When to start repair.

[j.d.roost]

Hello.
I have a mesa 2:90 that has a channel randomly dropping out (a-side). The volume will get halved + and the sound will go full treble. I have replaced all tubes (power and inverter) and cleaned all pots/sockets.
I have some slight knowledge of electronic circuits and have built a few stomp boxes. I also have been reading MANY posts/forums on tube amp repair (and the dangers involved with high voltage caps). Knowing that there are lethal voltages stored in those filters ect. I really want to give it a shot repairing this thing. I am auto tech (for over 20 years) by trade so it tears me apart having to send it out for someone to repair.
My question(s) would be..

Is there anyone here that started repairing amps without any formal training?

If so, aside from making sure the caps are drained (think I have read just about everything there is to read on amp repair safety) what other pitfalls do I need to lookout for?

I am not looking to blindly go sticking my meter into/onto in inner workings of my amp. I just want to find out what's going on with this thing and see if I can repair it.
I am going to need to check voltages on this amp (so now my chances of shock are much higher ).Any good safety tips that will help here?

I am only armed with "internet smarts " and a general good sense of mechanical repair knowledge. A also don't have time in my life to take an electronics repair class (don't think there are any in my area focused on tube amp repair/service anyway).
If there is a good web sight dedicated towards people wanting to get into repair/building (aside from this great forum) I am open to links too.
Also I don't have a scope so it's going to be things I can only check with my Fluke.

I know I am going to get a lot of "DON"T DO IT" posts here but that's not what I am looking for.. I am trying to do this as safely as I can.
Thanks for the help.
Jd

[AmpFix]

If you have a curious mind and are determined to do things right, I say go for it. I have an electronics degree that I haven't really used in practice in the last 20 years. I'm a little rusty and I've had to re-learn some concepts.

Tube amps are fairly simple. Basic electronics knowledge like what a resistor, capacitor, diode, transistor, opamp & inductor do are a good start. Learn about power supplies and rectifiers and Class AB vs Class A amps. A good book on basic electronics could be a good idea. Learn how to discharge the capacitors properly and get a good multimeter at a minimum. Try to read a schematic and understand what each section of the circuit is doing. There are 2:90 schematics available on the 'net.

Armed with that knowledge and good trouble shooting skills, you should be able to repair any tube amp. repairing stuff is all about experience and knowing what to look for so there is learning curve.

As for your Mesa 2:90, as you know, intermittent problem are a real pain to fix.

This could be as simple as a bad solder connection. One of the things that the experts suggest is to "chopstick" the amp. That can help in finding loose connections or bad parts.

[j.d.roost]

Yea,I did the old wooden dowel thing already. Wish it was not intermittent too.. You don't know how many folks cars we need to turn away because we cannot duplicate the problem.. That's another reason I want to go for it on my own ( with help from you all/internet/books). I can see dropping it off for repair and having the tech tell me it's fine.(talk about karma slapping me back in the face lol)
I think I am going to need to wait until the amp acts (does it about every 1/2 hour or so..not clockwork mind you) up and then check voltages against a baseline.
Another thought I had was switching the a and b output transformers (on the right front of the amp) and see if the problem switches to the b channel. Looks like I might be able to get away with that by just swapping a few wires.

I have tried every thing you can (all new tubes/cables different preamp ect.) without actually tearing into the amp.
Thinking it's either a transformer on the way out or a bad cap/resistor. If I cycle the standby or power switch I can get the channel to come back on for a bit.

[AmpFix]

With an intermittent problem, I always try to characterize the problem as best I can:
i.e. Does it happen at startup? Is there a time delay? If so, how long? Do I have to play the amp for the problem to happen or does it happen if I leave the amp idling for say 1 hour and then I start to play, the problem is there? When the problem happens, does it go away if I let the amp cool down?

When the problem does happen, it would be nice if you had the ability to check the voltages on all the tube pins of all power tubes.

You could just re-flow the solder joints, clean the jacks and see if the problem goes away.

[gbono]

Not that familar with the two x ninety but you might want to give some more details on the settings on the front panel modes you're in during the problem(deep, 1/2 drive, etc). Since it's a stereo amp are you using both channels (where's the Level A/B switch setting) ??

You should build a light bulb limiter circuit if you plan to make repairs - very handy and a safe way to test circuits that have a history of blowing fuses.

Does this design have any PCB's (I can't remeber now) or is it point to point wiring? Have you cleaned all the appropriate connectors/switches/etc in the signal path? Re-tightened the tube socket contact pins? Get a magnifying glass or loop and do a good visual on the signal path too.

[Euthymia]

Where would one go to get "formal training" in valve amplifier repair?

About the only place I can think of would be apprenticing at an experienced tech's shop. Or maybe one of those Gerald Weber boot camp things?

Anyway, the best source of tube amp repair lore I know of is in books and online (here and at AX84) and yes, on one's own bench.

What books? Dave Funk's Tube Amp Workbook is probably the one I've learned the most from, followed by Tom Mitchell's How To Service Your Own Tube Amp (good flow charts).

Those two will probably get your particular amp back up and running.

A classic, now back in print, is Jack Darr's Electric Guitar Amplifier Handbook. He covers tube and solid state.

Next is Tino Zottola's Vacuum Tube Guitar and Bass Amplifier Servicing, and then the Gerald Weber books, the Dan Torres book, and Pittman's The Tube Amp Book 4.1th Edition (sic).

Here are a couple of useful websites:

Index

Tube Amplifier Debugging Page

Also, a good used 10MHz or higher oscilloscope from eBay, Craig's List, or a ham radio club swap (with one 10X probe), is pretty essential.

The earlier advice to observe and describe the conditions surrounding the fault is very good:

Is it after the amp has been on for a while? Only after playing loud for a period of time? When it's in the volume drop condition, does control fiddling, pounding the top, turning the amp off and back on, or anything else bring it back? When it's in its fault mode, do the controls have any effect?

Hint: in my shop, Mesas are notorious for effect loop dropout, which is what happens when the switch on the effects loop return jack gets dirty. Dunno if your 2:90 has anything like that, but Mesa just loves to run their amps' entire signal through a switched jack.

If you're safety-minded enough to work on cars, then you can do tube amps after you read about the precautions.

My favorite rule: if the power's on, I have my left hand in my back pocket.

[j.d.roost]

Little more history on the problem.

Cleaned all tube sockets/jacks/pots tonight with deoxit.
Same problem. The a channel will just randomly give you about a 1/8th of its total volume no matter where the volume pot is (will get "louder"when you rotate the knob but not loud). I hit EVERYTHING with the dowel tonight trying to get it to pop back on but no luck. The only way it will come back on is to cycle the standby switch off and on a few times. Then it will work at random (sometimes for a 1/2 hour sometimes for 30 seconds).
Power amp only so no send/returns. Just3 12ax inverter tubes a octet of 6l6s a power and two output transistors.

Amp does not blow fuses.
Looks like I might need to get a cheap scope.
Thanks big time for the advice.
Jd
edit** Looks like the debug page has given me a few places to look.

Failure only happens when the amp is taken off standby: Since the B+ voltage rises in standby, this often means that the higher voltage is preaking something over. This may take the form of the amp only coming on slowly after a delay when the switch is thrown, or of a squeal or pop after the switch is thrown, or ugly sounding distortion for a while until it "gets better"
Preamp decoupling capacitors
Signal coupling capacitors
Dirty, contaminated, or arcing tube sockets.

Faulty phase inverter
If for some reason the phase inverter input side is good but the inverted side is bad, the power amp will still work, but power will be very low. This can be a bad 1/2 tube, a faulty socket contact, a broken or open plate resistor or coupling capacitor to the output tube, or a bad solder joint on any of these.


Guess I will strart there.Looks like the challenge now is finding those parts related to my power amp.I wish the schematics told you what the parts in the circuit were.Looks like i have some reading to do.

[j.d.roost]

Ok need a little more help.
Checked voltages Tuesday and from what I can tell everything checks out.
BUT.. a funny thing happened. I decided to toss in another 12ax7 tube (old penta tube) in place of the new mesa that was in the A channel inverter socket. Fired up the amp and played it for two hours with zero problems.
Chalked it up to two?? bad mesa tubes I tried in that slot.. Anyway long story short I was playing it again last night for about two hours and decided to try out the "modern" mode on the 2:90 (don't like it btw). When I switched back to normal mode I started getting a very strange "chorus" like sound from the amp (sounded like I had a pedal hooked up) and then 10 min.later the A channel crapped out again. Hit standby got the loud pop sound that I usually get when the a channel acts up and then the channel came back on for another hour with zero problems.
I don't think it's the switching section of this power amp that is causing the problem (tried all that stuff before when it would act up every 20 min or so) but I think it has to either be a cold joint or maybe a flakey coupling cap (because I also loose all bass responce when it craps out).
Maybe the slightly different internal ratings of the penta tube lets the resistors/caps operate for a longer time without issues? Fixing/testing/pulling the any caps ect. requires me pulling out the pcb.

Only thing I did not do was check the voltages when it had the problem.
I think this is key to fixing things. Problem is I can't get it to act up again.
Should I stick the mesa tube back in ?

Any thoughts on what else I can try?

Thanks for the help.

[booj]

Sounds like a solder re-flow session is in order. At least that would take bad solder connections off the list.

[Enzo]

Check each and every control pot. Is the back cover loose on any of them? A lot of Mesas get this problem. If something smacks the knobs out front, it puches the shaft back, which pops the rear cover loose - the little tabs bend out. Since the cover is what holds the wiper assembly in place, the wiper can lose contact with the resistor strip. Demount any suspicious pots and using pliers, squeeze the tabs back down to tighten up the pot cover.

Just a thought.

[booj]

agreed on the pots. I've changed a few that lost connectivity just in one spot because the trace cracked, or the wiper got deformed because the control took a hit, which made the control work for only half a rotation. What your getting could be another symptom. Play with the controls and see if you can get the problem to come and go. You might be able to squeeze 'em back together, or you may have to replace them. Mesa's good about sending new pots quickly.

[j.d.roost]

I will give the pots another shot.
I HAVE tried push/pulling them when the amp acts up and there is no change in the channel.
The board would need to get pulled for a re-solder so thats not going to be fun..
Need to weigh my options here. I don't know if a tech will sit with this thing running on his bench until it acts up (not knowing if I would get "lucky" and have it quit within a few minutes) without charging me an arm and a leg.

Guess I will keep plugging at it. I have gotten a pretty good education on tube amps in the past few weeks.
Thanks again guys.

BTW. I DID get through to a Mesa Tech last week.Super nice guy to offer his time to me for free. He had me check a few things (voltages ect.) over the phone and said that it all checks out.He said to give a once over to the smaller caps on the right side of the amp (brown caps) as they are prone to leakage (I did not see any). One other thing to note was he wanted me to try bypassing the half channel ldr said that it may be getting "suck" in a state of neither off or on ( I did not want to take up any more of his time asking how to do this and I said I would try it later that night ). I am not sure if I should cut a leg or jump a side to do this any ideas?

Jd

[j.d.roost]

Ok, wanted add to this.
I "jumped" the a channel ldr. by bridging together the non led side.
I was able to play for three hours with no sign of the channel dropping out.

BUT.. now the amp will swell the volume (higher) every once in a while.Usually after a pause then a chugging riff.
So I think it fixed the cutting out but caused another problem.. Wonder If I should cut a leg of the ldr instead of bridging.

[Enzo]

Shorting across the part isn't a repair, it is a way to determine if the part is involved in the problem or not. If jumpering past it makes the particular problem go away, then it is likely the cause of that problem. The fact that some other symptoms exist may or may not be related. Once a new LDR is installed in place of the bad one, THEN we find out if there are further problems or if the added funnyness was just due to our shorting wire.

[booj]

C'mon! wheres your sense of adventure? Pulling Mesa boards is not difficult. It just takes a little time. Just make a drawing of any wires you need to disconnect to get the board loose after you remove the knobs and loosen thier respective nuts. Be gentle, making sure not to force anything to happen, even if your only a 32nd of an inch away from getting it loose. Make a note of any ties you need to cut. The pots are a bit fragile, so take extra care with them. Have fun with it.