Okay, this project is complete now so here are a few notes for anyone dealing with this amp in the future.
The amp came to me with a number of issues but first off, the clean channel was not good. There was a lot of distortion and it made the gain pot unusable past 2 on the dial. The owner had also previously had it modified to include a serial FX loop switch, but that gave a very low output signal with the loop engaged; about half the gain of the dry channel. Another set of things which became apparent later on in the testing was that the old FX mix pot was basically inoperative - unity gain across the range of the pot - and also, the FX loop appeared to be engaged with or without the jumper added to the send and return jacks.
First thing I tried was checking and rebiasing the amp. This is where anyone who rebiases this amp (or the DSL201) has to be especially careful. Don't go off the figures in the Marshall bias instructions. Take the plate-cathode voltage, and use that in conjunction with the EL84 power dissipation figures to find your bias point. The notes Marshall supply on rebiasing can easily lead to redplating and tube failure - If you follow the Marshall figures you might end up rebiasing your tubes to 100% PD, class A. The last tech had biased the tubes around 40%, probably in an effort to tame the distortion, but I rebiased it to 60% PD. On the schematic, the bias point is identified as CON5 and it's set up to be biased using the cathode voltage method. Not the crossover distortion method using a scope. I have a feeling that this amp had been incorrectly biased at some stage because there was evidence of charring around V7 and V8 and especially around the grid resistors there. I suspect the grid resistors were also under-rated so I replaced these with 1W Holco H4Ps. There was also a replacement HT fuse holder that had been fitted before the current owner bought it.
So there were a number of issues that had developed and it had been opened up by at least two other techs before it got to me. I won't go into the troubleshooting that took place next but on removing the board and getting a look at the traces underneath it was clear the board had failed in several places and needed repair. Seriously affected were the grid resistor pads for V7 and a ground shunt pad at the input jack.
IMG_20230318_204614.jpg
IMG_20230318_171501.jpg
The jack pad could have separated because of the mechanical action of the jack, but the grid resistor pads and traces were casualties of extreme heat. Perhaps caused by some intense cycle of oscillation previously. Grid resistors don't normally need to spec'ed too high. We will never know what caused that - some rogue pedal perhaps - but probing with the scope when everything under the board had been repaired and the thing remounted in the chassis, there was no trace of oscillation that I could see. This was certainly some problem that had happened before the first tech saw it, because of the work that had been previously done under the board. The second tech didn't bother lifting the board, and this had led to other problems.
The second tech had tapped the feed to V3s grid with the send and return to his switch but rather than pull the board he had snipped the resistor out and soldered wires from above, to its remaining legs. What he couldn't have known though, was that heat conducted down the connection and created dropouts and dry solder joints on the pads on the reverse.
So, I used small diameter wire connects to bridge the gaps in the traces and with all the connects secured underneath the board, I remounted and reconnected everything up.
IMG_20230318_212238.jpg
At this point I was hoping the problem might be fixed but didn't really see how those problems could have caused the distortion so it wasn't a surprise when I switched it on and found it to be the same. We were reducing the number of possible failure points though so I asked the owner to leave it with me while I investigated further. There were now, no obvious failure points. Faulty connections on the board had been fixed, new grid resistors were on the power tubes, pots and caps had all checked out okay. I was starting to think it might be transistor or op-amp failure but in the meantime, I had to go do other things so I just kept mulling but didn't actually get back to it for several weeks.
Then, one night, I was laying in bed and I thought about the switch the last tech had installed. The answer, it seemed, had been staring me in my face all along. I had to get up and check the photo.
1677427607706.jpg
There was a lesson in switches here. The A side of the schematics shows how the tech had wired the switch. What was obvious, when I looked at it like this was that the dry side of the signal was not being lifted as intended, but bypassed the grid resistor through the switch allowing the dry signal to hit the grid of V3 unfiltered and unattenuated. Once i rewired the switch as in the B side, the clean channel was fixed. It now has a clean signal up most of the gain pot range. And what's more, with the FX loop engaged, the FX mix pot works like a volume pot for the FX. If I were to remove the jumper from the FX send and return jacks and engage the FX with the FX mix pot all the way CCW, I should theoretically get no output signal. That's not the way this circuit plays though. Neither 100% wet nor 100% dry are in fact possible. The FX loop circuit is also tonally weak. A common complaint in the Marshall community as far I can gather. I would go further and say that the OD1 and OD2 channels are not my cup of tea. But for an amp that cost about $500 new, it's best not to set your expectations too high. The clean channel tone is quite decent though and I can see why the owner is only really only interested in running in clean.
FX switch.jpg
An object lesson then in learning to apply Occam's razor. It's the simple things that can escape you. Not forever though, and in the course of the service, repair, rebiasing and diagnostic I also checked all the AE caps. For an amp approaching or around 20 years old, the caps were all okay.
I would finish this off by posing a general question here: if you had to do all this, to get an amp fit for return to a customer, how much would you charge him?
You do not have the required permissions to view the files attached to this post.
