choke size and choke vs resistor

[daz]

I don't so much have a particular question but i'd just like opinions on large vs small chokes and even more so, choke vs resistor and what you think the difference is in feel and tone. My final cap, a 50/50uf is due tomorrow. Already replaced the other 'lectros and when i replace the old can tomorrow i'd like to have a little better idea of what you all think about the feel and tone of big vs small chokes and choke vs resistor. I tried a 230R 15w in place of my 20H in my 6V6 marshal preamp style build and i think i prefer the resistor. Seems like the choke is really stiff feeling by comparison and i'm not so sure i get better dynamics with it. Seems the resistor sags more when i hit it hard so it actually seems more dynamic. I know 20H is huge tho and i wonder if say a 3H might give me that less stiff feel but with other advantages' Not sure what i wanna do but i know u all will get me thinking about it in ways i am not now. Heres the schematic again so u don't have to go look at the other thread if u want to see what i'm dealing with.

[daz]

Actually i DO have a question....if i decide to use a resistor, what wattage and resistance would you suggest? I have a ton of them both heat sink mount and cement in many values.

[pdf64]

The resistance of the choke in question may be significant. Early Marshall RS type 20H chokes are typically around 500ohms

[Mick Bailey]

My Laney AOR 30 (4x12AX7, 2x6V6) uses a 1k, 5W resistor instead of a choke.

[Chuck H]

One thing I've always suspected regarding chokes vs. resistors for the screen supply is that the choke can sort of act as an inductive signal load for the screens and minimize higher frequency voltage spikes and overdissipation. I'm not scholarly enough to back this up but it seems intuitive. I started to consider this after building an amp modeled after one that would typically use a choke, but I instead used a resistor and had some trouble keeping the screens from overdissipating under heavy clipping conditions.

[Mick Bailey]

Interestingly, the aforementioned Laney doesn't have screen resistors. You wouldn't expect that, but it works just fine.

[Chuck H]

Interestingly, the aforementioned Laney doesn't have screen resistors. You wouldn't expect that, but it works just fine.

Sure. I never did put a choke into the amp I mentioned. I worked it out otherwise and it now clips just fine without overdissipating the screens. It was just an observation that all things being equal a choke may provide better spike protection.

[Helmholtz]

... may provide better spike protection.

When/where do you see "screen spikes" and how do they look?
Do you mean screen voltage peaks?

I can't think of an effect that would increase the voltage at the screens in a typical amp.
At high output/clipping screen current strongly increases making screen voltage drop (slowly without screen resistors added by an immediate drop with screen resistors).
Additionally screen voltage drops by B+ sag at high output.

[Helmholtz]

My Fender Super Champ (early 80s) uses a 1.5k/2W resistor. No screen resistors.

[daz]

The resistance of the choke in question may be significant. Early Marshall RS type 20H chokes are typically around 500ohms

It's a hammond HA-193C and pecs say 181 ohms. If i recall correctly I initially bought it because an amp i was trying to copy used a 20H so there was not proper reasoning behind it and therefore i'm not sure why the resistance would matter unless someone with a lot more knowledge than me chose it. And even if that were the case, that was when it was a el34 amp. When i converted it to 6V6 i didn't change it.

[daz]

One thing I've always suspected regarding chokes vs. resistors for the screen supply is that the choke can sort of act as an inductive signal load for the screens and minimize higher frequency voltage spikes and overdissipation. I'm not scholarly enough to back this up but it seems intuitive. I started to consider this after building an amp modeled after one that would typically use a choke, but I instead used a resistor and had some trouble keeping the screens from overdissipating under heavy clipping conditions.

How would i know thats happening? Nasty sounding top end?

[Chuck H]

When do you see "screen spikes" and how do they look?

I can't think of an effect that would increase the voltage at the screens in a typical amp.
At high output/clipping screen current strongly increases making screen voltage drop.

I never saw spikes, I only suspected them. Isn't it true that it's under heavier sag conditions when clipping that plate voltage drops also? And that's when you see spikes at the plates (which I have seen). I assumed screen behavior might be analogous. But without the advantage of an inductive load unless a choke is used. Purely speculation, but I've also suspected vintage Marshalls survive better than average being cranked all the time partly because they use that whopper of a choke.

EDIT: More research is warranted but I still don't have a bench in my new digs after a year here. Plus I'd have to get my old Marshall back in shape as it's currently in a sort of half done project stage.

[Helmholtz]

High voltage spikes at the plates are caused by the (partly) inductive load when plate current abruptly changes as with clipping.
A sudden drop of current causes a flyback spike.

Screen are either terminated with screen resistors or directly connected to a filter cap whose low impedance doesn't allow for spikes to develop.

[Chuck H]

High voltage spikes at the plates are caused by the (partly) inductive load when plate current abruptly changes as with clipping.
A sudden drop of current causes a flyback spike.

Screen are either terminated with screen resistors or directly connected to a filter cap whose low impedance doesn't allow for spikes to develop.

Well chalk it up to fanciful thinking then. I guess I concocted this trying to rationalize what I consider to be poor performance in tubes seen on a few occasions. Of course the impedance of the PS cap is low enough to make it a moot consideration. I appreciate the correction.

[catalin gramada]

There is no too much reason for sag,or substantial one,as time the current draw after plate supply is pretty consistent. You may try a simply sense diode after the plates supply cap, between plate cap and screens cap nodes if you screens current is limited .