JCM 2000 bias question.

[catstrat]

I have in my shop a Marshall JCM 2000 that has the drifting bias problem, or so i thought. These have always had this problem or some of them do and i've worked quite a few of these. I did notice something that i was un-aware of. Almost everytime when i try to set the bias using the prongs on the back, you will see the bias start to drift after 15 minutes or so and it's almost always the right pair of tubes as you're looking at the amps back. I decided to use a bias probe instead and after the bias was set, it held. I pretty much concluded that it was the bias adjustment/bias monitoring circuit giving false readings. Does anyone have any insight on this? My next question, if the output tubes start drawing excessive plate amperage will that affect the output tube grid voltage? I'm asking because i've never really gave it any thought. Thanks to all.

[Sowhat]

Sounds like you have the old Marshall bias drift problem... for more on this see this thread: http://music-electronics-forum.com/t29893/

[Enzo]

Or you could possibly be seeing the start of C46 failing. That is the little disc ceramic between pins 3 and 4 of only the end tube by the power transformer.

One thing about the little bias pins. Are you aware that looking at the back of the amap and those pins, the pin on the left is for the tubes on the right, while the right end pin controls the tubes on the left. I have no idea why they didn't just turn the thing over, but they are all "backwards." So don;t adjust the wrong side. Due to B+ loading, the two sides are somewhat interactive, so adjusting the "wrong" one will still affect the other side tubes some.

[Sowhat]

Yeah, C46. In my extreme fix example C46 was shot as well... a dead short! that particular TSL100 I repaired really had some awful problems... like Enzo says... left is right and right is left. As I remember there were several cold solders on some of the spade lugs as well. A lot of little strange things for what otherwise is a very nice amp.

[catstrat]

jcm 2000 bias.

Thanks everyone. The question i had was does the bias votage at pin 5 change as the plate amps go up. Yep, it does. We pretty much know that as it heats up there is voltage bleed from one tracing to another throwing the bias voltage out. That is my understanding. Am i right? This is the old style board. I did notice, as i spray freeze spray across the board the amp draw went down. I removed that disc cap from the circuit and the problem is still there. I'm going to use a heat gun set to cool and see if that affects it. It seems to me they could have laid out the trim pots to make a little more sense. What do you expect from a country that drives on the wrong side of the road. LOL.. I'll do some more test and get back with you. Thanks.

[Sowhat]

Yeah, that's my experience with old DSL/TSL boards. Although the drift thing is most noticeable with the bias circuit this does not mean that it is isolated to that circuit alone... the whole board is thermally flakey. They fixed that problem with later revision boards but for those early revisions they are either bad or in the process of going bad. This is why I sometimes just relocated the entire bias circuit (low voltage) on a separate good G10 glass epoxy board all together which is thermally isolated from the main board. This does not completely fix the problem but it solves the bias drift problem which is the most dangerous aspect of this failure. I have not heard the exact scientific reason that causes this but I think it must be with the foil bonding adhesive or soldermask used in those old boards. As for that C46, you don't really need that thing but it's there to siphon off any high frequency noise that may be present on the output section. Those things are cheap and abundant so replacing in a known good one can't hurt but leaving it out don't seem to hurt either. The problem with these small filter caps on the high voltage section of the output is that if they go short, you got some real problems, not only in these Marshall's but in other amps as well... Sunn's come to mind because with their lack of circuit protection, you don't just blow an internal fuse as in the Marshall, you get smoke and fire for your entertainment as associated parts vaporize!

[catstrat]

jcm 2000 bias.

Thanks Sowhat. I was afraid you were going to say these boards are marginal at best. I don't understand nor can i explain the reason for the bias drift either on these, but we do know it's heat related. I was doing some more test on it this afternoon and when i started, the plate draw was at 17 ma. Within 30 minutes ,it had risen to about 40 ma. I decided to take my fan and directed the air flow into the bottom of the chassis with the amp up-side down, and it did cause the bias drift to come back down to an acceptable range and it didn't drift as much , but that's not fixing the problem. The drift problem is still there. So i'm going to give the customer 4 options. We can find a replacement new board or i can do the bias circuit mod you're talking about or he can pay me for the time i have in it and he can sell the amp asap or we can rig up a cooling fan to blow air over the board , and that's not a fix at all, that's just a bandaid. He just bought this amp and i hate it that's he's having problems already but it is what it is. Thanks everyone for all your help.

[catstrat]

I owe a big thanks to SoWhat for posting the link that led me to the first fix for the bias problem i,ve seen. Usually you just replaced the board. This fix involves modding the board. I removed the o/p tube sockets and used a countersink drill bit and drilled the hole bigger where pin #5 comes through the board. Now when you re-solder the sockets in place, number 5 pin is totally isolated from the board. Then you change the 220k grid stoppers to 5.6 k and you solder 1 end of the grid stopper directly to the #5 pin and the other to the board where it would normally go. I changed a few other things but this mod does work. So a big thanks to Sowhat!!

[dehughes]

Interesting thread. Been a while since I've been here, but I'm back, because...

I just picked a circa 2001 DSL 50 head and it does not seem to manifest the thermal drift issue, but one thing I notice is that the bias setup seems odd. Is it really the case that the bias adjustment pots are "reversed" in that the left one adjusts the right tube, and vice versa? This is the first place I've heard this mentioned...

[Leo_Gnardo]

Is it really the case that the bias adjustment pots are "reversed" in that the left one adjusts the right tube, and vice versa? This is the first place I've heard this mentioned...

Correctamundo. Why Marshall does this way, another mystery of the universe.

[dehughes]

Well good to know. Thanks.

One other thing I noticed is that I could get my EL34B STR pair to bias up around 30mv, yet when I installed a pair of JJ EL34s they would only bias up to 18mv. Is there a way to get more "throw" out of the bias pots, or even just re-wire it to a standard single-pot setup, so I can bias up a fuller range of EL34s?

[The Dude]

Before you go looking into modding anything, pull that little bias board out and check all of the connections, or better yet, just resolder everything- it's small. Those boards often have bad solder connections.

[dehughes]

Okay. I may do that. I didn't see anything amiss when I examined the boards, only noticed I wasn't able to bias up the tubes as I'd like them. I'll pull it out again next week and give it a look.

[catstrat]

It seems, at least from experience , the tsl 2000s 100 watts are the one's with the bias drift. I've never seen this problem with the 50 watt heads. I don't know why Marshall has the bias trim pots reversed from the tube location. Maybe they had a cup of tainted tea that morning? Some people do just put a single bias trim pot in place of the two trim pots. I'm trying to remember what amp it was but it was a 100 watt with a bias pot for each tube. I'll remember it sooner or later.

[TimmyP1955]

The Marshall TSL122 TSL100 thermal bias drift repair page