Panor / Dynaco ST80 - distortion one channel at higher volumes

Opening a new thread seeing as I'm seeking advice.  (Been posting to the ST80 / ST160 thread)

Just purchased a ST80.  As I discussed in the other thread, it had issues with the bias adjust display, which turned out to be a burned PC trace.  This was on the right channel.  I repaired the damage to that, as well as some collateral damage to bias adjust board.  Noticed last night that it had some distortion I couldn't put a finger on.  

Let the unit burn-in today.  Noticed that the left channel (not associated with my repairs), is distorting when increasing volume.  Did the usual to try to isolate: swapped power tubes and input tubes.  Problem stays with the left channel.  Looking at the LEDs for the bias while it playing, the  two LEDs for the left channel bias, start to flicker red as the distortion increases, indicating that the voltage (current) on those tubes are getting driven high.  

Unfortunately, there are no test points for the bias, so I can't readily track the bias voltage.  (For that matter, running this thing without the bottom cover is going to be a real issue, so trying to get ideas up front).

Can I get a 'reality check' on my trouble shooting plan?  Going to get a set of tube test adapters so I can probe the tube pins while it's operating.  I suspect first thing to do would be to check the high voltage supplies.  Each driver tube has it's own dedicated solid state HV supply (+475V) and a dedicated +480V B+ for the outputs.  

As the issue is only on one channel, if this is power supply related, I'd assume (looking at the schematic), that the .68uf / 630V caps are fine as they are common to all supplies.  

Now where I need a little guidance is where / how to check the amplification, if the power supplies check.  As I said, tubes were swapped channel to channel with no change, so it's not a tube issue.  I'm thinking if I can put my hands on a scope, check the output of the first stage of the 12AT7 going into the 6DJ8.  I guess I should be looking at pin 2 and 7 of the 6DJ8?  I suppose I can compare results with the right channel.  

Assuming inputs are distorted there, I would guess I need to check the passives (R7, R8, R9 R10,R11)  If those check then C1 and C2 would be suspect?  If there's no distortion there, then I guess look at pin1 and pin 8 of the 6DJ8?  If that's distorted,thoughts?  I'd assume I start looking at C3 and C4.  If it looks good there, then I guess C5A / C5B?

Having not done this before, can check the driver stage without the power tubes in place? Or do I need them to load everything down?   

Sorry to be wordy, but it helps if I talk out my logic, and I'm trying to see from the collective group if I'm proceeding reasonably?  (of course this all assumes I can get my hands on a scope).  

I've attached the schematics in case anyone wants to play along.

boeingpilot wrote: (of course this all assumes I can get my hands on a scope).  

PC based USB scopes, some of which work with a smartphone, are pretty inexpensive.
https://loto.0sc.dev/
http://www.loto-ins.com/en/

Downside is limited input range, so you need 10x and 20x attenuators to go along with probe 10x to work on tube gear.

Be doubly careful as you can easily fry a computer or smartphone with the wrong connection.

should be fairly EZ to find the problem since you have two identical channels and one is good.
Start by checking the plate and cathode voltages on each tube and compare to the good side.

I agree, easiest thing to do is compare voltages of the good and bad channel. Start with input tubes and work your way through. Any major differences isolate where the issue is. Leaky coupling caps can cause exactly the problem you describe. Measure voltage in input and output side of each cap. C103 and C 104 would be the ones to check. Remember you have bias voltage on the outputs of both of them, Bias could be the problem. You should have around 475 volts on the input side of them and whatever the bias voltage is on the other side. If you have over the bias voltage by even a few volts you may have a leaky coupling cap. Then start checking parts, making sure no resistors have drifted.

Found this workbook online for debugging M-125 build.
The M-125 is basically half of a Tubes4HiFi Stereo xx or any number of typical two tube input amps like PrimaLuna, Cayin, Dynaco, etc.

It's pretty useful for channel comparison to find swapped resistors, bad solder joints, confirm transformer health, etc.

https://docs.google.com/spreadsheets/d/1lotw6IwJmBPydUpCq7_JiYpKtGpgh-dJFlRcJSZ5UdM/edit?usp=sharing

It's been updated a couple of times since I found it a while back.

Ok - so, as much as I hate to just 'shotgun' things, the design of this amp is really hard to have it running and disassembled at the same time (disadvantage of a huge PCB with the power supply and amplification all on one). Plus I'm traveling a lot, so don't have too much free time to spend on it.

(as a review, problem is with left channel). Replaced C3, C4, C2 and C1. (as an aside, and I doubt it made a difference, the 12AT7 and 6DJ8 were replaced as the 12AT7's that came with it tested weak, and I didn't really trust the russian 6H1Ns that were in it. As expected, same issue). No change, same issues. Taking Tuninfork's suggestions, I guess when I get the time, next drill will be to check each resistor. I was only able to find on octal tube test adapter, really need to find a couple more so I can run the amp and take measurements at the tube pins. Guess I'm going to be search for them.

Thank you for all the good suggestions. I just need to find a free evening to really analyze.

Just to throw a stupid question out there.... I replaced the ultralinear / triode relay on the right channel because of the PCB damage. Mmmm, seeing as the entire output signal go through there, wonder if the contacts could be bad on the one I left in there. (Thoughts?)

Ok, New Year, now have had time to look at this.   So, now that I have a scope started looking at the signal from the input forward.  As expected, signal looks identical on pin2 on V1 and V5.  However, on pin 8 of V1, I have about 3 times the gain as I do on pin 8 of V5.  This increase in gain follows through V2, when compared to V6.  

Ok I say, let's check all the passives.  Checked all the resistors, R1 - R21.  They all check.  All Caps except for C5a and C5b were changed out (see above).

Any ideas?  The only other thing I can think of us there is a bad (open) trace on the PCB.  I guess time to trace each connection point-to-point.  But something is definitely wonky around the first stage of amplification (phase inverter).  I'm pretty sure I'm on the right track, because the amp does act as if the finals are being over driven on the bad channel

(follow-up.. checked point-to-point, all connections on PCB check ok)

boeingpilot wrote:Ok, New Year, now have had time to look at this.   So, now that I have a scope started looking at the signal from the input forward.  As expected, signal looks identical on pin2 on V1 and V5.  However, on pin 8 of V1, I have about 3 times the gain as I do on pin 8 of V5.  This increase in gain follows through V2, when compared to V6.  

Ok I say, let's check all the passives.  Checked all the resistors, R1 - R21.  They all check.  All Caps except for C5a and C5b were changed out (see above).

Any ideas?  The only other thing I can think of us there is a bad (open) trace on the PCB.  I guess time to trace each connection point-to-point.  But something is definitely wonky around the first stage of amplification (phase inverter).  I'm pretty sure I'm on the right track, because the amp does act as if the finals are being over driven on the bad channel

(follow-up..  checked point-to-point, all connections on PCB check ok)

Do you have a decent schematics ? Mine is unreadable.

boeingpilot wrote:Ok, New Year, now have had time to look at this.   So, now that I have a scope started looking at the signal from the input forward.  As expected, signal looks identical on pin2 on V1 and V5.  However, on pin 8 of V1, I have about 3 times the gain as I do on pin 8 of V5.  This increase in gain follows through V2, when compared to V6.  

Ok I say, let's check all the passives.  Checked all the resistors, R1 - R21.  They all check.  All Caps except for C5a and C5b were changed out (see above).

Any ideas?  The only other thing I can think of us there is a bad (open) trace on the PCB.  I guess time to trace each connection point-to-point.  But something is definitely wonky around the first stage of amplification (phase inverter).  I'm pretty sure I'm on the right track, because the amp does act as if the finals are being over driven on the bad channel

(follow-up..  checked point-to-point, all connections on PCB check ok)

Check the feedback chain. Is any signal coming around R19 ?

peterh wrote:

boeingpilot wrote:Ok, New Year, now have had time to look at this.   So, now that I have a scope started looking at the signal from the input forward.  As expected, signal looks identical on pin2 on V1 and V5.  However, on pin 8 of V1, I have about 3 times the gain as I do on pin 8 of V5.  This increase in gain follows through V2, when compared to V6.  

Ok I say, let's check all the passives.  Checked all the resistors, R1 - R21.  They all check.  All Caps except for C5a and C5b were changed out (see above).

Any ideas?  The only other thing I can think of us there is a bad (open) trace on the PCB.  I guess time to trace each connection point-to-point.  But something is definitely wonky around the first stage of amplification (phase inverter).  I'm pretty sure I'm on the right track, because the amp does act as if the finals are being over driven on the bad channel

(follow-up..  checked point-to-point, all connections on PCB check ok)

Do you have a decent schematics ? Mine is unreadable.

http://www.brainfartz.com/images/Stereo%20Stuff/dynaco/dynaco-st80-manual.pdf

peterh wrote:

boeingpilot wrote:Ok, New Year, now have had time to look at this.   So, now that I have a scope started looking at the signal from the input forward.  As expected, signal looks identical on pin2 on V1 and V5.  However, on pin 8 of V1, I have about 3 times the gain as I do on pin 8 of V5.  This increase in gain follows through V2, when compared to V6.  

Ok I say, let's check all the passives.  Checked all the resistors, R1 - R21.  They all check.  All Caps except for C5a and C5b were changed out (see above).

Any ideas?  The only other thing I can think of us there is a bad (open) trace on the PCB.  I guess time to trace each connection point-to-point.  But something is definitely wonky around the first stage of amplification (phase inverter).  I'm pretty sure I'm on the right track, because the amp does act as if the finals are being over driven on the bad channel

(follow-up..  checked point-to-point, all connections on PCB check ok)

Check the feedback chain. Is any signal coming around R19 ?

I'll reassemble as soon as I can and check. (If I don't get it together tonight it won't get looked at until the weekend at the soonest. That was my thought as well (plus I can compare to the other channel at R119)

Thx!!

Hello BoeingPilot
a few hints for you here. Yes, you can check the bias DIRECTLY on each output tube by checking the voltage across the cathode output resistors (R20,21,120,121)
and since those are 2 ohm resistors, you should see 80mv across them when biased correctly (this is also the voltage that the 4 comparators are looking for if you examine the schematic).
That equals 40ma per tube. You can also check to make sure each of the 4 B+ voltages (A, B, C, D) are correct at 475vdc.
Likely one of them is wrong if you have distortion that stays with a channel when you swap tubes from left to right (and vice-versa).
Much less likely but possible is a bad output cap (C3, C4, C103, C104) or a bad relay on the triode-ultralinear mode switch.
If you don't use that much (very few people do) you could temporarily jumper each relay for UL mode.

All those voltages are good. Will recheck bias once it’s reassembled. (Traveling this week) the relay isn’t necessarily not bad. Remember, I had to repair the PCB on the other channel due to overheating in the relay area.