TSL60 troublesooting

[goldenGeek]

I recently bought this TSL60 head for cheap hoping to be able to fix it and use it or sell it. Anyway, theres something funny going on in the switching-department, I think. The clean channel works like its supposed to, but the drive and "more" drive channels doesnt make any sound at all. I have tried other working tubes to rule out bad tubes. So, here goes: when I switch to channel 2 (yellow led), the relay clicks and the led change to yellow (the green turns off). When switching further, to channel 3 (red led), the relay clicks but the yellow led stays on also. No sound in either position. So i started messing around measuring some voltages, thinking there has to be something wrong with the switching. My first question is probably stupid, but I'll try anyway:

If you look at the schematics, page2 bottom left half of the page, RL1a and RL2a is what switches the relays but my measurements reads about 22.5V at the point marked 12V in one position and about 0V in the other position. At the "top" of the diodes (D14/D19) I get a reading of 22.5V in one position and about 12.5V in the other - are these readings correct? I never get 12V at the 12V point at the schematic... any clues?

Marshall-JCM2000-TSL60-60W-Schematic.pdf

[goldenGeek]

i have been trying to trace where I lose the signal and it seems like its fairly early in the signal chain. I've probed all the tube "inputs" (with a home made probe to an battery amp) and on channel 2 and 3 I only have a good signal at V1 pin2, and a really weak signal, barely hearable, at pin 7. No signal at any other pin 2 and 7... any clues to where my signal could get lost?

[xtian]

Kinda busy schematic, huh? Get out the highlighter!

tsl60-trouble.png

So you confirmed you've got signal at V1.2 (shorthand for V1, pin 2). Right after C63, you should be able to get a much stronger signal. And then the signal splits into two paths, marked in yellow and blue. If you have no signal at V1.7, where did it vanish? Follow the yellow path to find out.

[goldenGeek]

Thanks @xtian, The schematic is kind of overwhelming. Really good idea to trace with different colours. I will dig into it tomorrow

[goldenGeek]

I have a strong signal at the top of R90/R91, barely hearable at the other side. Could it be a bad resistor/cap there (R92/C84) thats bleeding too much to ground? It's a pain to work on/troubleshoot this PCB, I guess I'll have to lift some legs to be able to measure? I'll grab my LCR and scope ang get to it. There is obviously something wrong with the switching as well, but I think (as far as I can read the schematic) that shouldn't affect this part of the circuit.

[xtian]

Measure resistance across R92. If it is less than 470k, cap C84 is suspect.

[goldenGeek]

I had two minutes free between activities today so I got around to measure, R92 measures about 18,7 ohm. So I guess a good starting point would be to snip off that cap and measure again... eventually replace the cap and/or the resistor. I will let you know how it goes

My only concern is that if this cap is bad, how many more are dodgy...?

[goldenGeek]

I'm starting to think that the PCB is conductive...

I Completely removed R92 and C84, no change. I measured both components and they are within specs as far as I can see. I also removed R90 and R91 which also measures fine. Then I started measuring references to ground and shorts. It seems like all points measure between 5 and 26 ohms to ground, without the mentioned components installed. Could there be a logical reason for this or is it possible a conductive PCB we're dealing with?

[xtian]

Make sure all the tone controls are set at noon. That's the only other path to ground I can see.

Is the PCB darkened from heat? That might make it conductive.

[goldenGeek]

I did some more measuring and I think it might have to do with the switching circuit. If you look closely beside V1 pin7 it reads "FTSW4", which I guess is somehow connected to "FTSW4" at page 2 bottom right corner and thats probably the path to ground. V1.7 is tied to ground for sure. When I measure F8 and F9 I get no readings at the G-point in any of the switching positions, which leads me to think there is something wrong further down that path, not being able to forward a reasonable voltage to control the JFETs. Also, like I said in my first post, the orange LED does not turn off (just gets dimmed a tad) when going to the red channel. It's hands down the worst schematic I've been working on so far.

[xtian]

OK, good analysis. And, I see you mentioned the switching system voltages in the first post, which I promptly forgot.

So, can you start (again!) by confirming the highlighted voltages?

voltages.png

[goldenGeek]

Left to right:
+15,1V - OK
+20V - too low?
+15,1V - OK
+1,1V - way off, wrong polarity
+15,2V (but measures only 1,7V on the other side of R43)
Pin 4 and 8 at IC2 is totally off, no negative voltage on any pins of IC2, voltages is varying with channel-switching. I did not write down these voltages, but I can do it if its important.

[xtian]

Well, we've identified some trouble spots, primarily that the +/- 15v rails on IC2 (pins 4 and are not good. The negative 15v rail especially is suspect. So, trace that supply back to its source. See page 2, area C3. The +/- 15v rails appear here, supplied by Zeners ZD4 and ZD5. If you do not have -15v at the junction of ZD4/C96/R84, you know you're getting close to the source of the problem!

[goldenGeek]

Aaaaaand it’s alive Thank you so much for the support Xtian! The problem was indeed in the negative 15V supply. I had to remove one more board to access ZD4 (and ZD5) and there was no power present at ZD4. So I poked around a bit in the same area and discovered that one leg of R84 was just sitting loose on top of the solder-hole. After refitting (and measuring) R84 and refitting new components around V1b where I removed for testing, it is now working again like it is supposed to.

[xtian]

Ah, man, I love it when it works!