Can someone please explain this contraption?

[Greg_L]

I found this in a Marshall 1959 SLP....it's an older Plexi reissue before they had fx loops.



That's hanging off the PI node filter cap. The green wire - as far as I can tell - is the filament center tap. And where it's connected at the junction between the two resistors sits 50vdc. I can't tell the capacitor value underneath but it looks like maybe 40uf.

Is this some sort of elevated heater ground reference gizmo?

[Enzo]

SOunds like that is what it is from here.

[Leo_Gnardo]

LOOKS like that's what it is from here. Enzo is gifted with long distance hearing, one of his amazing qualities. You go guy!!!

[J M Fahey]

Same, and very much doubt it´s original.

[Greg_L]

Yeah I know it isn't original. It's not on any schematics. I just wasn't sure what it was at first glance. The amp works fine and the heaters are dead-on 6.3vac so I'll just leave it. Thanks for the replies!

[Chuck H]

Most likely an attempt to reduce the cathode to heater differential at the cathode follower that was causing problems in Marshall amps with some Rusky preamp tubes. Indeed, leave it there. It looks like it was implemented well enough and it won't hurt anything.

A peripheral observation... Many old Fender amps have a tremolo circuit that elevated the oscillator triode cathode with a 100k or even a 470k resistor (the 470k resulting in 200V or more on the cathode!) But I don't remember ever hearing about any problems with these amps. So why was it a problem with the Marshall CF?

[Leo_Gnardo]

A peripheral observation... Many old Fender amps have a tremolo circuit that elevated the oscillator triode cathode with a 100k or even a 470k resistor (the 470k resulting in 200V or more on the cathode!) But I don't remember ever hearing about any problems with these amps. So why was it a problem with the Marshall CF?

If memory serves, voltage on the Fender trem tubes' cathode, 100K to ground, sits at about 20V with trem off and oscillates around that value with trem on. Plus, that tube - theoretically doesn't affect the audio. Though in some rare cases a defective tube in trem osc function has definitely raised hob with the audio in some repairs I've tackled.

The venerable 5F6A preamp, copied by Marshall and so many others, EQ drive stage also has 100K cathode resistor. In this case I often see 100V, sometimes up to 140-150V. That's certainly enough to cause distress to a tube that can't handle it. Floating filament on DC can certainly help those tubes survive, also suppresses noises that happen when a high cathode voltage starts leaking randomly to filaments - pop crackle hiss. Side note, many amps with this feature employ a half watt resistor in that Rk function, barely adequate & often the resistor roasts or fails entirely - in spite of being theoretically OK. 1 watt resistor improves the situation.

FWIW some 60's Ampegs floated filament on a 0.1 uF film cap, no DC voltage applied. In some highly modded (extra gain for lead channel) amps I've tried that with mixed results. Sometimes more uF are needed and sometimes it doesn't help much at all.

[Chuck H]

If memory serves, voltage on the Fender trem tubes' cathode, 100K to ground, sits at about 20V with trem off and oscillates around that value with trem on.

Thanks. I hadn't considered the difference with the bug trem. Makes sense. But...

The custom shop Vibrolux Reverb uses a 100k with a non bug trem and does have 156V on the cathode with no filament elevation. The vintage 6g11 Vibrolux used a 470k. Schematic voltage indicated at 200V. I do think there were a couple of other Fender amps that used the 470k cathode oscillator as well. But even with this I guess it's not enough amps for any problems to become noteworthy compared to how many high voltage Marshalls with a CF were made.

[g1]

Many old Fender amps have a tremolo circuit that elevated the oscillator triode cathode with a 100k or even a 470k resistor (the 470k resulting in 200V or more on the cathode!) But I don't remember ever hearing about any problems with these amps. So why was it a problem with the Marshall CF?

I seem to recall nickb's research on the CF tube problems pointing at an issue that was current related (turn on surge?), maybe that factors in here.

[Helmholtz]

I seem to recall nickb's research on the CF tube problems pointing at an issue that was current related (turn on surge?), maybe that factors in here.

AFAIR, the problem with the CF Nick analyzed was not related to high heater-cathode voltage. He even did dedicated tests to exclude that possibility.
Rather he found that the issue was that the grid of the direct coupled CF is at full supply voltage before the heater has reached its final temperature, causing detrimental tube current with the cathode not being hot enough (i.e. no protective electron cloud).

https://music-electronics-forum.com/...-tweaks/47283-

[Greg_L]

So then is this something that is simply remedied by using a standby switch?

[Helmholtz]

So then is this something that is simply remedied by using a standby switch?

Yes, B+ should be applied only after the DCCF is fully heated.

Nevertheless, high Vgk can cause tube degradation, but that seems to be more of a long term effect.

[Chuck H]

Ok... Standby is fine with me since I use it as a matter of course after many years of conditioning. But there are also MANY threads on the forum that indicate that a standby switch is sort of a vestigial appendage on guitar amps and there's really no need to use it. So... Assuming there are players out there that AREN'T using their standby switch, there is a FET circuit indicated in the thread Helmholtz linked that should solve the problem. Though I agree with the following post on that thread that a more quick and dirty solution that's easier to implement would be better. I just don't have one at this time. But I'm thinking that something like series Zeners at the CF grid to dump excess voltage at startup, but valued such that they don't affect normal operation might be possible.?. I haven't played with it yet. To be honest I don't have the CAD savvy to simulate all the circumstances well enough to test this.

[Greg_L]

Ok... Standby is fine with me since I use it as a matter of course after many years of conditioning. But there are also MANY threads on the forum that indicate that a standby switch is sort of a vestigial appendage on guitar amps and there's really no need to use it. So... Assuming there are players out there that AREN'T using their standby switch, there is a FET circuit indicated in the thread Helmholtz linked that should solve the problem. Though I agree with the following post on that thread that a more quick and dirty solution that's easier to implement would be better. I just don't have one at this time. But I'm thinking that something like series Zeners at the CF grid to dump excess voltage at startup, but valued such that they don't affect normal operation might be possible.?. I haven't played with it yet. To be honest I don't have the CAD savvy to simulate all the circumstances well enough to test this.

I also use standby as just a standard course of action. I've seen multiple debates on it's it's pros and cons. It's just to me, it's such a simple effortless thing to do, why not have and use a standby? Especially on solid state rectified high power amps with near 500vdc on the B+ rail. I'd like my tubes to be ready for that blast of DC. I record and gig a lot. It's nothing to flip the amp on while you get other stuff ready. And when you're ready, flip it out of standby and the tubes are all nice and warm and ready to make noise.

[Greg_L]

I found a few more little weird things on this amp last night...not too weird but I'm wondering why someone did this.

There were 250pf silver mica caps going from grid to ground (pin 1) on each output tube. In the process of removing these things I noticed a gigantic blob of solder on pin 8 of one of the tube sockets. While removing this blob the pin just fell off. Broken. Pin 8 on this tube socket was held together with solder. Ugh. New socket ordered.

Also someone switched the .022 PI coupling caps to .01. Don't know what that's all about. Make the amp brighter?

And finally there is a .022 cap to ground off the ground side of the input jacks. I'm guessing this is to bleed off RF interference.