trouble shooting a Princeton Reverb Reissue

[g1]

In case we are looking at different schematics, here is the one I'm looking at.
65_Princeton_Reverb_schematic.pdf

Regarding V3, it is the reverb recovery/mixer tube, R36 and R37 are it's plate resistors. The only thing really common to both sides is the tube itself. As JazzP suggested, if you check it's cathode voltage readings (TP6&26), they are probably both low. Because the tube is not working as hard, the supply voltage Y (TP32) will be high.
It may be the tube itself that is weak, try swapping one of the other 12AX7's with it and see what happens to the voltages.
I'm only mentioning V3 because you were concerned about your plate voltage readings being high, I doubt it has to do with the sound issues though I suppose it's possible.

[MWJB]

Ooops! You're quite right G-one, R36 & R37 are not in the vibrato circuit at all, my bad.

Mike McCue - check both ends of R36 & R37, one end may be at B+ as indicated at "Y" 270vdc in your case), the other may be somewhat lower? If it's 270vdc at both ends there may be a bad connection, as previously suggested, checking the cathode voltages at TP26 & TP6 might show if there is anything amiss?

EDIT - on further observation, voltages at R36 & R37 look to be wrong on the schematic! They should be somewhat higher than the voltages shown, more like 2/3 of the B+ supply at "Y"/TP32. So in fact, your 175vdc at R36 & R37 looks to be correct & more in line with what I would expect, than those on the schem.

[mike_mccue]

Yes,

I had checked that this morning.

TP32 has 270vdc.

R36 and R37 see that 270vdc on one side and there is 175 and 170 on the other side connected to the plates.

I'll try another tube and I'll look at the cathodes.

edit to add: I see that you are suggesting that the schematic may be incorrect... I suspected this as well. I wonder what other people measure here on this model amp?








I have tried the new tubes the amps owner wants to use. They are NOS JAN Phillips 6V6.

Two things,

1) I can't seem to adjust the bias cool enough for them... the lowest I can get is 29ma so I didn't leave the amp on for more than the time it took for the measurement to stabilize. I guess I will have to adjust the range by swapping or adding a resistor so that the bias pot is effective for the tubes... or try different tubes.


2) One of the new tubes also seems to want to run away and draw excess current.



I was asked to look at the amp because the owner recognized that the new tubes ran hot when he first bought and installed them. He says he powered off immediately, put the old tubes back in, and sought out help. He's been playing since 1964 so I am inclined to think he wants the best for his amp and is telling me info to the best of his knowledge.

But what are the odds that a pair of matched tubes would be sent out and one would have a run a way issue? I keep thinkiing the amp must have something going on that needs diagnosis.




So now I have two sets of tubes that don't seem to work well with the amp.

I personally own several fresh NOS sets of 6v6s but I'm reluctant to test them in this amp as I don't know if I'm looking at a chicken or egg scenario.

The amp seems stable with no power tubes installed yet I'm concerned that the amp is hurting the tubes. Does that make sense?




I personally own a Silverface Princeton and Princeton Reverb that I play hard and maintain satisfactorily. I have maintained a few friend's old Princetons as well.

This Reissue has got me scratching my head.



Thanks for all the help.


best regards,
mike

[52 Bill]

I would suggest that you focus on what seems to be the main problem of the output tubes red plating and not worry about the preamp tubes etc., until you have the bias straightened out.

If you remove the output tubes and read the voltage at pin 5 what do you get? Now adjust the bias pot full on and full off what readings do you get? While monitoring the bias voltage try moving things around. Can you make the bias voltage fluctuate at all?

Have you checked the cathode resistors and diodes on the output tubes, are there any odd readings? You're using some sort of bias probe, does it use socket adapters? If it does, what happens without the adapters?

If either of the 0.1uf coupling caps gets leaky it could throw off the bias voltage and cause the red plating. Any loose connections in the bias or output sections? How about the output tube sockets, any signs of arcing?

[MWJB]

Hi Mike, I'm more than "suggesting" the schemtic is wrong, I'm garanteeing it. Check the AA1164 schem, 240v at the B+ supply (TP32) and 160v at the plates of V3 - 2/3 the voltage of the B+ supply, which is normal for a 12AX7 with a 100K plate resistor and a 1.5K cathode resistor. Many Fender schems have errors.

I wouldn't put NOS 6V6 tubes in this amp without using a lower voltage rectifier. 29mA per tube may work fine with NOS 6V6s IF the plate voltage is 425vdc or less at idle. If it was a problem with the circuitry, then the runaway would not have followed the tube in your earlier test, it would have stayed with the socket? Fit some current production, matched 6V6s, (they take the voltage in the RI fine at moderate/lower idle currents) and monitor plate, screen & negative grid voltages as they idle.

[mike_mccue]

Thanks MWJB,
That makes sense to me.

best regards,
mike

[mike_mccue]

I would suggest that you focus on what seems to be the main problem of the output tubes red plating and not worry about the preamp tubes etc., until you have the bias straightened out.

That makes good sense.

If you remove the output tubes and read the voltage at pin 5 what do you get?

I shorted out the tremolo jack as advised my MWJB and the voltages seem matched and stable at pin 5 on both sockets: -43.8vDC.

Now adjust the bias pot full on and full off what readings do you get?

I was at the end of the range. The range is -43.8vDC through -33.5vDC

While monitoring the bias voltage try moving things around. Can you make the bias voltage fluctuate at all?

No, it seems stable. I clipped to each socket's pin5 and tapped all over that part of the circuit with a wooden chop stick. Nothing caused any change.

The bias does drift very slightly over time but it is not fluctuating or wavering on the meter. The range seems to be between -43.3vDC and -44vDC

Have you checked the cathode resistors and diodes on the output tubes, are there any odd readings?

The Cathode resistors look ok but I had mentioned that yesterday after the amp was powered up for a while that they measured lower than when cold. I confirmed that this morning. That almost seems the opposite of what I'd expect. They are 1.1ohm (spec at 1ohm) flameproof, but they measure at 0.8+ohm after the amp was powered at idle for 10 minutes. ?

You're using some sort of bias probe, does it use socket adapters? If it does, what happens without the adapters?

Yes, it is a Compu-Bias meter with socket adapters. I removed it and all the measurements were made without it. It occurred to me that I should consider it a variable but I use it often and it seems to work fine with the work I have done. I can work without it but I do enjoy being able to see the readings from both tubes simultaneously... and it gives me early warning when the tube starts running away. Still, I can't help but wonder if it exasperates the problem.

If either of the 0.1uf coupling caps gets leaky it could throw off the bias voltage and cause the red plating. Any loose connections in the bias or output sections? How about the output tube sockets, any signs of arcing?

No signs of arcing. I tapped on the coupling caps a bunch and the bias supply seemed stable.







It seems as if the factory tubes are no longer useful because one is going bad and the NOS tubes are not a candidate without some other consideration being made. I was thinking that I can just swap the resistors but I think what MWJB is explaining is that doing this will not be enough because the plate voltage will begin to rise as I find a bias point and I'll end up out of a safe range for the output wattage on the NOS tubes.

I think I am beginning to understand why the symptoms were confusing. The old tubes are bad and the NOS tubes can't work yet without bringing the voltage down.

So, I think I'll advise the owner to consider a pair of new tubes. I'm not selling my tubes and I don't have any non NOS USA 6vs anyways... so I'll wait until the owner makes a decision. It seems like the amp is fine and it just needs some tubes it likes. Maybe the noise will go away and it will be that simple.


Thanks very much to everyone. Please add any ideas or comments that you think might be helpful. I'm always eager to learn.


best regards,
mike

[52 Bill]

The top end of the bias adjustment seems too low to me. Isn't there a 56 volt zener in the bias supply? I'd look at the entire bias supply starting at the rectifier diode and see if you can find some reason for the reduced voltages. Maybe a bad zener or a bad filter cap or even a bad vibrato intensity control.

[MWJB]

-43.8v at the high end looks OK though, colder than a BF/SF Princeton will run as stock?

[Tom Phillips]

The top end of the bias adjustment seems too low to me. Isn't there a 56 volt zener in the bias supply? I'd look at the entire bias supply starting at the rectifier diode and see if you can find some reason for the reduced voltages. Maybe a bad zener or a bad filter cap or even a bad vibrato intensity control.

Take a close look at the circuit topology and you will see that the Zener (D4) just limits the Bias voltage to -56V. The bias voltage is set by the voltage divider made up of R28 in series with R22 & R2. Therefore, during normal operation the bias voltage doesn't reach near -56V and the zener does’t come into play.

-43.8v at the high end looks OK though, colder than a BF/SF Princeton will run as stock?

Agreed. I would think that -43V would be sufficient to bias any reasonable 6V6 pretty cold at the plate voltage used.
There’s a lot of strange stuff going on with this amp even after the original 6V6 that has a tendency to run away has been identified. I suspect that there is more than one thing going on or a problem with the test meter used. (Or all of the above) For example the report in post #22 about the bias sense resistors changing from a high of 1.1 Ohm to a low of 0.8 Ohm as the amp warms up for 10 minutes is really strange. That’s way too much of a temperature coefficient for even the cheapest consumer grade resistors. Under what conditions was that measured? Is the measurement repeatable? What test equipment is being used?

Tom

[mike_mccue]

Hi,

I'm on the road for a few days so I can't check yet.

I'm using a one year old Fluke 179 with the factory supplied clip on leads to measure the resistor values.

I made the measurements several times in each condition to confirm what I was seeing.



I thought resistors got MORE resistive as they heated up.



Not sure what to make of it.


best regards,
mike

[mike_mccue]

I started working with the amp again today.


I powered it up without power tubes installed and measured R20 and R32 at 1.1ohm and they remained at that value for ten minutes.

I then placed the original tubes in and hoped the one would not run away. I powered it and monitored the draw. I was not using the Compu Bias but I could see the imbalance by measuring the voltages at the various spots. The values at R20 and R32 did not change when I warmed the amp up for a while. The only difference I can imagine is that when I measured the resistors at 0.8ohm that maybe having the tube run away a few times may have caused more stress.

At some point I noticed that the plate voltage was at 460vDC... which I think is the result of my having biased the tubes a bit colder on Friday.

My wall voltage is at 124.7vAC this morning... I did not measure it Friday.




It seems that the best course of action is to get a fresh pair of new production tubes. I will recommend a pair of JJ's. Or perhaps a pair of TungSols?


I bought a used 1973 Princeton Reverb a couple years ago and it showed up with brand new EH 6v6's. They lasted 3 days before one arced internally and I was lucky to be there when it happened. I've been running NOS tubes in it ever since.

I think I'll advise the owner of this amp to avoid anymore of those tubes.


I know he'd really like to use the NOS tubes, but I think it's important to get the amp 100% before I try to help with that.



I'd appreciate any comments or ideas you might have.

Thanks.



best regards,
mike

[MWJB]

JJ6V6 are very tough, good sounding tubes, though the EH are actually what come as standard fitment in the PRRI & are fine if biased moderately (<10W per tube). The EHs can't handle the same current as JJs/NOS (but can handle more voltage than NOS) so wouldn't be my first choice in an original PR without a bias pot.

You'll need a lower voltage rectifier like a NOS 5Y3GT to run NOS6V6, even then you may have to bias hot (30mA?) to keep plate voltage manageable. I'd ony consider this if the owner uses the amp at home as the drop in voltage & softer rectifier will drop power noticeably for live work.

[mike_mccue]

I placed a new pair of matched JJ 6V6 S tubes in the amp today and it was very easy to use the bias adjust to set the tubes to 10 watts.

They seemed to remain stable at idle for a half hour or so and then I played the amp and feel it sounds very much improved.

The noise is no longer apparent and the sound is clear and strong.

I'm going to play it hard tomorrow and hopefully decide it's ready to return to it's owner



Thanks very much for helping.



best regards,
mike