diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Having some trouble with an overheating amplifier on one channel (http://www.diyaudio.com/forums/solid-state/220658-having-some-trouble-overheating-amplifier-one-channel.html)

whubbard 29th September 2012 05:31 PM

Having some trouble with an overheating amplifier on one channel
 
Hi All,

I'm a little stuck so I'm turning to you all for some help. I have an Adcom GFA-5200 powering Monitor Audio Studio 2 Bookshelf Speaker. After playing for a little while the right channel is getting hot while the left stays fine. If I measure you resistance across the terminals of the speakers the right measures 35Ω while the left measures 61Ω. I'm guessing that's the problem, right? I'm more than happy to take everything apart, but I'd love to hear people's thoughts first. I'm planning on measuring the resistance of the wires as well, something may have happened to one that I'm not aware of (although I would find that doubtful). Is there anything that could be wrong with the amp that would cause this to happen?

The weirdest part is that once thermal protection kicks in it the left channel that seems to be shutting down, not the right. I just went and checked the way it's wired, and I'm sure on which is left/right.

Thanks

Edit: I should add that I'm getting the overheating warning from the amplifiers thermal protection, which in turn is shutting down the channel. There is only one heatsink in this amp and it's on the right rise, so I know the left side of the amp will be cold.

whubbard 29th September 2012 08:29 PM

Wound up playing around with it and reconnecting the wires at the speaker terminals and now the right speaker is down to 6Ω, but the left is still at 35Ω. Also, I was wrong about the left shutting down when the right is overheating. The right is overheating and the right is shutting down. (Duh)

It's all still very bizarre to me. What would cause this sort of overheating on one side of a shared heatsink and not the other? Should I put new thermal paste on the op-amps?

whizgeek 29th September 2012 08:50 PM

Are you really sure you have the correct channels on the heatsink?

Measuring speaker resistance only while disconnected, should get you somewhere around the 6 ohms, but must be the same for each speaker.

If you are still getting different readings, then you have a speaker problem. sort this first.
probably bad connection in the crossover, maybe causing amp to oscillate.

whubbard 29th September 2012 08:59 PM

Sorry, should have mentioned that. They are both around 6Ω disconnected. 6.2Ω for the left, 6.4Ω for the right.

Quote:

Probably bad connection in the crossover, maybe causing amp to oscillate.
Would I be able to see this on a scope if I played a sine wave through? I'd have no idea what I would be looking for or where to put the probe though?

whizgeek 29th September 2012 09:21 PM

Probably easy enough to see oscillation.
Just be careful with the scope, if one of the inputs is connected to mains earth, it may not be the best thing for the amp output pins. try probe tip, without the grounding first, on each output pin
Pick a fairly low frequency first, maybe 30hz so you dont destroy your ears, and compare the quality of the signal for each channel.
than try maybe 15Khz, and do the same

whubbard 29th September 2012 09:30 PM

Should I just hook up a line level input that isn't grounded? Like a laptop, phone or portable CD player?

Also, I'm a little confused where you want me to put the probe? The output terminals?

Thanks for the help too!

gootee 29th September 2012 09:54 PM

Is it actually getting hot? Maybe the thermal protection circuits have a fault.

Are you measuring the resistance when powered off? Do you short the output terminals first? Measure at the output terminals without the speakers connected. If the resistance measured at the output terminals is really different, then you should be able to trace it to something in the output circuits, or possibly the power supply.

whizgeek 29th September 2012 10:48 PM

dont worry about the input, the cd player will be great.
first measure dc voltage across the output terminals of the amp. should be in the millivolts range.
put the oscilloscope probe or input signal wire to the red output terminal of the amp, while playing something and look for a signal on the scope.
At high volumes you should get trace 3 divisions above and below center line with scope input set to 10V/div, try 20mSec/div for sweep time.
If there is no usable trace, connect the negative/ground scope input to the black speaker terminal of amp.

whubbard 29th September 2012 10:53 PM

Gootee: I'm actually starting to think that it's a protection circuit fault. I can still touch it with only mild discomfort. I've just never had an amp cut out before when I could still put my hand on it. I don't have anything to measure the actual temp though. I'll measure the output terminals too (playing or off?).

Whizgeek: Thanks for the clear instructions for the scope. I'll give that a whirl too.

Again, many thanks to all that are helping.

whubbard 3rd October 2012 07:04 PM

I haven't yet been able to check with the scope, but I'm leaning towards the amplifier itself having some internal issue. I left it on yesterday with no speakers plugged in and no sound input. The unit was just sitting on a countertop in a 72 degree room. Well this time the left channel went into thermal protection first. The amp is getting hot, but it still does not burn me when I touch it, which leads me to believe the circuit is unnecessarily sensitive.


All times are GMT. The time now is 03:41 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2