Need Help - Burnt Sanken

[p.heimpold]

Hello,

after modding my Musical Fidelity X-T100 I connected it for 2 minutes without heatsinking the SAP15N and P transistors (Darlington output stage as described in the Sanken datasheet)
and they burnt.

At least, that is what I think: they were very hot after the "smokey" event and the 4A fuses blew.

I modded the amp especially in the preamp stage that seems to work fine.
The only items that I changed in the power stage are the rectifier bridge (rebuilt with 8A stealth diodes K-K DC+ / A-A DC-) and two capacitors.

When I replace the fuses and turn of the amp, the amp lights on and the fuses do NOT blow, but there is no sound at the speakers.

Any idea and suggestion on how to proceed in trouble shooting? I think that I have to replace the transistors (need to check whether B-C-E are shorted), but if they were shorted the fuses would have to blow, no?

Any help and advise is greatly appreicated.



Patrick

[Nordic]

confirm speakers still work first.

[jaycee]

Without heatsinking, all bets are off. You have toasted the outputs, and if youre unlucky, more of the circuit.

If you had speakers connected, check them, because if you were unlucky, they are toast too.

[p.heimpold]

Hello and thanks,

I am assuming the outputs are dead.
I re-checked the fuses and they blow, connecting another speaker to the amp does not produce any sound.

For what concerns my main speakers, they should be OK as I have an impedance mathcing transformer between them and the amp but yes, that one might be cooked.

What I wonder: can it really take so little time for a pair of transistors to heat up so much?


Thanks


Patrick

[jacco vermeulen]

Originally posted by p.heimpold
I connected it for 2 minutes without heatsinking,they were very hot after the "smokey" event.

Eeeh,

[p.heimpold]

I know, I could hammer myself for that but my intent was just to see whether I had hooked up everything correctly...

[jackinnj]

Originally posted by p.heimpold
What I wonder: can it really take so little time for a pair of transistors to heat up so much?

the SAP16N, SAP16P are now unobtainium -- here's the Sanken replacement from Profusion:

http://www.profusionplc.com/pro/gex/...r%20darlington

I thought that these transistors had built in over-temp compensation -- there is a diode string in the P and N devices which change the bias current with temperature -- at least that's how they are supposed to work. I think that there might be other things going on.

FWIW, I haven't used the replacement devices yet.

[p.heimpold]

Hello Jackinnj,

my understanding is that the current sensing is meant to regulate "the temperature dependency of the idling current".

I understand your concern that something else might be going on.
I also would like to ensure this.
For what concerns the Power Supply of the output stage, I only rebuilt the rectifier bridge with 4 diodes. C-C to DC+, A-A to DC-.
Any suggestion on how to measure it is working fine?

I unsoldered the two PNP transistors and it becomes apparent that:
1) one has a nice crack
2) one has a "vulcan", looks like a burnt part inside the transistor

I measured resistance across B / C / E and they look OK (value pops up shortly and then gets back to zero) except for C-E on transistor 2), there seems to be a permanent short.

Measuring tension between the legs shows that BE on transistor 2) is about 970mV whereas the others are around 890mV.

My guess: P Transistors are damaged.

Will check the N Transistors now.



Bets regards



Patrick

[p.heimpold]

N Transistors also show inconsistent values amongst them, so my best guess is to change the two pairs.


Regards



Patrick

[destroyer X]

C'est la vie mon ami

Better luck next time.

Probable that output have melt and maybe drivers too.

regards,

Carlos