I will be away from the computer for awhile, for another three weeks. Also I will not have all the information with me. In the meantime, here is the plan.

Remove, mark and set aside:
Q507
Q509

IC401
IC402
Q423
Q425
Q424
Q426

Measure transistors in circuit with no power:
On the LEFT channel
Q427 NPN 2SC3944
Q429 PNP 2SA1535
If either appears suspicious, remove both NPN and PNP pair.

On the RIGHT channel
So the same with this pair
Q428 NPN 2SC3944
Q430 PNP 2SA1535

Install new R505, 6.8Ω

Power up. If no smoke, measure voltages at
R427
R429

R428
R430

TP401
TP501
TP502

 

Soo..

I have now removed, marked and set aside:

Q507
Q509

IC401
IC402
Q423
Q425
Q424
Q426

I tested Q429 and Q427 in circuit at first.
Q429 was suspicious, and Q427 measured short circuit in every way
So i removed them and tested out of circuit.
Q429 measured short circuit in every measurement.
Q427 seems too be fine! (-strange it tested short circuit in every way in circuit?)

Q428 and Q430 tested OK in circuit so i did not remove these.

Replaced R505 6.8Ω

Then i turned power ON!

R506 burned up immidiately.

I felt it was humming a bit, like something was using power, and there was slowly coming smoke from somewhere else, i did not see exactly before i turned it off.
But when i took a closer look, i believe the smoke were coming from R427... it seems to have been hot..

So since it was slowly smoking i did not take any measurements.

I know you wrote you will be away from the computer for a while so i'm not expecting any more information before you get back. No stress here

 

Ok, I am back at home base after being on the road for three weeks and travelling over 5000km.

At this stage, I think it would be a good idea to make a list of all the components that are off the board.

 

I didn't notice your respond until now.. glad to have you back

I have ordered some replacement transistors, SC3944 and 2SA1535 plus some 6.8Ω resistors

I will make a list of all the components off the board soon.

 

Here's the list over all components off the board at this moment:

Q501
Q507
Q509
Q503
Q505

Q427
Q429
Q423
Q424
Q425
Q426

C425
C426

R401
R402

IC401
IC402

 

Thanks for these.

Remove Q428 and Q430 for now.
Remove R505 and R506.
Replace R505 and R506 with 100Ω resistors.

We are going to look for shorts across the power supply without steadily blowing R505 and R506.

I do not expect the current through R505 and R506 to be much greater than 100mA but that is still dissipating about 1W.
R505 and R506 will get hot very quickly when you turn the power ON.

Set the DMM to 200VDC and connect between J402 and GND.
Turn on the power just long enough to get a stable reading, expected to be about -50V.
Turn off the power

Set the DMM to 200VDC and connect between J403 and GND.
Turn on the power just long enough to get a stable reading, expected to be about +50V.
Turn off the power.

 

Finally got to do something here.

I removed Q428 and Q430 marked and set aside.
Replaced R505 and R506 with 100Ω 1/4W resistors

I just looked at the two power transistors... Q515 and Q517.. i tested Q517 in circuit.. then i removed it.. it has a short in every measurement.
I also removed Q515.. this tested OK.(!!)
So now these are off the board too.

Then i connected my DMM between GND and J402 (200VDC) and turned power ON.

I did not get any reading.. R505 and R506 went up in flames instantly and the voltage dropped to almost nothing.. and i turned it off.

I feel like every component we install in the LEFT channel gets burned when we turn the power on, every time.
Q517 was installed new just recently...hmm

 

Looks like a new short has developed across the power rails. We will have to find the short with the power off and cable unplugged.
Remove the charred remains of R505 and R506.

Use the DMM set to Ohms and measure between J402 and J401 - both ways.
Do the same with J403 and J401.

 

So i measured between J402 - J401 and J403 - J401 both ways..

I have to set my DMM to (2000K) to get a reading, it seems to stabilize at 45 in all four measurements.

 

The Ohmmeter is showing 45kΩ which indicates no passive short.
If there is still a short, then it must be an active short, i.e. it only happens when the power is on.
That is going to be a tough one to find. Let's get a second opinion.

 

Measure resistance between J402 and J403.

 

Measured between J402 and J403
With + to J402 it measured 386 ohm
With + to J403 it measured 1333 ohm

 

Hi,
I don't know if this will help or even be relevant. Some years ago I repaired several of these amplifiers that had blown their output transistors and a lot of other components as well. From what I remember is they all showed a strange failure mode; R505 would burn up and C425 would usually blow up even after completely rebuilding the damaged amp channel. I eventually located the actual cause of all this destruction with the aid of an oscilloscope to see how clean the main power rails were. They had a very large AC component on the DC rail (positive). The actual culprit was D601 (bridge rectifier) breaking down when voltage was applied. So I replaced that and every other part that was damaged and success.
Good luck.

 

Thanks for dropping by. TS doesn't have an oscilloscope.
This sounds like excellent advice.

@Marius83 See if you can find a replacement for D601.

You might have to adapt something like these:

http://uk.farnell.com/multicomp/gbp...h-800v-module/dp/2674963?MER=sy-me-pd-mi-alte

http://uk.farnell.com/taiwan-semiconductor/gbpc2508w/diode-bridge-rect-1-ph-800v-module/dp/2677247

 

TS does actually have an oscilloscope!
Only "problem" is the seller did send me a kit instead of a "fully welded" one, like i ordered..
And i have not assembled it yet.. maybe now's the time

@MrChips, is the component in both links correct for my application?
I saw the one of them was named "+" at the end..

 

I only just noticed that the two components have the same part number GBPC2508W except for the extra + symbol.
I selected the two totally at random from two different manufacturers. I am only going by what I can see from the photographs. I do not know if the connections would work with the original D601, Panasonic SVDS10VB20F. Structurally, the Panasonic part looks sturdier.

 

I'm confused. Both bridges in the links have the appropriate markings. + - ~ and ~.

 

I selected the two bridge rectifiers totally at random from the Farnell UK web site, only looking at form factor (and connection style). I assumed 800V and 25A should be ok.
The part number from Taiwan Semiconductor is GBPC2508W,
The part number from Multicomp is GBPC2508W+.
I don't know what is the difference, if there is a difference. The two having the same part number is accidental.

 

According to my simulation, R505 and R506 should be passing 100mA. Power dissipation is 68mW.
I still cannot see what is causing R505 and R506 to smoke.

 

The + difference is the unit of issue. The + sells in a minimum package of 5 and the other one you can buy single units.

That is based on the ordering details on the two pages, not any specific findings related to it.

I still cannot see what is causing R505 and R506 to smoke.

Neither do I. I do know that if the load on their resistances drop to the equivalent of 300 ohms they are exceeding the power rating. With 100 ohms in their spot, the load equivalent drops to 275 ohms.