Power Transistors in Power Amp Getting Hot !

Thread Starter

BamBam57

Joined Nov 14, 2019
20
Hi Folks !
This is the Power Output to the speakers on One Channel on a Stereo Power Amp ! D100 and D200 are bolted to the heatsink along with another pair of D100 and D200 for the stereo other channel ! The power amp seems to be working fine for audio and at the moment I cannot hear a major problem but the Heatsink gets hot on the back even withour ANY signal ! I know this to be the case as I have a duplicate Power Amp which stays cool pretty much not even without a load but with a load, so I know this one is NOT normal ! I can feel from D100 and D200 just by touching them that two of them on the same channel are getting hot while the other two remain cool ! However this heats up the whole of the heatsink so for sure there is a problem somewhere ! Have ordered some new D100 and D200 power transistors but looking at the forums here, it appears that it could be caused by something else !
From the circuit below, can anyone offer any guidance as to what may be causing D100 and D200 to heat up !
Thanks Folks ! Best Regards Paul
0-02-0a-3e3ef2c90b61b738bc17ea933f46930e86afbd910220e4d076b4049f74c74d3f_65c3e674.jpg
 

ronsimpson

Joined Oct 7, 2019
2,989
In the middle of the picture there is a pot labeled "PR1/2 1k". It controls how much power is waisted with no signal.
Probably it is set somewhere mid range. Make a mark as to where it is set now so you can get back to that point.
If the adjustment was set to less resistance the outputs will be a little cooler.
How to measure if this is working? If you have a meter measure Emitter to Collector of A7.15. It probably reads 1.3 volts (with not signal). It you reduce that to 1.0 volts it will help.
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
So the two right power transistors on the heatsink seem to be getting very hot ! Although possible could be one of them but the fingers cannot differentiate ! haha !

0-02-0a-0c73b6b6c63ca670db6c6ff3d2f366f49bd7e547501bfd5f2539adb23820966a_b7f4c48c.jpg
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
OK ! Ill try that ! But this amp is 30year old British Minimalist Audiophile amp and I was guessing it would be a component failure rather than an adjustment to make. So if this doesnt solve it, what would be the next thing to check in terms of failure ? Please !
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
... Maybe try to take the transistors loose from the heat sink assembly and apply heat sink type grease to relevant surfaces. ... Nothing is visibly evident from the photo ... The grease could make a difference as far as the heat transfer efficiency, especially if none was applied at the time of manufacture.
Good idea ! But they are tight and the problem is they are getting too hot and are attached exactly the same to this heatsink as the other two on same heatsink ! So its not a problem with attachement but of those two actually heating up ! Other Amp is exactly the same as this and they ALL stay cool not heat issues at all ! All identical ! Just this pair playing up ! Best Regards Paul
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
In the middle of the picture there is a pot labeled "PR1/2 1k". It controls how much power is waisted with no signal.
Probably it is set somewhere mid range. Make a mark as to where it is set now so you can get back to that point.
If the adjustment was set to less resistance the outputs will be a little cooler.
How to measure if this is working? If you have a meter measure Emitter to Collector of A7.15. It probably reads 1.3 volts (with not signal). It you reduce that to 1.0 volts it will help.
OK ! A7.15 = (Q7.15) on the diagram are those transistors which are connected to those Black extension wires and secured to the top of the heatsink also as per picture ! So they must get hot or they wouldnt have gone through all that trouble to put them there ! Will have to strip the whole amp board out to get to the underside to check the voltages there ! So time for any other ideas and checks to make ! Need to order some parts so may as well get at it ! haha !
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
Check Q7/Q15 for open circuit. That would give the effect you describe.
Thanks Albert ! Ill do that also ! Creating a list of checks to do now ! So keep them coming ! Its my retirement project ! haha ! And Ill order two of those transistors also !
So far it seems cuplrit could be Q7/15 with a check for the Emitter Collector Voltage as per Rons suggestion also ! Getting there hopefully ! :):)
 
Last edited:

Dodgydave

Joined Jun 22, 2012
11,285
You don't need to check the track side, you can check the voltages on the White Emittrer Resistors for Zero volts at the speaker output, and the voltages across them.
Also the Quiescent current setting is the preset around Q7 .
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
You don't need to check the track side, you can check the voltages on the White Emittrer Resistors for Zero volts at the speaker output, and the voltages across them.
Also the Quiescent current setting is the preset around Q7 .
Thanks Dave ! I checked the voltage at those White output resistors R94/95 and across them and they measure 0v dc. Then checked the milliamps across the PR1/2 Both channels on the visible top terminals and they BOTH read 0.96mA. Then checked the Vs across same and they read 0.77V and 0.8V respectively ! Does this help ? Best Regards Paul
 

MrSoftware

Joined Oct 29, 2013
2,188
I did not study the schematic, but look at it and determine if it is a Class-A amplifier. Being an audiophile amp, it just might be. Class-A will give you the cleanest output, but will waist a ton of power when sitting idle, which means a lot of heat.
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
I did not study the schematic, but look at it and determine if it is a Class-A amplifier. Being an audiophile amp, it just might be. Class-A will give you the cleanest output, but will waist a ton of power when sitting idle, which means a lot of heat.
Thanks Mr Software ! But you havent read the details above about the fault above ! It shouldnt be running hot !
 
Compare the voltages across R47. Should be in mV range. Pr1 sets that value. he BIAS adjustment would have the vales that it's supposed to be.

Compare the voltage across the E-C junction of Q7. It should not be zero.

No signal, the transistors should be barely warm and those voltages above should not run away. The E-C value will change with temperature.
 

Ylli

Joined Nov 13, 2015
1,086
It is running hot, but the audio quality is OK? If the audio quality is OK, then the transistors (including the output pair) are most likely good. Running hot is usually caused by excessive idle current.

First let's check the output offset. With no speaker connected, connect your voltmeter between the speaker output terminals on amp. You should see less than 100 mV.

Now, connect your voltmeter between the emitter of D200 (point "a") and the audio out (point, "b"). IOW, across R47. Not having the amplifier specs, I can only guess at the manufacturers recommended idle bias, but setting it to 10 mV should be safe and enough to eliminate crossover distortion.

Use EZ-hooks and connect your test leads with the unit off so if you accidentally short something you don't blow out any parts. When you are comfortable with the test lead connections, turn the unit on and allow it to stabilize (maybe 10 minutes). If you see more than 10 mV, carefully adjust PR1/2 to reduce it to that 10 mV. Adjust carefully, as these adjustments can be touchy, and a few degrees of pot rotation can make a big difference in idle bias - even moreso if the pot is dirty or bad.

If you feel the other channel (which is working and *not* getting hot) is good, then you can measure the idle bias of that channel and use the same number for the 'hot' channel.
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
It is running hot, but the audio quality is OK? If the audio quality is OK, then the transistors (including the output pair) are most likely good. Running hot is usually caused by excessive idle current.

First let's check the output offset. With no speaker connected, connect your voltmeter between the speaker output terminals on amp. You should see less than 100 mV.

Now, connect your voltmeter between the emitter of D200 (point "a") and the audio out (point, "b"). IOW, across R47. Not having the amplifier specs, I can only guess at the manufacturers recommended idle bias, but setting it to 10 mV should be safe and enough to eliminate crossover distortion.

Use EZ-hooks and connect your test leads with the unit off so if you accidentally short something you don't blow out any parts. When you are comfortable with the test lead connections, turn the unit on and allow it to stabilize (maybe 10 minutes). If you see more than 10 mV, carefully adjust PR1/2 to reduce it to that 10 mV. Adjust carefully, as these adjustments can be touchy, and a few degrees of pot rotation can make a big difference in idle bias - even moreso if the pot is dirty or bad.

If you feel the other channel (which is working and *not* getting hot) is good, then you can measure the idle bias of that channel and use the same number for the 'hot' channel.
Thank You Ylli ! Will do that in the morning ! Thanks for the comprehensive instructions, can you just clarify what IOW means and what across R47 means. I think Im getting there ! And I think everyone who shares this post will benefit from a great learning curve also ! Thanks to everyone so far who have added their twopenneth worth ! Best Regards Paul
 

bertus

Joined Apr 5, 2008
22,270
Hello,

@Ylli, As we do not know the idle current and one channel is in working order, the two channels can be compared.
That way we can see if the idle current in the faulty channel is to high.

Bertus
 

Thread Starter

BamBam57

Joined Nov 14, 2019
20
Hello,

@Ylli, As we do not know the idle current and one channel is in working order, the two channels can be compared.
That way we can see if the idle current in the faulty channel is to high.

Bertus
Ill also open up my other duplicate amp which runs perfectly cool and check against that also ! As Ylli says, Ill try not to blow it up ! haha ! Best Regards Paul
 
Top