Could someone point me to a place on line covering grid stoppers and how to choose value for a certian cutoff. I want to use them to roll off high end but do not fully understand the process. Thanks.
I believe the concept is to counter feedback before the tube by shunting such HF against the miller C.
Liberal use of plate zobels can quash things as well.
Tim
Liberal use of plate zobels can quash things as well.
Tim
Thats what I want to learn, How to pick the correct resistance to work with the miller capacatance to make an RC filter to roll off high frequencys above 20Khz or so.
Aha! Well, if you want grid stoppers to do that, it's just a simple use of the formula f(-3dB) = 1/2piRC, where R is total source resistance (output resistance of driving stage plus grid stopper) and C is total load resistance (grid capacitance plus stray plus Miller capacitance).
I've done some pretty scary breadboard layouts and found that a 1K resistor as close to the tube socket as possible is a pretty good fix-all 😉
Using grid stoppers to cut down the treble might not be such a splendid idea (except maybe in guitar amps), as you would have to use a very big value, in most cases, to get any significant effect (I'm not talking about the subtle changes you might get with different smaller values...). Too much resistance in series with the grid will result in distortion due to modulation of the tiny grid current variations over a cycle; Morgan Jones cites ca 60k (if I remember correctly) as the value where he started to get a measurable increase in distortion (with the valve he used - may have been a 6S45P?).
I think an rc filter before the grid (and, preferably, a grid stopper) would be a better idea (except as noted in guitar amps, where the added distortion due to a huge grid stopper (esp. in overdrive) might be interesting.
Anyway - good luck; whatever works works. 🙂
I think an rc filter before the grid (and, preferably, a grid stopper) would be a better idea (except as noted in guitar amps, where the added distortion due to a huge grid stopper (esp. in overdrive) might be interesting.
Anyway - good luck; whatever works works. 🙂
The grid stop resistors are not obligatory but are nice anti-oscillating/parasitic solution in some circuits.
About 300 ohm decent ones (Dale RN, Riken) are ok.
About 300 ohm decent ones (Dale RN, Riken) are ok.
Grid-stopper values are very much "suck it and see." They aim to damp the resonant circuit formed by valve capacitances and inductances in the grid/cathode loop. The main thing is that the resistor body should be as close to the valve pin as possible. Carbon resistors are a good idea because the higher resistivity of carbon means fewer turns are needed and they have lower inductance. In general, valves with higher mutual conductance are more likely to oscillate, so directly heated triodes are almost immune, but frame-grid valves designed for UHF use (ECC88, 6C45 etc) are more susceptible. Good RF layout (short wires, ground planes etc) reduce the problem. With really good layout, it's possible to get the grid-stopper down to <100 Ohm, but Mullard used between 1k and 4k7 for EL34 and EL84 power pentodes in ultra-linear mode. Once you start paring the value down, you need a good oscilloscope to monitor the anode of the valve concerned whilst provoking it with a square wave to make sure that it's not oscillating.
Most people just use a larger value than strictly necessary in the knowledge that if it's too big it will definitely damp oscillation.
Most people just use a larger value than strictly necessary in the knowledge that if it's too big it will definitely damp oscillation.
You can say that again. I never take design rules for granted and have often found i prefer the sound without a grid-stopper. No idea if this is simply because another lousy resistor is not in circuit or the grid prefers seeing lower driving impedance but even on problematic valves like 6S45P i'd first make sure that grid-stopper is really needed.
Thanks for your help. I may look at a different way then. I,m just hearing some sort of distortion with some sibilance and real high picalo flute parts.
Tim Wyatt said:
Can you 'scope it?
This sounds more like HF parasitic oscillations not on the grid, but plate, causing your iron to saturate. Perhaps a NFB foobar?
Yes I can scope it. The negative feedback was acting weird, with a cap in parallel with the feedback resistor it won't work. I switched the output transformer wires on the power tubes and got that great screech, switched them back. I have gone over the wiring in this area and double checked the schematic and all looks well. the only way it will work at all is to take that cap out. What do you think it is?
P.S. Sounds great exept for this
P.S. Sounds great exept for this
Something is fishy in your NFB, or the pre-tube it's fed back to 
What size cap did you use and what is the FB resistor value?
What size cap did you use and what is the FB resistor value?
Let's see a 1, 5 or 10kHz (whichever gets the best detail on the ringing) squarewave scope pic, with and without Cnfb, please. 🙂
Tim
Tim
Hi,
He, he....You really have to live that far away, don't you?
As an experiment after reading a GA issue many years ago stating without much further context that gridstoppers would improve "definition", I once added gridresistors where none were really needed.
The result was shocking: less "definition" and a dull and sluggish sound...
Hi-hats almost sounded like wood blocks, yikes........
Just like screened interconnects, if you don't need them don't use them.
If you know how to layout your stuff and select your valves you'll hardly suffer from hum or wild oscillations.
If you nonetheless should, just use the smallest value resistor that does the job. Preferably carbon comp with the body of the resistor soldered as close to the grid pin as you can manage.
If you follow that and a dozen other guidelines you'll start to understand what the high-end mags mean when they talk about "air" around the instruments.
In fact you'll even be able to hear the acoustics of the recording venue and occasionally even be able to tell to microphone capsules apart; that and nothing else is what I'd call "definition".
Cheers,😉
He, he....You really have to live that far away, don't you?
As an experiment after reading a GA issue many years ago stating without much further context that gridstoppers would improve "definition", I once added gridresistors where none were really needed.
The result was shocking: less "definition" and a dull and sluggish sound...
Hi-hats almost sounded like wood blocks, yikes........
Just like screened interconnects, if you don't need them don't use them.
If you know how to layout your stuff and select your valves you'll hardly suffer from hum or wild oscillations.
If you nonetheless should, just use the smallest value resistor that does the job. Preferably carbon comp with the body of the resistor soldered as close to the grid pin as you can manage.
If you follow that and a dozen other guidelines you'll start to understand what the high-end mags mean when they talk about "air" around the instruments.
In fact you'll even be able to hear the acoustics of the recording venue and occasionally even be able to tell to microphone capsules apart; that and nothing else is what I'd call "definition".
Cheers,😉
- Status
- Not open for further replies.