Originally Posted by K_19
If anyone with the knowledge of the inner parts of the Amp and the power supply of the 252s can confirm this then that would be great.
I see that Elusive Disc's description of the 2170 specifically warn of doing this though, saying it can damage the amp and the headphones...
Let's not assume all the PSU's Stax uses are identical as they just aren't. That said I just measured my 100V SRS-2170 PSU and it gives 18.47V unloaded off a 100VAC source and 21.86V off a 117VAC source. Utterly unregulated and only people who want to buy new amps regularly would use the 100V transformer at anything other than 100V.
Originally Posted by DefQon
That's exactly what I'm going to do today. Since my DMM is borked and a replacement on it's way early next week. I'm going to do two things to identify damage and rule out whatever components is borked.
Originally I said the two ELNA 350v caps I measured (assuming one for each channel) one of the cap's measured 315vdc on the neg pole and 620 on the positive pole while the other ELNA cap measure 315vdc and 0vdc respectively, thanks to spritzers "replace/measure diodes in the output stage" idea, I looked at the PCB more carefully following the tracings from the R/L-input to the trafo ps and what looks like (assuming) is the output stage 4 diodes with bypassed ceramic film caps and 2 ELNA caps, the white bias 600v B+ wire connects to the 4 transistors and copper bridge comes from this section.
Now I'm going to first de-solder the 4 diodes in that output stage (?) and record the values down and see if I can get replacements if required. Next I'm going to resolder the pairs of the diodes and swap positions with the other two. This way, if the right channel is dead but the left channel is ok, this should give me 315/620vdc on the neg/pos poles of the ELNA caps respectively. And the right channel should read 0vdc on the pos pole with no sound.
This is one theory I have of what has happened.
The next option is to desolder the 4 big transistors and again swap a pair in for the other pairs position so just swapping places. I'll do it in in steps so if one thing goes kaputz once I turn it on I can narrow this down as a suspect and confirm my findings once I get the DMM in by measuring the points.
So it's pretty much a pair of transistors for each channel, pair of offset/balance trimpot adjustment for each channel, output stage has a pair of diodes (used to rectify AC to DC) bypassed by some ceramic film caps running in parallel with two ELNA 350 47uf caps (1 each channel).
If one component was half damaged then I would at least get some static out of the left side through L+/- pins. But the fact that it means there is no sound at all, could only mean dead transistor or diode. Upon close-up inspection nothing looks blown, all solder pads that look half-arsed has been reflowed. If a resistor was blown, my early measurements would've been out of line and physical burnt marks could've been seen.
One small flaw to all of this... there is only one PSU for both channels. Two caps stacked on top of the other, or one cap gives 315VDC and the other one on top of it also gives that too but since it's "ground" is 315V the output voltage is 630V.
No chance that there are Germanium diodes in this amp. The one I restored had bog standard Si from Toshiba which I replaced with something similar. Even the Stax amps from the late 60's has Si everything.
The problem is clearly in the amp section so look there. Measure between the channels but a simple test of the output voltages will confirm that it at play here.