Of pots, grid stoppers and RFI

[krikor]

I posted the following over at head-fi, but this is probably a more appropriate venue...

I've been contemplating ditching the Crack's volume pot and using an external passive attenuator (Bottlehead Submissive or the Tortuga LDR). As a trial, I've been doing this with another preamp I have on hand by simply cranking the Crack's volume all the way up to effectively bypass the pot (zero resistance in the signal path) and use my preamp for volume control instead. However, I've noticed some rising static/noise as I reach the very end of the Crack's volume knob travel.
 
I live about a mile from 13,500 watt radio tower transmitting 6 different stations' signals. This has caused problems for me with other  headphones/amps (most notably the Schiit Fulla/AKG K7XX combo) in which I've been able to actually "tune in" a station well enough to hear it depending on how I arrange the cords and tilt my head.
 
My theory... based my limited knowledge and some white papers I read at aikenamps.com... cranking the volume all the way up drops the "grid stopper" resistance (the resistance of the pot that is in series with the 12AU7) to zero. Since grid stoppers "act as a very high frequency low-pass filter," I'm basically opening my Crack up to RFI (wait, that sounds bad).
 
Do I have these concepts correct?
 
The long and short of it... are there ideal resistor values for the input voltage divider when using the Crack amp sans volume control? I've seen it recommended elsewhere here that a simple 100k grid-to-ground resistor is all that's needed (no series grid stopper after the grid-to-ground resistor). But is this the ideal situation? Or is mine a unique situation that requires a grid stopper resistor?
 
Thanks.

[Doc B.]

The grid stopper is so simple to add that you should just do it. Try a metal film, carbon film or carbon comp of somewhere around 100 to 500 ohms. Get the body of the grid stopper resistor up close to the tube pin. Try 50K for the grid resistor. 50K rather than 100K or higher might give you a slight advantage in terms of noise rejection.

[Paul Birkeland]

You can also add some capacitance from signal hot to ground in the Crack.  It would be best at the pins, but will still be somewhat effective if you do it at the pot.  Try something like 220pF to start, and see if that helps.

-PB

[krikor]

Thanks for the help! I may have to give this a try if the issues persist and I move forward with trying out a preamp in front of the Crack.

I also now see that I had my understanding of the circuit a bit wrong. The pot is a voltage divider pure and simple. The signal goes through the wiper resistance (assuming it is set less than full open) BEFORE the portion of the wiper resistance that goes to ground. A grid stopper is located AFTER the grid-to-ground resistance.

QUESTION: if I remove the pot and, per Doc's recommendation, install a 100-500 ohm grid stopper and a 50k grid resistor, does that mean my input impedance is essentially 50K? Does the grid stopper have any effect since it is after the grid resistor? Just curious, as 100K or 50K input impedance won't make any real difference for me.

Thanks again!

[Paul Birkeland]

if I remove the pot and, per Doc's recommendation, install a 100-500 ohm grid stopper and a 50k grid resistor, does that mean my input impedance is essentially 50K?

Yes

Does the grid stopper have any effect since it is after the grid resistor?

No

[krikor]

Thanks again! I'm getting there