TM36 Combo - Died after 2 hours play tonight (FIXED)

I've been jamming a lot recently and on Monday night my combo lost power and died down to almost nothing after 2 hours or so of playing.

I've had it about 6 years and assumed it was just a dead power tube. Foolishly, I trusted my assumption so much that I only spent a few minutes studying the TSC info before deciding to go out and drop $140 on a full set of 4 new tubes. As it turns out, I'd misunderstood the directions and misread the LED flashes. I believe they were actually doing just as they are now which is indicating no problem (12 flashes on all 4 followed by one single flash on 2 & 3 with pick inserted in for manual readout of bias points).

Went to the shop and popped them in and as soon as they were installed, the amp fired up crisp, clear, loud and good as new.

Then tonight at rehearsal, exact same thing happened. Wailing away for 2 hours followed by the big fade.

Let it cool off. Brought it home. Fired it up and nothing but pre-amp level volume. Nothing from the power tubes at all.

Determined to solve this but realizing I'm being a bit naive to stumble along on my own as greater minds than mine probably know what's going on here.

Not certain what to try next but would prefer not to waste anymore money chasing the wrong solution(s).

Any and all replies greatly appreciated.

Thanks

Hi Rob, good to hear from you.  Your post contains a couple of issues.  There is the issue of your problem itself but there is also the statement "...$140 on a full set of 4 new tubes."   I hope that is for a set of Power and Preamp valves though the 4 sounds like it isn't?  If it is just the power valves then have you tried a simple set of something quite ordinary and inexpensive like JJs or did you just take advice that "your amp will sound like a $5000 boutique amp if you just get a set of 1954 NOS Kirkakovich Gold Scrotums from the Posniavitch factory in Siberia"?  I promise you most people here are really not precious about our valves, that is because we know how little difference it makes in the real world.  How little difference in the long run can be demonstrated on a bench with test equipment far more accurate than your ears.  There is no significant difference tonally between valve makes no matter what the gurus say.  If there was it could be demonstrated on the bench and no one ever can when challenged.  The claim is always that their ears tell them.  Well your ears are amazingly delicate and sensitive but not in a comparative way if there is anything more than a couple of seconds between comparisons and I've never known anyone change out and run up valves in a couple of seconds.  I did this very thing to a guy online who claimed to have shown the differences and had plots to show it.  When I just asked politely if we could see the plots the thread went strangely silent.  And it stayed that way!  If you believe you are getting something special then good luck to you, but there is absolutely no evidence to show that any of the claims of valves having "a sparkling morning dew mountain spring top end" or a "powerful yet restrained and tight Kardasianlike bottom end" hold any truth at all.  That's all marketing hype to sell more valves at higher prices and "self-appointed guru speak" and actually means nothing.

Now, sermon over, to the problem.

The TSC showed your valves were working correctly as you thought so no problem at that point.  Which valves did you actually buy?  Did you stick to the correct EL84/12AX7 types?  You did not go for an "alternative" type such as 12AT/U/Y7 or Russian 6P14P did you?  Some of these alternatives are anything but and, in particular, the EL84 swaps often simply cannot take the anode voltages that an EL84 can even though they are recommended by the experts on some forums.  We have had someone here recently who damaged the rest of his amp by using one of these "equivalents" which apparently had magical mojo tonal properties.  It only took a quick search to find out that they could not cope with the HT inside the H&Ks despite the fact that people were raving about them on other forums.  A valve blowing out big time can take out an output transformer without any effort at all and that is expensive!!!

So to some tests.  First of all, when the amp is running with the fault, get a jack cable and put it into the Loop Send and back into the Return.  That takes out the switches on the Return socket and rules them out as a problem straight away.  These can be a problem on a lot of amps if they get just slightly dirty or corroded and they can be occasionally cleaned gently as a matter of maintenance.  There is plenty of advice online as to how best to do that.  Many loops work by routing the signal across the socket switches when no jacks are inserted and opening the wafers to route the signal through the cable/pedals when they are.

Next pull out the jack from the Loop Send socket.  Does the amp buzz normally when you touch the tip with your finger, or better still play when you feed in a signal from another amp's Send socket?  That would be with your guitar -> Working amp input -> Working amp Loop Send -> Your amp Loop Return.  The working preamp feeds its good signal into your own suspect power amp.  As a secondary test you could try that the other way around, feed your own Loop Send into the working amp's Loop Return and prove that your own preamp is still working up to the Loop.

If neither of those first two tests work, (lead in send and return and working amp send into your return), then you have a problem in your power amp section.  This could be simple such as the HT fuse blown but you say that you have "nothing but preamp level volume".  You may mean that you have tried that test already and you get a good signal out of the Send or you may just mean that you have some output from the speaker but at a very low level.  Can you clarify which you meant?

One other thing I would suspect for this is a fault in the PI valve which is V3 in the TM36.  If this has not been changed in your valve swap can you swap it out for a different 12AX7 and test again?  If you have a special balanced version in there make sure to mark it for identification so you can replace it if that is not the problem.  There's no sense in paying for a balanced valve then not using it where it is needed.  Marking the top of the valve with felt tip marker will not hurt it in any way despite what the internet says.  This is a plain glass envelope not a halogen light bulb!

Hi Bordonbert

Thank you so very much for your prompt and detailed response. I've just woken up and heading out to work and barely had time to read it all thoroughly but wanted to reply quickly with as much as I can.

I bought the amp at a local retailer here in Vancouver BC Canada and called them about replacement valves on Tuesday morning. They looked up all the specs and recommended the valves I bought. You are correct, I did not replace the 3 preamp valves, only the 4 power amp valves.

The ones they sold me are GrooveTube GT EL84SQ Mediums. I'm guessing you thought they were higher end valves due to the price I quoted and perhaps that's inflated due the weak Canadian dollar or maybe my retailer has a high markup and charged me far more than the set was worth. I'm not sure. Thankfully though, it sounds as if I bought and installed valves you approve of.

Although I do feel like I 'wasted' this money as perhaps the valves were not the issue to begin with, I'm hoping the ones I pulled out (the originals that shipped with the amp) are all fine and now I just have a spare set of backup replacement tubes for future use or to add value to resale should I choose to part with the amp.

Test results:

QUESTION wrote:So to some tests.  First of all, when the amp is running with the fault, get a jack cable and put it into the Loop Send and back into the Return.  That takes out the switches on the Return socket and rules them out as a problem straight away.  These can be a problem on a lot of amps if they get just slightly dirty or corroded and they can be occasionally cleaned gently as a matter of maintenance.  There is plenty of advice online as to how best to do that.  Many loops work by routing the signal across the socket switches when no jacks are inserted and opening the wafers to route the signal through the cable/pedals when they are.

No change Just a significantly reduced level of guitar sound that suggests preamp is fine but no power.

QUESTION wrote:
Next pull out the jack from the Loop Send socket.  Does the amp buzz normally when you touch the tip with your finger, or better still play when you feed in a signal from another amp's Send socket?  That would be with your guitar -> Working amp input -> Working amp Loop Send -> Your amp Loop Return.  The working preamp feeds its good signal into your own suspect power amp.  As a secondary test you could try that the other way around, feed your own Loop Send into the working amp's Loop Return and prove that your own preamp is still working up to the Loop.

No buzzing or noticeable sound change whatsoever when I touch the end of patch cable with finger. No time to send a signal from another amp at the moment, sorry (off to work in minutes).

QUESTION wrote:If neither of those first two tests work, (lead in send and return and working amp send into your return), then you have a problem in your power amp section.  This could be simple such as the HT fuse blown but you say that you have "nothing but preamp level volume".  You may mean that you have tried that test already and you get a good signal out of the Send or you may just mean that you have some output from the speaker but at a very low level.  Can you clarify which you meant?

When I power up the amp and plug in now, then crank the volume near full on 36w setting, I can hear the guitar signal at a range of about 0 - 5% volume going from Master Volume 0 - 10 on any given channel

Lastly...

QUESTION wrote:
One other thing I would suspect for this is a fault in the PI valve which is V3 in the TM36.  If this has not been changed in your valve swap can you swap it out for a different 12AX7 and test again?  If you have a special balanced version in there make sure to mark it for identification so you can replace it if that is not the problem.  There's no sense in paying for a balanced valve then not using it where it is needed.  Marking the top of the valve with felt tip marker will not hurt it in any way despite what the internet says.  This is a plain glass envelope not a halogen light bulb!

I don't fully understand this but, no, I definitely do not have a spare 12AX7 laying around as I'm not even sure what this is at the moment haha.

Thanks again and I look forward to any follow up you provide.

Enjoy your day

Good stuff, that's really helpful Rob.  You would be surprised how many people ask for help then don't report back with the info you ask them for!  The quick test you have done shows that the Loop sockets seem to be fine, (that was just a hope on my part as it would have been a simple and free fix).  So what does the rest of it show?

Well, as you have said, it looks as though the preamp section should be fine, that is V1 and V2.  V3 is usually the Phase Inverter valve for the output stages.  If this is already old news to you I apologise but here is how it all ties together.

The output stage is a class B/AB one which means that basically it has a pair or pairs of valves.  In each pair one valve is set to amplify the signal above the 0V line and it hands over to the other one which amplifies the parts below the 0V line.  Each valve just idles when the signal is out of its particularly territory.  The transformer takes the current from each of the valves, the +ve and -ve parts, and adds them back together to give a true representation of the original signal but amplified greatly in power.  If there are 4 valves as in our case they work together in balanced pairs, one of each pair for the +ve section and one of each for the -ve section.  At the point where they cross over from +ve to -ve there is always a dead patch.  If this is just ignored we have Class B operation but it sounds rough, particularly at low volumes.  In order to remove this each valve is permanently turned on with a DC bias current so it is already up and running when it receives control and does not lose any of the signal as that jumps across the dead space between.  Think of it a bit like relay runners handing over the baton.  It is pretty slow and clunky if the new guy stands still until he takes it.  He gets up and running and then the transfer happens in the shortest time.  This mode of operation is Class AB, it is not Class B with no bias and it is not Class A where the valve amplifies the whole of the signal with a whopping big bias, it is between the two.

The output pairs need to be matched in order to cancel out the DC current which flows through each valve in the transformer.  Matching means selecting valves which will have the same anode current for the same grid voltage. Transformers become less efficient if they are magnetised with a DC current running through them all the time which would happen if the valves were not matched. With matched valves the bias currents cancel each other out.

The +ve and -ve signals fed to the two or four output valves need to be supplied in opposite polarity so a Phase Inverter, (usually just a 12AX7), is used to take the signal from the preamp and give it out in two opposite forms, one an inverted version of the other.  I suspect the PI valve V3 may be faulty and the output valves are not getting their correct drive.  It's easy to check, simply swap the V3 with V2 and check again with the lead into the Loop Return.  If that valve is faulty then the power amp should come to life. Make sure to swap the valves back afterwards.

And don't worry about the valves you bought, Groove Tubes are a respected make.  I thought the price may have been a touch high for them but here in the UK a GT EL84 is currently around £15.50 which is fairly standard.  Your Can$140 is about £80 and that's not too far off the mark.  At Amer$ rates it is getting a bit "they saw you coming!" but that doesn't apply of course.  Just remember that anyone making claims about valves being "sonically superior" is flying in the face of most technical engineering data and should be able to demonstrate the difference clearly and unambiguously to you.  Do not be afraid to say "I can't hear it!!!"  Remember the Emperor's New Clothes?  That is the case here and guitarists in general are ridiculously easy to convince in that manner about technical stuff like this.

If it were me and I wanted to keep the amp I would definitely get a set of the 12AX7s.  I would stick to 12AX7 and not be seduced with promises of better sounds out of the other 12A*7 types in the "family".  There is absolutely no family that those valves fall into as they are not related in any way.  They are totally different valves designed for different jobs and are all a very bad fit in a stage designed for a 12AX7 by a competent design engineer.  I wouldn't waste my tiome trying to figure out which make sounds best, they will all sound pretty much if not exactly the same under live band conditions. I have never heard anything which convinces me that I can hear any significant difference at all despite what people have claimed, and I am a retired 50yr long guitarist, hifi nut and electronics design engineer used to investigating differences between components both aurally and on the bench. You are absolutely on the money that your original EL84s will be a useful backup.  You need to do one thing though...

When you get a little time fit the original four back in the amp and run the TSC plectrum test.  Check out which is the best match into pairs in terms of LED flashes and separate them into two best matching pairs.  Mark them with a permanent felt marker on their tops.  Maybe pairs 1 and 2 or A and B.  Always use them in those pairs so you can always pop in a matched pair when you need it.  And having one pair of originals with one pair of GTs is not a problem as long as each separate pair is matched with its partner in the same pair.  It is not necessary to always use a matched quad where both of Pair A matches both of Pair B.  As long as the total DC bias current on each side of the transformer adds up to the same then it will be happy.  And you have TSC to automatically help you here too.

Give this a go when you have the time and get back to us with the results.  We may well be able to tie this down a bit further before admitting we need to send the amp off to be fixed. If some of it isn't clear, (I do write out as clearly and unambiguously as I can in full and that sometimes becomes a chore for today's Twitter generation to cope with ), don't be afraid to ask and I'll try to make things clearer.

Thanks again sir!

So please correct me if I’m wrong but you’re suggesting I replace all 3 of the preamp tubes with something like what’s available here https://www.long-mcquade.com/departments/117/Guitars/Parts/Audio_Tubes.htm for $25/each?

As for the old tubes, I did keep the two pairs isolated by removing them separately and storing them in the packaging from the replacement valves so I suspect I’m good to go in that regard and would not need to put them back in and label them as suggested unless I’ve missed a point.

Thanks again. Really helping me see my way through this far better than I ever could have on my own!

Good work with pulling those old valves and keeping them paired up that was the logical step. Sorry for assuming you wouldn't have done that.

Try swapping V3 for V2 and rechecking the input at the Loop Return to see if that fixes the fault. The amp won't play properly as the suspect valve is now in the V2 slot but you would get the power amp up and running if its Phase Inverter (PI) valve was the faulty one. If it does work then we know the original PI V3 should be binned. If there is no change then put V3 and V2 back in place and we will think again.

Wait to see what you turn up with that test before buying new preamp valves but in the meantime...

When it comes to choosing preamp valves there are a couple of recommendations. If you are playing high gain stuff with the amp then it wouldn't hurt to pay a little bit more for a low noise version of the valve in the input position, and for the PI valve it is sometimes useful to pay a few dollars more for a "balanced" valve. In a 12AX7 and other preamp valves there are actually two separate triodes. Usually it makes no difference how well matched these are as they would be used in separate stages which don't rely on each other at all. In a PI however, the two triodes can be built into a combined circuit, (called a long tailed pair), and it can help the action of that if they are matched together so each side of the output gets the same level of signal from triodes of the same characteristics. That internal matching of a double triode valve is known as "balanced", while the term "matched" is kept for two separate valves which are checked to be the same.

If it turns out you need a 12AX7 and it were me, I would get a set of three 12AX7s with one of them balanced for the PI in V3. If you can't get a balanced one it isn't the end of the world and any decent quality valve will do the job correctly, it's just a little refinement. I would replace the originals one at a time starting with the PI. That way you can tell which of the valves you are taking out will be faulty and bin it keeping the other originals for spares.

Hi again

So just home after a long day out and about and preparing to swap the preamp tubes as discussed but am considering holding off for now because I went to google up the manual and couldn't find what I was looking for (a diagram indicating exactly which of the 3 tubes is which - I think I know but wanted to confirm).

Then I stumbled upon this video: https://www.youtube.com/watch?v=6GmJpX99KV0 which gave me reason to pause on a couple counts and I was hoping you could help me confirm or refute

1 - This gentleman swapped out and replaced his tubes with new matched pairs matching 1 & 4 and then 2 & 3. That's not what I did yesterday. I did 1 & 2 and then 3 & 4. Wondering how serious this could be and if it could be responsible for my problem reoccuring? Should I remove them and replace them in the correct order?

2 - I got my grubby little hands all over the springs, all 4 tubes that came out and all 4 of the new ones that went back in. Should I remove, clean and replace them or is it not as much to worry about as his method suggests? Is it okay to use rubber gloves or should I use cloth as shown in the video?

3 - Since his video doesn't mention it and I can't find it in the manual, going from right to left looking at the back of the amp, is the one with the cylindrical metal protector over it the V3 (P1) you mentioned and the other two are V2 in the middle and V1 on the far left? I don't want to botch your suggested test as well as I seem to have likely botched much of the process thus far.

Thanks

Viewing from the front the valves are, V1 (input) on the right under the aluminium can, V2 (second preamp gain valve) in the middle, V3 (PI) on the left.

The guy in the vid is right that there is a correct procedure for opening up the amp to work on the valves.  However it is safer to do this as I will describe though it should amount to the same in the end.  On each end cheek do the following in this order:

1) Remove completely both of the top screws.
2) Remove completely the middle level rear screw on the silver handle and ignore the front screw as it only holds the handle to the end cheek.
3) The bottom two screws should be loosened only enough to let the end cheeks gently pull outwards so the top cover can be removed upwards.  These bottom screws have an anti tamper device inside the amp.  If they are removed too far or completely, this is altered and shows up to a service tech when the bottom cover is removed.  This can void your warranty if it has been done without authorisation!
4) Remove the top cover taking care not to damage the LED strip which sits under it on the top edge of the front Plexiglass panel.  When you refit this make sure the clips holding the LED strip are in the correct place and the wires on the end are dressed correctly and not pinched in any way.

You have got the placement of your new valves wrong, they should have been put in slots 1+4 and 2+3.  You can tell this because it will be the same as the TSC lights when you put the amp on 18W.  The inner two lights come on to show that valves 2+3 are switched off.  The TSC lights match the positions of the valves they are monitoring.  Don't worry, it won't have damaged the valves nor your amp at all.  Just remove 1 & 2 and swap them leaving 3 & 4 in place and you have them back in pairs again.

As I have already mentioned there is nothing dangerous about handling valves, they do not have any sensitive coating on them like a halogen bulb which must never be touched.  There is nothing for grease or dust to react with.  This is a common piece of internet bullcrap.  I wouldn't let dirt build up on my valves and if it is obvious when looked at I would clean them but I do not bother at all with gloves or cloth when I touch them.  I just make sure my hands are reasonably clean.  Nor does anyone else that I know of who works with valves regularly take any special precautions.  None of us have ever had a problem.

If you look at the top posts in the responses to the vid you will notice there is one from an Albert Burton.  That is me!  I discovered this vid when I first got my first H&K and it was so obviously posted by someone who was not a tech savvy person that I put in the correction to his advice.  Note he says: "...and I'm told, even though I haven't done this myself because I've been good enough not to touch the valves, that if you do [touch them], apparently you can damage or ruin or break them".  That "and I'm told" is how this crap gets started.  Someone else who told him has been told by his mother's milkman's sister's boyfriend that this is so, and it gets passed down the line becoming "truer" every time.  It got an expected standard response from someone else who was obviously not a tech savvy person and his ridiculous arguments put forward in his defending of the inaccurate info just shows how ingrained the "mojo myths" become.  Trust me and the thousands of other engineers out there, it makes no difference to a valve that you touch it with clean fingers.  It is just a hard glass tube with a vacuum inside, how is it damaged by having a smear of grease on it?  If it is true then how is it not damaged by having its manufacturing info printed on it?

The V1 valve is the most sensitive to noise of the three.  Any noise picked up or generated at this stage is amplified by the following valves so becomes more significant.  If there is only one covering can, you can bet it will be on that V1 to make it more quiet, that is generally your clue.  If you have a low noise valve it is almost always best used in there for the same reason though there is no need to get prissy about buying anything special, an ordinary 12AX7 should be quiet enough if it is up to spec.  Always remember valve guitar amps are amplifying a very low level signal up to a high power.  They are inherently noisy pieces of kit, even the best ones would get nowhere in the hifi world based on their noise performance.  Newer people coming into guitar work nowadays expect them to be absolutely silent at even dimed levels.  That 'aint goin' to happen, ever never not nohow!  That's why the noise gate was invented.




EDIT: ...and hey, go easy on yourself. Everyone starts somewhere and we have all made the sort of mistakes sometime that you have, (well I know I definitely have though maybe some of the guys here may not have? ) You can't learn anything new without getting things wrong, that's the only way forward. You're keeping an open mind, and hopefully using your own intellect to think your way through what I'm saying so you don't just follow it blindly. And you should constantly be questioning it. Never stop questioning whatever anyone tells you in this game. If they are genuine they won't mind at all if you suggest what they say may be wrong and why you think that.

You're doing just fine!

Good morning Albert! I was hoping to wake to a response and once again you've not let me down.

Sadly, I'm again in a morning rush as I'm off to the dentist in an hour so let me reply to each part as I can then I will move on with the test and hopefully respond before leaving home.

First, the 'bad' news..

In the spirit of your kind words encouraging me not to fear questioning given advice. I'm forced to suggest that I believe you've made a mistake in one part of your post (2 actually but the 2nd one is more of an inconsequential oversight). That second one being that you got caught up in the aspect of the video related to removing the cover of the TM36 head to access the valves. Mine is the TM36 Combo so this piece was irrelevant to me and is why I'll be describing my procedures from the perspective of the rear of the amp with the back removed from the combo unit. Sorry to have led you astray there as it resulted in your generously clarifying a part of his video which has no bearing on my situation.

I'm happy to hear that incorrect placement of the power valves did not cause any problem. Big relief there so thank you for that. But your suggested simple solution to fixing the incorrect order strikes me as itself being incorrect. I'll outline why in as much detail as possible to ensure I'm making myself clear and you can confirm or refute in kind so we're sure we're on the same page...

*******************************************

For simplicity and clarity, I'll call the 4 new valves I bought VA1 and VA2 VB1 and VB2 where the first 2 are a matched pair from the same box and the second 2 are likewise, a unique, matched pair.

Looking into the open rear compartment of the combo unit, when I pulled the old tubes, I pulled from left to right, the first 2 valves (1&2).

I then replaced them with VA1 and VA2.

Next, I pulled, from left to right, the last two valves (3&4).

I replaced these with VB1 and VB2.

The resulting configuration then became, going from left to right, VA1 - VA2 - VB1 - VB2

Please do correct me if I'm wrong but, were I to follow your suggestion of simply swapping VA1 with VA2, I'd still be left with mismatched placement.

It strikes me more so that what I need to do is swap valves 1 and 3, resulting in the following:

VB1 - VA2 - VA1 - VB2

I think that's correct but will await your confirmation

*******************************************

On to the test...

bordonbert wrote:Viewing from the front the valves are, V1 (input) on the right under the aluminium can, V2 (second preamp gain valve) in the middle, V3 (PI) on the left.

In the combo and, again, from the rear, the V1 input valve under the can is on the right side as well so I suspect the order you've listed here holds true for me with the second preamp gain valve being in the middle of the three.

bordonbert wrote:Try swapping V3 for V2 and rechecking the input at the Loop Return to see if that fixes the fault. The amp won't play properly as the suspect valve is now in the V2 slot but you would get the power amp up and running if its Phase Inverter (PI) valve was the faulty one. If it does work then we know the original PI V3 should be binned. If there is no change then put V3 and V2 back in place and we will think again.

I will leave V1 where it is, under the aluminum can, swap the other 2, run the test and report back...

Not the result I was expecting...

After swapping the preamp valves, but before even adding the patch cable to the fx loop, the amp has begun working again.

I have not yet made any changes to the power valve configuration.

I was expecting it to not play properly as the suspect valve had simply been moved but that we'd get different test results. Colour me confused. Despite my gratitude at hearing the amp sound as it should, my fear is that it will now play fine for a couple hours before dying again.

Off to the dentist and awaiting your reply.

Cheers and thanks yet again

Rob

Hahaha!     Yes I got myself in a bit of a tizz there didn't I.  Of course yours is the combo, you actually posted that in the thread title!  I just plain forgot, my sincere apologies.  (Must have been...  )

So for the combo, the valve with the screening can on it is the V1, the one next to it is V2 and the V3 PI is the one at the opposite end to the can.  That should help whichever way you look at it but I believe from H&Ks own pictures that looking from the back the V1 screened valve should be on the right.

Also my apologies for mixing up the output valve swap, that was just plain old stupidity on my part, no excuse.  With your matched pairs incorrectly fitted in 1+2 and 3+4 you would need to swap either 1 & 3 or 2 & 4 to get one matched pair in the outer slots and one in the inner slots.  (I've played this out on paper this time just to make sure my brain hasn't gone to sleep again.  )

I'm not surprised that the amp came to life when you simply swapped them, some valve problems can be like that.  It's a pity but the most difficult problems to deal with are the intermittent ones which come and go at random.  The thing to do now is to get your hands on a single spare 12AX7, anything from a working amp would do.  Just put the original valves back into their correct slots and play the amp for a while until the problem occurs.  Then turn it off, wait for it to cool, and swap the replacement test valve into slot V3, the PI.  The fact that the cable in the Loop Return did not buzz when touched shows that the problem is after the loop so it can only be around the PI valve V3 or in the output valves themselves.

I would put my money on the PI valve being the fault but it could still be something like a bad joint around the PI valve base.  I have experienced that one myself.  Along with joints on high power resistors which can get very hot, those valve socket joints are the most stressed in the amp as they go through a high heating and cooling cycle along with the valves and that can weaken solder.  Once again I would caution you against the current advice that solder joints are a thing of the Devil and PCBs are his Hellspawn.  I would take a well designed modern PCB over handwiring anyday!  PCBs are easier to produce, cheaper to produce, take less time to produce, and show themselves as extremely reliable when you consider how many billions of them chug away doing whatever they do without a single problem for decades.  They are in everythng and so few go wrong.  The only ones that techs see and report bad things about are the statistical few which do turn up their toes and even then it is usually a component and not the PCB itself which is at fault.  There are so many more PCBs around than hand wired rat's nest types that that is inevitably going to be the bulk of what they see.  I do not look down on the old style at all but it is not better than the modern either in reliability or in sound.  Techs like them simply because they are easier to work on!  How many aircraft and even spacecraft are kept in the air with handwired electronics?  Why is that?

And the logical next step to that thinking says that if PCBs are a "bad thing" then they must be riddled with dry and crystalline joints.  Well they aren't, bad joints are an absolute rarity in modern PCBs!  They do happen of course but in the context of how many PCBs have how many joints which keep working for ever, problems are an absolute scarcity.  Often you hear of people fixing their amp by reflowing every joint on a PCB and it starts working.  There you are, it MUST have been a bad joint.  They forget that they have broken down the entire amp to get at that PCB and so many other things which were far more likely to be the problem have been altered in doing so.  And they never come back and own up that it wasn't a fix when it happens again later.  But, that said, the issue of heating and cooling of joints can be a problem, both on PCBs and in handwired gear, and joints which have a high thermal cycle during use are worth keeping an eye on for issues.  It only takes a reflow of each joint with a soldering iron with a little new solder added to solve it.

That would probably be beyond what you can do yourself I would guess but there is every chance that the valve swap would fix the problem, it's much more likely, and it is the logical thing to do in the first instance.

I'm at home all night working on some other projects so I'll try to run it for hours and heat it up to replicate the issue if possible. Won't bother buying another valve if I fail to replicate it but if it does occur again tonight, I'll buy one before rehearsal Monday and be prepared to swap it out.

Personally, I suspect it is just as you suggest and has to do with a faulty joint. I would not attempt to fix that myself but we have a pretty reputable repair tech here in town who I could take it to if needed. Pretty certain it's long ago warranty expired so I just hope the final bill isn't too ghastly if needed.

Thanks again and I'll report back with updates over the coming days.

Cheers

Quick question...

Do you know if it would make a difference which settings I run on to try and get it to peak heat in order to replicate fault?

By that I mean, I can only play so loud here at home but if I play at a low Master gain setting on 36w or a higher setting on 5w, do you imagine one or the other would drive the tubes harder and increase likelihood of failure?

That's an excellent thought!

Use it on the 1W setting on your Power Soak even if that sounds not quite how you like it. You are right, that way you are driving the power amp hardest and then throwing away most of the power as heat in the power soak. I would also use it on the higher gain channel too as that will bring in the additional gain stage which is switched in for the Lead channel. (This may be present in the Crunch channel too but I'm not sure for the TM36.)

Excellent

Well...

Not sure if this is good news or bad news but I really put it through it's paces for over an hour without a full fail.

But then eventually a new issue arose.

I noticed that, rather than dying altogether, it was getting slightly quieter at times and seemingly losing bottom end strength. It just sounded weaker / wimpier than usual overall.

Started adjusting EQ buttons etc and many were crackling with static as though the pots were filthy (worth noting they've never been cleaned?).

The more I messed with them and then the level knobs on both hot channels, the more it seemed to fade in and out of full power/volume. I hate doing this to any gear but I gave a few moderate thumps of the fist on the top and it would immediately clean up and sound full again but then would fade off again as I played more. Definitely some dirt and/or decaying connections in there.

Then, just now as I was typing this up, I realized it may have been sounding 'weak' cuz I was playing in 1w mode. Popped it into 36w and as soon as I turned up just a little and hit a note, the poor thing wailed like it had been shot. Really wish I'd been video documenting it for ya haha

Turned it off immediately and suspect I may need to swap valves 2&3 back again before proceeding cuz it seems to have gone from dead to dying. While that just suggest progress, it's not the path I was hoping for lol. Ah well...

The adventure continues!

Everything seems to point to you having a duff 12AX7 in there. The one you swapped into the V3 PI slot seems to have sorted out the power amp side of things but, don't forget, that means you have swapped a suspect valve into the V2 slot. I reckon your next step is to swap out that valve for a new one and it may be the solution to your problem.

Do you have another amp you can scavenge a 12AX7 from to try it out for an hour or two?

I wouldn't worry about the crackling pots, that would be a secondary issue even if it was a separate problem at all.

Good news!

Bought a Mesa 12AX7 (ECC 83) today for $25 and swapped it out with the middle preamp tube first and was back to no gain.

Replaced the middle preamp tube then swapped the Mesa in for the 1st position tube and the amp sounds and appears to be behaving normally once more.

Do you imagine that, if I run it hot for a couple hours tonight and don't run into any more problems, that the issue is resolved or should I buy 2 more valves to do a full swap just to be on the safe side?

You could buy a couple more 12AX7s as they will need to be changed out at some time but the two decent ones I think you probably have may have a lot of life left in them still.  Assuming there is only a single valve faulty you could stick with the one you have bought but I'm confused as to how it fixes things when you put it into the V1 position.  That's the one under the metal can right?

With the fact that the amp did not kick into life when you tried the Loop Return test, (fitting a cable into the loop return socket and touching the tip of it with your finger), it points to the PI in V3 as the problem.  With that test you are injecting a signal into the power amp after both V1 and V2 which are also isolated by the socket switch, so if it does not improve things then V1 and V2 are probably not the cause and certainly something else is involved too.

Others would recommend just swapping out all three valves and have done with it.  I can see their point, that is the easiest way to go forward, but it does not really prove what the problem is.  You may even find out that you have a recurring problem because it wasn't a valve and then you are three valves worth of $ out of pocket too.  It takes longer but it would be better to know what it really was and that you have corrected that.

Can you confirm that it was V1, the input valve under the metal can, that you substituted and it fixed things because if that is the case I don't think you have fixed it, I think it is just temporarily working.  I still lean towards V3 being the problem, (though I could always be wrong, it has been known    ).

I would also try putting V1 back in place and replacing V3 (PI) with the Mesa valve. Does the amp work in that configuration?

Your instincts are bang on sir

I did not replace the one under the can. It was the one on the other end of the three.

My poor description in last post must have thrown you off. I should've said third position instead of first.

Good good good!  Really sounds like it was the PI valve after all.  The test evidence all points towards it and the successful fix suggests it would be a permanent one.  Where you go from here depends on how you feel you are placed moneywise, (and that often involves how old you are so it's worth putting in for others of all ages who may read this  ).  If you are feeling pretty flush then get two more 12AX7s to make up a full set, they don't cost the Earth.  If you aren't and money is an issue just get your hands on a single replacement and you will have a spare for one which may go toes up on you in the future.  Current attitudes amongst "doods" and "bros" may not agree with that moderate approach but we all have a limit as to how much we are able and how much we think it is right to throw at our gear.  We're not all professionals, we just sometimes see ourselves that way.

I would put your new valve in the PI V3 slot, (the end one without the can), and trust it now.  Personally, I always keep a full set of valves with me on gigs as a backup just in case.  That said, mine aren't a brand new set, only guaranteed good ones which I may have swapped out previously and a couple I have picked up at one time because they may have been going cheap.  You should mark the valve you took out as a suspect one but keep it for a while until you are absolutely sure the amp is fixed with the new one.  If it happens again then we have the wrong cause and the valve may not be at fault after all.  Once it is proved fixed then bin the suspect!

You already have a new set of EL84s which weren't the culprit so the old ones will give you support there, (in pairs of course).  The PI valve is brand new so should be ok for a good while.  Then you would be covered whether you get a couple of new replacement 12AX7s and keep the two you then remove as spares or whether you get by with one.

Have you noticed any real change, (note I said REAL change), to the amp's sound with the replacement output and PI valves?  (No, I'm not trying to point you towards any particular direction, just to get you to be honestly critical of that aspect and to get your honest unbiased view as to anything which may have changed and how significant you think that is    ).  I would expect some change as the faulty PI valve could have been colouring the tone of the amp for some time and so should make a difference when replaced simply because the new one works properly.

As a last thing the hardest pushed valve in your amp is V2.  It acts as a standard gain stage followed by a Direct Coupled Cathode Follower (DCCF) stage and those are often run massively beyond their capabilities for a short time.  I haven't got the TM36 schematics but I have access to the GM36's and it is definitely set up with a DCCF and I believe the TM36 is based on the same basic configuration.  The H&K Standby only cuts off the output valves via the TSC circuitry so HT is connected to the V2 stages immediately at switch on independent of how the Standby is set.  That means that until the valves warm up and start conducting current, maybe 5-10secs, the DCCF has the full HT across its grid and cathode.  That is extremely bad!!!  There is a lot of folklore surrounding DCCF valves.  There are views that some valve makes are better in this position than others.  For once I could agree with this and it does not surprise me.  Some manufacturers' designs will take this abuse for longer than others but in the end they will all be the worse for it, sometimes drastically.  There is nothing you can do unless you want to solder in place the fix for it which is very simple and cheap but beyond most guitarists' abilities.  I plan to put this up in the "Coffee Shop" area we have for random topics so keep you eyes open there. Just keep an eye on the sound of your amp and watch out for anything untoward in V2.

Thanks Albert.

First full rehearsal with the new valve in place later this evening so we'll see how it goes. I ran it hot for about 6 hours Saturday night but only playing intermittently so not sure that would generate the same load on the valves. But it sounded just as sweet when I shut it down as it did the day I bought it so am proceeding with a good deal of optimism.

Likely to be trying to sell this baby once I profile it with my Kemper but will be sure to advise the new owner of this thread and encourage picking up a couple more 12AX7's as backups.

Can't thank you enough for all your help. Not only simplified what might otherwise have been a complicated and costly process for me but was educational and enjoyable along the way.

My only regret is not video documenting the whole damn thing. That ghastly wail it made on Thursday night was a sound I hope to never hear again unexpectedly but would've made for an entertaining upload haha.

Rehearsal went great with no repeat of old issue.

Oddly, I was getting a lot of unwanted noise from my trusty old Line6 wireless pod but that's an adventure for another day haha

Have a great week Albert!

That's great Rob.  It looks like a fix to me.

The important thing as an owner fixing your own amp is always to look at the symptoms and use some simple tests to get more information to be able tie it down to a particular area.  We see it a lot but coming to a place like this and asking "my amp died, what could it be" is a daft thing to do.  Firstly it doesn't give the guys who know a bit about it anything to work with so they won't help other than to say "give us some info (with an implied 'dumbass')".  Secondly it invites every dood who doesn't really know anything except what they have been told by their bros who don't know anything either to give their "opinion", (which of course is supposed to be respected as much as everyone else's and given equal importance nowadays).  The usual advice given to approach this is a) to swap all of the valves, b) swap every capacitor, c) reflow every soldered joint, then with their "knowledge" exhausted they drop out and move on to the next guy with a problem.

That isn't how a decent tech approaches it.  Faced with an unfamiliar problem, he tests various aspects of the amp until he finds something odd and then uses his powers of deduction to come up with new things to test to rule out or confirm some of the things which may be causing it.  Often with a common amp a tech can go to the most likely cause straight away which makes the wannabee boys think that he is just picking something out of the air.  He isn't, he may have years of experience of amps like your own with the same common fault to tell him the correct place to look.  Every amp model has its own weaknesses and common faults which he may see time and time again.  You and I aren't in that situation so, rather than just swapping out stuff at random without knowing whether it will fix it or not, you have to take a logical view.  And that isn't hard.  We all have a brain, it just needs to be put in gear.

For the record, your running the amp and using it intermittently over an extended period is likely to have been a pretty good test.  Preamp valves don't run like output valves.  Do you remember I bleated on about Class A A/B and B types of circuit?  The output swaps the+ve and -ve halves of the signal between two (banks of) valves then adds the two sides back together in the output transformer to make an amplified version of the original signal.  Your output is Class A/B but your preamp valves run in class A.  That means they work alone with a constant steady standing current through them all the time.  When they have a signal put through them that standing current increases and decreases from its steady value as it follows the signal.  They are effectively always running under stress.  It isn't quite as simple as that but for our purposes a 6hr run in would test them pretty well even if they were just idling.  And it also helps to run new valves in for a period too.

Well, we had a great gig on Sunday outdoors in the sunshine and with another indoors to come on Friday it should be a good week for me.  Playing is the thing and the older you get the more you appreciate the time you can still spend doing it.  Good luck with the band, hope the amp holds up.  If you get any more problem you know where we are.

Again, no problems but am considering swapping out the other two valves just to see if it improves the sound. Getting a lot more of that loud hissing sort of sound when my guitar is turned all the way off and it's the same with guitar cabled or using wireless pod (maybe worse with pod but couldn't get other guys to shut up long enough to do a decent test and I'm leaving the combo at the lockout for now.

Will report back but bottom line is it's great to have it cranked loud and proud from beginning to end of rehearsals without failure.

Playing IS what it's all about!