DAP Palladium 900 getting hot

[saint gillis]

  I'm trying to fix this amplifier (Schematic attached),  the 900 has 3 power NPNs and 3 power PNPs.

  The Right side was dead. The power PNPs were all dead, and 2 of the 3 power NPNs were dead. The other medium power transistors seemed ok.

  So I've changed all 6 power transistors. Now the amplifier works, the sound is good. But it is very hot. All the power transistors are getting very hot.

  My knowledge in power amps is a bit limited, so I'm asking here for some help/ideas...

 

[mjrippe]

I'm going to guess that the Left side is not hot?  Since you have two channels, start by comparing voltages on the transistors of both sides.

 

[Khron]

Might be worth looking into the bias as well (VR1 mainly, but possibly also 2/3).

 

[saint gillis]

Yes the left channel is cold. VR1 doesn't seem to act much, but I suspect VR2/3 to set some kind of idle current point... I'll try

 

[gyraf]

When doing idle current adjustments, expect thermal runaway - i.e. hot transistor drawing more current, getting hotter, drawing more... etc..

Can be good to put two 10R carbon resistors in + and - power lines while adjusting, so they can burn in stead of the expensive silicon

Don't forget to listen. Very often an amplifier running hot will sound significantly better than a cold one of same type. Not suggesting boiling it, but a couple of watts of idle power is nearly always worth it..


Jakob E.

 

[saint gillis]

Good advice, unfortunately I've just shorted a transistor on the good working channel...  So I'm now screwed to compare left and right..

The right channel is getting REALLY hot, so I guess there's something to fix..

 

[saint gillis]

  VR2 & VR3 seem to set a start where the transistors start to amplify at their correct level, for instance with a 1KHz signal I got like 5Vpp at the output, and at a certain point of bias we have directly something like 60Vpp.
  So I've set the 2 trims at the begining of this zone, and now the tansistors are running something like 40mA each (before it was more something like 600mA each !)
  VR1 doesn't seem to act much... for distortion or symmetry maybe? I didn't notice any action on the waveform..

  I'm gonna let it cool down and test again.

 

[saint gillis]

So I confirm, this was it

 

[saint gillis]

In fact it is a bit more complicated :
VR2 & VR3 set the starting point of operation of the transistors. And VR1 sets somehow the idle current for ALL the transistors...

 

[Khron]

VR1 makes total sense that it sets the bias for all the transistors.

https://www.electronics-tutorials.ws/amplifier/class-ab-amplifier.html

(A blend between the "adjustable" and the "class AB" versions)