Twin Reverb [UL]- rebuild help.

[Sea Chief]

Good folks of the MEF,

I bought a dead 135w UL TwinR with tubes OTT'ing, bias problem & some bad mods done. Sent off to Rift amps/ new tubes, mods undone, & after much patience & excellent thouragh tests.. alas the intractable bias issue remains.

Symptoms: after on for ~1hr bias 'runs away'; the conclusion.. a dreaded board rebuild (a conductive board/ the only last thing it can be I'm told by Rift on expert authority/ the xfmrs being defo signed off as good).

I can't afford the £400+ for their pro rebuild, so amp's back here. I've scratch-built a vg DlxRev, so I'm thankfully just competent enough to strip out this board, keeping bits Rift say to, ie blue caps etc > & rebuild with a new board. Then hope to send back for double-checks (esp pro bias meter checks- important) to sign off.

But what route should I go? AB763, or put back as stock UL 135w? if so, where on earth do I get a board specific to this circuit from.. maybe this factor might determine what circuit I choose for the rebuild-?

Opinions most welcome. Thanks, SC

[Jazz P Bass]

Have you tried using a hot air hairdryer to dry out the board.

They typically go conductive from moisture.

[Sea Chief]

Hi JP hope all's good in JPland..

moisture likely- yes Rift concurred with this. Hair dryer, seriously?

so hang on what's the idea: strip all out, then what 1. use a DMM to ascertain that I -do- get continuity acroos the board (I assume this is way to proove a 'conductive' board-?).. then if so/ 2. scrape of waxy oomschka from board (apparantly they were dipped in such gubbins? why to prevent moisture ingress I assume?).. 3. hair dry it > hope to see no continuity/ ok board?

[pdf64]

You'll likely never measure board resistance with a DMM, the issue generally only manifests when there's a lot of voltage.
The amps are excellent and worth a bit of time and trouble, if you're so minded.
Taking your report of Rift's findings at face value, I don't perceive that it's worth putting any further work into the old board.
You may be competent to build a new board using proper material, and use turrets rather than eyelets, several options here Amp Maker: Guitar amp kits and parts :: Turret board + tag board
I think that the HT of 135W is too high for the TR AB763 circuit to work reliably, and anyway a g2 HT node with choke and stacked caps would need to be added.
Best just to get the amp back working as intended.
The addition of an overall bias adjustment control, and 1 ohm cathode current sensing resistors would be beneficial.
Don't expect tubes that have red plated to run stably again, ie you'll probably need a new 6L6 set, get your vendor's advise about the most suitable type to cope with the voltage in this amp. Sovtek 5881 used to be able to take it.

[Chuck H]

Things I'd like to know, but were't mentioned...

Were the power tubes replaced?

Were the bias circuit capacitors replaced? (they aren't always addressed with a "cap job")

Has the tremolo been examined as a possible culprit? Since it affects the output tube bias any problems from an old tube, roach, leaky coupling cap, etc. could likewise affect the output tube bias.

If the answer is no to any of these there are still avenues to explore before rebuilding the board.

[Sea Chief]

Hi there Chuck- yes bias caps replaced (& a new wee bias circuit to a pot on back too- usual mod I think), power tubes replaced too with matched quad of JJ's. Also all filter caps replaced, new 25/25 caps in the preamp area of board too. So all in all quite extensive revisions, & a huge ammount of time spent testing xy & z over 6 months it was there (a favour to me).

As to trem circuit: I'll ask but I'm absolutely sure would have been tested too; basically a rebuild was their diagnosis & next step -they- would take.. so I'm totally adamant every facet you mentioned would not have been overlooked. Therefore it is a rebuild I'm looking at.

Thanks, SC

[Chuck H]

Well, I just revisited the schem and the trem notion was a mistake. At a glance it looked to be coupled to the bias circuit, but it's just some ground leads shared on the schematic. So, unfortunately, you may be back to considering the board

[Tom Phillips]

...Symptoms: after on for ~1hr bias 'runs away'...

...Were the power tubes replaced?...

...power tubes replaced too with matched quad of JJ's...

In situations such as this it is prudent to try another set of power tubes. One can never be 100% sure that new parts are good parts. New tubes can have defects. One tube within the quad can have an intermittent problem that loads down the bias voltage and cause a runaway condition.

[66 Kicks]

Replacing the main board seems a bit heroic, especially since I'm not seeing anything in your posts that convinces me that the board is faulty. Is there anything else that malfunctions that could possibly be associated with a conductive main board?

The bias circuit is spread over three boards. The bias winding, the 1.2K resistor, and the diode are located on the little board with the bridge rectifier. From there a wire goes to the doghouse where the 2.2K resistor and the bias capacitors are located. From there a wire goes to the center tap on the tube matching pot. where a wire also goes to the 2.2M resistor on the main board that biases the tremolo oscillator way into cut-off. The outside lugs have wires that go to the power tube grid leak resistors on the main board. If you can get the bias circuit without the power tubes installed to malfunction long enough to get voltage readings, you should be able to pinpoint where the problem(s) is/are.

So there are only two places that the bias circuit touches the main board. If all four tubes red-plate (OTT'ing?) at the same time, it is unlikely that the board is causing the problem at the grid leaks, although not impossible. The wire from the pot. center tap to the 2.2M resistor can be temporarily lifted at the 2.2M end to see if this eliminates the problem. All four points at the grid leak resistors can be raised off of the board and temporarily tacked together in mid air to see if this eliminates the problem. If neither eliminates the problem, replacing the main board will not cure it.

[Chuck H]

I inferred that the bias problem existed before the amp had the tubes replaced. But the opening post doesn't say that. It only says that there was a problem with the bias, not necessarily the same problem. So in this case I shouldn't have qualified the tubes yet. Yes, "prudent". As in maybe NOT have to replace the whole circuit board!

[Sea Chief]

I inferred that the bias problem existed before the amp had the tubes replaced. But the opening post doesn't say that. It only says that there was a problem with the bias, not necessarily the same problem. So in this case I shouldn't have qualified the tubes yet. Yes, "prudent". As in maybe NOT have to replace the whole circuit board!

Good point Chuck.. I should've said I got amp with the prob & a very bad set of tubes inc. 1st thing Rift told me to do was buy a new set & send with amp to them. Cheers for looking at schematic for me btw.

66 Kicks, Rift would have done every possible test to pinpoint if they could, as they were going to get up & running for me.. & concluded that only a rebuild etc. Knowing their [his] huge wealth of knowledge as being both an excellent builder & as one of the best repairer both here in UK & in the EU.. I just take his word, that a conductive board/ rebuild job it is. Even all 4 pwr tube sockets were replaced & new R's put in, plus new R's in various places on board ie bias circuit as well as many board caps, all doghouse recapped. All in all its had a tip to toe service/ everything plus the kitchen sink has been thrown at it.

[Chuck H]

Well, unless the amp has other issues related to a conductive board, tonal or safety, you could also rebuild the bias circuit on it's own wee board and elevate any connections from the PI coupling caps to the power tube grids. Not pretty, but it would probably solve the problem and it's a lot easier (and cheaper) than replacing the whole circuit board.

[66 Kicks]

A little touchy about Rift, eh?

If you are only interested in replacing the board, Fender used the same old Twin Reverb board with a couple of extra eyelets punched in it. You can do the same with a standard two channel w/reverb ab763 board.

[Tqi]

https://tubeamplifierparts.com/turre...=0&sort=normal

Here's a source for AB763 circuit boards and additional eyelets, if you don't want to build your own from scratch. Check it against your existing boards to find the extra holes you need to create for the UL135 changes. Best to replace all three boards - if one's gone bad, best not to trust the other two that were probably cut from the same batch of fibre.

Save all the capacitors if you can, they're expensive and it sounds like most of them have been replaced at least once of late. The resistors and diodes are literally pennies, you can get a good supply from Barry at AmpMaker (Amp Maker: Guitar amp kits and parts) which is a lot cheaper than buying them 100ea at Mouser, or buying fancy NOS or Reissue ones from TAD or similar. All his resistors are 5%, so you can ignore any 10%/5% distinction on the schematic. If you can salvage the caps, this can probably be built for £70-£100. Double that if you need to replace the caps - Barry's selection of those isn't perfect, import the missing ones from AmpRepairParts.com. Mouser's great, but you'll be buying 10 of everything and they'll all be radial. ARP has everything except the tiny Mica picofarad ones in Axial form. They also have replacements for the Bias pot and Heater balance pot as well, if you need them - both original CTS parts in Fender packaging.

Replace the high voltage hookup wire with Barry's PTFE kits while you're at it. The wires from the pots are probably fine, but the wires between the boards and to the tube/output sockets should be replaced. The best advice when fixing something is to fix one problem at a time, but if you're doing a complete rebuild anyway, then you're better off replacing as much as possible in one go and treating this as a new amp build with a few NOS or used parts.

This should be a fun project for you, but I'm sorry it's an 'involuntary' one. Best of luck.

[Sea Chief]

A little touchy about Rift, eh?

If you are only interested in replacing the board, Fender used the same old Twin Reverb board with a couple of extra eyelets punched in it. You can do the same with a standard two channel w/reverb ab763 board.

Not at all- I was just saving you(s) from time suggesting 'try this, try that'.. as I'd only say 'been done' bc they've (pros) tried every little thing & conluded > the board. I could go round in circles for a year trying this/ that, which has probably already been tried, or, take their advice & crack on.

I will see (tentatively: I can't ask QQQ's if they've concluded -board- already) if they think splitting the bias section off into a new wee board.. but are we talking cut the board next to the 2nd 0.1uf far LHS of board, cut here? And I will ask if the trem circuit components have been checked as it feeds into the bias circuit. But any more Q's than that & they'll be a lid flying off @ Rift!