Tubemeister 36, TSC and resistor blown

Hi everyone!

After my Tubemeister 18 that I still own and I found great, I bought an used Tubemeister 36 30th Anniversary to have a little more headroom in some situation where my band and I have to play with basic amplification and guitars not mic'd.

I took it home, powered up and bang... one of the four TSC leds (the outer right) always on. I'm damn finicky, so, even the amp was playing good (I noticed a little background noise) I wanted to understand why the led was on. Strange, 30th Anniversary edition should be something like four years ago, a tube should live for years... anyway.
I opened the head and tried to swap the two outer tube and the issue seemed to be confirmed: now the outer left (where I put the suspect tube) was lighted and the outer right was blinking. According to the TSC system, that should mean that the suspect tube was faulty and the twin one was shut down by the system. Fine.
I ordered four JJ EL84 to replace them, once done I powered up the head again: the outer right led on again. The amp was playing great, less noise (maybe none at all), the tube was receiving power and it was lighted up (seen by eye, I mean), but why the led on? Maybe a TSC issue or not properly "calibrated"?
I took the suspect original tube again and tried it . Three JJ and one original. I was quite smooth, the TSC would bias everything.
Powered up the amp, smoke and smell of burned. Shhh.....t! Instant power off, replaced the tube and tried again: the amp was working good. TSC outer right led still on. Why the smoke?!
Because I'm mad and silly, I tried to test the inner couple replacing the brand new JJ with the faulty tube and another original working tube. Smoke and smell again, then a small "pop", all TSC leds on and no more sound by the amp.
F....ck.

I let it cool down, I disassembled it, now I've found four 2W resistors, I think what they call "anode fuse" (or cathode, I don't remember), three of them visibly burned.
They are R134, R93, R92 and R133. The only one that seems fine is R93. The problem is, while I can see the color of stripes on three resistor, I can't really read the ones (especially the central stripe) of R133... I think that could be the issue, but I don't know how to replace it since I don't know its value.



I searched everywhere on the Internet looking for a service manual... that I found for the Grandmeister, but not the Tubemeister... I tired to mail the H&K support with no luck, then tried to contact them via Facebook, but they answered me they don't send service manual or values directly to customers, but invited me to take it to an authorized service (that's far away from my town)... this used amp would cost too much for a servicing.

R134 seems weird to me, too... it seems to be a "zero value" resistor?!


Anyone could please help me?

Thanks, thanks, thanks to everyone.

Rob.

Sorry to hear about your troubles.  There are a few advisories here.

Robyhood wrote:I took the suspect original tube again and tried it . Three JJ and one original. I was quite smooth, the TSC would bias everything.

NO!  Don't muck about with your gear like this.  What did you expect to find improved by going back to the original valve and setting everything up wrongly?  You now have a pair of unmatched valves in there with one of them a possible suspect for a problem.  If you wanted to see the result of testing the new valve in that slot then you would have been better swapping the two new ones in the pair as you originally did.  The new valve is far less likely to be the source of the problem than the old one.  Would you put your old brake pads back into your cars brakes to check whether it was them that were at fault?  I'll bet you wouldn't!

Anyway, to the problem.  You are right, the TM36 schematic is not out there in the wild, but the GM36 is.  And what do you know, those resistor references you gave match with that, full marks for giving us useful information without us having to pull teeth to get it, people don't always make it this simple.  The match from TM to GM doesn't surprise me as there will be certain shared "family characteristics".  Therefore, it very much looks like these are the Screen Grid resistors which are 1k 2W devices.  (Should of course be 'Brown Black Red' where it matters but some Chinese resistors seem to use a code of their own nowadays.)  It is common for those to be taken out with some valve faults.  IF YOU ARE ABSOLUTELY SURE YOU ARE ABSOLUTELY CAPABLE OF DOING THIS, (WITHOUT JOKING DEATH IS A POSSIBILITY), then change them all out.  You have no way of knowing whether the others are damaged or not, it isn't always obvious externally. The parts are not expensive and it wouldn't do any harm as long as your desoldering and resoldering technique is up to scratch.

There is NO tonal value in messing around with resistor types.  Do NOT be tempted to replace with some concoction of NOS carbon composition crap which you have read on a forum will give you Santana's tone and which should have been consigned to the bin many years ago!  Things don't work that way except in uninformed peoples' imaginations.  Get a decent quality metal film or preferably metal oxide.  The object here is to dissipate heat and withstand high voltages.  (Yes, I know that screen grid resistors don't dissipate massive amounts of heat or have high voltages across them in use but they sure as Hell can when a fault occurs in a valve.)

As you have just bought this amp and don't know its history, while you are in there I would also check the value of all of your fuses, especially the Anode Fuse!  I would have thought that that should have blown by now.  You have a mains fuse in your plug.  If you are on the East side of the Atlantic in the UK (220V) that must be a T800mAL (T= time delay/slow blow).  For the West side in the USA (120V) it should be a T1.6AL.  There are two fuses on the PCB inside the amp.  These are labelled FU1 and FU2.  FU1 is the anode fuse and MUST be T400mAL universally.  It should be the one next to the diodes and furthest away from the capacitors and pots.  The other is the low voltage fuse and MUST be a T800mAL universally.  If these are the wrong values which is common with musicians who "know" that putting a 3A fuse in place of that hard to find 400mA one will do no harm, (their electrickery guru told them so), then you lose all the carefully calculated protection and might as well not have the damned thing in there at all.  Or perhaps you should have a wrap of silver foil, that would do just as well "for now".

With that work done and tested, let us know how things go as this may not be the entire problem of course. It is possible the TSC area has damage too and that isn't home fixable.

Bordonbert, thank you so much for your answer.

You are right, I shouldn't mess with new and old tubes, but since I found the same exact TSC led on with new tubes... I didn't know what to do! Maybe just enjoy the amp... instead I fried it...

Anyway, those "screen grid resistors" are four, one per tube of course. Two of them (the central ones) are 1KOhm (all four are 2W). But I didn't know the other two... so you say to watch in the Grandmeister 36 schematics? I understand the values isn't important for tone, but I think it's important to make the respective tube work fine, right?
I'm fine with soldering, I build effect pedals, too, from schematics... I'm just a quite scared about "lethal voltage" of capacitors... I mean, everybody talks about them with scare... the amp is disconnected from days from mains...

About the fuses, I noticed (have to check again) two fuses on the board, I checked both and continuity is ok.

What I noticed are bad soldering of tube sockets, it seems manually soldered by someone not very skilled. This is strange, I thought everything was soldered by "robots"...

I can set your mind at rest over a couple of things and worry you with a couple of others.

These 4 resistors will all be the same value.  If the two inner are 1k then that matches up with the H&K schematic.  It might be that the outer two have been changed out before which could be why they look different.  Replace all four of them with good modern 1k 2W components and you will be set to go.  Certainly the correct value is important for the amp to work as designed but there are some stupid beliefs out there as to how different types of component affect tone.  I defy anyone to be able to tell the difference between resistor types in a guitar amp with around 2% clean distortion levels when they don't know which they are hearing!  This is not hifi.  Blind listening tests show this piece of "common knowledge" to be a non-starter.  The different types are important from the point of view of noise, stability and reliability and in all of those the modern types win hands down.

As to the lethal voltages in the amp.  The major problem is having to work in there when it is live!  You grasp, you die!!!  You aren't in that scenario.  H&K do what everyone should do in their designs, they fit drain resistors to the main PSU caps.  These slowly drain the charge out of the cap and bring the voltage down to safe levels.  As long as you give it a few minutes between switching off and unplugging from the mains you will be ok.

The fuses also MUST be checked for their value not just to prove they still have continuity.  If someone has put a higher value fuse in there that could be part of the origin of your problem.  The fuses' values will be stamped into their end caps and you will probably have to remove them to check it.  Carefully prise both of the PCB fuses out of their holders with a small screwdriver or similar, one at a time so you can't mix them up, and check they have the correct value as I posted.

So far as I am aware, everything IS soldered by robots.  It sounds very much to me like you have got your hands on an amp which has had problems before and they have only been partially addressed.  I would gently reflow those valve sockets yourself with a little fresh solder, that costs nothing.  Being subject to heat from the valves in some amps they can benefit from a touch up like this every few years (I said years people, not months).  It sounds from your original post as though the TSC LED was on from the moment you first fired the amp up.  That could be because there was an existing valve fault of course but it has been known to be due to a problem in the TSC circuitry itself.  If so that would be a "new board" scenario as that area is not reworkable.

I would get this set up as soon as possible with the new resistors and valves as that is cheap, see whether the TSC then goes clean, and if it stays lit up then have a think about whether you want to talk about it with whoever sold you the amp.

As long as you can get to grips with this problem and solve it without too much cost and hassle it may not be too much of an issue and you could end up with a decent amp at the end of it so don't panic.

Thanks again for everything!

However, could you confirm the "strange" resistor (all black bands) is a 0 KOhm resistor?!
Anyway, I will replace all four ones with 1 KOhm 2W.
About fuses, I understand your point, but how could they be the issue? I mean, maybe they won't "act" like fuses (that's dangerous), but if the inside filament is okay and there's continuity the amp should work.

About TSC... if after replacing the resistors everything was ok but the TSC led... should I worry or just keep playing?
I read a discussion about a Tubemeister 18 (or 36, I don't remember well) with the same issue, with TSC led working ok after 1K resistor replacement... Anyway it shouldn't be THIS amp, if I'm right it was somewhere in the USA, I'm in Italy.

Man, you gave me great help! I'll let you know as soon as I do the work!

Yes, the technical value for the resistor is 0ohms but it shouldn't be in that location. It is possible that these are just 1k resistors which have been picked up as cheap Chinese knockouts. It could be out of the backdoor as mismarked or a number of ways. These aren't the H&K originals I would think, they are most likely a botchy repair. The 4 numbered resistors you gave us should be 1k.

You are correct, the fuse shouldn't be the current issue but it could have been the cause of the original damage. If it is say a 2A fuse instead of a 400mA one then that represents a really serious overload under fault conditions before the fuse blows and stops the current flow. If say a valve went out on you and the fault current maxed out at only 1.5A that could take out any number of items and the fuse would do nothing to protect them. You really need to check the value to know that you are protected by it in the future. That's the only reason it is there in the first place.

If the TSC LED is still lit with everything sorted then you have either not got everything sorted or the TSC itself is damaged. It really should be looked into as it could be something important. There could be damage to the MOSFET in the power valve cathode or the TSC control circuitry with its dedicated micro-computer chip could be damaged. That would need more specialist equipment to diagnose, and getting replacement parts for the TSC control means basically a new board from H&K. You have already seen they will not supply customers, only service agents. So you're up against it that way.

Anyway, don't worry, let's cross that bridge if we come to it.

I'm back... as a winner!!!!

After your advices, I bought those 1K 2W resistors. Meanwhile, H&K answered me confirming those four resistors are 1K 2W.
The resistors arrived today, so I spend some minutes desoldering the old burned ones and soldering these. I retouched a bit some tube socket solder point, as said some was ugly.
Anyway, the old resistors was really strange, they seemed original, but just two are 1K, at least one seems really 0 ohm according to bands. I didn't know a 0 ohm resistor existed.
So, I put it back everything together (without doing the worst error: tightening all the screws... you know, if you assemble everything as it was ok for sure it's the time that something goes wrong...!). Turned the amp on... no burn smell... good. All valves lighted up. Fine.
Switched stand-by to play mode... all TSC led off! Wonderful! Tested TSC with the pick... quite good, even if the flashes are not all the same... strangely I heard some "clicking" sound... mhmm... tried to plug a guitar... no sound. Sh.........t.
Tried to plug the guitar into FX return... and the sound was there! Ok, the power amp was ok... then I remembered I had difficulty to put back the preamp tube with the "cylinder metal cover"... I powered off everything and remove that tube... found! A pin was bended and not inserted in the respective hole. I straightened it with pliers... inserted the tube back with more care... switched the amp on... and there the sound!!!!
Now it works great! Great sound, it works perfectly!
The only thing I noticed (but I noticed that even the first time I tried it after purchased), there's not so much volume difference between 18 and 36 watt modes... Yes, it's a bit louder, but not as I was expecting... maybe it's normal...

Well... in my foolishness of frying it for those no sense tests, I was very luck, the repair costed me a couple of bucks (I bought ten resistors... just in case...) and no major components like microprocessors or FET or whatever was damaged.

Bordonberd... than you SO, SO, SO, SO much for your great help! For your patience, availability and time spent to write here and help me. REALLY.

 

...and all is well with the rest of the world!  Well done for sorting that out.  Daft mistakes are the name of the game when you start learning a new field.  It's only when you do things wrong that you learn and move forward.

It's a testament to the TSC system that it took this in its stride and came out still doing its job.  I would keep an eye on the LED flash count but as long as they match within four flashes you will be ok.  You might even be able to move them to put them into better matched pairs if the two valves of each of the outer and inner pairs have different counts at the moment.  It isn't important to get both pairs matched, only the two valves in each of the pairs.  It is sometimes said that your tone can improve if the two pairs are not matched to each other as distortion can start in one pair earlier than the other so the onset of saturation is a little more gentle and progressive.  I reckon that would depend on your style of music.

Now, surprisingly you won't notice much difference in volume between the 18W and 36W settings, that's just the way your ears work.  As a rough guide, it's often quoted that a perceived doubling of volume requires about a tenfold increase in power so dropping from 36W to 18W is not much of a step to your ears.  Nevertheless that is the first step in setting up the amp with a single pair of valves so they can then be dropped further by bigger steps with the power soak.

A little advisory here.  Keep an eye on that valve that suffered the bent pin.  It is a slim possibility, (very slim), that in bending it and straightening it you may have broken the seal between the glass and metal at the microscopic level so letting air slowly leak in.  That's very unlikely but just possible so watch out for any gradual change in either the sound or in the look of that valve, (I know you can't see it under its screen but just check at infrequent intervals when you fiddle with the valves for some other reason, [Oh yes you will!   ] ).

And as to our help here, you are more than welcome.  It's a pleasure getting another great amp back on the road and just doing a couple of minutes of typing to do it is no real chore.  But don't give me all the credit, you did the work yourself!  Full marks for getting the real stuff right first time.  There is nothing really difficult in any of this as long as you just apply a degree of common sense, a dollop of logical thinking and a boat load of humility as you learn.  Anyone can get to grips with things at this level really if they apply themselves.  Oh, and as long as they don't just swallow the illogical ideas prevalent nowadays that many sites will allow to go unchallenged in the name of 'respecting conflicting opinions'.  Opinions are opinions are open to interpretation, facts are facts are verifiable and cast in stone!

It is always very satisfying to take something you bought with a problem and make it into a full blown bargain with just a little effort and even less cost.  Now rock out on that thing!

...Ohhh! Did you check the value of those fuses?

No?

That would really have been a good step as it is surprising for the amp to have this problem and NOT have the HT fuse blow. If it has blown then it has been replaced. And if it has been replaced it was by the same idi... person who put in a 0R resistor!!!!!

Would it take long to do it and set your mind at rest just in case another valve goes down and the fuse really needs to be the correct value to blow as it should?

Good morning!

Yeah, great satisfaction to have solved this with little money, I'm not against professional technician of course, it's just I bought this at very good price and I needed a chance for a "low cost" repair... otherwise no bargain, I would have paid at least half of the amp price if brought to a repair centre...

I forgot to mention the fuses, sure I double checked them, but the values are correct, the same you said and the same printed on the board.
I could try to change them anyway with fresh ones... but they are really intact!

Well, about the preamp tube I was really careful when straightened it, I tried to inspect around the pin right because I was thinking the same... glass is "weak". Well, it seems a little "stressed", but it seems good. I will check it again from time to time.
What could happen in case of that tube failure? Just no sound or could it damage something?
Seen that they are China tubes (so I think they are cheap ones... or I'm wrong?) I could replace them with JJ counterparts, just to be safe.

About TSC, it seems there are four flashes between the two tubes of outer pair, while the two in inner paire flashes the same. Around 12 flashes overall. The first left stop flashing, then the inner two, then about two more flashes of the outer right. I could try swapping positions to have more consistency of flashes... hoping I won't fry it again...

Good news to hear you completed it just as advised.  Once again it's amazing how many people we give advice to who either don't even read it completely (cases of 'Twitteritis Extremis' where anything more than ten words means a Homer Simpson moment) or they read it but don't take in the importance of each step we recommend.  As long as the fuse is the correct value and it indicates continuity on a meter, (just visual inspection is no guarantee), it will be fine so well done again for listening and doing the background stuff as well as the exciting bits with the glowy heaty things.

The valve will not always show anything visible in terms of cracking around the pins.  It can be separated at almost the molecular level with tiny amounts of gas creeping in over extended periods of time.  Best not to fret about it as it is either good to go for years or it will slowly degrade probably over a lengthy time.  If there was a problem you could notice a gradual drop off in the gain of the stages it's used in or a lowering of frequency bandwidth.  That wouldn't be the subtle kind of "is it or isn't it" issue, it would eventually become obvious that something is wrong.  It won't do any damage to the rest of the amp, this happens to a huge number of valves naturally over time.  Visually you could even see the top of the valve discolouring or getting white and crusty as the gas creeping in reacts with the 'getter' at the top.  That's a sign the valve needs replaced.

As to the Chinese valves.  Many people hate these and will swap them out for even yoghurt pots immediately they find them.  Others (who are usually the silent majority) actually like the sound of them.  A few (like me) don't think there is much difference between them and the JJ/EH level valve.  I will admit to thinking there is a slight difference in the top end roughness but it's slight and far and away less important than speaker selection.  I use JJs myself as they are decent (can't hear any difference between them and EH or anything else I've toyed with) cheap and readily available.  It won't hurt to swap out the preamp valves when you have the funds and can be bothered but there is no current absolute need to do so.

One other point.  I recently came across a guy who had actually done bench tests on the effects of using 12AX7s in a circuit known as a DC Coupled Cathode Follower circuit.  There is one in the H&K amps as in the majority of others too.  This is a well known "valve eater" configuration for various technical reasons.  He showed (with proper science/engineering) how and why this use kills valves and he found a useful thing.  Chinese valves are relatively immune to it.  We don't know exactly why as that would take electron microscopes and supporting equipment, but yes, those cheap Chinese valves are BETTER for the DCCF slot in your amp.  And the cathode follower has a job less to do with tone generation and more to do with just reducing the impedance out of the stage, (not exclusively true but in general it is). Trouble is there are two valves in each 12AX7 bottle so if you stick with one for the DCCF you often have to put up with keeping the other Chinese triode in another part of the circuit.

You pays your money and you takes your choice.  You could swap out the other two 12AX7s in the V1 first two stages (under the screen can) and V3 Phase Inverter (other end) slots for something like JJs, (or the platinum based NOS types you are keeping under your bed for the coming of the rapture).  You then could leave the Chinese one in the DCCF and Stage 3 slot.  That way you get better life out of the Chinese than you may out of a new JJ.  Personally that's what I would do.  I'm lucky as in my other working amp, a Marshall JVM205H, there are two DCCFs and they are deliberately both in the same bottle.  No contest!

The TSC is showing you have the best setup for your valves as they stand.  The only thing you might consider is swapping the inner and outer pairs.  The reason I mention this is that when you select 18W and below on the power soak it is the inner pair which are switched off so the better balanced pair would be in the outer slots and used most of the time.  That said it is akin to arguing about how many angels can dance on the head of a pin.  It isn't a huge issue by any means to just leave them as you have them.

Don't worry about doing anything 'stoooopid' with things like swapping the valves around now.  You had your moment with that and it won't happen again.  I know of a world champion bike racer who had a horrendous accident caused by a nut not being wired off before he raced.  When he was setting his pit team up for his comeback he chose the guy who had missed that nut saying that he knew that was one guy who would definitely be sure to get those things right from now on.  Just view it like that.

Yes, I did all my homeworks! :p
I checked the fuses with a multimeter, not just by visual inspection!

I'll swap the inner tube with the outer as you suggested, that make sense. Better to have the exactly matched pair always in use. Nice tip!
I'll do a "cross switch", I mean, I'll put the outer right (more flashes) inside left and outer left inside right... just to check if the number of flashes changes someway. Just because I remember that the TSC indicated more flashes on the outer right tube both with original NOS tubes (okay, one was faulty) and with fresh JJ ones. Maybe the fix led TSC led will appear again... hehe...

About the preamp tubes... I didn't understand well about DC Coupled Cathode Follower circuit... V1 is the one with the screen can (the right tube if seen by front of the amp)? Then V2 the central and V3 the left one?
I can't understand well the stages... a single tube "drives" two stages? What would be the best from right to left, just to make it simple? Sorry, I don't know how tubes work... maybe I should read something on the web to improve my culture about them!
... Done! So a 12ax7 is a dual triode tube... ok.
Anyway I didn't understand "You could swap out the other two 12AX7s in the V1 first two stages (under the screen can) and V3 Phase Inverter (other end) slots for something like JJs ------ You then could leave the Chinese one in the DCCF and Stage 3 slot."
I don't understand which slot or tube is part of which stage...

The TM36 schematic is not in the public domain yet so it is a little bit of logical guesswork to say what its valves do but it is so similar in layout and even component numbering that it must pretty much be the same general configuration as the GM36 whose schematic is available out on the internet.  The GM36 has three valve gain stages each using a single triode.  The first and third of these are used for all of the channels and the second is only added in for more gain in the Lead/Ultra channels, (remember the GM36 has 4 channel to the TM36's three).  We can assume the TM36 is similar to that but with maybe just the Ultra channel missing.  That means the two triodes in the first valve and the first triode in the second are fulfilling those roles.  The second triode in the second valve is used as the cathode follower which helps to drive the tonestack more easily.  The third valve uses both of its pair of triodes to act as the phase inverter stage of the power amp.  So we most probably have something like:

Triode	Job	When used
V1a	Gain Stage 1	Used in all channels
V1b	Gain Stage 2	Used only in Lead channel
V2a	Gain Stage 3	Used in all channels
V2b	Cathode Follower	Used in all channels
V3a + V3b	Phase Inverter	Used in all channels

If you look at the amp from the front the valve on the right has the metal can on it.  That is V1 (or as H&K say T1).  As it has two triodes inside it these act as the first and second valve amplifying stages.  This valve might benefit from being swapped out for a JJ or EH or anything else you decide to get your hands on.

The next valve in the middle is V2 and its two triodes act as the third amplifying stage and the direct coupled cathode follower I was talking about.  That one has the stressful job which seems to be better handled by Chinese valves so it could be left as the original Chinese 12AX7 in order to help the cathode follower cope with the additional strain its job causes.

The third valve on the left uses both of its two triodes connected together as the phase inverter in the power amp.  It also could be swapped out when you get a chance.

If you decide you really like the sound of the amp as it is with the Chinese valves in there then keep them as long as you can, there is nothing "wrong" with them.  It's only that people usually think that more expensive valves must sound "better" and, if they think that then guess what they hear?  That is despite the fact that the curves we can measure on the bench with ultra-accurate equipment much more sensitive and reliable than our ears show the valves to be the same in the same circuit. That is why the most common advice on the internet is to get them out of the amp as soon as possible before they "rot their way through the PCB".  My own view is that there is no urgency for any changes at the moment.  If you wanted to try a different valve make in there then V1 would possibly be the place to start as that was the one with the bent pin.  Changing V1 is possibly going to offer more obvious differences as it is at the start of the signal chain and any changes it makes to aspects such as noise and added harmonics are amplified by the other valves after it.  Then V3 is the next to swap.  Knowing what I do about the cathode follower I personally would leave the Chinese valve in V2.  It is only my opinion but I haven't heard any really significant change that I couldn't mostly dial out with the controls by putting a different type of valve in anywhere.  Perhaps a tad less roughness at the top end with "name" valves but the Gain and Treble controls can take the edge off that if you have the courage to turn them down a touch.

I have no words again, you made a table to understand the stages, too! Great!
Well, for now I'm leaving the preamp tubes as they are, maybe in the future I'll follow your tip and change only V1 and V3 tubes, leaving the Chinese of the cathode follower thing.
BUT...
maybe I declared victory too fast. I've another "minor" (I hope) issue now. I have no idea if it's related to the bended-pin-tube. When I power the amp on, in play mode, there's a bad noise, a mix of hum and crackling. Like a sort of fireplace sound cranked up. It happens in all three channels, the send/return on or off have no influence on it.
I can start to play the guitar and the sound is good, the noisy sound is in the background. Then, after some minutes (maybe from 2 to 4 minutes, perhaps it's faster if I play), I could say when the amp has warmed up, this noise goes away and the sound of the amp is great as always.
I don't remember I had this problem on first attempt when I re-assembled the head.
No difference if I switch from 36 to 18 or less.
TSC leds when in use are ok, TSC pick test is ok(perfect match when I swapped the tubes as you suggested. Both couples have the same flashes between own tubes, just the inner pair a flash more than the outers).

What could be the issue? Bad welding points on tubes sockets? Bad FETs? Leaked preamp tube?

As said, this is a "minor" issue for now since the amp plays perfectly when warmed and this noise is gone, but I'm a bit worry.

I had a TM36 with a similar problem, which I'd discovered by noting no change in output between 36W and 18W power soak. My local amp tech traced it down without too much difficulty.

Do you remember what that was caused by?

Heeeelpppp!!!!