Marshall TSL:100 bias drift problem

[lowell]

Oh I didn't know that.. what is the issue with C46 if it's not the bias drift?

Also, do you have any tips on drift?

[Leo_Gnardo]

Oh I didn't know that.. what is the issue with C46 if it's not the bias drift?

Also, do you have any tips on drift?

check post #3

[The Dude]

^^^^ Yes, that for bias drift. The problem with C46 is that it's rated voltage is too low from the factory and often burns or shorts. I always replace it with a higher rated cap as a matter of course- even if it's not bad. Others have clipped them out and left them out without issue. It is there to prevent oscillation.

[Enzo]

It is generally good practice to allow and amp to warm up at LEAST 20 minutes before adjusting bias.

[lowell]

Thanks guys...

Leo.. I should have mentioned that mod has already been done but did not fix the bias drift. It truly seems as though removing C46 fixed it. Odd.

[Leo_Gnardo]

Thanks guys...

Leo.. I should have mentioned that mod has already been done but did not fix the bias drift. It truly seems as though removing C46 fixed it. Odd.

C46 is undoubtedly a liability. I've seen 'em charred - a big clue even I can appreciate. The TSL100 is an amp with multiple failure modes, and you've addressed 2 of them. Board? Cap? Yes, both. And... once you get 'em working, they still sound terrific.

[g1]

I think the terminology may be the problem here. You were actually measuring idle current at the cathode (90mV) rather than bias voltage at the grids, correct?
In that case, it was 'idle current drift' rather than 'bias drift'. Bias drift will always cause change in idle current, but idle current drift is not always caused by bias changes. Hope that makes sense.

Change in plate or screen voltage can cause change in idle current, this cap could have caused that. Hum/noise could also be seen as change in idle current by the meter, and this cap may have caused that too.

[lowell]

Agreed.. BUT, now I am back to where I was before... months ago. I took a break from this amp cause I couldn't figure it out.

I put it back in the cab and it is now running away really bad... I cannot get one of the pots below 100mv and it's still climbing after an hour. I then remembered that it went away when I pulled the PI. Pulled the PI and it absolutely goes WAY back down. Also, over an hour of leaving it on, there is a hum that gets louder. I did notice some bulging B+ caps, and figured I'd replace them when I fixed this problem... but NOW I'm thinking they ARE the problem. My theory at this point is that those 2 caps are failing, and failing more as they heat up, producing more hum, which seems to come thru the PI and then on to the PA. So my theory is that the hum is what's causing the HOT bias readings, not thermal runaway or drift. Gonna replace those caps and see what happens. It absolutely makes sense being that the problem goes away when the PI is pulled. I can't come up with any other hypothesis. The 2 caps are the main high V B+ caps.

[The Dude]

I'm wondering if you have an oscillation problem. Did you replace C46 or just remove it? Do you have a scope? If so, scope the output and see if you spot any oscillation.

[lowell]

This was happening before I removed C46... and after. I am just now recalling all the tests I ran on this amp months ago. Usually I have notes on these things... but have been out of the routine.

[Enzo]

Did you replace C46 or just remove it?

Did you scope the output of the amp or at least around the power tubes to see if there was high level RF oscillation?

You say pulling the PI drops the tube current. OK, monitor MAINS current draw, does that go up when the PI is installed? Does it look higher than average?

[lowell]

Unfortunately due to circumstances my scope is not available at the moment.

I removed C46 altogether.

Running a mains current test now... at power on it's sitting at 1.16A. Seems very high... in standby it's at 560ma. I'm way out of practice... but that seems high.

If I recall... Filament secondary draw is about 3.6A.

3.6A x 6.3V is about 23W. 23 / 120vac primary is only .19A - So in standby it should only be pulling .19A... save for some HT current charging the main filter caps.

Thoughts?

[Enzo]

OK, make an RF detector in place of a scope. Your meter cannot measure high frequency AC, so take a diode and put in series with your meter probe, and add a smallish cap to ground from the probe tip as well. Not critical, maybe 0.01uf? And measure the output as it sits there. Normally an idle amp makes zero output. If it is sitting there making 50 watts of 50kHz, this diode thingie will rectify it and the cap will smooth it, and your meter will read it as a DC voltage.

There are various ways, here is one:
https://www.youtube.com/watch?v=QUmPBBnzkRE

I fogure an amp like this at idle should usually draw ROUGHLY 40-50 watts from the wall. Half an amp sounds close enough. But your half an amp was in standby, in operate, your 1.16A is about 140 watts. So either it is biased way hot or it might be cranking HF oscillation.

[lowell]

As always Enzo thanks for the help. Wouldn't this just show ANY noise the amp produces? IE If there's hum anywhere in the amp that's being amplified to the output, wouldn't that show as DC voltage with the probe you describe? How do we know it's an oscillation?

[lowell]

Lemme answer my own question. The smallish cap won't register lower frequencies. This will only yield rectified high freq.