Is there a way to do this?

[daz]

Say you go from a 100k anode to a 220k, and your cathode is a 1.5k/.68uf. And after changing to the 220k you want to adjust the cathode R to bring the stage's gain back to where it was with the 100k. Can you determine what you would have to raise the 1.5k up to assuming you want to keep the .68uf in order to have about the same amount of gain ?? (and yes, i know the cap will create a brighter sounding stage with a bigger cathode R) Or will the cap keep you from being able to bring the gain back down to where it was without raising the cathode R much too high?

This will be much more complicated, but what if you raise the anode to 220k and want to bring not only the gain back to where it was with the cathode R adjustment, but you also want to retain the same tone? I assume the cap must be changed in value, tho i was also thinking what about the same cap but with a resistor in series limiting it's influence to about the same as it was b4?

[eschertron]

It is my humble understanding that changing the anode resistor *changes* the gain. Do you have any other motivation for the switch? Changing the cathode resistor can bring the bias point back to the same voltage level that you had before, but this is not the same thing as bringing the gain back to where it was. I recommend that you hit some websites where the theory and practice of load lines is explored. Data sheets with the specs on whichever tube you are using can be found, and you may find it enlightening to draw your current and anticipated load lines out. I'm very much a visual person and it has helped me to see exactly how gain and bias and current all relate in the tube stage.
Happy hunting!

[kleuck]

Using a 220K plate resistor in place of a 100K one will yeld to more gain, whatever you use as a cathode resistor, the only things you will change via cathode resistor change are bias and therefore the symmetry/asymmetry of the clipping to make it simple. Added to that, the harmonic distortion (before clipping) should be lower, according to the datasheets.

[Gingertube]

Gain with bypassed cathode resistor is = mu (RL/(RL+rp)

where RL is the load resistance equal to the anode load resistor in parallel with the load presented by the next stage (typically 1M)

rp is the internal tube impedance = about 65 KOhms for a 12AX7 at typical guitar amp operating points.

And mu is approximately 100

So changing the anode load resistor from 100K to 220K would (for above example) change the RL from 100K || 1M = 90K to 220K || 1M = 180K

Working thru' the gain equation that would change the gain from about 58 to about 73

Note in many high gain preamps you will sometimes see a lowish value resistor from anode to 0V (after the coupling cap), this resistor reduces the effective RL and hence the gain and is often called a "gain dumping" resistor.

(Example: check out the Trainwreck Express schematic, the Rg1 of the the 3rd stage can be anywhere between 56K and 330K, this resistor changes the Load of the 2nd stage and hence its gain which in turn determines the level of overdrive of the 3rd stage, want the overall sound a bit fatter and warmer then use the 56K, want more agressive overdrive then push that value up to 150K or in extremes even that 330K value)

If you are not concerned with frequency shaping using the cathode bypass cap then just deleteing that cap is the easiest way to drop the gain.

Cheers,
Ian

[daz]

What i'm trying to do is change the dynamic of a given stage by creating the gain it has in a different way. Same amount, but differently. IE: more gain via upping the anode and then reducing it at the cathode back down to where it was or visa versa. In other words, i don't want to change the amount of gain, only the way it's implemented. Seems to change the whole character of the amp. So i asked in order to see what values would do this. It seemed like changing a cathode bypassed stage with 100k/820R to 220k/1.5k with the same bypass had about the same gain. But i'm curious to know how close i am in that and would like to be more accurate in choosing a cathode R to get the gain close.

[kleuck]

If the cathode is bypassed, changing the resistor won't change the gain.

[daz]

If the cathode is bypassed, changing the resistor won't change the gain.

I've heard that, but it sure seems to. I suppose the way then would be to change the resistor then add a resistor in series with the cap to lower the gain it contributes.

[kleuck]

Yes, but the difference (max difference) you can achieve is about 6 db... A more useful trick imo is to split the plate resistor, you can have a 200K load with two 100K resistors, and taking the signal in the middle, have only half the amplitude, but the sound of the 200K load. Often used in old Hi-Fi stuff for line out or the like (not half the load then, but way lower)

[daz]

Yes, i've used that before but forgot about that. Sounds like the better way. Thanks.

[Chuck H]

The split plate load is a neat way to reduce gain and impedance when 1) You don't have a voltage divider between the coupling cap and the following input (could be a mixing stage, etc). 2) When you want to take two different amplitude signals from the same plate. Or 3) When you just need a lower level/impedance signal AND the stage will be run clean. I'll qualify that last one. The suseptibility to power supply noise goes up dramatically when a split load is used. If that stage is clipped it affords noise levels that can reach into the audio threshold.

I actually don't see any sense in increasing gain with a higher plate load only to dumb it back down with a split load. There are better ways to shape tone than increasing the effect of the cathode bypass cap with this sort of circuit. JMHO.

Also, daz's achematic affords post plate attenuation at each stage. Ergo, a split plate load doesn't really fit into any useful criteria. ALSO...

You can't use a split load on the stage feeding the cathode follower because it'll throw the bias out the window.

Daz... Try this: Cathode Bypass Capacitor Calculator

This is just one page. The site offers a bunch of calculators that'll tell you what you want to know if you navigate around.

[eschertron]

...and there's the question of how critical is this gain change to the next stage? By Gingertube's example, the gain has increased by 73/58, or *just shy* of 2 decibels. Is it possible that the audible effects are a result of something other than the gain change?

[daz]

Great site, i'll check it out well.
I was only thinking of this because it seems like when you switch things around like creating gain in one place and dropping it elsewhere or visa versa, sometimes you find a little gem of a tweak that is a bit of the missing puzzle. Like adding or subtracting high end....some places it works better than others. I don't know the theory like you guys do, but then again if i listened to everything here i might not try some things that end up working great. When i used a 220k anode on V2 and a 1.5k on the V2A cathode it was one of those moments. It truly transformed the amp in a good way, like that was something it needed all along. So i decided to try this kind of thing throughout the 3 gain stages and see if anything else clicks. Tried a number of things in that vein and so far looks like it's best as is. (tho i did shitcan that cap across the V1B plate) I have a new turret board, wire, and all components plus filter caps on the way. So i'm just trying a few ideas i have before they get here to minimize any alteration after they are installed.

The split plate load is a neat way to reduce gain and impedance when 1) You don't have a voltage divider between the coupling cap and the following input (could be a mixing stage, etc). 2) When you want to take two different amplitude signals from the same plate. Or 3) When you just need a lower level/impedance signal AND the stage will be run clean. I'll qualify that last one. The suseptibility to power supply noise goes up dramatically when a split load is used. If that stage is clipped it affords noise levels that can reach into the audio threshold.

I actually don't see any sense in increasing gain with a higher plate load only to dumb it back down with a split load. There are better ways to shape tone than increasing the effect of the cathode bypass cap with this sort of circuit. JMHO.

Also, daz's achematic affords post plate attenuation at each stage. Ergo, a split plate load doesn't really fit into any useful criteria. ALSO...

You can't use a split load on the stage feeding the cathode follower because it'll throw the bias out the window.

Daz... Try this: Cathode Bypass Capacitor Calculator

This is just one page. The site offers a bunch of calculators that'll tell you what you want to know if you navigate around.

[Gingertube]

When you look for tube datsheets try to find one with examples of various operating points on it. You can infer quite a bit from these.
Example: This is one of the better 12AX7A datasheets (see the bottom of page 2)
http://www.mif.pg.gda.pl/homepages/f...3/1/12AX7A.pdf
Cheers,
Ian

[daz]

When you look for tube datsheets try to find one with examples of various operating points on it. You can infer quite a bit from these.
Example: This is one of the better 12AX7A datasheets (see the bottom of page 2)
http://www.mif.pg.gda.pl/homepages/f...3/1/12AX7A.pdf
Cheers,
Ian

Thanks, but thats well over my head. I'm no tech...just ask Chuck.

[Chuck H]

Thanks, but thats well over my head. I'm no tech...just ask Chuck.

Hey now! I wouldn't know a tech if I saw one in the mirror!

FWIW I have never figured out how to follow data sheet plots. Not that it matters near as much as everyone makes of it. At least not WRT guitar amps. Half the time you'll want to be biased hot or cold anyway. Then there's the fact that once the tubes are clipping, power supply is sagging and voltages are tanking and spiking it all goes out the window. Being able to plot tube charts is a good idea if you want center bias and don't have a scope. I'm sure I could figure it out if it were ever a priority. For now, if a stage is to be biased hot or cold I decide how much so by listening. And as long as you stay within safe operating parameters that is way better than plotting. If I need a stage center biased I just use my scope and bias so that the tube clips symmetrically. Plot schmot.