It's possible something happened, I had a problem soldering with the amp powered up (mains switch did not release as expected) and maybe I was soldering something on the input, don't remember now though.
I also also ordered a ready made pentacon to balanced xlr cable for my Zen dac and when I connected it to the amp, it went crazy.
Turned out the idiot company who built the cable did not connect ground in the pentacon and connected ground with cold "-" in the xlr!
The amp is powered by smps so maybe I had some serious ground voltage differences in both scenarios.
As your test tone of 500Hz is a multiple of the power source frequency, try a non-harmonic
tone like 60 or 65Hz instead. If the level of harmonics is then substantially different, this may indicate whether the source of interfering harmonics is power supply related. As you may be thinking, it could well be the SM power supplies that introduce the noise and the amp just doesn't have enough PSRR to squish it.
tone like 60 or 65Hz instead. If the level of harmonics is then substantially different, this may indicate whether the source of interfering harmonics is power supply related. As you may be thinking, it could well be the SM power supplies that introduce the noise and the amp just doesn't have enough PSRR to squish it.
So the 2.2ohm signal to power ground was shot on both channels (no burn marks, not sure how that happened) so something had gone very wrong at some point (either the shitty cable or me soldering in a "live" amp).
Fixed that but the amp was still sensitive when connecting the dac or adc, turned out I had 40 volts ac between the DAC and ground in the AMP, figured it was the PC and flipped the ac phase on the PC inlet and it was down to zero.
BUT now the distorsion was back! So changed the input back to + again and it was gone.
Really confusing and I checked bias and dc offset, dc offset was off so adjusted that down to zero AND the distorsion was back again! I dont know if the distortion came back after adding the resistor or changing the phase on the PC.
I assume this means I am sort of back where I started or does this make sense to anyone?
Fixed that but the amp was still sensitive when connecting the dac or adc, turned out I had 40 volts ac between the DAC and ground in the AMP, figured it was the PC and flipped the ac phase on the PC inlet and it was down to zero.
BUT now the distorsion was back! So changed the input back to + again and it was gone.
Really confusing and I checked bias and dc offset, dc offset was off so adjusted that down to zero AND the distorsion was back again! I dont know if the distortion came back after adding the resistor or changing the phase on the PC.
I assume this means I am sort of back where I started or does this make sense to anyone?
Last edited:
I am obviously not very smart or quick for that matter but started thinking if the negative side of the amp don't work it would sound horrible when I cranked it up so..
Tested adjusting DC offset a bit more into negative voltage and guess what the distorsion disappeared again and it turns out at the same negative level as the positive.
Would this then indicate a bias problem as many of you very kindly and persistently have suggested?
If so and given that the bias test point are on spec or running higher what could be wrong, another resistor broken or maybe T9?
Tested adjusting DC offset a bit more into negative voltage and guess what the distorsion disappeared again and it turns out at the same negative level as the positive.
Would this then indicate a bias problem as many of you very kindly and persistently have suggested?
If so and given that the bias test point are on spec or running higher what could be wrong, another resistor broken or maybe T9?
Unfortunately the schematic isn't completely - one don't have the whole diagram. I miss parts for global NFB, at least the DC rail (I recall this is a version without AC global NFB - input stage differential amp have only a gain factor of 2 through high resistance emitter resistors) so as the parts for DC-servo.
According your explanations I guess, there is a transition resistance anywhere in the plugs/jacks for amp-input, speaker relay contacts, output terminals and input terminals of speakers so as solder joints in the whole signal pad.
To check whether the dc conditions without abnormalities check the DC-voltages on follow resistors without connected speakers and shorted inputs (compare both channels and check for plausibility):
1) R10-11-3-8-18-20-21-22 (quiescent current for input stage)
2) D1-2, R23 (voltage and quiescent current for Zener diodes)
3) R26-R27 (quiescent current for buffer of VAS, T1B and T20
4) R17-19 (quiescent current for voltage amplifier stage T12-10)
5) R228-2-1-RE1-6 (quiescent/idle current of each from four-stage resp. double darlington buffer stage)
Unfortunately I don't understand the theory of operation of DC servo. Please upload the whole schematic include power supply and GND-management-schematic (very important, unfortunately mostly not to find) - then I will give you additional advices.
Last edited:
I started doing some of what you suggest but did not find anything obvious and could not really let go of the fact that it sounds ok meaning if something was really broken it would be more obvious when listening to it.
As it was affected by the DC offset I focused on anything close by so to speak and eventually I disconnected the power-supply resistors (R42 and R45) to the DC servo AMP.
Did not think it would make a difference as the DC Servo jumper was open but I was wrong, distortion was gone!
I still don't understand why, could it be that the DC servo amp started ringing/oscillated and somehow propagated small amounts of this back on the aux power rails?
As it was affected by the DC offset I focused on anything close by so to speak and eventually I disconnected the power-supply resistors (R42 and R45) to the DC servo AMP.
Did not think it would make a difference as the DC Servo jumper was open but I was wrong, distortion was gone!
I still don't understand why, could it be that the DC servo amp started ringing/oscillated and somehow propagated small amounts of this back on the aux power rails?
It certainly explains that you got lots of high harmonic distortion, especially at low signal levels. Assuming that the DC loop was still marginally stable, for very low levels, the DC loop tried to suppress the signal. As it has a limited control range, especially taking into account the low-pass filter after the op-amp output, it failed to do so for anything but very small signals. So basically you got crossover distortion from the faulty DC bias loop.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.