cathode R value

[daz]

I've already tried the latter option.

[Chuck H]

You know what? Your at the point where the limitations of simple tube amp circuitry is showing it's failings. This is where other external devices can often help. As many players in the past have already learned. I think an attenuator might be in your future. As a way to keep your amp closer to it's "sweet spot" rather than relying on the master volume. You may or may not know that I and others have some pretty good attenuator circuits available here in forum threads. I can't remember if your amp is a 100W or 50W model. But I'm sure I can draw up an attenuator circuit that you would enjoy. A lot.

[daz]

You know what? Your at the point where the limitations of simple tube amp circuitry is showing it's failings. This is where other external devices can often help. As many players in the past have already learned. I think an attenuator might be in your future. As a way to keep your amp closer to it's "sweet spot" rather than relying on the master volume. You may or may not know that I and others have some pretty good attenuator circuits available here in forum threads. I can't remember if your amp is a 100W or 50W model. But I'm sure I can draw up an attenuator circuit that you would enjoy. A lot.

Thanks Chuck. Maybe, but not really sure if thats the way i wanna go. I have often thought of it. (it's 50w by the way) But i have a belief that i CAN get more of what i want as is. For example, i have had better tone in certain aspects than i do now. It's more than just wanting that, it's also the fun of the chase like i was saying. I want to play with the higher voltages for a while anyways. If i can't find a way to retain the desirable aspects it gives while eliminating the bad i will go back to how it was. Then maybe i'll hit you up for the attenuator schematic.

[daz]

Major find here. I suspected the negative aspects of the tone associated with the higher voltages was happening at V2 or PI. So i tapped off the B+ before the PI and sent it to V1 via a 2.7k dropper and a 22uf filter. The tone is much better that way. Better than it was with higher voltages throughout the preamp that is.....I still have to determine if i prefer it over the old PSU with 10k droppers. But it no longer has that bright harsh top with high voltages all around, and it pretty much has the best aspects of both ways. ALMOST. So more testing. But this is a whole new ball game now. Oh, and V1 plates are way up there at around 275 VDC ! Isn't 300 max for a 12AX7? Should i drop it a bit lower to be safe?

[Chuck H]

Well I don't know what your rail droppers are, but...

If a standard Marshall is: rectifier to plates, choke to screens, 20k to PI, 10k to V2, 10k to V1 then...

You could always do something like: rectifier to plates, choke to screens, 20k to PI, 4.7k to V2, 2.2k to V1.

This is just an example, but it demonstrates that you can still work with the rail and increase preamp volts without changing the PI volts. And 275Vp at V1 is high. Some Fenders ran that way but it's bound to be too stringent and harsh for a Marshall type circuit. So something like the example would give you a bump without going so high.

[daz]

Well, what i have always found is the CF at V2 does not like voltage or filtering increases at all. So i can't really do that. But tapping the B+ from the same source at the PI before the PI's node works fine. It's just a matter of whether i can find a way to fatten and smooth the tone so it's the same as it was before the PSU change. In other words, retain the higher voltage's harmonic complexity and dynamics with get the same degree of smoothness and fatnes as before. If not, no big deal. But it's something to keep me busy ya know.

Well I don't know what your rail droppers are, but...

If a standard Marshall is: rectifier to plates, choke to screens, 20k to PI, 10k to V2, 10k to V1 then...

You could always do something like: rectifier to plates, choke to screens, 20k to PI, 4.7k to V2, 2.2k to V1.

This is just an example, but it demonstrates that you can still work with the rail and increase preamp volts without changing the PI volts. And 275Vp at V1 is high. Some Fenders ran that way but it's bound to be too stringent and harsh for a Marshall type circuit. So something like the example would give you a bump without going so high.

[Chuck H]

You might be surprised at how much the actual plate voltages on V1 affect the overall tone/feel of the amp. You could always split the B+ rail after the PI if you want to experiment with voltages below 275Vp. Where you were at 125Vp for V1 up to 275Vp is a huge jump. Lot's of experimental room in between. I think you'll like dialing in this parameter because it IS so effective for fine tuning the feel and tone.

[daz]

You might be surprised at how much the actual plate voltages on V1 affect the overall tone/feel of the amp. You could always split the B+ rail after the PI if you want to experiment with voltages below 275Vp. Where you were at 125Vp for V1 up to 275Vp is a huge jump. Lot's of experimental room in between. I think you'll like dialing in this parameter because it IS so effective for fine tuning the feel and tone.

I have already to some extent. I've had it at 200, 240, and of course as i said 275. The original voltage was about 125. The differences that I'm finding are the most noticable is that higher voltage give a heavier feel and thinner tone. The 10k droppers give the strings a lighter feel and a fatter tone. Higher voltage gives a punchier tighter feel. I can't make any decisions right off the bat. This will take time for my ears to become acclimated to each and be able come to a decision. I like both, they're just different flavors that are both good. But it's mostly a feel thing more than tone. I also need to try different tubes with the higher voltages and see if they react better/worse than they did before. By the way, are there any components that i should consider tweaking because they will make more of a difference now? I noticed for example that with 125 V a 220k plate on V1 a or b didn't seem to do much, where as in the past i recall it making a considerable difference, and that may have been when i had higher voltages in the amp.

[Chuck H]

With the lower Vp on the first preamp tube the plate resistor is a go to for altering gain. Dynamics are almost not in play at low voltages. With higher voltages I would stick with the typical 100k plate resistor. As you recognized, it's more about the feel. Headroom goes up with Vp but so does gain!!! So when you increase Vp on V1 you no longer need to isolate it from decoupling (which is what the larger plate resistor can do). So, with the higher Vp you get gain AND dynamics. I'd guess that a Vp of about 210 to 220 will be to our liking in the long game. Not at all pushing the tube but still exploits it's potential and dynamics. A tad of clipping at V2b can be achieved at this level and it contributes to the ubquitous "touch sensitivity" that is often chased. This, as it applies to Marshall topology.

[daz]

Ok, i'll see if i can get it to 200V off the B+ node where it is coming from now. As to V2, I will have to check the voltage and see where it is. By the way, is this why matchless uses parallel nodes, so they can fine tune every stage voltage wise? I would think so, but then i have to wonder why no one else does, or at least i haven't seen others so there can't be a lot. From playing with filter cap values i found each stage prefers a different value, so i would think given that each stage seems to prefer different voltage and filtering, parallel nodes would be ideal.

[Chuck H]

There are other amps that do it. But it isn't common. I have a build going right now that uses a split rail. The AC30 also uses a split rail. Most preamp tubes have wide operating parameters as to the voltage and most amps simply aren't tweaked in this area because it wasn't needed for function. Also, You DO wan't to use a node for V1 that is beyond the main plate supply on the rail. There is still some noise and ripple early in the supply and this can be amplified in the early stages. Grounding for the filters may also get tricky to eliminate hum and noise. So try to keep a perspective on whether your amp is humming or more noisy than usual.

[daz]

There are other amps that do it. But it isn't common. I have a build going right now that uses a split rail. The AC30 also uses a split rail. Most preamp tubes have wide operating parameters as to the voltage and most amps simply aren't tweaked in this area because it wasn't needed for function. Also, You DO wan't to use a node for V1 that is beyond the main plate supply on the rail. There is still some noise and ripple early in the supply and this can be amplified in the early stages. Grounding for the filters may also get tricky to eliminate hum and noise. So try to keep a perspective on whether your amp is humming or more noisy than usual.

No problem there, it's the same as it's always been which is pretty quiet. But is what you are saying due to the lesser filtering since it no longer have V2 and 3's filtering upstream from it? Which brings me to another question.....does the filtering in the previous stages contribute to V1's filtering value? In other words, lest say we have a total of 40UF upstream in the 12ax7 stages (2x20uf) from V1 and i now am tapping from before those 2 stages. And lets say i have 20uf at V1 too. How does the earlier caps affect V1's cap value needs? In other words, if it works great with 20uf, does it now need a larger value to receive the same amount of filtering?

[Jazz P Bass]

I tend to think of the caps as decouplers.
In a typical amplifier, where there is a string of voltage drops (resistors) for the separate stages, each 'drop' gets it's own capacitor (decoupler).
So, it is not that they add up, so much as they act on each separate drop.

[daz]

I tend to think of the caps as decouplers.
In a typical amplifier, where there is a string of voltage drops (resistors) for the separate stages, each 'drop' gets it's own capacitor (decoupler).
So, it is not that they add up, so much as they act on each separate drop.

So then no matter what values of cap you use on the nodes before V1, they won't influence the tone/feel of V1 at all?

[Chuck H]

Everything is relative. If, for example, you used only 1uf at all stages prior to V1 then certainly there would still be hash and ripple at the V1 node. And since electrolytic caps are waayyy less than perfect this would mean that even a high value cap to decouple V1 might not be enough to reduce noise adequately. Of course this is never the case. And at some point the along the rail the power is so close to pure DC that any further filtering is pedantic. At that point it's more about decoupling. I was just saying that I don't think I would derive the V1 supply from the same node as the power tube plates. Since V1 is the most sensitive and can audibly amplify even tiny amounts of noise I just think it would be better to branch from the PI node.