Gigrac 1000st power drop out

[bnwitt]

I've got a Gigrac in for service and the problem I am having is it will run for a few minutes and then completely shut down all power. All functions including cooling fans die. If I turn it off for a minute and back on it will go through this cycle again. It does this with or without a signal or speaker load connected. During the time it runs, all functions work properly including output.

I have attached the power supply schematic. Unfortunately no voltages are indicated on the schem. Solid state is not my forte so any suggestions as to where to start are appreciated. There are three pc boards in this unit. The power amp board, the power supply board and the preamp/control board. I have scoured over the power supply board and cannot find TH1 or TX1-B thermal overloads. I have found TH3 on the power supply board. The unit when it came in had that smell to it. You know, the smell of over heated components. I'm wondering if TX1-B is an overload internal to the main transformer. Any ideas?

[bnwitt]

Well after a little study of the psu it seems that IC5 is receiving a shutdown pulse and that is what is killing the power to the unit. My question is what is causing this. I'm no solid state guy so I'm struggling with the logic of this circuit. It appears that OPT1 opto transistor is engergizing Relay RLY2-B and closing contacts RLY2-A bypassing the resistance of R7 and TH3 and subsequently activating TR23 and pulsing pin 10 of IC5. Now I may be completely off base so if I am wrong I hope someone will set me straight. If this is correct, I'm now trying to find the source of energizing OPT1 which I don't see on the schematics. Pin 1 of OPT1 doesn't seem to really go anywhere but from one connector to another. I'm stumped.

[bnwitt]

further study

Ok, I have now found that opto 1 is controlled by TR39 on the power amp board which appears to be engergized by overload current of either power amp's output. So, I would assume excessive current in either power amp would cause the shutdown pulse to IC5. Looks Like I need to test each of the 8 output transistors. Any Feedback is appreciated.

[bnwitt]

I have tested all 8 power transistors in circuit for beta and they seem fine. Given the symptom of 2 or three minutes of operation before shutdown, I am assuming one or more of the power transistors has excessive collector/emitter leakage and is allowing excessive output current after it(they) warm up. To my knowledge, the only way to test for leakage is to remove them all from the circuit and then test them individually. So, at this point it would seem I should tell my customer that his $499 powered mixer is not worth the additional labor to make this determination and repair. Any of you gurus have a comment? Am I totally off base on this diagnosis?

[bnwitt]

Wierdness

so I put this thing back together and ran it at room shaking levels from both channels for 4 hours and it wouldn't cut out again. Checked all speaker cables again and speakers and all are fine. Very weird.

[deanplmr]

Power Dropout

You probably have this figured out by now (it's March 17th...), but it sounds to me like there was something getting overheated in your unit. After a time, a thermal interrupt is happening somewhere and signaling the power relays to open. A bad connection can also cause these things to happen. Taking things apart and putting it all back together disturbed something that was causing the problem and now it begins to work again. These intermittent kinds of problems are a real hassle to find. I had a TV recently that had a similar problem. It would just shut off after running for 10 minutes or so. Never did find the problem. I put it back together and returned it to the customer telling him I could not not find the problem. He calls me the next day saying, "What do you mean? It works fine now!..." Guess I got lucky.

Shutting down after warm-up:
Some power amps use thermistors that can go bad, especially if they have a constant voltage on them. Look for one (or two) of these on the transistor heat sink. Others use a diode to sense temperature, or even an active circuit that puts out Hz/KHz pulses that vary in frequency depending on temperature. These are common in systems that use a microprocessor to monitor things. Also, I've seen some PA heads have a power supply monitor circuit where if any of the various system voltages go out of tolerance, the circuit shuts down to prevent further damage. It's a lot of stuff to check out. Sometimes it best to just start at the relays and work your way back till you find the offending part(s).