DSL 100 with Shorted Power Tube

[dorrisant]

Back in May I got this one in with a complaint that it just started blowing HT fuses. When the second fuse blew the amp was shut off and brought to me. I tested all tubes and found weak preamp tubes and a couple of really weak power tubes. I replaced all of the tubes with new Mesa tubes. Not my choice, at least it was a new set. Checked all the voltages. Rebiased to factory spec at 90mA per side... tested and burned in. No problems. Gave it back.

A month ago it comes back to me with the same complaint... Two fuses burnt. This time i found a shorted power tube... all pins. I had heard lots of complaints about the TSL and DSL "run away bias" or "bias drift" problems. I have done the mod that entails drilling out the PCB at #5 pin to isolate the pin from the board. Inspired by this site:

http://www.hullerum.de/Marshall/TSL122repair.html

No slam to the author of that site, but when I've done it, let's just say I leave it looking like the factory did the mod. Same results though, never had those (2) come back yet. I didn't do this mod at this time... I was torn between the concept of the board being the issue or just heat coefficient resistors in the bias circuit, so i just changed all of the bias resistors to 1w metal films. Replaced the 1R bias measurement resistors with new 10w. I've seen these fried before. I replaced the bad tube, rebiased to 80mA just to be on the safe side and let it cook for about three hours with no signal. Then I plugged in and played for about 45 minutes loud enough to require ear muffs. Good enough, or so I thought...

It came back today. Same thing with the HT fuses. One power tube (same side - other tube, so it doesn't seem like it is the socket) was completely shorted, all pins. Checked voltage at the plates (513), screens (509) and bias (-39.9 and -41.6) with the power tubes out. Each side of the push-pull is biased separately. They all looked reasonable, a little higher bias one one side than the other. The higher biased side (-39.9) was the one with the shorted tube. I'm still not sure if the bias ran away under the heat or not. I don't really trust Mesa power tubes... I mean they beat on them under load. No thanks, I don't treat any tube that way.

Gonna isolate the bias pins (#5) for now. I know I should have done it before. Probably gonna do it to every TSL\DSL that i work on. My question is do you think that isolating the bias pin from a possibly conductive board will really solve this? Has anyone else ever tried it and with what kind of success? Any other suggestions? I have it here close to the bench, ready to check whatever you folks can think of that I haven't.

Thanks
Tony

[Richie]

Might ask the owner which output jack he uses. Check this info. I think on these amps this needs to be done. I fixed one a while back that melted.

The repair fix is easy. But Seems this is another one of the problems that can take out tubes, or even the OT .
Its good to also look at th other info posted, but the output jack problem is down the page a bit. I think if you also google it, it will come up and show pics etc..
http://music-electronics-forum.com/t12900/

here is another link with fix info.

http://www.marshallheads.com/viewtopic.php?f=8&t=48

[Stevem]

If the owner of the amp is driving a higher impedance load then that can arc over the OT and or the tubes and make a mess over and over!

If the amps where cars they would have been recalled for that crappy bias circuit!!!

[dorrisant]

Thanks for the links Richie, much appreciated. Looks like I'm gonna hit up the local parts store and stock up on some 22pf 2000v caps.

Steve, I totally agree. Someone stated in another post that if these were Marshall's first run of amps the company would have went under from the cost of warranty repairs, and the TSL/DSL line was not the only problem children they had.

I'm not trying to complain about it too much, I just want to be more aware of how to keep them from coming back. Thanks guys, I appreciate the direction.

Tony

[Neeno]

Hi Tony
I fixed a broken TSL100 almost a year ago.
It had the OT shorted, and two EL34 on one side had the getters completely cleared up (one tube was shorted)

The problem was the PCB, to be specific the material of the PCB.

It seems that the high temperature generated by the tubes compromised the insulation capability of the pcb material at the point that the bias and the plate traces had less than 1MOhm between them.

That made the bias drift, the power tubes shorting and the OT fail.

I replaced the OT, the tubes and I've got a newer revision of the PCB from a UK company (I don't recall the name, but I can find it again if you are interested) the new board is made with a better material which is more stable at high temperature.

Hope this helps.
(sorry for my bad english)
Nino

[Stevem]

I had one last year that the owner had pulled out all the boards and unplugged all the dam board jumpers, that thing was a nightmare to get all those jumpers back where they should be, I should have charged the guy three times what I did!!!

[dorrisant]

Thanks Neeno. That sounds like the sane kind of gremlin here...

I have another question...
The PT is a little warm... Around 125℉. Am I being paranoid? The was after running about 4 hours with no signal.

Tony

[Stevem]

Is this 125f with the chassis in the cabinet with the back on?

[dorrisant]

No Steve, chassis sitting in open air, no cab.

Tony