TSL 100, does it have "the drift"?

[tdlunsfo]

So the store sold and shipped this amp only for the customer to have an issue with it (don't remember the original problem), and their tech told them it had bias drift. The amp was refunded and sent back, and my task is to check that it does in fact have bias drift. So I installed my two plate current probes on the outer tubes played it for a few mins, unplugged the guitar and let it set for 5. I then started a timer and took bias readings for an hour at 15 min intervals. Next I put a signal through it at 10W and took readings again. Here are my results. Initially there appeared to be some drift but was surprised how steady the bias was while playing it through a dummy load. I guess I expected it to drift while being played as well. Obviously these tubes are not very well matched but thoughts on readings and if you think it has bias drift? Any changes to my test setup? I was going to test the inner tubes tomorrow to see if I get a different result.

Idle
00: 43.4, 25.7
15: 42.1, 27.0
30: 41.5, 28.4
45: 41.2, 29.4
60: 41.0, 30.0

@10 W
00: 42.5, 38.6
15: 42.4, 40.0
30: 42.3, 40.2
45: 42.5, 40.4
60: 42.2, 39.6

[The Dude]

Bias is an idle condition measurement. You cannot check/set bias with an operating amp producing signal. Of course tube current will increase when the amp is making sound.

Edit: Also, the TSL100 has a bias adjustment pot for each side of the push pull. You should be able to get the two sides more closely matched through adjustments. Those readings are a bit low for a pair of tubes if I'm understanding your measurements. I'd be biasing each side around 70mA (about 35mA per tube).

Edit again: My bad and apologies. I reread your post and see you were using a bias probe on single tubes, so ignore my statement above about bias being low. That said, the amp has built in bias test points, so it's really not necessary to use a bias probe. If your tubes are not well matched, put the closest matched ones together on each side and use the bias pots to match current from side to side.

[tdlunsfo]

I'm well aware that setting bias is without signal, and that's the first set of readings. The 10w test was more out of curiosity to see if the tubes would also drift while played. I've done one other bias drift Marshall and that one would slowly drift into a red plate condition... hence the signal test. Also aware of the bias pins but I prefer my probes.. just habit I guess. I'll rebias tomorrow and run more tests.

[Jon Snell]

The bias test points are the standard way to measure dissipation.
There is no point in measuring the dissipation current whilst running a signal through it, sounds like you were just having fun.
As long as the left pair matches the right pair in overal dissipation, at idle, that is sufficient to meet the requirements of the test and measurement procedure.
Your readings at idle could be better but there is only a reading for one valve shown, if the other valve in the pair mirror your readings, the average idle current will null.

Best to measure things like this as per the service manual otherwise there will be issues.
I would like to see the reading taken with the standard measurements before deciding but at this point, I would say the customer, just didn't want it and there is no issue with the bias.

[tdlunsfo]

I was having fun but also trying to see if I coud SEE it drift into redplate like the previous amp. Agree I am only looking at two a tube in each side and the bias points would be more definitive. I'm leaning toward your assessment as well but I think I'll hold judgement until I retest. Thanks for the feedback.

[Mick Bailey]

What's the date on the amp's bar code and the board revision? Amps dated 1998 to 2003 are the main problem area, but many have had warranty replacement boards, or may have been done since by an independent. 1998 amps I think started at rev. 3. Marshall extended the original warranty period as a goodwill gesture.

I've had these amps where the drift is quite slow - sometimes a couple of hours before they begin to hum due to high current draw. The bias issues are not only confined to the main board - you always have to check the two trim pots on the bias board because the solder joints break up and initially cause intermittent bias issues but in the longer term fail and give instant red plating. I kept a set of tubes to show customers where the glass had softened and the vacuum pulled them into funfair prize shapes. Those were from an amp with no drift, just a bias board failure. While I was at it and for the price of a couple of trim pots I kept these in stock and replaced them while the board was removed.

[g1]

but at this point, I would say the customer, just didn't want it and there is no issue with the bias.

Without knowing the original complaint, I would say this is a good possibility. Or maybe the tech did not want to work on a TSL or was bracing the buyer for 'worst possible scenario'.

[tdlunsfo]

I was able to get the amp biased pretty equal from side to side and then let it sit for two hours. The bias moved around some the first hour, so I rebiased and it was rock steady the second hour. I told the store to resell it and he said he would keep it local just in case.

[Mick Bailey]

I've had a few of the early TSL/DSL series from 1998 - some of them regularly used, high mileage examples - that had their original unmodified boards and didn't drift. For some reason, they just survived without issue. A bit like cars that all rusted away and within 6 years until you thought there were none left, and then you came across one that's never had any issue at all and in regular use.

[tdlunsfo]

I've had similar experience with the Fender IC caps. Customer brings me a first gen 30 year old hot rod deluxe with original caps that test bang on and no leaks. A week later, an 8 year old 3rd gen comes in with a completely open 47uF. They all get changed but makes you wonder...

[Valvehead]

I've had a few of the early TSL/DSL series from 1998 - some of them regularly used, high mileage examples - that had their original unmodified boards and didn't drift. For some reason, they just survived without issue. A bit like cars that all rusted away and within 6 years until you thought there were none left, and then you came across one that's never had any issue at all and in regular use.

Ive seen many many Jcm2000 with the Drift. I think Ive had 2 or 3 come through that DIDNT have that issue. Ive got one here now i just went through and no bias drift ! I think its a 2002