Long-tailed Pair mods and what they do

[tubeswell]

Hi - I have seen at various times and places, different resistor values in the tail of LTPs in some schematics

The 5G9 I have recently built has a 10k and a 470R in the tail. It doesn't have an NFB.

I have seen schematics with 28k and 820R respectively. I suppose there are other variants

I gather lifting the resistor values in the tail changes the tube bias. Would increased resistor values increase the gain or just make the AC output voltage signal of the PI cleaner?

[Chuck H]

http://www.aikenamps.com/

Go to "tech info"

Go to "advanced"

Go to "long tailed pair"

Good article, and It saves me alot of explaination.

Cheers

Chuck H

[dai h.]

if you wanted to, for the 470 ohm R you could replace it with a C (reverse audio) taper pot (wired as a variable resistor) to vary the gain (of the PI thus the amp output). There is some info on this in the first Kevin O'Connor Ultimate Tone book and also Mojave Amplification uses this control (labeled "Power Dampening" IIRC).

[MattT]

To answer your specific Q: it's the cathode resistor (the 470ohm or 820ohm, etc.) that sets the bias. Larger values will have less current through the tube and a little less gain. Increasing the tail will increase the headroom a little and increase balance. With a 10K tail you pretty much NEED the 82k/100k mismatch at the plates...but with the Vox value of 47K you can just use 100k on both plates and it will likely balance fine (probably even better than the 10K tail with the 82k/100k mismatch). That's just generally speaking...NFB or lack of it will affect things too.

[gdgross]

There was a fantastic thread on here a few years ago which discussed a lot of the sonic and electrical properties imparted by various resistor changes in the LTPI. If anyone wants it, I saved a copy and I can email it to you. gdgross at gmail.