Reverb Noise

[Daver]

I just built an amp that uses a reverb system similar to a Blackface Fender. It has its own filter supply node for the reverb driver stage and a 1MA Dwell control. The amp also has a pre PI master. When the volume is high enough (3/4 to full volume) and the Dwell is high enough (past half) I get a farty, electrical kind of noise only when playing. It can sometimes sound ring modulator-ish. If you lighten up on the strings the noise is not as bad or non-existent. If you turn the Dwell all the way up, the noise isn't as bad. So it seems to be signal strength dependent. If I turn down the Mix all the way, the noise is still heard, so it may be coupling through the power supply. So far I have...

Substituted a different driver tube and reverb tank.
Substituted a different filter cap.
Resoldered all solder connections on the board and tube socket.
I don't "think" it's the Dwell pot because it is dead silent when rotating it.

Could it be the reverb transformer? Any other tests you guys can think of trying? Thanks!

Dave

Well, I tacked in another reverb transformer. The noise is less but still there. It's more of an unpleasant distortion now. It doesn't fade out near the top of the Dwell rotation as it did before. Weird.

[stokes]

Are you sure you have the proper reverb tank?They have different input/output impedences and are made to be mounted in different positions i.e. some horizontal,some vertical,open side up etc.

[Daver]

stokes,
Yes, it's the correct tank. As I mentioned I subbed a tank (from a Twin Reverb) with the same results. Both reverb transformers are regular old Fender style. Thanks.
Dave

[Daver]

I put a scope on it today. The signal into the reverb driver stage is clean. If I look at it on the input to the tank (output of the reverb transformer) the signal gets very distorted looking. If the input signal or dwell are high enough the signal distorts so much it appears to double in frequency. I also can hear the signal coming from the chassis and a buzzing when the really bad distortion occurs. I'm using a load and no speaker. Maybe I'll try using a blocking cap on the tube's plate and scope that on the primary of the reverb transformer. Any other ideas? Thanks.
Dave

[Daver]

Well, if I scope the grid of the driver stage I can see it, too, is distorted. I tried adding a 100K resistor in line with it. That did nothing. I removed the cathode bypass cap (22uf) and that reduced the distortion but didn't eliminate it. I'm at a loss to explain how this one stage is distorting so badly. It's almost 100% classic Fender stuff that I've used before. I can't imagine I have two bad reverb transformers, but I guess anything is possible. Anyone have any ideas? Help!
Dave

[Daver]

OK, I tried another 1M Dwell pot with no help. If I replace the reverb tank with a 10 ohm resistor I have no distortion on the scope. It's as close to the 8 ohm input of the tank that I have. Still, another reverb tank exhibits the same problem and putting the original I have been using into two other amps gives no problems at all. So somehow, some way, when a tank is connected it makes the reverb driver stage distort. Any and all suggestions are really appreciated.
Dave

[tubeswell]

What about playing with the size of the cathode resistor(s) in the driving stage(s), and/or adding more padding, in order to reduce/clean-up the signal swing into the RT driver stage?

[Daver]

tubeswell,
I did try removing the cathode bypass cap on the stage that feeds the driver. It helped some, but the reverb was weaker. I'm trying not to use any crutches to make this work. What makes me crazy is that this "should" work as is. It's almost 100% Fender and I've used this set up in three other amps. The other puzzler is that it doesn't distort with a purely resistive load but does so with a reverb tank. I appreciate the help!
Dave

I just tried using an 8 ohm speaker in place of the reverb tank. That works fine too. No weird distortion. I can't believe I have two bad reverb tanks that work fine in other amps and two bad reverb transformers. This is too odd.......

[Enzo]

Exactly what reverb pan are you using? It SHOULD be an AB type pan, not an EB. Is it? Don;t tell me what it ought to be because of where it has been, just verify the number on the pan.

yes, your circuit should be working. And by connecting it to a speaker it does work, right? SO either the pan is wrong, the jack or cable to the pan is shorted, something.

Have you tried a different set of RCA cables?

Disconnect the wire from the drive jack, and conect it directly to the input of the reverb pan with clip wires. OK or still bad?

[Daver]

Enzo,
Thanks for the help. The original tank I'm using is 4AB3C1B. My test tank is from a circa 1976 Twin Reverb and bears the same number. I have tried different reverb cables. It works fine with a speaker or resistor as the load.
So, I disconnected the reverb transformer from the jack. I measured the center conductor to ground and got infinite resistance. But, if I use a jumper to the reverb cable bypassisng the jack I get no distortion! So it seems like there is something wrong with the jack. What I don't know. When I used the speaker or a resistor as a test load, I clipped a test lead on the center conductor of the jack to run to the test load and had no problems. If the jack is bad you'd think I'd still have the same problem. I'll try replacing the jack and see what happens. I'll report back. Thanks!
Dave

[52 Bill]

...If the jack is bad you'd think I'd still have the same problem.

Could it be the location of the jack? It could be that you are creating an ultrasonic oscillation in the circuit due to lead dress, output too close to an input.

Just a thought.

[Daver]

Bill,
Thanks for your thoughts. The send and return jacks are mounted on the tube side of the chassis. They are each very close to their respective tube socket. They are 1+1/4" from each other. My Fender's are 1" apart. The wires from the reverb transformer are very short.
I tried replacing the send jack and I got the same results. So taking into consideration what Bill mentioned I attached a load resistor to the tank end of the reverb cables. This leaves any possible wiring problems inside the chassis exactly the same as before. With the resistor the distortion is gone. WTF? I tried removing the return cable just for grins and I still have the same problem.
Next I disconnected the send cable from the amp. I used a jumper connected to the center terminal of the reverb jack and clipped it to the center of the reverb cable. The reverb transformer is still connected to the jack. Doing this the distortion is gone! It appears that when the cable shield is grounded this weird distortion crops up only when the tank is used. It also attenuates the signal. If I attach a jumper to the shield on the reverb cable I can watch the distortion come and go as I ground and unground it. What am I missing here???
Dave

[Daver]

I now tried moving the ground for the reverb transformer to several different spots. It was originally connected to the cathode of the driver. I tried connecting it at the send jack and at the recovery stage cathode as well as random places on the chassis. No difference. Just to complete the insanity I tried a third reverb tank. Same crap. Any other ideas??? I must be overlooking the obvious or something. Thanks.
Dave

[jmaf]

Dave, can you do the test again, removing the tank and adding a dummy 1k resistor load to the secondary as if it were the reverb tank?

I'm curious if you get the distortion across that resistor.

EDIT:
Some disconnected thoughts here:
1) Are you sending only high frequencies to the tank? Sending low frequencies has caused me trouble in the past, switched to a much lower value cap and my problem was gone.
2) If the above test results in distortion across the resistor, it means that coupling transformer is bad but it only presents the problem with a load.
3) If the above test results in clean sound, then you've got an impedance mismatch with the tank primary solenoid or something is wrong with the cable + tank side of things.

Just some loose thoughts, I hope it helps somewhat.

[Daver]

jmaf,
OK, I did it again. The resistor is 10 ohms, not 1K. The tank input impedance is 8 ohms so it should be close. No distortion with the resistor. I'm using the "standard" 500pf cap to couple to the driver stage.
I realize this thread has gotten long and wordy, so to recap... I tried a different reverb transformer, two sets of different type cables and three reverb tanks all with the same rsults. D'oh! I appreciate your help and thinking out loud. This is one of the weirdest amp problems I've come across. Thanks!
Dave