Something went wrong with my amp, big flash, burned PCB

Привет, ребята. Я использовал tube meister 18 около 1 месяца. Я приехал в деревню и родители прислали мне по почте мой усилитель. Когда я включил режим ожидания, из области под левой силовой трубкой вылетела желтая искра. Подумал, что трубка плохо подсоединена, вытащил и вставил обратно. При включении режима ожидания на чистом канале все было ок, а при переключении на 2 канал - треск и зеленая вспышка с дымом на том же месте. Я не могу отправить в сервисный центр, потому что я из Украины. Где может быть проблема? Это фото прогоревшего места. Вы можете видеть, что анодный разъем расплавился. Также TSC показал, что все ок.




Bordonbert - Translation (care of Google):

Hi guys. I used Tubemeister 18 for about 1 month. I came to the countryside and my parents mailed me my amplifier. When I turned on the standby mode, a yellow spark flew out from the area under the left power tube. I thought that the tube was badly connected, pulled it out and put it back in. When you turn on the standby mode on a clean channel, everything was ok, and when you switch to channel 2, there is a crackle and a green flash with smoke in the same place. I can't send to a service centre because I'm from Ukraine. Where could be the problem? This is a photo of a burnt place. You can see that the anode connector has melted. Also TSC showed that everything is ok.

That area is the main connection from the anode of output valve V3 to the output transformer. The circle of soldered terminals on the right is the valve base. On the PCB it should be a direct track from the valve pin to the spade terminal we can see in the picture with no other components connected to the track at all. It certainly looks as though there has been some sort of arcing of the HT in that area. The connection to the terminal is a very short one on the other side of the PCB from the valve pin just under the letters "R7" on the right passing under the letter "e" of "Anode 1" to the spade terminal.

It looks as though a piece of dirt or maybe an insect of some description has managed to get close to the Anode 1 HT terminal, (390V so you must be very careful), and the HT has arced across, possibly to the MOSFET transistor nearby on the left. There may be no service centre near you but I would suggest you perhaps find an electronics repair shop near you who can simply open up the amp and inspect and clean this area. If you are not used to working with LETHAL (that means DEADLY) high voltages I would not recommend doing this yourself. The arcing from the HT line is not necessarily going to be a problem as that comes directly from the power supply with no other sensitive components to be damaged. What it has arced to is the question. The nearby MOSFET transistor you can see is the TSC control in the cathode circuit of the valve. Some areas of this are very sensitive. The TSC says that everything is ok but if that has damage it may just not be registering and reporting correctly.

Can you answer a few questions please?

Is the amp currently working? You do say that you tried it after the first problem and it was fine on Clean but you had a second problem when you changed to Channel 2. Does it still do that or is it now working but you are worried it may still have a problem?

You have obviously had the base of the amp off to take the picture. Did you clean any dust or dirt from the blackened area before taking the picture? It may have been an insect which has been destroyed by the arc and the amp may now be working fine.

Can you not wait for the current political situation to stabilise a little so you could get it to an authorised service centre under warranty?



Я надеюсь, что это переводится правильно.

Эта область является основным соединением от анода выходного клапана V3 до выходного трансформатора. Круг припаянных клемм справа — это основание клапана. На печатной плате это должна быть прямая дорожка от штифта клапана к лепестковой клемме, которую мы видим на рисунке, без каких-либо других компонентов, подключенных к дорожке вообще. Конечно, похоже, что в этой области произошло какое-то искривление HT. Соединение с клеммой очень короткое на другой стороне печатной платы от стержня клапана сразу под буквами «R7» справа, проходящего под буквой «е» «Anode 1» к лепестковой клемме.

Похоже, что кусок грязи или, возможно, какое-то насекомое сумело подобраться близко к клемме HT анода 1 (390 В, поэтому вы должны быть очень осторожны), и HT зажглась дугой, возможно, к MOSFET-транзистору поблизости. слева. Рядом с вами может не быть сервисного центра, но я бы посоветовал вам найти ближайшую мастерскую по ремонту электроники, которая может просто открыть усилитель, осмотреть и очистить эту область. Если вы не привыкли работать со СМЕРТЕЛЬНЫМИ (то есть СМЕРТЕЛЬНЫМИ) высокими напряжениями, я бы не рекомендовал делать это самостоятельно. Дугообразование в высоковольтной линии не обязательно будет проблемой, так как оно исходит непосредственно от источника питания без каких-либо других чувствительных компонентов, которые могут быть повреждены. К чему это привело — вопрос. Ближайший MOSFET-транзистор, который вы видите, является регулятором TSC в катодной цепи клапана. Некоторые области этого очень чувствительны. TSC говорит, что все в порядке, но если у него есть повреждения, он может просто неправильно регистрировать и сообщать.

Не могли бы вы ответить на несколько вопросов, пожалуйста?

Усилитель сейчас работает? Вы говорите, что вы пробовали это после первой проблемы, и все было в порядке на «Чистке», но у вас возникла вторая проблема, когда вы перешли на канал 2. Он все еще работает или теперь работает, но вы беспокоитесь, что проблема все еще может быть? ?

Вы, очевидно, сняли базу усилителя, чтобы сделать снимок. Убирали ли вы пыль или грязь с почерневшей области перед съемкой? Возможно, это было насекомое, которое было уничтожено дугой, и теперь усилитель может работать нормально.

Не можете ли вы подождать, пока текущая политическая ситуация немного стабилизируется, чтобы вы могли отнести его в авторизованный сервисный центр по гарантии?

More details about turning on at the first time : I forgot to turn off switches (on and standby), and just powered it already with all switches on. Instantly it flashed with big sparkle near the left power valve. I turned amp off. 1 Hour later I turned in on second time just to see if amp is dead. When i turned on standby and looked at valve socket. I saw that there were little sparkling under it, and it sparkled in the rythm with noise in speaker. I bought it used and I sure that previous owner never did anything with pcb in amp, he just replaced stock tubes once. So next day I reconnected tubes, looked at pcb, everything was okay. Turned amp on again. Standby on - no sparkles, no noise in the speaker, waited a few minutes, sounds okay on clean, Im already believed that amp is fine. But then i switched to distortion channel, !and when i hitted string on guitar, right at that moment! - boom, flash and smoke with noise in speaker. Anode connection melt after that 3rd attempt, and i think that smoke is from melted rubber on this connection. I did not cleaned anything, that how it was. I have multimeter, maybe I can check something with it? Is anode voltage depends on which channel is selected and how i hit the strings?

That is much better info to say what is wrong. I would not use the amp any further as it is now. You said "I saw that there were little sparkling under it, and it sparkled in the rythm with noise in speaker." That is a sure sign there was some foreign material acting to short the anode HT. The sparkling is where it was simply burning away after the initial flash. That would cause the noise you heard in the speaker.

The anode voltage does not depend on which channel you select. The power amp takes no part in that, it is always set up the same after the preamp no matter what control selections are chosen. The anode voltage will change when you hit the strings, it has the large output signal to the transformer on it and that swings up and down when you play. There is nothing you can do to help with your multimeter I am afraid.

The best I can suggest is that you simply gently clean any surface dirt or carbon from the area with a dry cloth. Then make sure that nothing from the Anode 1 terminal and its cable is making contact with anything else around it where it has been damaged. Then check as to whether the PCB has been so badly burned that it is now conducting the HT. Can you get another picture of the transistors from above and the rubber cover of the Anode 1 terminal with its cable from the side so we can see how bad the damage is?

It will depend on what the HT arced to but it is possible that the TSC circuitry has been damaged by this. If that is true then there is nothing you can do except get a technician to fix it for you. It could require replacing of at least one of the cathode MOSFETs, maybe even both, and perhaps even the whole TSC PCB as well as that is usually not serviceable if it is damaged.

It is also possible that the output valve V3 has now been damaged by too though I think that is less likely. Can you get a spare pair of EL84 valves to test the amp with?

Can you get a picture like this one showing the two transistors and the terminal:

У меня во втором усилителе только один el84. И да, на левом транзисторе просто копоть, я его немного почистил после фото, выглядит хорошо. Индикаторы TSC мигают, когда усилитель находится в режиме ожидания, и гаснут, когда усилитель не находится в режиме ожидания. Насколько я знаю, режим ожидания на этом усилителе отключает катод, а не анод. Так почему же анодное соединение расплавилось, когда я выключил режим ожидания?



(Bordonbert: I have only one el84 in the second amplifier. And yes, there is just soot on the left transistor, I cleaned it a little after the photo, it looks good. The TSC indicators flash when the amplifier is in standby mode and turn off when the amplifier is not in standby mode. As far as I know, the standby mode on this amp turns off the cathode, not the anode. So why did the anode connection melt when I turned off standby?)

I looked underneath pcb through little hole and find out that its totally fine at this area
Usually I never read forums, but i see that you write smart thoughts and helped many people here. I felt myself hopeless when something went wrong with my amp, because I dont have anyone to ask about it. Thanks so much!

I'm not sure if you mean you only have one EL84 plugged into its socket.  Do you mean you only have one actually fitted in place and the other one out of its socket?

I think you are worrying about nothing with the burning on the metal Anode 1 terminal.  It looks as though something has been burned next to it and it has only been blackened by that but it doesn't look badly damaged to me.  As I said earlier, the problem is not the Anode 1 terminal where the HT is, it is what the voltage jumped to.  The nearest things are the other PCB tracks or the transistor.

The PCB tracks nearby are all to do with those transistors so it comes down to them.  Some of their terminals are more sensitive than others.  The transistors are 600V devices across their two main terminals, (the drain and source), so they should have some resistance to the 390V HT supply.  However the control terminal, (the gate), is much more sensitive.  Unfortunately that terminal is the nearest to the Anode 1.  However, your TSC says everything is ok so it may be that there is no damage there either.

I would guess the only thing you can do yourself is make sure there is no dirt or residue left on the PCB.  Clean that small area well with a simple solvent which will evaporate completely such as meths, (don't use soap and water).  Give it time for all the solvent to evaporate or you could cause a fire.  Then get your hands on at least one other EL84 which you know is working as it should, and try that in place of the one giving the problem in slot T3.  It will not matter for testing that the valves are not matched now.  Sometimes problems like this can be caused by mechanical faults inside a valve, though that would not cause the external burning on the PCB.  It may be the other way around and the arcing fault may have actually damaged the valve.  We need to eliminate that possibility.

Can you borrow either a pair of EL84s or a single one just for testing from another amp?

I have 2 el 84 in hk amp and one el84 in my vox amp. Should I put this one tube in my hk?

I have a little good news perhaps.

I've had another look at the PCB layout this time in the drawings for it. I can now see that the problem could very well be that the Anode 1 terminal has just shorted with the ground area. If that is true it is not so serious. Under those transistors is a large grounded area which they are mounted on to keep them cool. That is certainly close enough to the Anode 1 terminal to short across if something is there which should not be. If that is the case then the transistors would not be involved. Let's hope that is correct. You need to clean off any residue between the transistor case and Anode 1 terminal. If there is still carbon from the burned material it could still conduct HT and short the supply down to ground again. Try that then test the amp again.

I don't think I have asked this before and I should have done really. Is the amp currently making any noises when it is turned on and not in Standby even a soft background hum or hiss, or is it totally silent? If it is totally silent then the HT fuse may have blown. If that is the case then you would need to source some of the correct fuses.

["I have 2 el 84 in hk amp and one el84 in my vox amp. Should I put this one tube in my hk?"]


Yes, that is a good idea but only once we know the HT fuse is ok. I need to know whether the amp can be heard making any noise at all first. Please answer that before you do any further testing.



Once we have the answer to that, make sure you can identify all three valves so they can be put back in their correct place after testing. Mark them on their tops with a small identifier using a felt marker. Just a number will do as long as you can remember what it means. It is always best to keep them where they started if that is possible. On the internet you will hear people telling you you must keep your valves clean or they will be damaged. That is utter nonsense! Engineers have been using dusty valves with fingerprints and even labels stuck on them without any problems for decades. They have all of their identifying information printed on them already in thick paint! Valves are very robust and are only hardened glass envelopes, their surface is not delicate like halogen bulbs are in any way. Of course it makes sense to keep them clean but the idea that labelling them with a felt marker will damage them is ridiculous. This is what I do with every one I use so I can track how it is working and I have never had any problem, nor has anyone I know who does the same.

Make sure you have cleaned the PCB as best you can then swap the Vox valve for the T3 valve in the H&K and test again. Hopefully there will be no problem. I would also try the H&K valve in the Vox too. If it is not currently blowing the HT fuse in the TM18 it should not have any destructive problem inside it which will hurt the Vox.

Are that fuses you talk about?

Yes. The HT fuse is labelled FU1 and it should be the one on the left in your picture. It is a 200mA fuse and should never be replaced with anything higher than that. It looks ok from here though they can be broken inside where you can't see. I think we have to assume it is ok and work from there.

Do you get any guitar sounds or background noise from the amp when it is switched on? By background noise I mean any hum or hiss through the speakers? If you do then the fuse is fine and there is HT. If you just get total silence then it could be a sign that that fuse has blown.

Okay i removed power soak knob, now i will try to clean pcb

TM18 Break Down

Have a look at the above picture in the link.  It shows the TM18 being broken down.  The Power Soak knob pulls off and the nut on the switch then unscrews.  You can see in the picture that the knob is only a push on type with splines.  There is no screw nor should there be any glue used to hold it on.  If it is stiff try levering it off with two strong spoons on opposite sides.  Support them with a couple of pieces of paper or a cloth so you don't scratch the panel.  Make sure to keep them directly opposite and push evenly on them both.  Another way to remove stubborn knobs is to take a piece of string and put it under the edge and bring it out of the other side.  With the two ends of the string supporting the opposite sides pull squarely outwards and the knob should pop off.  You want it to be fairly tight when you push it back on or it may keep on slipping off.

Yes, this definitely looks like something odd has been sitting there and has been fried by the HT. The terminal is fine and the cable to it is also undamaged. Beyond that we can't really say. Try to get the area as clean as possible with no black carbon dust left and try it again with your Vox valve in place. We have to hope that there is no serious damage to the PCB. It should be possible to clean the area so the voltage does not track across it.

Transistor legs are black and like you said sparkle touched them surely, because it is not just dust, it is burned metal

Can I use acetone to clean it?

Can I disconnect this anode terminal? Looks like that there is a little hollow because of burned pcb. I also found that cathode connection not that wide as it has to be. It is burned, but not complitely, i checjed it with my multimeter, from transistor leg to tube leg, almost zero resistance. May anode short to cathode at that spot? It may be a reason of flashes when i turn off standby. 

I would be careful with acetone, it's a very powerful and aggressive solvent and will attack most plastics. You could try it on a blank area of the PCB to see what effect it has. It will most probably take off any ink such as the PCB screen print and could remove the PCB lacquer coating. I'm not sure but it might even attack the PCB substrate itself, I have never tried it.

Don't go over any tracks or pads with it until you are sure it will not damage the surface in any way. Definitely Keep it off any components and plastic parts. You should need nothing more than a drop soaked onto the end of a Q-tip so it doesn't pour around on the PCB. Then you can rub gently with the tip and any residual debris should come off. It won't necessarily need to be absolutely as new, a little blackening left in the PCB may not be a problem, as long as there is no powdered carbon track there on the surface. If you can I would start with something a little milder such as Methylated Spirits, (methyl alcohol/methanol). You could also consider isopropyl alcohol (isopropanol) such as you get as a general delicate cleaner.

Save the acetone as a last resort and test it on a blank area first.

That's a good picture!  Yes, you are right, there is damage under the spade terminal.  Certainly ease it off gently and inspect and clean it up.  If it is a loose fit then gently squeeze the cable terminal to close up the gap a touch.  There isn't a massive current going through there, valves tend to work at high voltages and low currents, only the output anode currents and they are fused at 200mA.  It certainly looks like the fuse hasn't blown so the surge should have been less than that and not for a prolonged period.  So as long as the contact that is left is clean and the terminal is a fairly tight fit I would guess that it will be ok.

The location of the burning is not what I wanted to see really.  I was hoping the arc would have been down towards the two pink resistors.  Looking at where it is it will possibly have involved the PCB tracks it passes over.  It is also possible that the problem was a poor contact there in the spade terminal itself.  If that gradually burned away the surface in use, the contact would get worse and worse.  Turning it on after a jolting postal journey might just have caused an arc from the cable spade terminal to the PCB terminal.  If that has happened then we may be fine as it would not have involved the transistors.

That connection will definitely need to be cleaned up too.

I will also point out that the PCB track which passes under the terminal nearest to it does actually narrow at that point.  It has the thick part reducing down to thinner exactly how the other track does near the TP12 lettering.

I can't believe, it is works! Cleaned everything as good as I could, putted in old tubes to test, and turned it on upside down without base to have better look at pcb. No sparkles, no noise, all channels works fine.  Through this situation I learned how tubes and mosfet transistors work. Thanks for taking the time to help me!

Hahaha! I'm really glad to hear that. You must have done a great job.

Keep an eye on things from time to time. Inspect what you can see of the area you have worked in to check that nothing odd shows up. Now that you have had the amp open don't get too confident. Remember that in that area THERE ARE DEADLY VOLTAGES. It's best to not play around unless there is a real reason to do so. It's also a good indication of how well the amp was designed by H&K, that nothing has been damaged even in such a powerful event. People complain about the TSC control system and I have even known them to remove it!!! But this has shown it is pretty robust.

But, well done to you on fixing your problem and for staying safe while you did so. Now get on and PLAY!!! We all need a way to forget our troubles for a while.

Remember, we're all thinking of you out there.

Yes, I seen post where guy just removed tsc system for no reason. I think that everyone would prefer to have all designed functions in amp than 2 forever extinct leds at the back of it. Are you administrator of this forum? I see that someone moved our messages to a new topic (cause when I found this site, at panic, I didn't notice that I wrote it right under someone's question) . Have you ever worked in Hughes&Kettner or you just like this amps and know physics very well?)


P.s. I found out that google automaticly translated multiple of my messages from english back to russian and you had to translate them back to english, lol.