Metro JTM45 tube intermittently redplating in V1 socket...

[Heliotricity]

It just occurred to me, I was looking at this other guy's amp again and I saw it was modded to have an added master volume control on the back (which my amp does not have). Does the 47nf cap have to do with that mod? That might explain why it is not showing up on the schematic.

[Heliotricity]

Update: Amp has been running good since I last posted 7 months ago. After the symptoms went away I swapped out the Shuguang tubes and put in some KT66s, going back and forth between a pair of Valve Arts and a pair of Tung Sols. Amp has been sounding great, no issues. Just the other day though, I hear the signal going weak, very similar to the sound when one of the tubes started to redplate before, sustained notes start dropping out in an uneven unnatural way and there are a couple unusual sudden swells in the volume. Before shutting it down I take a look at the tubes, no redplating...a little later I take the amp out, drain the caps, check voltages, all are in spec. Negative bias voltage on pin 5 of V4 and V5 are at -44.

Figuring maybe the tubes just died, I put the shuguangs back in and check bias. They bias fine at about 36ma and are consistent but after about 4 minutes the bias starts to rise slowly up to 40,50, 60 so I shut the amp off.

I try a different set of tubes I have and the exact same thing happens. Stable, then it slowly rises until I have to power the amp off.

Any ideas what could be going on here? I checked the resistors off the bias pot (2 resistors in parallel which = 37k) and it is in spec, the 15k resistor between the two bias caps is in spec. I did notice that the bias supply dropping resistor is 124k on my amp where the metro schematic indicates it should be a 220k. Could that be part of the problem?

Any help appreciated.

[Jazz P Bass]

When you are using the bias meter, check the bias voltage (-44Vdc) and see if it goes lower when the meter shows a rising current.

[Heliotricity]

When you are using the bias meter, check the bias voltage (-44Vdc) and see if it goes lower when the meter shows a rising current.

Yes it does. Although today the amp acted a little different: instead of the tube bias point being stable for a few minutes then rising, it would be stable then rise about 10ma up to around 37ma then come right back down, stay for a couple minutes then do the same thing, finally it stayed stable. I had the tubes biased pretty cold (V4=25 and V5=27) so that if it started to rise I would have time to check the bias voltage.

When the tube bias went up +10ma the bias voltage went from -44 down about to -35, it was hard to catch because by the time I saw the tube bias going up and switched my meter over to test the negative voltage it would be back on its way down again and the negative voltage would be moving up again to the -44.

Since it finally wasn't moving for a chunk of time I decided to raise the bias more to where I would want to leave it were I to use the amp (37ma). After doing this within a minute my bias meter got no readings on either the bias or the bias voltage and I saw the tubes were both redplating, so I shut it right off.

[Heliotricity]

Check the tube socket. Look at from the top where the tube plugs in. Does the contact for pin 5 look bent or dirty? Is it flooded with solder? Try to find something that is the same diameter as a tube's pin. Do you feed equal tension on all the pins when you insert and remove the pin?

From the top of the socket nothing looks out of the ordinary, tension felt normal. The sockets are by Belton which from I understand are well made. Some from the underside on pin 5 were bent a little so I straightened them, plugged into a different outlet. Today the bias was generally in a continuous state of up and down flux within 2-3ma but with little spikes where they jumped up as high as 10ma and then back down.

Could really use some input. I'm running out of ideas and really don't want to spend more on another tech.

[Jazz P Bass]

If the bias voltage is changing, you may want to explore why.

The bias supply capacitors may not be up to snuff.

[Heliotricity]

The bias supply capacitors may not be up to snuff.

Last night I went through and checked the values of all the caps and resistors and nothing was out of range. The symptoms of whatever the problem are unusual in that not only do they vary slightly, but they are stable for periods of time and then destabalize, either slightly or to the point where I have to shut the amp off.

That said, I'm at the point where the only way to proceed seems to be swapping out parts. So today I swapped out the bias caps and tested but no difference. Tried 3 sets of tubes. All did the thing where they were stable for 4 minutes then the bias started rising so high I had to power down.

What would be the next logical suspect, filter caps? Coupling caps? Also, is it possible something under the turret board has a bad connection? That would be a pain in the neck to remove on the off chance that is what is causing the problem.

[g1]

Are both tubes red-plating at the same time? It is unlikely that both coupling caps would fail at the same time. Usually with a bad coupling cap one side will red-plate first. So if they're red-plating together, most likely something common to both sides, the bias circuit.
For the bias supply, you need to check all the connections. HV winding to R32, D1, R29, through to R26/R27 junction.

[Heliotricity]

Are both tubes red-plating at the same time? ...For the bias supply, you need to check all the connections. HV winding to R32, D1, R29, through to R26/R27 junction.

Yes, both tubes are responding in unison. With the exception of any connections under the turrent board which I cannot see without taking that off I have checked over all the connections, I'll go over them again specifically throughout the spots you mention.

Aside from running the amp, the only things I can't test are the filter caps and the rectifier tube. What are the chances of either of these being the problem? I've seen cases where brand new filter caps are bad. Since the tubes are both redplating as g-one mentions it is probably not the coupling caps, then my next thought (unless I find a bad connection somewhere) is to change the filter caps.

Thanks everyone who has offered some guidance on this by the way, I appreciate your help. With some luck I'll track down where the problem lies before long.

[loudthud]

It is unlikely that the rectifier tube or the B+ filter caps could be causing the problem. The likely cause is the caps in the bias filter or one of those wires on the back of the board

[Heliotricity]

It is unlikely that the rectifier tube or the B+ filter caps could be causing the problem. The likely cause is the caps in the bias filter or one of those wires on the back of the board

When you say the caps in the bias filter do mean the 16uf electrolytic caps at C19 and C20? I haven't quite mastered circuit terminology and want to make sure I know which ones you mean.

[loudthud]

C21, C20 and C19 are the B+ filter caps. C21 being the first cap after the rectifier tube (V6) is sometimes called the "Reservoir Cap". C17 and C18 are the filter caps in the bias supply. Your build looks very neatly done, but you need to check any wires soldered to the backside of the turrets to make sure none of them has come loose or is not soldered completely.

[Heliotricity]

Your build looks very neatly done, but you need to check any wires soldered to the backside of the turrets to make sure none of them has come loose or is not soldered completely.

I think I got it. Detached enough wires on the turret board so that I could get under it and check connections. Sure enough, the wire on the underside of the post where the R20 and R23 v3a and v3b plate resistors meet was loose. It was in the eyelet but loose. Strengthened the connection as well as a couple other iffy looking ones. Put it all back together, checked voltages with the power tubes and preamp tubes out, then with the preamp tubes in and finally with the power tubes. All is in spec. The bias was rock solid, no up and down crap. Played through it for a bit, sounds normal. Still have the Shuguangs in it which I don't like. Tomorrow I'll play some more, if it holds I'll put in the KT66s, rebias and keep my eye on it.

I'm guessing probably when the amp was sitting upright in my cabinet that wire which was loose could potentially make enough of a connection to make the amp work, but over time if I moved the amp the symptoms of the bad connection would show up at random times. Which would explain why the amp would work for a couple weeks and then act up, then be normal in the techs shop.

Keeping my fingers crossed that the issue is resolved. Thanks again for the input everyone.

[g1]

The R20/R23 connection problem could not have affected the bias.
But you mentioned touching up some of the other iffy looking connections. Hopefully one of them will have been in the bias circuit and the problem will not reappear.

[Heliotricity]

Update: Been playing the amp regularly since last post and it is running great with no issues whatsoever, bias is solid with no wavering. Put back in the KT66s and it sounds dynamite. Checked all voltages throughout the amp the other day and it is all right in spec.

The R20/R23 connection problem could not have affected the bias. But you mentioned touching up some of the other iffy looking connections. Hopefully one of them will have been in the bias circuit and the problem will not reappear.

It occurred to me at the time that the loose wire may have not had anything to do with the earlier issue. One connection I did remember that seemed questionable which I strengthened was at the base of the 220k bias supply dropping resistor. I did strengthen up a few others, focusing particularly around the bias section but perhaps that was it. Thanks again everybody for all the input.