Sonic effect of changing the screen grid R from 1K to 100?

[rp]

I built AX84P1 (SE EL84 / 1-12AX7) for a friend. Works as it should but I'm not too impressed, not wild enough for me, kind of SS sounding - I'll take a 5F1 or AC4 any day! What I learned, and had suspected, is that 5W amps are throttled by a full tone stack and a MV. Sad but true.

I noticed that most vintage EL84 amps used 100R on the screen grids, this design specs 1K. Will I gain a little more vitality from the amp by going to 100R or is this a trivial change?

Since this is not my amp will it make the amp less reliable? It's not very loud either (255V plate) so it's always run flat out.

I had asked about this on the AX84 forum but never really got a good answer.

[Darkbluemurder]

Soundwise there will be a difference (lower screen resistors will make the amp a bit brighter) but not much.

For reliability it is important that the screen dissipation is not exceeded.

[Alexo]

You'll get less compression with 100r, but keep an eye on the screens and make sure they don't start glowing under heavy signal!

...so yes, 1K is a little extra insurance. NOt looking at the schematic, can't say if the extra insurance is needed. But SE amps are already generally pretty un-compressed sounding, so it's hard to say if reducing what little comp you have will help you out any.


Have you swapped tubes around? Tried different brands/countries/vintages? Sometimes these things can go a long way.

What is the negative feedback situation in that amp? Reducing NFB can liven things up sometimes.

[rp]

Have you swapped tubes around? Tried different brands/countries/vintages? Sometimes these things can go a long way.

No, I don't want to know - my buddy gets a Sovtek, the stash of Mullards, RTC, and Dynaco branded 6BQ5s (heavenly joy in a matchless) are MINE ALL MINE! But I guess I can try in secret.

What is the negative feedback situation in that amp? Reducing NFB can liven things up sometimes.

Nada. I think this thing needs 300-320V on the plates, but what do I know, and too late now.

[Cliff Schecht]

Making the screen resistors smaller will lower the miller capacitance and reduce gain in the output stage. It shouldn't change the distortion sound too much but with less gain and a lower miller capacitance you should expect a brighter output stage that clips a bit later. The extra brightness can be taken care of with a strategically placed small value cap.

[PCollen]

I noticed that most vintage EL84 amps used 100R on the screen grids, this design specs 1K. Will I gain a little more vitality from the amp by going to 100R or is this a trivial change?
.

You see quite a lot of EL84-driven amps with 100R on the screen grids, and many with no screen resistor (Valve Jr.). But the consensus I found, gathered from the Valve Jr and 18-watt forums, is to go with 1K's. I did that in my VJ head, but my Fender BJ still has the stock 100R's. When I put in the 1K in the s/e VJ, I heard no difference at all..maybe my ears need de-waxing...

BTW, both amps have around 300 Vdc on the plates, +/- 5% thanks to Florida Power and Light.

Edit 9/02/10: Corrected EL34-driven to EL84-driven.

[Alexo]

Does the screen resistor really change the miller capacitance? Are you guys thinking of the grid (g1) resistor?

...depending on your load, changing the screen resistor may not change anything, if you aren't drawing much current through it.

[Andy Le Blanc]

might thinking of the other other screen grid

Are you gonna put a pentode/triode switch in it?
put the 100r on the triode position and a 1k on the pentode.

Its a high transconductance, high sensitivity tube, G1 is 10 pf to the rest of the tube.
Some paraphase and see-saw inverters will provide the conditions for some
nasty sounding grid current. But its SE anyway.

Its more important to have a screen grid resistor in the first place.
There's some RF things that are reduced with the stopper as well as reducing
the potential excess current on G2 in some conditions.
The tone does change, smoother in the top end crispies maybe.
There's a lot of discussion about the qualitative tonal value of changing
the screen grid resistors, any value from 1k to 3.5k, get a handful of
values and give them a listen.

[Structo]

I have often heard that 6L6 tubes like 470-500R on the screens while 6V6, EL84 and EL34 tubes like 1K on the screens.
Screen resistors are known to increase tube life, possibly by making the plate more positive than the screen (not at the same potential) thereby not overloading the screen.

[brewdude]

I have a 6V6 equipped VJ, and after a post in another forum, which gave a critique of different size screen grid resistors in an SE 6V6 amp, I removed the screen grid resistor all together. The result was a much more open sounding amp, less constricted constipated sounding and more lively.

I was lead to believe that the 6V6 may be more robust than an EL84 though.

[Merlinb]

Changing from 1k down to 100R will have a negligible effect on the overdriven tone, and no effect on the treble. It will, however, put the screen grid at risk of frying.
If you want a more 'tubey' sound then a larger resistor would be the direction to go.

[Cliff Schecht]

Does the screen resistor really change the miller capacitance? Are you guys thinking of the grid (g1) resistor?

...depending on your load, changing the screen resistor may not change anything, if you aren't drawing much current through it.

Changing the resistor on G2 will change the voltage the screen sees for a given amount of power. Changing the potential between G1, G2 and the plate (think tetrode here) always has an effect on frequency response. Merlin is right though, this will not have a noticeable effect on frequency response (unless you are doing RF ).

[rp]

Andy Le Blanc: Are you gonna put a pentode/triode switch in it?
put the 100r on the triode position and a 1k on the pentode.

No, I want another pentode in there! God forbid 1/2 of one. I might as well plug into my clock radio.

Structo: while 6V6, EL84 and EL34 tubes like 1K on the screens. Screen resistors are known to increase tube life...

My 5F1 and 5E3 have no G2 resistors (as in the schematic) my Lighting just 100r. I don't know if they eat tubes, I don't gig w/them. But they all sound truly wonderfully glorious as they are. If they wish to eat tubes I will feed them tubes. Will have to see, my stash of NOS and pulls is not infinite. I think Leo would have been the last person on Earth to ship amps that would have quickly burned up and there are a lot of original champs and deluxes out there with the original tubes.

brewdude: I removed the screen grid resistor all together. The result was a much more open sounding amp, less constricted constipated sounding and more lively.

That's exactly what I'd hoped to hear, and I still want to try it on the P1, though after reading the posts here I might not bother and I'm sure gonna pay attention to Merlinb. I know it's not the lack of G2 resistors that make the deluxe and champ dynamic and wild - it's the lack of tone stack, among other things. But purely anecdotal I once owned a fawn AC10, real early with Elacs, that was totally on the verge of madness, in fact it was mad! Plug the guitar in and it would almost play itself! Was like what I figure an Express to be. (Wish I'd blueprinted it) That thing had no G2 Rs but had 100Rs in series w/ the plates - WTF! Like I said I understand enough to know that low value or lack of G2 rs by itself won't make an amp sound better but, with the exception of a Marshall, all my great amps have had no or very small value G2 Rs. Just a coincidence? That's is why I started up this post.

Merlinb: Changing from 1k down to 100R will have a negligible effect on the overdriven tone, and no effect on the treble. It will, however, put the screen grid at risk of frying. If you want a more 'tubey' sound then a larger resistor would be the direction to go.

Not looking so much for tubey which to me implies warm and full and not congested and not so even as ss. I just want more explosiveness (the kind w/o smoke) That quality that kinda scares you until you learn use to it. Even my little 5f1 has it. The p1 not a trace.

BTW how does one tell if G2 is glowing? Isn't it inside, behind the plate?

[Alexo]

Why don't you just put a ground lift switch on the tonestack if you think it's what's throttling the tone?

...and if you want another pentode, you can swap the 12AX7 with a 6U8 or some such triode-pentode combination tube. The AX84 guys were pretty into that last I checked (couple years ago). Just be sure that the tone stack follows the triode and not the pentode.

[Merlinb]

with the exception of a Marshall, all my great amps have had no or very small value G2 Rs. Just a coincidence?

Yes, coincidence. Of all the differences between those amps, the screen resistor is a puny contributor. You're comparing F1s with station wagons by comparing their tail pipes.