question about bypassing preamp cathodes

[iknowjohnny]

I have a build that loosely similar to a JCM as far as the 2 tube cascaded pre. But i'm not using that 10k cathode stage like they do. Anyways, i recently put 820R's on both V1A and V1B. I've tried nearly every value on earth, but for some reason this just sounds good. But i only bypass V1B because if i bypass V1A too theres too much gain and too much tonal change. Also, i chose V1B instead of A doe the bypass because it sounds bets that way. It's a .68uf and it really puts the amp's drive tone right where i want it.

That said, heres the *problem* if you want to call it that. I was told 820R is the lowest value you's ever want to use there, and that if you don't bypass it you're asking for trouble in the form of blocking distortion. Now, i dion't hear that happening, but the suggestion i'm asking for it makes me wonder if maybe it's on the verge or very slight and could be affecting the tone negitively. So i thought about bypassing it. But the problem is, i have the gain controlled via grid leaks on both stages, and they are already as low as they can go b4 the tone suffers. So bypassing V1A will cause too much gain, especially if i use a high value like 22uf, which i would have to in order not to change the tone. So the question is, how can i bypass V1A w/o changing the gain or tone? I was thinking of a 22uf in series with a resistor, but not sure what size resistor to use or whether it will even help the situation at all if the resistor is high enough value to not change the tone or gain enough to notice. Any thoughts?

[Andy Le Blanc]

One way to look at a cathode bypass cap is that it gives you a control over
degenerative feedback. Take a look at the RC table in the back of a receiving
tube manual. Read the general circuit considerations about the use of Ck.
That being said, You might want to experiment with the Rp and Rk values.
You might find it desirable to go with larger values of each to achieve the same ends.
Increasing Rp to get more gain to a point and increasing Rk to get more
bias reducing the gain to the desired operating point.

Now I know this may not give the tone you want but its away to get around
using a bypass cap, a bypass can introduce distortion. Large bypass values
are more to reduce heater hum. You can use lower values 820R, but it
remember its your source of bias, you need to pay attention to it.
820 is the "traditional" fender value, golly gee-wiz.
If you have the pre about where you want it, you can choose criticaly the
Rp values and fine tune, and do the same with Rk, scab in a pot. to explore.
You can also scab in a pot on the ground side of Ck to fiddle with it criticaly.

[markr14850]

Interesting. What is the mechanism by which a large bypass cap reduces heater hum?

[Andy Le Blanc]

I assume that the cap creates a more steady voltage condition the larger it is.
Another assumption is that the larger the value the lower the frequency it will
pass.

All the RCA figures are for DC volts applied to the heaters, as low as .5uF.
which appears contrary to practice.

[iknowjohnny]

Increasing Rp to get more gain to a point and increasing Rk to get more
bias reducing the gain to the desired operating point.

See, thats where the problem is. I've experimented with every value imaginable on both sides of V1 and nothing sounds as good as 820R on each. And a .68uf on V1B gets the tone right where i want it. The only reason i want to do anything is that someone told me that with no bypass on V1A there would be the possibility of blocking due to the feedback. Now i'm not hearing blocking, but then again i don't know if there is a small amount, just enough to cause the tone to not be as good as it otherwise would. So i figured i shoul then add a bypass there to insure i'm not getting any kind of blocking or other artifact like oscilation or such. So then i arrive at this problem....how do i add bypass w/o changing the tone or gain to any noticeable degree? Not being a tech i think in basic terms that may or may not make sense in the real world of tube circuits, and i therefore came up with the idea that MAYBE adding a large cap and resistor in series would allow a full range frequency to bypass so as not to affect the frequency response, and the resistor to allow no more than necessary to alleviate any artifacts w/o changing the gain to any noticable degree. I may not be clear enough but thats the best i can describe what i'm trying to do. If that won't do it and theres another way, i'm all ears.

[Andy Le Blanc]

You might be there, blocking distortion is usually associated with large
values of coupling caps.
You can also bypass the plate resistor with a small value ceramic or mica.
100pf or less.
Shouldn't effect the tone if its small enough. There's a mention in the wreck
pages with the universal gain stage.

[iknowjohnny]

You might be there, blocking distortion is usually associated with large
values of coupling caps.

ok, well according to what i was told, NO cap at all with a 820R cathode was a recipe for blocking. Thats the only reason i thought to do something.

I just now tried a 22uf on V1A and reduced the gain back down to approximately how it was with a split load plate. the result is that i can't really hear any notable difference. But then if there is no difference i guess i was ok as it was ! so i'll probably just go back to a 100k plate load and no bypass on V1A.

[Andy Le Blanc]

Don't agonize yourself, there does come a point where it is what it is.
If it performs to expectations you've got it. Keep your other solution in mind
if it does pop up as an issue later on.

Matchless doubled up the first stage and had the bypass as an option.
chieftain maybe. Its hard with such a simple form of electronics,
there's so many "traditionalist's" saying that the amp has to be this, or has to be that.
Its your amp, its your voicing.

[iknowjohnny]

i know, and i have to admit i DO agonize because i am a perfectionist. Not a good thing to be in a field where you aren't very knowledgable. But what drives me is that i know theres going to be particular values that will sound best to me. But between your ears changing from one minute to the next and a million other factors, it turns into a never ending process. I was happy with it for a long time, then all of a sudden i find i will be doing some gigs after 6-7 years of having quit. And when rehearsing i realized i needed to change things to get the tone to sit well in the band mix. And it now does, but i'm still trying to round off a few rough edges and along comes this guy with "if you don't have bypass....."etc etc. anyways, i think there was a movie named after the art of amp designing...."the neverending story".

[Andy Le Blanc]

Sometimes the only reason I go out is to try out an amp or a tweak.
Its nice like that, get a handle on the practical result of a question posed,
How's that gonna sound ? What if I try that? Its the best way to R&D.
After a while you begin to look a schematic and go "That'll be fine",
Strum one chord, turn up the gain and play three notes and know it's good enough.
But you have to keep playing out to know for real.

[iknowjohnny]

True, and that has always been one of my weaknesses. I would find something that sounds fab at home and try it at the gig even tho i have been thru that a million times before and i know chances are it's gonna suck. LOL! It's the one lesson i will never learn.

Anyways, i have been going back and fourth between how i had it before i posted and that tweak i mentioned with a 22uf on V1A and a split load plate to keep the gain in check. I now think it sounds better like that. Smoother and rounder. So maybe the guy was right and i was getting just enough BD to cause it to sound a bit harsh and not as round as it now does. But it will still have to pass the rehearsal test. It also has a bit less punch in the highs too, so that could translate to a sound that doesn't cut thru well. On the other hand it may balance out the harshness that often comes in a bit at gig/rehearsal volumes and be just right. time will tell.

[xk49w]

I'm still curious about how an unbypassed cathode resistor causes blocking distortion. I suppose I could see how it could happen if a stage's cathode bypass cap were to get charged up by a lot of grid current flowing then the tube's bias point would shift toward cutoff. But I am talking about a case with a bypass cap, not one without it.

[iknowjohnny]

He didn't say it like that exactly, it was in respect to a small resistor. He said with such a small cathode resistor thats not bypassed (820) you could get blocking distortion. In other words, the combination of no bypass and a very small cathode resistor.

[Merlinb]

He didn't say it like that exactly, it was in respect to a small resistor. He said with such a small cathode resistor thats not bypassed (820) you could get blocking distortion. In other words, the combination of no bypass and a very small cathode resistor.

It sounds like your friend is getting mixed up.

Blocking is made worse by adding a cathode bypass cap. Leaving a cathode resistor unbypassed actually goes a very long way to getting rid of blocking, because it is similar to adding a grid stopper equal to Rk*mu.

In any case, you should not worry about it- if you can't hear it then you don't have a problem.

[iknowjohnny]

yeah, thats kinda what i first thought. But i usually consider anyone else on these amp tech forums to be wiser than I about tube circuits. So instead of assuming he was wrong i asked others to be sure. In any case it was a lucky thing for me he said that because it caused me to change the amp in several ways and it now sounds better than it ever has. And thats amazing considering how long i have been tweaking this thing. (Read: forever....it's a sickness:)