marshall jcm 900 4100 blew a fuse, which tube is opv(2 or 3)?

[georage]

[g1]

should i be thinking about using 6l6gc instead? or do these amp have to have sovtek tubes?

Either the sovtek 5881/6L6WGC or some other 6L6GC should work for this amp.
But not a 5881/6L6GB (psvane, tung-sol, NOS).

[georage]

thanks for that advice g1. if iI go with jj 6l6gc's. the data sheet lists 500v for the limit of these. 472 is really really close to that.
plus, is the jcm 900 4100 really designed for 5881s or 6l6gc's? i dont want to over current the output transformer.

i know this thread is getting old but I'll surely be asking more questions.

[georage]

as far as i can tell the two tubes that smoked are opv 3 and 4. those are the bottom pair on the schematic, they dont share a cathode ground, they share the bias resistor. If the bias was too hot, (which it was), i would expect one of the cathode pairs to go instead of a pair with shared bias resistors.
this amp has the series bias resistors which was later revised to be parallel bias resistors...im talking about r26, r25, and (undesignated 1.5k) that go to the control grid of opv4...v7 on g1's shematic
btw the schematic i provided is different from the schematic G1 provided and has different tubes designated as v4-7...g1 provided the more acurate schematic and so from now on that is the one i will refer to.

[g1]

thanks for that advice g1. if iI go with jj 6l6gc's. the data sheet lists 500v for the limit of these. 472 is really really close to that.
plus, is the jcm 900 4100 really designed for 5881s or 6l6gc's? i dont want to over current the output transformer.

The 500V is not a problem. Lots of amps run above the max. recommended plate voltage, usually they are biased fairly cool. In my experience, I find most amps that are not known tube eaters usually come out of the factory in the 50 to 60% range. The 70% you see written about is a 'do not exceed' for class AB which has somehow been misinterpreted as the target bias.
The amp was originally designed for EL34 but then modified to use the Sovtek 5881, which is not a real 5881. I would say it is more designed for a 6L6GC to stay on the safe side. You don't have to worry about the OT as it is the same as used in the EL34 version.

as far as i can tell the two tubes that smoked are opv 3 and 4. those are the bottom pair on the schematic, they dont share a cathode ground, they share the bias resistor. If the bias was too hot, (which it was), i would expect one of the cathode pairs to go instead of a pair with shared bias resistors.

In amps with more than a pair of power tubes, when one fails, it will often pull down the bias of the other tube(s) attached to it's grid, causing it to fail as well. I think that is what happened here.
Either that or the coupling cap from the PI is leaky and pulling down the bias voltage on those 2 grids.

[Greg_L]

I was thinking coupling caps leaking making the bias go beserk, but I don't know. Do these amps have conductive boards like their JCM2000 cousins?

[The Dude]

thanks for that advice g1. if iI go with jj 6l6gc's. the data sheet lists 500v for the limit of these. 472 is really really close to that.
plus, is the jcm 900 4100 really designed for 5881s or 6l6gc's? i dont want to over current the output transformer.

i know this thread is getting old but I'll surely be asking more questions.

OT current will be primarily determined by tube bias regardless of which tubes you use.