Volume/Gain pot values?

[yunger]

why are interstage volume/gain pots more often 1M? the amount of signal passing through the pot is based on the ratio of the voltage divider, right? at low volume/gain, the series resistance of the pot, the possible grid resistor and the internal capacitance of the following tube makes a significant low pass filter. guess that's why there are often bypass caps on a 1M pot.
the reason i ask is, i am using a 250k pot on my amp and at anything after about 50/50 ratio is enough to drive the following triode crazy. so what's the point in a 1M....... in for example, a marshall 1959?

[tubeswell]

so what's the point in a 1M....... in for example, a marshall 1959?

Higher input impedance for that (following) stage.

[yunger]

Higher input impedance for that (following) stage.

please excuse me while i expose my ignorance for a moment.....
so a 1M would allow the signal to hit the following stage harder, earlier on the dial vs. a 250k? does this increase the likelyhood of greater hum/noise in the amp?

[daz]

Noise doesn't seem to be part of the equation. I've been experimenting with 1M vs 500 and 250k and so far no big difference as far as noise or anything else for that matter. I'm sticking with 1M.

[yunger]

i found this chart showing loading effects on gain in a 12ax7 on the valve wizard site. so a 1M pot will pass more gain at full rotation vs a 250k or 500k.
in my case, a 250k at full CW rotation sounds really ugly. i wouldn't want to broaden the range of that uglyness by reducing the load.

[pdf64]

Bare in mind that the vol control forms a high pass filter in conjunction with the capacitor from the preceding plate that feeds it - if you reduce the pot value, the cap would have to be increased by the same ratio, in order to retain the same low end rolloff frequency.
The ugly sound described is likely to be blocking distortion. That can be reduced by rolling off the low end at a higher frequency, eg reducing the value of the vol pot. Adding a grid stopper resistor between the vol pot wiper and the next grid can help also (100k-470k).
daz put forward a great idea IMO, of putting a 100k resistor load on a 1M pot - as the pot is turned up, it's effective value reduces and so the low end rolloff frequency moves up (ie ~90k on full), thereby keeping blocking distortion and muddy tone at bay.

[daz]

As much as i'd love to take credit for that, i am blind to the theory you describe and ignorantly put that 100k there just to lower gain ! But it certainly does help with low end roll-off too. However, i no longer have that there because i've changed the preamp so much it doesn't need it anymore.

[tubeswell]

please excuse me while i expose my ignorance for a moment.....
so a 1M would allow the signal to hit the following stage harder, earlier on the dial vs. a 250k?

All other things being equal yes. Generally the aim of impedance bridging is preserve as much of the signal as possible between stages. Rule of thumb if you want the signal strength preserved is to have the output impedance of the previous stage to be less than 1/5th of the input impedance of the following stage (up to a point*) for a good signal. If you lower the input impedance of the following stage, then you dump more signal to ground. So the ideal size if the grid load resistor really depends on what the output impedance is of the previous stage (and that depends on what the plate resistance is and the plate resistor as well as the following stages' grid load resistor - the whole thing forms a voltage divider).

* if the grid load resistor is too big, then the stage will be noisier, and if it is miles too big, then it won't allow grid current to bleed away.

[Guitarist]

It seems that , to my ears, having a pot, not at full, often worsens the sound due to "drag". This is why amps usually sound better at 10. Well, and...

Adding a CF prior allows you to use a smaller pot, eg. 100K or 250K. Now you don't need a brightening 270+pF cap. The control moves more transparent but yet does what you want. I can't recall doing this specifically on a JCM800 for instance, but I'm pretty sure more players would love the change and spend much more time around 2 or 3 on the preamp. 10 would also slay.
In the stock case, the 1 Meg preamp control acts more like "a tube worn out" simulator

[wizard333]

putting a 100k resistor load on a 1M pot - as the pot is turned up, it's effective value reduces and so the low end rolloff frequency moves up (ie ~90k on full), thereby keeping blocking distortion and muddy tone at bay.

What? Schematic?

Adding a CF prior allows you to use a smaller pot, eg. 100K or 250K. Now you don't need a brightening 270+pF cap. The control moves more transparent but yet does what you want. I can't recall doing this specifically on a JCM800 for instance, but I'm pretty sure more players would love the change and spend much more time around 2 or 3 on the preamp. 10 would also slay.

What? Schematic?

[pdf64]

I searched back through Daz's postings and found
http://music-electronics-forum.com/t15227/
Message #1 shows the schematic I was thinking of, which I hope makes things clearer.

Re the CF thing, my take on it is, if R31 (between V3 and V4 on the high gain pre-amp) was replaced with a 100k pot in this Marshall 6100 schematic.
http://www.schematicheaven.com/marsh..._6100_100w.pdf

[wizard333]

That makes it clear, thanks!

[stevesamps]

Uneven voltage dividers can be good

The 250k thing or even 100k pots can be used as a
less component way of dumping signal to ground on high
gain amps,

If you are just cascading stage after stage you see a lot of that
like a 1m in series and 100k to ground, although I do
see a lot of designs where they seem to be more "concerned with
supporting the signal with a higher Z.

I am not a high gain builder per say, but I end up
dumping lots of signal to ground between stages on purpose
when I do cascade them,

I guess it helps to build harmonic distortion by slowly building up the
gain over stages.. but it also helps keep you from slamming the grid and
getting ugly blocking distortion

anyway my 2 cents

steve