Heat damage on audio amp MOSFETs?

[Guille]

Im taking care of this massive 90s amplifier before we sell it because its been giving us some heat problems (thermal protection would shut it down after a while but the problem would dissapear if a fan wasdirected towards its intake).
Upon inspection I noticed 2 things: most of the MOSFETs on the power stage have developed some blisters that I've seen before on some toasted microcontrollers and BJTs but Im by no means an expert and would like to get your opinion.
Is this heat damage and how serious would you guess it could be?
I've also noticed only transistors on the previous stage have thermal paste, all of the perhaps damaged transistors lack that paste. Im thinking on applying some, it cant hurt right? 

Thank you for your time 

EDIT: I've attached the service manual in case someone can confirm the potentiometer noted on page 3 is indeed the bias adjust.

[TheMG]

Looks like there's a sil-pad, so thermal paste not required.

I dont see any damage. Looks like some flux or something spattered onto the transistors. They are not really in the right spot on the transistors for "blow holes".

With regards to overheating... have you checked the basics such as the bias adjustment and such?

[dietert1]

Lilkely the power transistor pins were soldered by hand. You can wipe the flux using alcohol.

Thermal run-away can be an indication that something is wrong with the amplifier and it's a good idea to have it checked. Something i found difficult to understand with many amplifiers is bias current control. Typically you will see a transistor as a sense element. With mosfets in the output stage that sense element should be a mosfet, too. If there is a mixture of mosfets and bipolar drivers, things get complicated and there won't be a stable bias current. Maybe some freak adjusted the bias current high to use it as a class A amplifier..

Regards, Dieter

[Audiorepair]

Mosfets can fail open circuit, you would never know.

BJTs tend to short and blow stuff up and the amp stops working.

It could be that some of these are sitting there doing nothing, and the others are taking the strain and overheating.


Edit:  You could try putting on a lightish load and probing with your fingers to see if you can feel any transistor significantly hotter or colder than the others, though this is a pretty rough and ready test really.
Thermal camera might show something.
I believe there is a way to short out the gates of individual transistors to detect those not working, but I was never brave enough to try it on a huge power amp with expensive Mosfets.

[Guille]

Thank you all for your replies, Im certain now that those spots werent blisters, with regards to adjusting bias I've noticed theres a potentiometer that was purpousedly placed so that it can be easily adjusted without dissasembling the chanel but I dont feel knoweledgeable enough to tinker with it.
I've attached the service manual to this comment with a note to where that potentiometer is placed in the circuit (page 3) in case someone wants to take a look and confirm thats its use.
Should I also check if what I assume are biasing diodes D2, D3, D4, D6 (right by the pot) are good?

Many thanks to everyone, this forum is pretty helpful

[dietert1]

Yes you marked the correct potentiometer for bias adjustment, but i don't see how you can measure bias current. There are no source resistors in your circuit and even the switchable supplies (IRF540) don't have anything, neither any jumper to wire in a current meter. Maybe that's the reason why bias current is off.

You can try to turn the amplifier off and set the potentiometer to "no bias current", the short circuit potentiometer position, need to check with ohm meter. Then without audio you can slowly increase the potentiometer until you see a slight decrease of supply voltage, like 1 V. Such a decrease indicates that some bias current started to flow. This isn't very precise, but it doesn't need to be.

Warning: Fooling with the bias adjustment can destroy the amplifier, so think twice and move the potentiometer in very small steps. Also you should know that your amplifier is dangerous due to its sheer power and makes a bad test object for learning these things. You certainly don't want to drop a small screw driver inside the life circuit.

Regards, Dieter

[coromonadalix]

that look like crus deposit / dust on the solders,   just need a cleanup

For the biais, an mosfet based amplifier is adjusted in current readings,    sometimes you pull the main amplifier voltages rails fuses to put an ammeter in place ...

[richnormand]

Also if the bias controls are old  move them a bit with the power off (after marking their positions).
A scratchy pot can go intermittently open and zap the amplifier when initially moved with the power on. on old units

Next use a resistive load and a measure the DC offset at the output before adjusting the bias, not a speaker that could get damaged if something goes wrong.

Unless the output devices are too warm at idle there is no need to mess with the bias adjustment specially if you do not have the factory specs.

[Audiorepair]

I second that.

Unless the output devices are too warm at idle there is no need to mess with the bias adjustment specially if you do not have the factory specs.

[David Hess]

I have only seen blistering occur on the surface directly over the semiconductor die so that looks like something else, like rosin splatter.

[CaptDon]

On one particular model of Carver amplifier you turn the bias to zero as a starting point
and measure the A.C. line current at room temperature and then adjust the bias on each
channel for a specified increase in A.C. line current!! I often adjust a slightly different way,
at a cold temperature of around 50f I slowly increase the bias for zero crossing distortion
at low output levels (2w to 5w on a 1kw amplifier). This results in good operation with no
distortion when cold and generally results in less than specified bias current allowing the
amp to sound good and operate safely. Doing outdoor sound jobs in the scorching sun when
everything painted flat black is too hot to touch makes you do unusual tricks to keep everything
running with a safety factor.

[David Hess]

Doing outdoor sound jobs in the scorching sun when
everything painted flat black is too hot to touch makes you do unusual tricks to keep everything
running with a safety factor.

One of my early audio amplifier designs went into thermal runaway when left inside a hot car.