Fender Twin Reverb with crackling noise on the Reverb

[nevetslab]

This 2008-vintage Fender Twin Reverb (65 reissue) came in with a nasty buzzing crackling noise on the reverb circuit. The tank itself appears fine. I went thru a few V4 tubes, no change. Removed V4 to listen to see if that noise was still present, and it wasn't. Swapped out the PI tube V6, no change. Not affected by the Vibrato Intensity control, which sums it's modulation signal with the output of V4B. Removed V2 just to eliminate it's contribution to the signal chain, and no change. I also unplugged the tank, and turning up the pot, I was getting the same noise, just not modulated by the crash of the tank. Injected signal into the Reverb return jack, and get the same noise & crackling.

So, removed the chassis to the service cradle, and, since I was hearing some noise on the Reverb pot as well as both Volume Pots, I removed the front panel PCB, removed each of the pots, cleaned / lubed the resistance tracks and put them back onto the PCB, and touched up the solder joints on the input jacks while I had the board out. Put the board back in into the chassis. I also cut the tie wraps Fender had used to bundle the AC mains pri/secondary wires, all laid out straight to where they plug into the main PCB, and twisted each secondary pair, then twisted the three pairs together, laid them back down and plugged them back in.....something I normally do on these amps, just to lower the radiated stray field from the wires. Powered it back up and gave it a listen. Now it's quiet, no more crackling noises or hum buzz. Put the chassis back into the cabinet, tightened up the mtg hardware. All sounded fine again, no crackling. I had it running with burst pink noise while I began screwing down the upper rear panel, and suddenly the noise was back.

Removed the panel, then removed the three rear chassis flange screws that hold it tight against the cabinet roof foil. Amp is quiet again. Pressed the rear of the chassis up against the foil, and had the noise again. So, pulled the chassis back out to see what I missed.

Nothing obvious inside the chassis. All the rear panel hardware was tight. I gave the grounding screws a tweak, just to be sure they were all tight, and did the same with the xfmr mounting hardware. The AC pri/secondary wiring bundle was bulging a little above the surface of the open chassis (sighting with a straight edge), so pushed it down as low as it would go, and did the same with the wires from the Doghouse (power supply cap housing). Opened the Doghouse for inspection, all looked fine there, though had to tighten down the PCB mtg screws a bit. Powered back up, and all is quiet like before. I put my aluminum shield plate onto the top of the chassis to simulate it being mounted, holding it down firmly. No change with that, still quiet.

I turned to the cabinet, finding two large patches of aluminum roof foil torn from the large sheet applied to the cabinet roof. I peeled away some of the edges that were protruding down, then, after turning the cabinet upside down (handle removed), I placed a couple strips of 2" wide gaffers' tape to the roof foil, and cut away the foil that had been folded up onto the side of one wall. The chassis had been really hard to remove from the cabinet to start with, and did require having a thin metal panel slipped in between the roof and the chassis, to run from one end to the other to prevent the chassis from peeling away roof foil, as they just love to do.

So, feeling confident I'd addressed everything, short of ripping off the entire sheet and replacing it with a stapled-down 10mil sheet of aluminum, I re-mounted the chassis. Didn't even bother with the rear flange screws...just wanted to see if it was quiet again.

It wasn't....same noise just as I had when I began. Bloody amp sitting there laughing at me, cackling with it's nastiness. Sigh.........

As I didn't find any components in the Reverb return path respond to prodding with my chop sticks to try and make it misbehave, I'm not sure what I missed. I'm ready to remove the roof foil and cut/install a new sheet of aluminum (roof flashing), and staple it into place. That's an hours task, as you have to gut the cabinet so there's nothing in the way of the staple gun during the installation.

I haven't run into THIS problem before. What have I missed?

[nevetslab]

Further checking.....I loosened the top mounting screws to drop the chassis down and away from the roof foil. No difference. I unplugged the reverb cables again, and this time, the noise was gone. Tightened the top chassis screws back up, still quiet (with the cables unplugged). It wasn't like that before the first time around, so at least the work on the front panel PCB made a difference. Plugging the cables back in brought the noise back, so, switched back to S/B, removed the tank and cable. I found one of the two long tank springs had broken it's termination wire, while the other end was still intact, and the spring had unhooked, now laying in the pouch. I was out of 4AB3C1B tanks, but have a bone yard of broken ones.

While the tank was out, I fetched a replacement reverb cable, though it was a 4ft stereo cable, not the shorter one for the Fender amps. That hooked up fine, reverb worked, no noise, so the cable appeared to be the culprit, or one of them. Now, having never tackled trying to repair a tank's termination wire, and seeing they're held in place in the brass tube with both the rubber sleeve and solder at the back end, after being folded over, I unsoldered the one that had broken off, still leaving a stubble of the wire. Gripped it, heated up the joint and extracted the wire. Did the same with one of the broken tanks, straightened the end so I had a chance of forcing it back thru the rubber sleeve. That was a fumble session, though succeeded. Folded the wire over, soldered it into place. Found the smaller size spring and stretched it out, hooking it at both ends. I didn't have any epoxy mixed up, so I dabbed a touch of RTV on the two hook joints, checking to make sure I didn't have them touching the brass sleeve.

Connected the cable back up, switched back out of S/B, to give a listen. Still quiet, and the Reverb works, noise not there. Re-mounted the tank into the pouch, installed it back into the cabinet, folded the excess length of the cable up, and secured the cable to the side wall, plugged it back in. Now the tank sounds a bit dark, and not like the new tank I just installed in the previous Twin Reverb.

Has the RTV I applied on the two termination lead hooks made its' way onto the brass sleeve to damp out the spring? That's my guess. That end of messing with reverb tanks is uncharted territory for me. But, the noise and crackling problem is at least gone.

[Chuck H]

Now the tank sounds a bit dark, and not like the new tank I just installed in the previous Twin Reverb.

Has the RTV I applied on the two termination lead hooks made its' way onto the brass sleeve to damp out the spring?

Maybe? I wouldn't have thought so. Perhaps it will improve as the adhesive continues to harden with age?

A new tank was only avoided because you didn't have one on hand. So this repair is a likely win anyhow. You could still test the "tone" of a new tank in this amp if/when you get them both in the shop again.

[nevetslab]

Maybe? I wouldn't have thought so. Perhaps it will improve as the adhesive continues to harden with age?

A new tank was only avoided because you didn't have one on hand. So this repair is a likely win anyhow. You could still test the "tone" of a new tank in this amp if/when you get them both in the shop again.

Well, I have one more Twin Reverb in the shop to service, though.....it's in there with issues on the tank. Usually, it's just one of the thin wires between the RCA connectors and the 2-pin connector, or the wires that have been directly soldered as those IDC connectors always have their wires snap off. So, I'll see how it wounds in tomorrow.

[g1]

Any time I've replaced those little 'cores' that the springs hook on to, the tone is never the same. That's even with using nail polish rather than RTV at the hooks. So the RTV may or may not have made the situation even worse.
I always figured the rubber dampers (in the brass tubes) to be more like foam; and I figured no matter how careful you are, the heat probably impacts it's density resulting in a change in the damping factor.

[Helmholtz]

The magnet-to-spring hooks must be joined rigidly to facilitate high frequency coupling. If the magnet has a round wire hook, a rigid glue (e.g. thick Super Glue) or cement is needed. Silicone is too soft and elastic.
Old Hammond tanks used a flat wire magnet hook that didn't require glueing.

Another concern is the orientation of the magnetic axis of the ferrite magnet which influences transducer efficiency. If the magnet was turned out of optimal position /angle during repair, reverb signal may suffer.

[nevetslab]

The magnet-to-spring hooks must be joined rigidly to facilitate high frequency coupling. If the magnet has a round wire hook, a rigid glue (e.g. thick Super Glue) or cement is needed. Silicone is too soft and elastic.
Old Hammond tanks used a flat wire magnet hook that didn't require glueing.

Another concern is the orientation of the magnetic axis of the ferrite magnet which influences transducer efficiency. If the magnet was turned out of optimal position /angle during repair, reverb signal may suffer.

Ah...two mistakes made then. I was about to pull the tank back out, get the Silicone glue off, change the spring I had selected, having mixed it up with the one that came off. I wasn't aware of the orientation of that tiny ferrite magnet. I had removed that from another tank, managed to get it fed into the rubber sleeve inside the brass housing, and, it may be in the same orientation, but no certainty of that. I'll see what it sounds like with the other spring before gluing it into place. It was an attempt to restore operation without having to order more tanks, which I have to do anyway.

OK, a silly question. I have a collection of broken 4AB3C1B tanks from Fender Twins, Deluxe Reverbs, where I thought one day I might take a crack at cobbling working tanks out of them. How would you go about optimizing the orientation of that fine magnet wire/ferrite magnet without soldering it into place first? All have at least one of the ferrite magnets/termination hook for the spring missing (broken off).

[Helmholtz]

OK, a silly question. I have a collection of broken 4AB3C1B tanks from Fender Twins, Deluxe Reverbs, where I thought one day I might take a crack at cobbling working tanks out of them. How would you go about optimizing the orientation of that fine magnet wire/ferrite magnet without soldering it into place first? All have at least one of the ferrite magnets/termination hook for the spring missing (broken off).

No silly question at all! I kind of expected it and hope my limited English suffices to explain this.

The magnetic axis is perpendicular to the cylinder/barrel axis. So one half of the magnet is a north the other a south pole. You want the magnet axis to be parallel to long core legs for max. torque. The magnetic axis is the line between/through north and south poles.
To determine the position of the poles you may use a needle compass or just let the magnet attach to another magnet. The sides that attach are poles. Opposite poles being on the opposite sides. Doesn't matter which is north or south as it will be excited by an AC field.

When you have identified the pole sides, you may mark them at the front of the magnet with a tiny dot or just observe the relative orientation of the magnetic axis to the plane of the hook.

magax.pdf

The picture in the middle of the page shows double-arrows indicating exciting AC field (1) and magnetic axes (2) of the barrel magnets for most effective operation. If AC field and magnet axes are parallel to each other, torsional force is zero.


I hope this is clear enough.

[nevetslab]

No silly question at all! I kind of expected it and hope my limited English suffices to explain this.

The magnetic axis is perpendicular to the cylinder/barrel axis. So one half of the magnet is a north the other a south pole. You want the magnet axis to be parallel to long core legs for max. torque. The magnetic axis is the line between/through north and south poles.
To determine the position of the poles you may use a needle compass or just let the magnet attach to another magnet. The sides that attach are poles. Opposite poles being on the opposite sides. Doesn't matter which is north or south as it will be excited by an AC field.

When you have identified the pole sides, you may mark them at the front of the magnet with a tiny dot or just observe the relative orientation of the magnetic axis to the plane of the hook.

I hope this is clear enough.

Your English is definitely clear. I did pull up some diagrams of cylindrical magnets to help get my bearings, as well as placed a smaller tank on the desk to look at the mechanics. I see the tiny ferrite cylinder magnet oriented nearly symmetrical within the core laminations' thickness, which when I had positioned my replacement wire/magnet, I did my best to repeat, seeing those others in the same relative position. I wasn't aware of the pole markings, but in looking at a tank that's complete in front of me, I see (with my loupes and surgeon's headlight) the cylinder magnets are marked with a line at one end, like we'd see on a diode. And, looking at each spring, both magnets are oriented N-S-Spring-N-S. And the adjacent coil is oriented opposite it's neighbor, S-N-Spring-S-N. Not sure if all the tanks are set up that way, though this Accutronics tank I'm looking at is.

I hadn't caught your reply here prior to my opening up yesterday's attempt with that tank, and removed the silicone glue, removed the spring, cleaned out all of the leftover glue strands, and installed the other spring. I gave a listen to the tank while still unglued, and it sounded again like a normal 4AB3C1B tank, while not having a second amp set up to compare it to. Didn't have the darkness I had inadvertently caused with the damping of the silicone glue. I added a tiny dab of JB Weld 2-part epoxy to the spring/hook joint, removing all excess glue threads so it will harden to a similar glue dab as we see on these production tanks. The spring I removed and the one I installed today were the smaller dia wire in the springs relative to the thicker wire used in the spring that was still intact. That much I was aware of. The thinner wire spring tends to droop more than the thicker wire spring.

As the tank is now back in the pouch and mounted, I've no idea if I got the magnet poles oriented correctly with regards to it's mate at the other end. 50-50 chance on that. Now that I know, and would be having to cobble other tanks based on what I can salvage, hopefully I'll be able to get the replacement wire/magnets extracted from them to rebuild what I can. Most of the boneyard are just missing one spring from the two, though some have open xfmr coils and both springs intact, so they'd serve well for the magnet/wire to repair those with broken ones.

Many thanks for your explanation!

[Helmholtz]

Please see the illustration I attached to my previous post #8.

Pole orientations of the magnets at opposite ends of the springs don't influence each other. What matters is the magnet orientation within its transducer core.

[nevetslab]

Please see the illustration I attached to my previous post #8.

Pole orientations of the magnets at opposite ends of the springs don't influence each other. What matters is the magnet orientation within its transducer core.

Ah, you must have added the attachment after/while I was replying. I didn't see the attachment until now. OK.....all of these Accutronics 4AB3C1B, being built for horizontal open-side face down, they're NOT in the optimum mounting orientation. So, now looking again at a dead 4AB3C1B's magnets and their markings....I see three of the four magnets with a tiny mark on one end of the cylinder that could be marking North. One has no marking that I can see. The two that do at one end of the tank, their markings are facing each other, as to oppose each, perhaps.

Somewhere in my collection of instruments I have an Annis Handimag and companion 5 Gauss Magnetometer, which might be suitable for this application. Goes back to my studio tape machine maintenance days....it's probably hanging out with my Tentelometer in storage. I can't think of where I have a pocket compass. No doubt lost it with what camping gear I had decades ago.

Well, this is an education to get it right the first time (when I tackle the bone yard of broken tanks to cobble working tanks out of. Somehow, I suspect the compass needle will be more sensitive than the 5-Gauss Magnetometer. You can't determine North with that instrument.

[Helmholtz]

Again, it does not matter if you have a north or south pole. Just use the "let 2 magnets attach each other" method I mentioned above.
The point is that the AC magnetic field of the core and the line from north to south pole (or vice versa) of each magnet (= magnetic axis) should not be parallel, but at right angles instead. This ensures max. torque.
The relative orientation of north and south poles of different magnets is irrelevant.

It might be confusing that the attachment is originally about mounting orientation. But this is not what I'm speaking about. Just used the picture for convenience. I wanted to show that the magnet poles should not face the core legs, because in this case they would not be caused to "rotate" and not generate torsional waves.
The magnetic forces tend to align the poles with the magnetic field between the core legs, just like a compass needle rotates and aligns with the external field. If the magnet poles already face the legs there will be no (torsional) movement.

[Leo_Gnardo]

it's probably hanging out with my Tentelometer in storage. I can't think of where I have a pocket compass.

Stand back, nevets has his Tentelometer! Instead of measuring tape tension, you can make sure the spring tension on your Frankensteined reverb tanks is correct, once you have the magnets & hooks all sorted out. I had the same idea, many years ago. Collected a good 4 dozen wrecked tanks then started discarding bad ones. One day about 20 years ago I gave it everything I had, spent a day trying to restore a tank. Because I had noticed the longevity of our good ol' USA made Accutronics tanks had gone to blazes, and like you I had the idea I might be able to restore them to health the nuts & bolts way. What I discovered, it was an immense waste of time.

A couple years ago Juan related his story about trying to manufacture his own tanks. I think his review went something like "it's an exercise for people who want to drive themselves insane."

I still have a big box 'o tanks in the attic. AS IF they'll ever get fixed. I better pitch 'em in the dumpster. What is good, we have some OK reverb tanks from China now. I'm not much happy with the ones from Korea. And the price per tank is less than 20 years ago.

Something else to keep in mind, reverb tanks sound different one to the next, a fact I ran across back when I bought Accutronics a dozen at a time. Some were outstanding, most OK, and there was always a dog in each batch. Even those eventually found homes.

[nevetslab]

Again, it does not matter if you have a north or south pole. Just use the "let 2 magnets attach each other" method I mentioned above.
The point is that the AC magnetic field of the core and the line from north to south pole (or vice versa) of each magnet (= magnetic axis) should not be parallel, but at right angles instead. This ensures max. torque.
The relative orientation of north and south poles of different magnets is irrelevant.

It might be confusing that the attachment is originally about mounting orientation. But this is not what I'm speaking about. Just used the picture for convenience. I wanted to show that the magnet poles should not face the core legs, because in this case they would not be caused to "rotate" and not generate torsional waves.
The magnetic forces tend to align the poles with the magnetic field between the core legs, just like a compass needle rotates and aligns with the external field. If the magnet poles already face the legs there will be no (torsional) movement.

Ah, I see. Well, the first set of magnet/spring-hooks to try this on will be the tank that has an open coil. It's magnets/hooks and springs are all intact. I might take a whack at that in the morning, having finished up with this last Fender Twin. Some of these tanks are like a bad penny....they keep coming back with broken tank leads, or connections to the tank xfmr pin connections. Ain't got nothing to do with the tank surgery you've helped me out with. I like the idea of letting the un-mounted magnet/spring-hook parts define their mounting orientation!

Many thanks!

[nevetslab]

Stand back, nevets has his Tentelometer! Instead of measuring tape tension, you can make sure the spring tension on your Frankensteined reverb tanks is correct, once you have the magnets & hooks all sorted out. I had the same idea, many years ago. Collected a good 4 dozen wrecked tanks then started discarding bad ones. One day about 20 years ago I gave it everything I had, spent a day trying to restore a tank. Because I had noticed the longevity of our good ol' USA made Accutronics tanks had gone to blazes, and like you I had the idea I might be able to restore them to health the nuts & bolts way. What I discovered, it was an immense waste of time.

A couple years ago Juan related his story about trying to manufacture his own tanks. I think his review went something like "it's an exercise for people who want to drive themselves insane."

I still have a big box 'o tanks in the attic. AS IF they'll ever get fixed. I better pitch 'em in the dumpster. What is good, we have some OK reverb tanks from China now. I'm not much happy with the ones from Korea. And the price per tank is less than 20 years ago.

Something else to keep in mind, reverb tanks sound different one to the next, a fact I ran across back when I bought Accutronics a dozen at a time. Some were outstanding, most OK, and there was always a dog in each batch. Even those eventually found homes.

LOL! I never thought about using the Tentelometer for something like that. I guess you'd have to place the calibration tape strip with the weight (forget it's weight now) so the tape passed thru/between the springs, then lift up and......calibrate your arm force first. I think Juan's right, as far as trying to manufacture his own tanks. This past exercise was more of a challenge to see if I could even get one working again, not realizing the trick Helmholtz revealed on optimizing the magnet orientations.

I haven't yet landed on a use for the long tank chassis.....I think that's mostly why I haven't tossed them out, since there's enough volume and real estate inside to build something into them.....what, don't know. Free metal work, even though cheap. If I tossed them, that's when I'd find a use for them.

I've never listened to a batch of them, looking for the best one. With all the amps in our inventory for rental, and how often they fail, you'd think there would always be at least a half-dozen of each type on hand. Just like having a well-stocked spare new tube stash. Maybe some day. Having multiples of popular amps tends to save us from having too many down and out of service.