Preamp Cathode resistor values

[topbrent]

Overall, what are the net tonal changes when tinkering with the cathode resistor values in the preamp?

Specifically, I am curious about the tonal changes in a AB763 Fender Super Reverb preamp (for simplicity lets use the normal channel-no reverb). 
- Stock preamp values -
- cathode resistor values are 1.5k(effective) on V1a-V1b.
- 100k plates sitting at about 250v +/-
- 22uf bypass caps

What would happen tonally and theoretically if I jumped that cathode resistor number up to 2.7k or 3.3k.
- I would be keeping those bypass caps on both triodes somewhere between 10uf and 22uf

[Fresh_Start]

I don't have the cutoff frequency formula at my fingertips, but the capacitance and resistance are both in the denominator.  Therefore, if you double the resistor value you need to halve the cap value to get the same cutoff frequency.

Raising the cathode resistor value increases the cathode voltage with respect to the control grid.  All other things being equal, that means a cooler bias and less gain.  I can't remember if that pushes the bias more toward cutoff or saturation, but it moves the bias point away from a happy medium between cutoff and saturation.  A lot of high-gain preamps seem to use cold biased stages.

I can tell you from experimenting with an AB763 that a 220K plate resistor and 2.7K cathode resistor paralleled by 4.7uf adds a little nice Tweed or touch of crunch to the tone.

BTW a 10uf or more cathode bypass cap coupled with a 1.5K resistor puts the cutoff point well below the 80 Hz +/- of open low-E.  In my limited experience, a 10uf cap will reduce some flabbiness in a Blackface circuit without really changing the tone.  At 4.7uf, you start to notice a little difference.

That's it for now,

Chip

[bigdaddy]

Webervst makes a nice little box for experimenting with that. It has a variable resistor and different caps on a switch. You can easily build one for yourself.

From my experiences it depends on the tube type and what part of the preamp that tube is in. Some times what you think you're going to get and what you get are very different. That's the whole fun of modding and building, learning how to dial an amp in for a specific sound or player and their rig.

I also have to say that the deception is great. A little mod that you think made a nice change in tone made the amp sound like crap in a bigger venue playing with a band. That goes for any mods, the difference between playing in your work area and at a gig is tremendous.

[Merlin]

What would happen tonally and theoretically if I jumped that cathode resistor number up to 2.7k or 3.3k.
- I would be keeping those bypass caps on both triodes somewhere between 10uf and 22uf

The headroom and distortion tone would change, but the frequency response would be hardly affected. Even with the resistance changes, the bypass cap is so big that it is fully effective down to sub-audio frequencies in every case.

[rzenc]

I hope this helps:
I tried to plot the conditions as per S.R.AB763  here http://www.schematicheaven.com/fenderamps/super_reverb_ab763_schem.pdf

410V of  B+ supply, 100K anode resistor - maximum current consumption: 4,1mA's.
Quiescent conditions: kinda center biased @ -2V for control grid. which drags ~1,5mA's with cathode resistor @ 1K5. which is the same current drag figures when you calculate quiescent plate conditions:
((410-260)/100K) = 1,5mA's.
Changing cathode resistor value yields changes in quiescent conditions. If t would be possible to keep current constant while changing only cathode resistor, we get different cathode voltages, which changes bias voltage for the stage with respect to control grid:
i - 2700*0,0015= 4,05V
ii - 3300*0,0015= 4,95V
However, when you change bias voltage you also change quiescent conditions. I draw 100K loadline and plotted the new quiescent conditions, which alter anode conditions and consequently cathode current, so current thru tube will change too. At -4V of control grid bias, we see 0,4mA's crossing tube, which yields quiescent anode voltage of 375V. At -5V of control grid bias, we see 0,05mA's of current and quiescent anode voltage of 395V. So you estimate how much bias you need for the stage and then check your quiescent conditions. For a given bias voltage you will have how much current can flow thru the tube and what size cathode resistor is supposed to be in order to achieve the target bias voltage.
From the interactions between amplification factor and transconductance we get anode resistance of: ((amplification factor)/(transconductance))= anode resistance = ((315-215)/(2,45-0,75))=~58K.
Considering only anode resistance (58K) and anode resistor (100K) we get Zout of 58K//100K = ((58000*100000)/(58000+100000))=~36K7 ohms.
For the effect on frequency response due to cathode capacitor:
f - -3dB frequency
Rk - cathode resistor
Ck - cathode capacitor

f = 1/(2*pi*Rk*Ck)

1K5 and 25uF: 4,24Hz
2K7 and 10uF: 5,89Hz
3K3 and 10uF: 9,64Hz

I hope this helps ans if I did some silly mistake, feel free to say what is wrong and point in the right direction, I will be glad!!

Best regards

Rzenc