EL34 and 6550 pins 1 and 8

[Randall]

He only chose EL34s because it needed tubes, and they are less than the pricey Sovtek 6550s. Is leaving the 150Ks in there really going to make that much difference if I have them biased where I want them? He said it's had 6550s in it since 1983, so there is going to be a difference I'm sure. Convince me the grid resistors need to be changed to 220K. He has been playing it with failing filter caps and worn tubes for a while now.

[Helmholtz]

Convince me the grid resistors need to be changed to 220K.

Why this?

The EL34s are fine with any value below 500k. Higher values allow for some more PI gain and power tube drive.

But I'm not your customer.

[Leo_Gnardo]

Convince me the grid resistors need to be changed to 220K. He has been playing it with failing filter caps and worn tubes for a while now.

If I was you, I'd be "lazy" (haven't I done enough already? one more prying up of that board is gonna drive me kooky!) and leave the 150K's right where they are. Customer is bound to notice an improvement, what with new output tubes and fresh filter caps.

I know you're not lazy for real. So write up the bill & I hope you collect compliments as well as cash.

Next time you run across a similar decision point, you're better informed what to do. All to the better I'm sure.

[g1]

I think you are fine the way you have done it.
My concern was more about the bias feed resistor than the grid leak resistors. I was trying to find the answer for why you would have to reduce the 47K (in series with bias trimmer) when the schematic shows a 56K there.
The only reason I could think was that it must have the 150K bias feed instead of the 220K. The other 220K's (grid leaks) were just incidental.

edit: link to schematic with 6550 values:
https://drtube.com/schematics/marshall/jcm800pw.gif

[Chuck H]

FWIW I would expect the el34's to break up a little sooner than the 6550's even if the 150k bias resistors are left in place. Though I couldn't say about those worn 6550's I suppose. But that's not Helmhotz point. He's just considering maximizing the circuit for the new tube type and what would be expected from it by many guitar players. That's not a bad thing and if the amps still guts up then the change wouldn't take long. Better? I dunno.?. I'd leave it with the 150k's under the circumstances. If it were MY amp and an hourly rate wasn't on the line I'd probably try it with 220k's and see what I thought.

[vintagekiki]

I was trying to find the answer for why you would have to reduce the 47K (in series with bias trimmer) when the schematic shows a 56K there.

With 56k (with series bias trimmer at min) bias voltage is high, the quiet current is low.
With 47K (with series bias trimmer at min) bias voltage is lower, the quiet current is bigger.

[Randall]

Yes, under the circumstances, I think this is splitting hairs, and I am going to leave it as is. I have a good feeling all is going to be just fine. Thanks boys!

On to the next!

[drewl]

FWIW, I have a 73 MKII 50Watt that had the 150K resistors as it probably had 6550's originally.

Someone had already installed EL34's when I got it in 1988.

I changed them to 220K, and changed the bias resistor years ago to give the classic Plexi sound, and it sounds great.

[Randall]

So the question here remains, did it NOT have "the classic plexi sound" when it had the 150Ks, and if so was it those resistors or the bias resistor that made the difference? Do we really know?

[Helmholtz]

So the question here remains, did it NOT have "the classic plexi sound" when it had the 150Ks, and if so was it those resistors or the bias resistor that made the difference? Do we really know?

Bias resistors are not in the signal path. No signal across them. C19 shunts any signal to ground, making the center point between the grid leaks a signal (AC) ground.

[vintagekiki]

So the question here remains, did it NOT have "the classic plexi sound" when it had the 150Ks, and if so was it those resistors or the bias resistor that made the difference? Do we really know?

For "the classic plexi sound" replace C10/470p with capacitor 250p, or in series with C10/470p on treble potentiometer solder capacitor 470p.

[drewl]

So the question here remains, did it NOT have "the classic plexi sound" when it had the 150Ks, and if so was it those resistors or the bias resistor that made the difference? Do we really know?

I'm sure there are some out there that say they can hear the difference !
Not me!

Kind of an incomplete sentence on my part.
I did some other tweaks and set it up like a '69, just to get it as close as possible.

It always sounded great

[vintagekiki]

What defines the Marshall "plexi" sound?

We started from "EL34 and 6550 pins 1 and 8" so by changing the extremely irrelevant resistors that do not participate in the sound formation, we are lost around plexi sound.

Question
What defines the Marshall "plexi" sound?

[Chuck H]

Wait... What? How is the bias resistor value "extremely irrelevant"? And the statement "not in the signal path" is also misleading. For all intents and purposes this is primarily a load resistor. So it's value is not irrelevant and whether it's in series with the signal or not it is certainly having an affect on gain, impedance and time constant. In my opinion these could certainly affect an amps ability to achieve the "plexi sound". Or ANY specific sound for that matter. These are operating conditions that are affected by that components value in the circuit.

This is mostly a devils advocate argument. I don't actually have a dog in this fight.

[Randall]

"This is mostly a devils advocate argument. I don't actually have a dog in this fight. "

Me neither. Talk among yourselves, I'm out. This ain't a plexi, and he didn't ask for it to "have that plexi sound". He wanted it serviced. Case closed, good night and thank you.