• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Spec 4 with Problems

I just thought of something. I have seen flaky speaker relays give me fluctuations in the DC offset. Try measuring the offset from the resistor across L1; it looks like maybe R19 on the schematic, hard to read on my copy. Measure from there to the power supply ground, right at the filter caps. If you still get a varying offset I have some more ideas to try.
 
If that has no effect, and the DC offset drift is still there - let's go back to the source, the positive input of the differential pair. There are two effective inputs contributing: the DC leakage of C7 and the offset compensation voltage.

I would at this point remove C7 and lift one end of the 120k resistor R6, at the end that connects to VR1, and ground that disconnected end of the resistor.

Then I would observe the DC offset. If no drift, I'd hook up r6 again and see if it started to drift. Then C7.

If it continued to drift, I would monitor the ratio of voltages between the top of the R22 (18k) R18 (470) feedback voltage divider and the R18 R22 junction.

What did you use for C13? It's DC leakage could be coming into play as well.
 
Warren - It is R19. I took the readings across R19 and ground and the drift is there also.

Mark - C13 is the original. I only replaced the electrolytics.
I'll disconnect C7 and R6 and report back. I'll do that tomorrow night.

Thanks for the help guys.
 
I'll wait to see what you find before I make further suggestions. We are homing in on the problem!
 
I didn't get a chance to pull C7 tonite, but I'll do so on the weekend.

I did check the voltages on the bases of Q1 on the amp boards. In the service manual, it indicates 0v. My readings are R chnl 0.1v and left channel 0.06v. Any problem with that? The left channel still has the post production diodes mentioned in a previous post. The R Chnl has a new Q1.
 
miklemke said:
I didn't get a chance to pull C7 tonite, but I'll do so on the weekend.

I did check the voltages on the bases of Q1 on the amp boards. In the service manual, it indicates 0v. My readings are R chnl 0.1v and left channel 0.06v. Any problem with that? The left channel still has the post production diodes mentioned in a previous post. The R Chnl has a new Q1.
Click to expand...

Base 1 to base 2 voltages count more than common mode voltages.
 
I pulled C7 and disconnected and grounded the one end of R 6. Still had the drifting. It went from 140mv to 200mv.

Where exactly should I monitor the voltages between R22/R18?

Thanks
 
miklemke said:
I pulled C7 and disconnected and grounded the one end of R 6. Still had the drifting. It went from 140mv to 200mv.

Where exactly should I monitor the voltages between R22/R18?

Thanks
Click to expand...

First monitor the voltage at the junction of R15, R22 & R18 to ground for stability.

Then with no signal I would remove or replace C13 if it is NOT a film type cap (if it IS an electrolytic cap), in case it was leaking, since this ostensibly IS a current input albeit with a rather high impedance. I don't think it will oscillate without it, or swap them (C13's) between the good and bad channel and see if the symptoms move. You could do the swap with R22 as well....

If you have 2 dmm's I would watch the voltages on both sides of R22.

The idea is that the voltages at the bases of Q1 should be identical and stable, while the voltage at the top of R22 (that's the amp's output) is most assuredly not stable - so what's happening between those two points?

THAT is our area of inquiry.
 
Mark,

Should I put C7 back in and reconnect R6 before I monitor the voltages outlined in the previous post?
 
miklemke said:
Mark,

Should I put C7 back in and reconnect R6 before I monitor the voltages outlined in the previous post?
Click to expand...

Reconnect R6, but leave C7 out, thus we KNOW we have no input to contend with, and using VR1, we can try to minimize the offset for the next steps - but then we have to watch VR1 for a while to be sure it's input to the circuit is stable.
 
The voltage at the bases of Q1 was stable, but did vary slightly from base to base. One was 69mv and the other was 67mv. On the other channel the same thing, one was 106mv and the other 109mv.

The voltage does vary at the top of R22, from 140mv to 190mv.
 
miklemke said:
The voltage at the bases of Q1 was stable, but did vary slightly from base to base. One was 69mv and the other was 67mv. On the other channel the same thing, one was 106mv and the other 109mv.

The voltage does vary at the top of R22, from 140mv to 190mv.
Click to expand...

Yet the bottom of R22, the junction of R15, R22 & R18 is stable?

If so:

Swap R22 channel to channel and see if the suspect channel stabilizes.
 
If that doesn't make a difference I'd like to make a suggestion:
I want to remove the gain at DC, to see if the drift is occurring in the output stage or in the input stage. So lift the side of R18 that goes to ground and install a 220uF electrolytic cap (any voltage will do) in series with R18 and ground, with the negative side going to ground. Now the gain is 1 at DC, and we have 100% feedback for DC. If the offset does not improve, I think the problem is in the input stage, because the 100% feedback should cause the diff amp to correct for the drift. If the offset does improve, I think it is likely being caused by something in the pre-drivers or driver stage drifting around. The next step will probably to open the feedback loop, so the diff amp has no control of the output stage, and see what drifts. What do you think, Mark?
 
I agree, I had thought about shorting out R22. doing the same thing, DC gain of one.

Then if the output has a stable but non - zero able offset, that warrants further investigation.

We're expecting the voltage divider in the feedback loop to have a constant ratio, which it doesn't.... while the amp assumes it does and regulates to the divider's output. Quite frankly by now I would have swapped out the two resistors with something close to see if that stops the drift after I had ruled out capacitor leakage.

Most everything else I expect to be compensated for by the feedback loop.
 
I'll have to order a 220uf cap.

I did monitor the voltage on the bottom of R22 and it's stable at 4mV.

Just a clarification, the drifting is on both channels.

Thanks
 
Instead of that, just short out R22. This will make the amp unity gain at all frequencies, including DC. Now measure the offset at the output.
 
I missed that.... both channels, hmmm.......

That means a common element. Not many common elements in there :

How stable are the power supplies, and how well do they track both under temperature and each other under temperature?

The offset adjust is a big voltage divider, everything else may just common mode reject out.

OK, just so I am clear, the bias setting is stable under temperature, while the offset increases badly under temperature?
 
The issue I'm having is the "DC balance" adjustment on page 14 of the service manual. I should be able to adjust VR1 for 0V at the speaker outputs. I can't, the voltage drifts on both channels.

The idle current adjustments (VR2) are OK.

Thanks
 
You must log in or register to reply here.
Back
Top Bottom