Mesa Boogie Dual Rect Rebuild

[CJ]

this thing does weird things, EL34's are blinking, tube rectifiers are popping like popcorn, tearing my hair out trying to find the problem, must be the pc board, never a good idea to mount power tube sockets to boards, have seen nothing but trouble with this approach.

so we rip out the board (seen at top) and put in some nice ceramic sockets, will report back to tell if this is a fix..

 

[musipol]

Fully agree. Chassis mounted sockets and flying leads........more work, but less issues. I have seen similar problems on the old Peavey Mace amps with pcb mount sockets and too little room between traces to the power tubes. In those cases, it was typically beer spilled on the amp that found it’s way to the power tube pcb. The power tube pcb in some cases was carbonized and I had to toss the pcb and rewire with new sockets. Just venting as I havent had to deal with one of these in a long time.

 

[CJ]

then there was the Marshall's with the conductive pc boards allowing HV screen supply to leak over to pin 5 and send the tubes into meltdown, Marshall sold replacement boards for 200 but required customer to pay> Marshall DSL100 conductive pcb

 

[mikeyB]

My oh my CJ, repaired one of these years ago, turned out to be a shorted valve which drew BIG current through the resistor, which set on fire and NUKED the PCB.
Luckily my mate had just powered up in the live room ready to go to the control room and play, the damage was immense, even after switching off within 2 or 3 secs!!! Replace those 2W or 3W with self extinguishing ones!! I'll try and find the pics from my backup disc, but yes, those PCBs are totally shite (totally charred/powdered) - beware!! I think it must be a similar thing to the Marshall DSL above!!
Totally agree - I will never socket a power valve onto a pcb!! You'd have to have a super design pcb otherwise and i'd still be constantly waiting for the smoke!

 

[john12ax7]

I've seen power tube sockets with really long pins, that should help with the heat issues.

A lot of these pcb tube amps are nightmares to work on.

 

[Matador]

Mesa's have always been my favorite amp as a player, and the bane of my existence as a repair person. None of them were designed with maintenance in mind, with hidden bolts, blind traces, and obfuscated/incomplete/incorrect schematics.

 

[Tubetec]

Apparently Marshall pcb boards of a certain vintage were prone to absorbing moisture and hence leakage between pins . Ive seen similar issues with certain Boogie amps ,the F50 was one . I had a customer with one of the lower end 50 watt Boogies , he insisted on having me fit a set of set of Tesla JJ tubes where really the only tube you want in a Boogie is a Sovtek 5881 mil grade . Likewise I always reccomend the Sovtek 5U4G for Dual Rect's . Of course Boogies dont have any bias pot either , so tube choice tends to be more critical to satisfactory opperation than most amps. The long and short of it is in my experience Boogies eat bad tubes alive , either pre or power stage.

 

[CJ]

there are a couple of wafer switches that go on the back panel for manual channel switching etc,
i forgot about them, went to install them after wiring the tube sockets and there was not enuff room, had to unscrew the tube sockets, push them over a bit and the switches fir with just enuff clearance, thank my lucky stars!

i bet a few people have switched that bias sw on the back to EL34 while running 6L6's,

yeah mesas are a pain to work on, even the first one was like a super reverb circuit stuck in a princeton cabinet.


certain Fender's can have problems with those black turret boards going conductive, but at least they don't wipe out the power section. CBS made the problem worse when they skipped on the wax dip step during assembly.

OT here is a Fender factory tour from the 50's. a guy spraying enamel with no mask and another guy sweeping the floor with no shirt trying to impress the ladies. no guards on any of the machinery, Cal Osha would have a field day! not as bad as the BC Rich shop with the beer bottles rolling around the floor.

 

[okgb]

pretty thin traces for " made in America " last time I worked on one. Many models seem to suffer from the switching fets going bad. It's rumored that when they get one back for service they immediately replace all the fets. And of course randall claims they've learned something that no other company does. If I'm ever tempted by one of their ads I just think of the manufacture and the problems if one one goes bad. i think it's karma for mesa ripping off Soldano for the rect and mentioning it, there are similarities between some Dumbles and early mark series [ lead drive positioning, who came first? ]

 

[musipol]

What about Mesa ripping off Fender, as in the original boogie amps? Basically just an AB763 circuit with an extra stage on the front and the LC based bipolar EQ circuit kludged in. Then again, if you really look at it, Marshall ripped off the old Fender 5F6 bassman circuit for their first designs. Soldano certainly "borrowed" heavily from Marshall in his stuff(2203/2204 with an extra stage in front and 39k instead of 10k on the cold clipper stage).
Having worked on a few mesa's, I would never buy one. I have not had every model they make on my bench, but they are the embodiment of 5 lbs of Sh*t in a 1 lb box...............no thanks.

 

[CJ]

some of their smaller stuff is not as bad IIRC,

 

[Matador]

The last Mesa Lonestar I worked on had the EL84's biased to almost 18 watts EACH. That was the factory bias point! You could feel the heat radiating from the back.

Four, 10 cent cathode resistors later, it was down to a manageable 10W and still sounded great, and these will actually last! I've installed countless bias pots in Mark series amps, because despite what Randall thinks tubes are getting harder and harder to come by, and running them 50% overspec makes owning one an expensive proposition.

 

[john12ax7]

The lack of bias pot was always kind of annoying. Think the dual recs are typically way under biased which contributes to the harsh aggressive sound. So then people use the EL34 switch with 6L6 tubes, which is then often running too hot.

 

[soapfoot]

OT here is a Fender factory tour from the 50's. a guy spraying enamel with no mask and another guy sweeping the floor with no shirt trying to impress the ladies. no guards on any of the machinery, Cal Osha would have a field day!

I once read an interview with Leo Fender where he talked about a visit from the fire safety inspector.

The inspector came in and a guy was stretching MOTS over lap steels over an open flame...next to several 55 gallon drums of acetone and nitrocellulose.

The inspector didn't say a word. He just turned around and walked right out the door... to a phone booth down the street, where he called to tell them they had to shut it down

 

[Disco Volante]

Also no earmuffs anywhere. Those wood-shop workers would not have been getting the full enjoyment of the sound-gear they were building...

 

[CJ]

even with the new tube sockets the plate current is starting to run away after it settles at 80 ma on the right bank, had a JJ arc over at this point, so a bias pot is going in next. s/b closer to 70 ma a pair for 460 B+ (solid state rect). the boss is buying EHX tubes now which i hope turns out better than JJ EL34's. although they both could be from the same factory i do not know. maybe put some NOS russian power tubes in there, forgot the part number but i know there are a few good ones out there. maybe Mesa made a deal with JJ and EHX to cook tubes = sales.

 

[abbey road d enfer]

The lack of bias pot was always kind of annoying. Think the dual recs are typically way under biased which contributes to the harsh aggressive sound. So then people use the EL34 switch with 6L6 tubes, which is then often running too hot.

I tried to sim the bias circuit, but I never came close to the published values. I guess the schemo is not correct. It is not facilitated by the fact the circuits are different between the domestic and export version.
can anyone check the voltages, before and after the rectifier, after the (supposedly) 820r, after the 3.3k resistor(s), in both positions of the bias switch?

 

[okgb]

This guy is looking for you tubes views of course but I like how he's not afraid to look bad, he gets 4 episodes out of trying to fix a mesa output tube problem before giving up & stripping it. here's the 4th episode

 

[CJ]

oh man, that is hilarious! at his expense unfortunately, i am going through the exact same thing!

one thing he did not do was change the bias voltage or lower the B+.

i think i might add some cathode resistors of maybe 50 ohms or so on each pair to guard against runaway. once the left or right pair of power tubes starts to conduct more than the other, it gets warmer, which lowers the plate resistance, which means more current, more heat, which also drags the pwr supply nodes down like he experienced, i do not think today's tubes can be run at old school specs, if they do not fail then their life will certainly be short compared to NOS RCA etc.

i usually check output transformer ratios with the variac on the primary side as to no generate dangerous voltages, and i load the secondary with a resistor. if that guy would have done that, his voltage on the secondary would have dropped, thus yielding a different impedance ratio that might have been easier to shop for. he might have got 5K:8 or 5.2K:8 instead of 5.8K:8.

thanks Greg!

 

[Matador]

Aside from his Libertarian rants, Brad is certainly entertaining!

I don't like his proclivity for shotgunning parts: he often guesses at root causes rather than testing and proving prior to repairing. Why rebuild an output section if you don't even know why dissipation is running away?