QSC Power Amp gets hot

[Harald Kapp]

As the transistors and the heatsink get hot, the thermal transfer from transistor to heatsink seems to work. So the thermal paste is probably o.k.
I think the section of interest is here:

Measure the voltaeg at the base of Q1 vs. ground (red arrow) and the voltage from the base of Q2 vs. ground (blue arrow).
Also check the resistor (NTC) R17 (green circle). It needs to be electrically connected, no lose ends and also should be well fixed to the heatsink near the output transistors. The purpose of this resistor is to reduce the bias voltage for the transistors when they get hot to reduce power consumption and thus reduce the temperature. If this resistor is not well coupled to the heatsink, it cannot sense the rise in temperature and thus cannot counter the heating up of the transistors. It is possibly the component marked red here:

 

[WHONOES]

The output voltages look ok. Can you do the transistor voltage tests directly across the base and emitter pins. I'm not interested in the power supply voltages in this instance, just the Base Emitter voltage. Both output transistors on both channels.

 

[gerryt]

The output voltages look ok. Can you do the transistor voltage tests directly across the base and emitter pins. I'm not interested in the power supply voltages in this instance, just the Base Emitter voltage. Both output transistors on both channels.

Hope I did this right for you , base pin to emitter pin in quiessent mode:
Channel one:
.706 DCV and -.645 DCV
Channel two:
.864 DCV and -.580 DCV
I did notice that the numbers for all four power transistors decreased slowly as I held the multimeter leads on the pins.

 

[gerryt]

As the transistors and the heatsink get hot, the thermal transfer from transistor to heatsink seems to work. So the thermal paste is probably o.k.
I think the section of interest is here:
View attachment 50885
Measure the voltaeg at the base of Q1 vs. ground (red arrow) and the voltage from the base of Q2 vs. ground (blue arrow).
Also check the resistor (NTC) R17 (green circle). It needs to be electrically connected, no lose ends and also should be well fixed to the heatsink near the output transistors. The purpose of this resistor is to reduce the bias voltage for the transistors when they get hot to reduce power consumption and thus reduce the temperature. If this resistor is not well coupled to the heatsink, it cannot sense the rise in temperature and thus cannot counter the heating up of the transistors. It is possibly the component marked red here:
View attachment 50886

You are correct in identifying the thermistor attached to the heat sink. I tested it by adding heat from a soldering iron using my multimeter to see if impedance increased as the thermistor warmed up. The resistance increased as it warmed up. I also checked to be sure it was well connected and was making good contact with the heat sink. But the overheating still was happening. I will try and to the voltage test you have suggested on Q1 and Q2. I have noticed that the little Q2 heatsink gets quite warm as compared to Q1. Thinking that Q2 may be faulty. Thanks very much for your help.

 

[WHONOES]

Is the channel that gets hot the one that measures .706 DCV and -.645 DCV?

 

[gerryt]

Is the channel that gets hot the one that measures .706 DCV and -.645 DCV?

No, it is the other channel

 

[Harald Kapp]

The resistance increased as it warmed up. I also checked to be sure it was well connected and was making good contact with the heat sink.

Acc. to the schematic that is an NTC with Negative Temperature Coefficient. The resistance should decrease with rising temperature, not increase.

Channel two:
.864 DCV and -.580 DCV

That 0.864 V looks suspicious. It could mean the respective transistor is getting too much bias and therefore doesn't turn off when it should, thus creating a high current through both transistors in the transition region which in turn can heat up the transistors.
There is a trimpot (TR3 in post #21) parallel to the NTC. Note the position of the wiper and slowly turn it towards lower resistance to reduce the 0.864 V to a value nearer to 0.7 V as in channel one.
Does that help reduce the temperature?

 

[gerryt]

Thank you so very much for helping me with this, I really appreciate it. I will try reducing the trim in channel two and see what happens. I just tried this and channel two still gets hot. I could only turn the wiper a small amount and it was at the end of its travel. I have also noticed that the heat sink for Q2 in the schematic gets considerably hotter than Q1. Thinking I might need to replace the transistors as a matched pair for Q1 and Q2. Thoughts?

 

[Harald Kapp]

I would put the transistors on the back burner. They seem to work (although a defect cannot be excluded), in my experience a defect would show as either an open circuit or a short circuit.
Return to the NTC. Check its value (you'll need to de-solder one end to avoid influences of other components to your measurement. Check the resistance at room temperature. You should measure ~ 50 Ω. Heat up the NTC (e.g. with hot air from a hairdryer) and check again. The resistance should fall with rising temperature.

 

[Audioguru]

I think the trimpot wiper makes poor contact with the resistive track.

 

[Harald Kapp]

A reasonable idea. Looks like these trimpots are sealed, so this idea is not easily tested. @gerryt : you'd have to de-solder the trimpot tr3 from the pcb and measure the contact resistance to verify this thought.

 

[gerryt]

Thank you very much for your suggestions, I will check the bias trimpot and the NTC.

 

[Audioguru]

Simply turn the trimpot a little and see if the voltage and current change.

 

[gerryt]

I have adjusted the bias trim pot for the channel that was getting hot and the problem seems to be resolved. Hopefully this will continue. Thanks to everyone for your help with this.

 

[Harald Kapp]

I have adjusted the bias trim pot for the channel that was getting hot and the problem seems to be resolved.

Hopefully just so much that the overheating disappeared. This pot sets the operating point of the output stage. If you turn the pot down too much, distortions of the signal can result.

 

[gerryt]

I adjusted the trim pot a very small amount, from about 10 o'clock to 11 o'clock. I will monitor the amps performance in my recording studio, I use it to power my mid field monitors. Amazing how a very small adjustment can make such a difference.

 

[Harald Kapp]

 

[Audioguru]

Was one channel always too hot because the bias trimpot was set wrong or did the trimpot get bumped and accidently changed?
Or did a transistor's conduction change?

 

[gerryt]

Was one channel always too hot because the bias trimpot was set wrong or did the trimpot get bumped and accidently changed?
Or did a transistor's conduction change?

I am not really sure why the channel was so hot. I had the amp in my portable rack since I bought it used a few years back, so I don't know if it has been running hot the whole time I have had it. I took it out of my rack to put in to my studio to power my mid field monitors and that is when I noticed it got very hot. I had never opened it up but not sure if the previous owner had opened it. When I did open it up I did not see any signs that it has been worked on previously. Not sure what the reason is for the the need to adjust the bias. I will keep an eye on it.