ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Bypass caps for bypass caps

:

8/17/2000 4:05 AM

Bypass caps for bypass caps

Hi,

You guys are way ahead of me. I'm still stuck on the 'cathode resistor and bypass cap' thing from a week or two ago.

Someone in there, I think Peter, mentioned bypassing a large value electrolytic with a smaller value film cap.
I experimented with resistor/capacitor combinations and I like to use 33 or 47uf caps with 1.5k cathode resistors on both channels. I guess I'm pretty used to a bottom-heavy sound.

My question?
Is the purpose of the film cap to compensate for some minor high end muffling caused by the large electrolytic? (Big dielectric)?
What value would be a good starting point? Today, I bought some .039uf film caps, but I don't know if they might be too big for this purpose. Maybe .001uf would be better. (Something tells me that it wouldn't make much of a difference in this application).

What do you think???

Thanks for the help!

Gord

8/17/2000 3:55 PM
Gus	Gord try panasonic V series for the bypass no electros. Try 1uf and 2.2 uf being film they work well. A 1uf pan v will have 1.5ohm ESR. Pan does make 2.2uf 50v V series and it is small. With a fender 1.5k and 1uf about 106hz high pass,1.5k and 2.2uf about 48hz. With smaller films the amp should stay "tighter" as you increase volume. This works like decreasing the cap into the PI.

8/17/2000 7:21 PM
Bruce	Not that it matters much, but I think your about an octave off there Gus, that is if you mean in an active circuit like a Fender preamp. Lower -3dB points: 1k5\|\|1uF in a 12XA7 with 100K plate load would be a little over 205Hz. 1k5\|\|2.2uF same tube... abt 90Hz. & 1k5\|\|22uF same tube ... abt 9Hz. Bruce

8/17/2000 7:30 PM

quote:
"Not that it matters much, but I think your about an octave off there Gus if you mean in an active circuit like a Fender preamp."

Bruce, Gus--
The method for calculating the LF cutoff involves not just the cathode resistor, but the output impedance of the cathode. You put this impedance in parallel with the cathode resistor and use that combined resistance as the R in the standard RC formula. I suspect (don't have a reference book handy) it's about 1k2 or so for a 12AX7. This would cause a difference of around an octave, as you observed....

C ya,
Dutch

8/17/2000 7:34 PM
Bruce	I think Randall A. told me when using a 12AX7 and a 100K plate load, it's about 1610 ohms Dutch. And the -3dB Fs I gave was using that 1k61 in parallel with the 1k5 ohm cathode resistor. Bruce

8/17/2000 7:44 PM

quote:
"it's about 1610 ohms Dutch."

I knew it was around there somewhere....

I guess that's the way my brain is going. The more I learn and understand about how stuff works, the less of the look-up-able data and stuff of that sort I retain....

I also suffer from the "pronoun disease", i.e., "Go get the thing over there. You know, the one sitting on the thing, with all the things on it...."

8/17/2000 9:39 PM
Gus	Bruce I don't understand. I know that the cathode plays a part in the gain but the 1uf and 1.5k are in \|\| and the total Z Of the RC should start to decrease at 1/2piRC so the gain should start to change at that point? Please name a book(s) or web site(s) Thanks.

Page 1 of 2