Qud EL84 Screen Resistors

[gingertube]

I posted this in another forum but did'nt get much in the way of responses so I'll try here.

I did a repair on a Strauss Amp this weekend past and noted that only 2 of the quad of EL84 had screen resistors. I note that the Peavey Classic 30 schematic shows the same thing. Also, in both cases the screen resistors which are fitted are only 100R.
This seems "dodgy" to me. Surely all 4 tubes need screen resistors and 100R is way too small.
On the Strauss the 4 EL84 are all in a line and it is the centre 2 which have the screen resistors.
Is this typical?
Is there any reason for it?
Since it is the center 2 tubes which have them I thought maybe it might be a heat management thing but the 100R value would argue against that.
B+ was 420V which also seems abusive (to the tubes).

I ask because I looked on my old amp chassis "donor" shelf last night for a new project and noted I have an old Mono PA Amp which had 4 x 6BQ5 (EL84) so I have power and output trannies to suit a quad of EL84 Amp.
Tempted to build a Vintage AC30 copy.

Cheers,
Ian

[billyz]

What is the screen voltage ? Sounds like you need 7189's . The book calls for 300v max on the screens. But we all know they can do more ! Super size me.

I just adjusted the screen voltage on a MAZ jr. Could not keep tubes in there long , I did add screen resistors. They are not needed actually if the screen supply is setup right. It had a GZ34, I plugged in a 5u4 and with the screen resistors, it looks almost RCA Tube manual ish. He was complaining about the brightness. The lowered voltage browned it out and he likes the tone better too. Really did not loose too much volume either.

[Jana]

Pure speculation on my part.

1. It could be that Peavey copied from another design that was not an ideal design.

2. Why 100R on two and not all four tubes? I suspect that if there was math involved and the designer decided, because of the 400R between the plate node and the screen node, that the screens didn't need a limiting resistor. But, because of the quad of tubes, the designer also decided that there needed to be some resistance in there someplace to prevent spurious oscillations. So, I am thinking those 100R are there more as a preventative measure for oscillations, errant behavior, etc. than they are for limiting screen current.

But, this is all conjecture on my part.

[gingertube]

Screen Voltage was same as anode voltage, actually 1 or 2 volts higher at +422 Volts.
The Strauss had Chinese EL84 but with "Ruby" printed on them.
It came to me with the complaint that there was a burning smell. I traced this to the output tubes dissipating over 50 Watts per push pull pair which was almost setting fire to the sticky test labels which had been left on them. I had to modifiy the resistor above the bias pot to get some more bias voltage to get their dissipation down to 25 watts per push pull pair. Also had to measure the idle currents via the voltage drop across the measured OT primary 1/2 resistances since the %%%% did not include current monitor resistors in the cathodes. This also showed 50% imbalance between the push pull sides. THis was a brand new amp - some serious quality control issues there.
I then also changed the tubes to a matched (by me not the factory) quad of JJ EL84 which did not require as much bias voltage.

I think that the 100R on teh inside 2 tubes (one each of the push pull sides) was definitely parasitic oscillation prevention (screen grid stop function) but why not on all 4 tubes. Are they thinking that a screen grid stop on one of a parallel pair will also supress oscillation on the other? That seems like wishfull thinking
Cheers,
Ian

[Jana]

Guessing again.

I think the 100R on one tube and not the other is to slightly change the operating point of one of the tubes. Otherwise, the screens, plates, cathodes are all connecting together. I don't know, I have a weird way of visualizing this--with the situation I described, it is like the two tubes are in a balancing act all the time trying to stay happy with each other. If one tube has a "hickup", then that upsets the balance. The other tube reacts to that slight voltage variation and tries to find its new operating point. By doing so, it then upsets the other tube and on and on it goes. The 100R on one tube, it seems to me, would be a way of having the tubes at slightly different places that they like to be. I don't know if my twisted logic makes any sense. If it does, please explain it to me, lol.

In short, I think it is to stop oscillations between the tubes, not from something other than the tubes.

But I agree, why not just put 100R one all four tubes. Answer--probably to save some money. Bean counters.

[Jana]

Or, it could be that the "designer" was a forking dumbass!

[gingertube]

Well the Strauss was almost certainly copied from the Peavey.
What the Peavey was copied from I don't know.

I've also seen a case of anode stoppers on just 2 of the 4 output tubes on a HiFi Amp, although in that particular case it was the outside 2 tubes of the 4 in a row along the back.

Am wondering if they knew something I don't or as you suggest the designer was a "dumbass".

Might be time to fire up B2Spice and see what a simulation tells me!

Cheers,
Ian

[matt h]

(deleted)

[Malcolm Irving]

... --i wonder if we're looking about that as a "screen resistor" is the wrong approach, and we should instead be viewing it as a "de-coupling" resistor between the two screen grids. ...

Yes, I'm wondering if there can be an RF instability between the parallel pentodes involving: the inductance of the short length of wire joining the two screens, the internal capacitance of the screens and the voltage/current relationship of the screens. If so, one resistor would provide the necessary damping. (Individual resistors for each screen would work just as well - but if the designer was confident that one would fix the problem, then why use two?)

[Leo_Gnardo]

By tossing that resistor in there, you create an imbalance, but it becomes a more (proportionately) consistent imbalance.

"A fine natural imbalance" J. Lennon Revolution #9 Aha so that's what he was talking about?

Lets also remember lots of early 100W Marshalls had 2 tubes with and 2 tubes without screen grid R's. Perhaps this is what they were up to but I still think it's best for each output tube to have its own SG resistor. We're not trying to save a couple pennies per amp to fatten the profit margin. Not much of a splurge to "do it right."

[matt h]

(deleted)

[Colossal]

Number 9
Number 9
Number 9

Screen resistors are cheap insurance and help keep the screen voltages lower than the plate (I'm looking at you Fender....). But it probably killed Peavey's accountants to spend the $0.03 for two more and so they said no to the designer.

[JazzGuitarGimp]

The only sextet design that springs to mind is the svt. A quick look at both the 1969 and 1970 schematics reveals a 22R, 1W screen resistor at all six tubes.

[darryl_h]

You could always bypass the two 100Ω resistors, to find out what happens...

[matt h]

(deleted)