Reverb howls with 12AT7 but not with a 12AX7 in the driver position...

[bruno]

I have a super reverb at home, that howls even if I set it to no volume at all. It howls when it has a 17AT7 in the driver position, however it doesn't do so with a 12AX7. I'm going to measure the voltages around that position this weekend and redo some of the reverb lead dress. The amp sounds wonderful in both channels with not noises.

But the reverb doesn't work with the proper tube. The 12AX7 sounds really like a cathedral reverb, which is not so usable.

Any ideas before digging in.

[phsyconoodler]

Get another 12AT7.It's gone microphonic.

[bruno]

Well that would be just sheer bad luck, since I've tried 3, with different brands all new and all with the same results... JJ, electro-harmonix and sovtek!

[phsyconoodler]

Check to make sure the reverb cables are shielded and the shields are intact.If one is askew,it may howl too.
  I like that word...'askew' 

[Fresh_Start]

I've built 3 Blackface Fenders and two of them were pretty difficult in terms of eliminating the reverb howl.  Different 12AT7s from different manufacturers ad nauseum.  Sometimes fiddling around with the reverb cables can solve the problem.  Don't let them touch each other if possible.  Make sure the "return" end of the reverb tank is as far away from the PT as possible.  Use a shielded cable to and from the Reverb pot.  Ground the reverb transformer on the turret board along with the circuit or at it's filter cap if possible.  Make sure the lead from the return jack to the grid of the recovery stage is away from and/or perpendicular to anything else.  Put plate leads down on the chassis and keep them away from grids - using vertical separation may help.

Bottom line - this isn't unusual, it's just a pain to solve the problem.

Cheers,

Chip

[bruno]

thanks guys... this is actually an amp I built from scratch, I'll post more photos of it this weekend. And I'll go through the reverb too.

[Dave]

Sometimes the least likely culprit is the culpable one after all. I had an amp one time that was a Marshally type with reverb (famous brand name). I always had problems with the reverb. I had tried everything in the book to try to figure it out to no avail.
The reverb worked fine, but it sure was sensative and at certain settings it would howl and also, it would cause hum problems.
I finally just resorted to only using it on settings that weren't problematic.
Months later, I decided that the amp would sound better if I changed it to use 6L6's instead of EL34's.
It not only sounded better with 6L6's, but the reverb problems vanished into thin air.
The point is, there are the usual suspects, and then there is everything else. Sometimes, you have to try everything.

Here, by the way, is my hunch on what the problem was...

With the EL34's, the power transformer (heater windings) were pushed pretty close to rated capacity. The power transformer itself would hum slightly when operated. With the change to 6L6's, it wasn't worked as hard and as a result, the humming stopped.
I think that while the trasformer was working so hard, it may have been producing more EMI and interfering with the reverb circuit.


Either that or the EL34's themselves were causing interference. Either way, the change to 6L6's was the fix.

Come to think of it, there were also problems with the built in EQ. If you set the bass too high, it would cause problems with annoying oscillations. That was fixed with the power tube change as well.

Dave

[bruno]

Ok so this is getting pretty frustrating... I've tried everything you sugested, rewired the reverb driver and recovery, did some chop sticking, measured voltages, I even swap output tubes. Moving some wires around does reduce some of the noise but it's always there.

As soon as I pop a 12AT7 in a turn the power on the amp starts whining loudly imediatly even if all my volumes are off, actually volume has no action over this noise. I tried removing the recovery stage tube, which reduced loudness, but the noise was still there, I remove V1 and V2, then I went further and removed the tremolo tube and the phase inverter tube. Leaving only the reverb driver 12AT7 in place. The noise is still present althought less loud. Then I put a 12AX7 in, and no noise.

Any other ideas? This doesn't sound like a microphonic noise, I have close to 460 volts on the plates of the reverb driver tube. And it only does this with the proper tube 12AT7. I've even used shielded wire to the reverb pot.

[67polara]

why would you have 460 volts on the reverb driver?  Don't you think just maybe that is too much voltage for an at7?  It's not a final so why?

[sluckey]

Did it ever work?

[Willabe]

I wonder if your reverb trans. is "ringing" when you put the 12AT7 in? The 12AT7 has more current gain than the 12AX7, maybe your amps reverb circuit is on the verdge of "whining" and the 12AT7 is enough to put it over the edge. Some guys put a 100pf cap across the 12AT7 driver tube, hook it up from pins 1&6, to pins 3&8. It might fix it by bleeding off the very high end in/at the driver tube and verb trans. Some guys always use it.

The other thing is your verb tank, maybe the 12AX7 doesn't have/give enough "ummph" to push it over the edge, but the 12AT7 does. Maybe the tanks gone wonkie, how is it mounted, where is it mounted? You said it happens even with the volume off, but with the circuit "on" something is still happening. Even removing some of the other preamp tubes, its still happening.

But it doesn't make sences that its still whinning with the PI tube pulled, to me. Should be NO output when that tube is pulled. Seems like something is feeding back (forward) to the OP stage.

67polara is right, 460vdc on the verb driver tube is too much. I didn't even notice that, I've got to read more carefully.           If you post all the voltages for the tubes, it will help to figure this out. 

slucky's the guy you want, he's very good at this. 


           Brad         

[rzenc]

How about hi-res. pics?? Maybe we can find something.
Also, take tubes out and measure resistance direct from tube pins.

why would you have 460 volts on the reverb driver?  Don't you think just maybe that is too much voltage for an at7?  It's not a final so why?

What are your voltages on the B+ rail? and on the reverb socket pins?


Best Regards,
Rzenc

[sluckey]

460vdc on the verb driver tube is too much.

460v is spot on for the original AB763 SR. May be too much for current production tubes though.

My limited experience from the past few years show that the driver tube selection is finicky. Maybe using a lower B+ is the answer. I don't know.

[bruno]

Hi guys, thanks for taking some time to read my issue...

The only reason I can think of is the 460V value being to high, but then again just like sluckey pointed out it is spot on on blackface values.

You also asked whether it has worked before, in all honesty I believe it has, but I can't really remember if I used a 12AT7 or a 12AX7 the first time around (when it worked) and I remember I also changed a lot in the amp, including mixing the reverb for both chanels.

The driver plate feed of the node before the choke, so I really have no way of controlling it unless I add a dropping resistor and extra filtering, right?

[Fresh_Start]

I'm not sure if this has any relevance, but reading Merlin's first book this morning I noticed that the 12AT7 has a much greater potential for grid current limiting than the 12AX7.  Could that affect the "clean" signal even if there is no reverb tank and/or recovery triode in place?  IOW if the signal going into the reverb circuit is enough to push the driver tube into grid current limiting, then it actually affects the voltage signal across the 1 meg grid resistor BEFORE the reverb driver.  Could that be causing the howling noise?  If so, maybe increasing the cathode resistor on the driver triodes could move the bias point and reduce grid current limiting.

Just a wild assed guess...

Chip

[67polara]

Can you post us a schematic of what you have I am sure these great minds here will see the problem in 1 second or sooner.  You changed the circuit and when ever I do that I always have problems.

Tony

[bruno]

Despite the fact that some blackface schematics show 460v for the reverb driver plates, I think the problem lays there, the 12at7 tube is giving me problems due to the high plate voltage.

I could try changing the bias point by changing the cathode resistor like sugested, but I feel I'm going to back up on this one and the drop the overall voltage of the amp as this was one of the many things I changed on one of my iterations through the amp after it was built.

[phsyconoodler]

I have built several AB763 amps with no reverb troubles.I'm betting it's a lead dress issue.We still haven't seen any pictures yet.
  However,I did have a problem with a reverb squeal with a Silverface Super Reverb that I never found the issue with.I eventually found that the board had gone microphonic and the owner didn't want to go any further with it.
  you need to recheck your work very carefully.The voltage is not your problem,believe me.I've had as much as 490v and not had them make any noises.

[DummyLoad]

...including mixing the reverb for both chanels.

undo that mod and see what happens?

[bruno]

Hi guys... I'll post some pictures during the weekend! Thanks for the tips.

Actually ISOTone I haven't tried undoing the mixing. That could be one of the changes causing this, although I don't understand why?

[DummyLoad]

I don't understand why?

point of resonance? - too much gain? - did it work beore you added the mix mod?

[bruno]

I'll get to try some more things this weekend. I can't quite recall if it worked properly before mixing it in both channels, because I might have tried it with a 12AX7.

However I think it's worth noting that even if I remove V1, V2, reverb recovery driver tube, tremolo tube, PI tube and leave only the reverb driver.  The noise is still there with a 12AT7, but not with as 12AX7. So I'm guessing this could be a problem with either the voltage on plate or the value of the cathode resistor (which is unlikely since I believe I'm using a 2.2k value).

[sluckey]

However I think it's worth noting that even if I remove V1, V2, reverb recovery driver tube, tremolo tube, PI tube and leave only the reverb driver.  The noise is still there with a 12AT7, but not with as 12AX7. So I'm guessing this could be a problem with either the voltage on plate or the value of the cathode resistor (which is unlikely since I believe I'm using a 2.2k value).

So, the only small tube that's plugged in is the reverb driver. And it still howls with a 12AT7. There's no logical way for the 'howl' signal to get to the speaker except thru the B+ rail. I'd check filter caps and grounds. I'd also look at wire routing. Maybe even look for a wiring error.

I'd love to see some hi rez pics of this amp. That could give some clues.

[bruno]

Hey Sluckey, yes with only the reverb driver tube in place (and both 6L6s and gz34 too obviously), it still howls, with a 12AT7, but not with a 12AX7.

As for the photos I'd really appreciate if you could go over them too... I'll post some on sunday!

Thanks guys

[Willabe]

But it doesn't make sences that its still whinning with the PI tube pulled, to me. Should be NO output when that tube is pulled. Seems like something is feeding back (forward) to the OP stage.

 


           Brad         

[bruno]

I read that, sorry, I just haven't had a chance to go near the amp this week...

[Willabe]

bruno,

You don't need to say sorry, sorry for what? You did nothing wrong. I am not very good at posting,     I type with one finger and I can't spell worth a darn so I can't seem to type what I mean to say half the time without tripping over my fingers.        I just would like to see you get your amp up and running again.

I did not mean to sound like I know or told you so, and for that I am sorry.


         Thanks,    Brad              

[bruno]

not at all brad, I realize you guys, who are answering to this post are doing so because they genuinely want to see the problem resolved. That's what so cool about this forum and a couple other. But hoffman has got to be at the top of my list.

Thanks

[bruno]

Hi guys, I finally managed to take some photos... I've been trying out everything with the amp. Different tubes, I used shielded wire in the reverb circuit, then I removed it and went back to regular cloth wire, I redid the original reverb mixing circuit for one channel only instead of two, I changed ground positions, did chopsticking. Everything I could remember.

The final result is the same, whenever I put a 12AX7 of any kind on the reverb driver the amp works, but obviously the reverb is too deep. If I use the proper 12AT7 (any kind of AT7) it howls as soon as I hit the standby switch, I can't figure out, why, or if the signal is bleeding throught the power supply.

Even if I remove all pre amp tubes, except for the reverb driver (PI tube included), the noise is still there, though less loud.

Here are the photos, any help is greatly appreciated, the amp looks chaotic right now as I've been messing with lots of wiring trying to figure out what is happening.

[sluckey]

See the node B filter cap? The one that supplies the screens, trem, and reverb driver? Move it's negative lead to the PT bolt where the node A caps are grounded. Maybe this will help.

[bruno]

thanks sluckey... I will try that and report back!

[bruno]

Hi, I've changed the cap ground. Made a sort of pyramid, using those 3 caps, and the one you mention sluckey is now referenced to the PT bolt. The problem persists. With 12AX7s the amp works fine, with 12AT7s, the amp howls from the minute I hit the standby off. Any ideas?

[Fresh_Start]

Here's a really dumb idea, but at this point what have you got to lose?

Disconnect the negative feedback, either at the speaker jack or the PI.  Does the howling stop?

Chip

[bruno]

went ahead and tried that too... still no luck!

Sluckey's sugestion also did not do the trick, however I did enjoy the way the first three caps are laid out after changing the ground to the PT bolt, so I'm going to leave it that way.

Any other tips? I still can't figure out why it only happens with 12AT7s

[sluckey]

Just exactly how many 12AT7s have you tried? Do you have a 12AU7 you could try?


Crazy symptoms may require crazy testing. Check your grounds resistance using this method...

Connect the black meter lead to the PT bolt where those 3 caps are connected. Now connect the red lead to each and every ground point on the chassis and board (I know, there are many). They should all read close to zero. Did you find any oddballs?

[bruno]

I tested 4 or 5 different tubes I think... I also tested one 12AU7 but the AU7 really disliked the high plate voltage (479V), and I could even smell the lettering burning away, so I quickly removed it.

I'll try measuring each ground. Thanks sluckey.

[tubenit]

I read thru the thread, but I'm not clear with it?  Did you ever lower the voltages?

I realized others have said the voltage isn't the problem, but have you tried a lower voltage like less than 400 volts?  

You've tried alot of other stuff that hasn't resolved the problem. Why not experiment with lower voltages?  What can it hurt?

I have built the one  tube reverb (12AX7) with the driver having voltages as low as 230v and was happier with it then the reverb that was stock on my Princeton Reverb. I realize it's a preference thing at some level.

I can't imagine your reverb being inadequate from using somewhat lower voltages on it?

With respect, Tubenit

[Fresh_Start]

Grasping at straws...

Can you try grounding the reverb transformer at the negative end of the filter cap for it's node on the power rail?

Is there any way you could have a "branch" off the power rail with a separate isolation resistor and filter cap just for the reverb?  That would let you try lower voltages without affecting the rest of the circuit. 

Can you swap either the primary (red & blue?) or secondary (green & black) leads on the reverb transformer?

Sorry if this isn't helping. 

Chip

[bruno]

I guys... I will get to experiment more with the amp on friday, I did lower the voltage by changing tap on the power transformer, and I basically managed to drop the voltage on the tube driver plates to 400 volts (other than the current 479V I'm reading), the problem persisted. I have moved the reverb transformer ground to a different location, even used a small connection wire to wire it directly to the chassis, still no luck. I haven't given it it's own filter node though, I might try that (what have I got to lose?). I also haven't swapped the leads on the reverb transformer.

[bruno]

Hi guys, I'd like to thank you, very much, for all the help and great support... I finally managed to track down the issue, and the amp is up and running with no issues. When I built it I was lacking a 2.2K or 2.7K resistor for the reverb driver cathode, so I used a 6.8k in parallel with a 4.7k, to get close to the value.

Turns I mischecked it because the 4.7k was open, giving me a total of 6.8K, that's why the 12AX7 worked but the AT7s howled.

I managed to track it down thanks to you guys, getting back to it with a cool head, and following sluckey's advice of checking all the grounds, which I did and in the process found the culprit.

Thanks again, no where else would I get this kind of help going.

[Willabe]

WooHoo!         

How's she sound? BTW, you did a nice job on the build, looks real nice, IMHO.

One thing tho, I'd be worried about the wires that are coming through the chassis without rubber grommets. Dougs got'em in his store, small price to pay  for safety sake. Just a thought.


            Brad               

[DummyLoad]

I finally managed to track down the issue...

good to read this! congrats!