads for Q5-Q8. I also removed +B so only ground and REM power connected. With 1 Fet on both sides (1 on the Q5-Q8 side and 1 on Q9-Q12 side) installed the voltage to the gate on the Q5-Q8 side is ~2.3v if I connect the transformer lead back to the board the voltage jumps to 6v then it gets hot.

I have attempted to find this "short", but everything I've tested seems to be OK. I have even gone to the extent of removing the transformer to test the leads which i didn't find any shorts weird resistance readings. Tested Q17/Q20 (and they were replaced with new), have also tested D16, and i'm not fully sure how to test a switching diode for D15. U14 voltage on pin 13 was 5.45 and pin 16 was 5.30. Pretty sure I'm overlooking something, or not doing something correctly.

 

[Perry Babin]

With only remote and ground connected and one FET per bank, what is the voltage on the gate (red probe on the gate leg and black on ground)?

What is the value of all 8 gate resistors? (no power applied)?

Did you check the driver transistors?

 

[86boy]

Q5-Q8 bank gate is 2.4v
Q9-Q12 bank gate is 2.4v

10.5 ohms

Q17 and Q20 appear to be good as well.

 

[Perry Babin]

You could have a shorted transformer winding.

Do you have a current limiter for your 12v power supply?

 

[86boy]

I can remove it again to double check the transformer windings. I pulled it out once, but didn't find any weird readings as each winding didn't seem to cross anywhere. Maybe it's shorting when it's installed? As a side note I can have the gate and drain connected on the Q5-Q8 side with +b and rem connected, but as soon as I connect the source it pops the fuse.

I also do not have a current limiter. It's just a 12v 50A capable power supply.

 

[Perry Babin]

For future reference, pulling an intermittently shorted inductor or transformer can make it much more difficult to find the fault. One of the first things you do is to push/pull/twist it while operating through a limiter (you should have one if you're going to do repair work) to see if the current draw changes.

Operating through a limiter with the FETs clamped makes it virtually impossible to damage anything or blow fuses.

 

[Perry Babin]

You need to insert an insulator in-between the windings at the point where they are shorting. It can be as simple as piece of cardboard with a bit of adhesive to keep it in place.

The next time you order parts, buy a 2 ohm 100w resistor like the one shown below to use as a limiter. If you buy 2, you can use them in series for a 4 ohm dummy load. Fan cooling from the end (through the hollow core) will significantly increase their power handling.

 

[86boy]

Ah yes that makes sense. I'll do that for next time for sure.

I did make a mistake, and I really screwed it up... It was starting to look promising although disaster struck. When I was putting in an output fet to see with no power to the amp at all not even the cables connected. It basically blew out a fet as soon as it touched the pads. In turned screwed up something else as I lost half the power supply again. I lost all power on the gate of the power supply fet. *sigh*

 

[Perry Babin]

Like many class D amps, you have to discharge the rail caps every time the amp is powered down. If you do not, the rail voltage will remain.

If you have to troubleshoot the audio circuit, it's best done with no rail voltage (red/black, or both wires disconnected on the rectifier end of the wires).

 

[RockfordFIx]

Perry is correct intermittent short of winding is hard to find. I had a few where one of the windings were shorting to the board itself on the bottom of the transformer. Another place to check is the MESHA board. Sometimes the powersupply gets will short on the MESHA if it damaged.

 

[86boy]

Thanks for the tips, but I was unable to find exactly where the short is on the transformer or maybe i'm just losing my mind haha! I can't get it to power up with a fet in the Q5-Q8 side again. With +B and ground only the fet gets really hot super fast, but didn't pop the fuse. with +B, REM, and ground it pops instantly. Then with REM and ground nothing happens. So i've again reduced it back down to the same issue as before, but no matter how much I twist/turn/push/pull the transformer nothing changes with the voltages on the pads and nothing makes any noise. I validated everything from U14 through the Q17-21 and everything around it... so I'm definitely back to this transformer...i think. Only weird thing (outside of some short i can't seem to find) I noticed was that U14 Ouput A (pin 13) has ~2.45v whereas Output B (pin 16) has ~3.15v.

 

[Perry Babin]

With only B+ and ground connected only, you should have 0v on the gates of all of the PS FETs. What do you have?

You stated that the drive voltage was identical for both banks (2.4v). Has that changed?