Fosgate 1000a2 with blown output

[81300sd]

They all read open

 

[Perry Babin]

Pull the drain leg of all of the 300 series IRF640s out of the board and power up the amp. Do you still have DC on the output?

 

[81300sd]

No change

 

[Perry Babin]

I don't see how that's possible. The IRF640s are the transistors that pull the power supply down to produce negative voltage on the speaker terminal. With them out of the circuit, There isn't much that can produce the DC.

Are you sure that there are no solder bridges anywhere in that channel?

You're placing the black probe on chassis ground when measuring the voltage, correct?

 

[81300sd]

Yes, using chassis ground. I just found that I have ~20ohms between the gate resistors and source resistors. Something is shorting. Searching for it now, could one of the irf9640's be shorted?

 

[Perry Babin]

If there was a problem with the 9640s, I'd expect to see positive voltage. That doesn't mean you should ignore the area near the 9640s.

 

[81300sd]

That 20ohms is coming from the splice on q323


Checking between the gate and source legs on the 640's I get 100 ohms, the 9640's I get 110 ohms.

On the good side I get 25K and 28K.

 

[Perry Babin]

That shouldn't cause it to read 20 ohms. There's virtually nothing near 20 ohms in that area. Remove the jumper temporarily to see how it changes.

Is it possible that the legs of Q323 are touching the MEHSA strip?

 

[81300sd]

Now it is 85K, q323 was not touching anything

 

[Perry Babin]

Before you removed the jumper across the collector and emitter of Q323 and the connection to ground, you had the following (correct me if I miss something).

Q322, Q324 out of the circuit
D305, D307 out of the circuit
All 640 gates out of the circuit
All 640 drains out of the circuit
-4v on the output (chassis ground reference)

 

[81300sd]

I put the gates back in, but everything else is correct. I will remove the gates and reinstall the jumper and retest

 

[81300sd]

Ok, I have .10 volts on the output now, but q312 has 47v on the gate leg with the resistor out, so that may be my problem.

 

[Perry Babin]

With the drains out of the circuit, where is it getting the voltage?

Recheck that transistor from the gate leg to the other legs to see if it still shows open (meter set to ohms).

 

[81300sd]

Its showing 65k to both the other legs so its shorted.

 

[Perry Babin]

If the leg is out of the circuit, it's definitely defective.

If the rest are OK, reinstall everything, remove the jumpers and see if it will power up without DC on the output.

 

[81300sd]

Ok, I have a burn in the laminate between the gate and drain on that transistor. I beleive I did it when removing the original.

Can I use standard epoxy to repair that, or does it take something special?

 

[Perry Babin]

Can you post a photo of the damage? Its likely that the best solution would be to cut away 100% of the burned fiberglass and leave it with a the small piece of fiberglass missing.

The burned fiberglass is often conductive and could be part of the problem.

 

[81300sd]

The burn is directly between the gate and drain legs on the backside of the board. I can probably cut it right there without messing anything up.

 

[81300sd]

Looking at it today, I think the easiest, best way would be to cut the trace and solder a jumper from the gate leg to the gate resistor. Do you see any problems with that?

 

[Perry Babin]

No.