reverb issue

[JCSIFU]

Hi everyone,
I have a bandmaster reverb, all original, that hasn't been used in a few years and it has a reverb squeal issue when reverb control is turned past 4. Sounds like feedback that breaks out into a deafening high pitched squeal almost immediately after it starts to feedback. I've already replaced with known good tubes, already tried different (same model) reverb tank. The issue is present even when no tank is connected though. I dabble in amp work with the stuff I buy, so I somewhat know my way around an amp...just that nothing looks out of place when I opened it up. All elec. caps are original, but sound is strong and full-filter caps good probably, and the cath bypass caps for both rev tubes don't really see that much voltage/stress so I wouldn't thinke they would need replacement-no visual signs of bulging, etc... I have tried disconecting the reverb footswitch as well as the actual tank conections, and it will still do the same thing. I took out the 12at7 driver tube and it still howled, took out the return 12ax7 and no howl...or reverb of course. Would anyone be up to taking a crack at giving some suggestions as to where to start looking?

thanks,
jc

[Chuck H]

It's probably a parasitic oscillation due to cross talk with the reverb return circuit (from ???). There are several "Bandmaster Reverb" models. Which one is this?

Make sure the ground connections are good and take voltage measurements on the return tube. If this is a CBS model there should be a cap from the input grid of the return tube to ground. If this cap was removed it would be a "mod" that is easily missed. The layout on the CBS amps necessitated extra measures against these sorts of things. Also, check the 10pf cap across the 2.2M to 3.3M series resistor to the summing stage. If this cap is shorted it can cause problems too. Point is, something is wrong with the return stage curcuit based on your observations so far.

Chuck

[JCSIFU]

Make sure the ground connections are good and take . If this is a CBS model there should be a cap from the input grid of the return tube to ground. If this cap was removed it would be a "mod" that is easily missed. The layout on the CBS amps necessitated extra measures against these sorts of things. Also, check the 10pf cap across the 2.2M to 3.3M series resistor to the summing stage. If this cap is shorted it can cause problems too. Point is, something is wrong with the return stage curcuit based on your observations so far.


It's a TFL5005D, I had looked it up before and I think it's a '72. Great suggestions for starting points, thank you very much. So if the voltage measurements on the return tube are off by a large enough amount, how would that cause an occillation? Maybe if a limiting resistor is much lower than spec it would make the tube run higher bias and that would be the start of it? I'll see if I need to throw another ceramic 10pf in there tomorrow as well as start on the voltages. Thank you again for taking an interest in my thread, I like to be able to learn about electronics by experiencing it and people like you make it both accessable and fun.

jc

[pdf64]

I think it may be more likely to be the shared cathode bypass cap on V4A and B, pins 3 and 8. If that goes bad, then there's a positive feedback path via the cathode between the reverb recovery V4A and reverb mixer V4B.
The reverb driver 12AT7 V3 has a really hard life in that era SF, as the cathode resistor is such a small value. Best to change it to 2k2, the BF value. Your call whether to bypass (22uF 25V) it or not - not bypassed the reverb drive level may be reduced (not necessarily a bad thing), if bypassed there may be oscillation, in which case put a the old cathode resistor in series with the bypass cap.

[JCSIFU]

I think it may be more likely to be the shared cathode bypass cap on V4A and B, pins 3 and 8. If that goes bad, then there's a positive feedback path via the cathode between the reverb recovery V4A and reverb mixer V4B.
The reverb driver 12AT7 V3 has a really hard life in that era SF, as the cathode resistor is such a small value. Best to change it to 2k2, the BF value. Your call whether to bypass (22uF 25V) it or not - not bypassed the reverb drive level may be reduced (not necessarily a bad thing), if bypassed there may be oscillation, in which case put a the old cathode resistor in series with the bypass cap.

Thank you for taking an interest in my situation. So from what you're saying I gather I could try replacing the 25/25 cath bypass cap for v4, and change the shared 820 on v3 to 2k2, and then just remove the 25/25 cap which would still probably be plenty of reverb available, or if keeping the bypass try putting the resistor in series with the bypass cap. Would this be a high or low pass filter in funtion when in series as stated?

[Chuck H]

or if keeping the bypass try putting the resistor in series with the bypass cap. Would this be a high or low pass filter in funtion when in series as stated?

It would be an ALL pass (filter?). Putting the cap in series with a resistor would just reduce it's effect of bypassing AC to ground. Bypassing AC to ground, in this application, reduces local negative feedback resulting in increased gain. The 25uf value passes ALL audio frequencies so the effect is not limited to a filter application but rather a quantity of gain. FWIW, if the oscillation is due to interaction from a bad shared cathode then a bypass cap should DECREASE the problem. In this case omitting the cathode bypass cap would make the problem worse, not better. Also, omitting the cap will change the gain structure of the dry signal in the normal channel rendering the amp other than stock in performance (lower gain).

JM2C

The amp should work properly with the circuit stock. I'm fairly certain it did from the factory. Changing the circuit is treating symptoms, not the problem. Take voltage readings for the return tube and post them.

Chuck

[JCSIFU]

It would be an ALL pass (filter?). Putting the cap in series with a resistor would just reduce it's effect of bypassing AC to ground. Bypassing AC to ground, in this application, reduces local negative feedback resulting in increased gain. The 25uf value passes ALL audio frequencies so the effect is not limited to a filter application but rather a quantity of gain. FWIW, if the oscillation is due to interaction from a bad shared cathode then a bypass cap should DECREASE the problem. In this case omitting the cathode bypass cap would make the problem worse, not better. Also, omitting the cap will change the gain structure of the dry signal in the normal channel rendering the amp other than stock in performance (lower gain).

JM2C

The amp should work properly with the circuit stock. I'm fairly certain it did from the factory. Changing the circuit is treating symptoms, not the problem. Take voltage readings for the return tube and post them.

Chuck

Got it, makes sense right...anyway, here are the voltages:

[Chuck H]

Ok... It would also help to know the voltage on pin 8 of the recovery/mixer tube. Since your driver tube plate voltage is high from spec (normal with todays higher wall voltages) it should follow that the recovery/mixer tube plate voltages would be high, but they're not, they're a little low. This could mean too much current is flowing through that tube on one triode or both. Perhaps due to oscillation from PFB as suggested or from a bias shift caused by a leaky coupling cap.

Chuck

[JCSIFU]

Fixed!

I pulled the cathode bp cap off the return tube and threw it on the cap meter...1.9 uF measured. The send measured 34uF. Replaced both with new atoms and no more noise. Will do a complete re-cap when tube depot order gets in.
Thanks so much Chuck H and pdf64 for taking an interest in helping me out!