Runaway Thermals on an Amplifier

[Bobertsawesome]

Hi all,

I've got an issue on this subwoofer/amplifier combo I just recapped. It was causing a noticeable hum at idle, so I replaced all the previous no-name caps with Nichicon Gold Audio Eletrolitics. It was running just fine when I was testing. Now, the two tansistors are running at extreme temperatures on idle. I've troubleshooting and double-checking my cap placements, no luck. I do find this circuit rather odd though because on my -60v and +60v rail they test open circuit, I can test for resistance and it shows they are connected somewhere, however, I attribute this to my meter possibly testing with a high frequency and misinterpreting the isolation transformer. When testing with my LCR meter (DER EE) at 100hz, however, the resistance is in the 100s of megohms. I'm sort of stumped beyond this, because after decapping the main capacitors to see if that was the issue and recapping them after testing, it was not heating up and only drawing 200mA. But, after switching it on and off again, it was back up to drawing .5A.

Could this possibly be the transistors gone bad? The thermal camera shows nothing else is really heating up that would tell me there's a fault elsewhere causing this.

The trannys are Toshiba 2SA1943 and 2SC5200, with the 1943 appearing to be the hottest.

[bob91343]

Your message is so confusing that I can't figure out what you are doing.  Read it again and try rewriting it so as to make sense to someone who knows nothing about your setup.

[oPossum]

The bias may have drifted too high. Measure the voltage between the bases of the output transistors and the driver transistors. And also across the two large emitter resistors.

[Bobertsawesome]

Hey Bob,

Thanks for the reply. Sorry its late here, would explain why it might not make sense  . Anyways, to reiterate on the issue, basically, the transistors pictured are being short-circuited (I think) and causing them to reach extreme temperatures. I have done some troubleshooting and nothing seems amiss from my knowledge. This board is the power amplification part of the active subwoofer, meaning it's built into its enclosure. At idle its drawing significant current, as well as heating the heat sink up to sizzling temps.

But I think I just found the problem, I used thermal grease for thermal conductivity to the heatsink rather than mica or thermal pads. This was a first to me, but I now know tranies can short out when on the same metal backplate. So far this issue has not occurred when switching to thermal pads.