Voltage and tone and current and feel and amp design

[smzinno]

yes, just consider it a warm-bias for your 12ax7. Marshall uses this, Mesa did also on the old MK series. I've also seen it recommended often by the gurus over at 18watt if there's too much fizz in the overdrive. You could also try 680 if you wanted to stay closer to center.

I noticed your earlier description of the 6G3 type PI in your amp was using a 1k5 cathode resistor. The original value here should have been 820r, could that have an error in your build or just a typo ? The 1k5 in that position would be a pretty cold value.

[dehughes]

yes, just consider it a warm-bias for your 12ax7. Marshall uses this, Mesa did also on the old MK series. I've also seen it recommended often by the gurus over at 18watt if there's too much fizz in the overdrive. You could also try 680 if you wanted to stay closer to center.

I noticed your earlier description of the 6G3 type PI in your amp was using a 1k5 cathode resistor. The original value here should have been 820r, could that have an error in your build or just a typo ? The 1k5 in that position would be a pretty cold value.

Right on. Thanks so much. I'll definitely be tinkering with the PI bias and tail resistors. I'll post back after a few changes...probably Monday evening...

As for my current values, I'm using a 1k2 as the PI bias resistor. I think it is a bit cold even at that. I'll go way the opposite direction and see how it goes....

[paulster]

I was just going to suggest changing the bias on the PI to warm it up a bit and then I saw that it's already been suggested! Definitely worth a try for the sake of a single resistor swap (or just tack another resistor on in parallel that'll get you down into the right range while you test).

I use a 750R in a Vox-style PI and it's much better than the regular 1K2 in my opinion so it's worth a try.

[Andy Le Blanc]

Here's one way that it was used, Ive found other references that the
practical circuit has too much distortion for HI-FI. But if the concept is used with
a high Gm tube as an inverter.........

[Andy Le Blanc]

Here's the O.H. Schmitt, referenced by the previous article.

[dehughes]

Very cool. Thanks guys!

So as I understand it, lowering the bias resistor in the PI will drop the plate voltage as more current is being passed. This will cause the stage to have more (?) headroom and distort later (to a point), and warm it up, tonally. Given that this stage is usually run at higher plate voltages than the preamp tubes, it's going to run stronger regardless, and closer to its maximum.

Running a higher value Rk on the PI will lower headroom, raise the plate voltage, and cause it to distort earlier and in a (possibly...) less desirable fashion.

The "Vox" values (1k2 Rk, 100k plates, 1M grids, 47k tail) will "clean up" the PI and let it distort earlier. "JTM" values like a 470 Rk and 10k tail will run the PI hotter and cause it to distort later, and the lower tail resistor provides less of a constant current source and necessitates an 82k/100k Ra setup to balance the outputs of the stages.

Am I at least close?

[dehughes]

SO, get this:

After doing some reading and thinking and reading and thinking, I decided to change the PI to this:

Rk: 820 ohm, to a 10k to a 4k7 tail. Unused side's cap meets where the 10k and 4k7 connect. AND, at this same joint, I ran a 56k off of the speaker tap for some feedback. Totally different feel. The "quickness" and harshness is gone in the upper midrange, at the sacrifice of some high end articulation and low end depth. It's like the amp is "more emotionally stable" than it was before. I'll crank it up more tomorrow and Thursday, but I think I'm headed in the right direction! Now I'm considering adding a presence control, or some sort of variable feedback resistor...maybe a 50k pot in place of the 56k resistor? Hmmm.....

I'll post back after more tinkering. Thanks for the help!

[Structo]

Yeah, see I think you are doing what I did.

I tried making several Deluxe type circuits work but in the end it went more in a plexi direction as I like the tone and feel better.


I'm sure you've seen this schematic.

[dehughes]

Very cool!

I'm in the middle of some tweaks on my build...I'll post back when I'm closer to settling on something....but in the mean time...

Two questions:

1) Where in designing the PI is the point where one needs to reduce the size of a plate resistor?

2) When is a small value (47-100pf) cap necessary across the PI plates, and what purpose does that serve?

THANKS!

[gearhead]

Very cool!
2) When is a small value (47-100pf) cap necessary across the PI plates, and what purpose does that serve?
THANKS!

https://tubeamparchive.com/viewtopic.ph ... +capacitor

[dehughes]

Very cool!
2) When is a small value (47-100pf) cap necessary across the PI plates, and what purpose does that serve?
THANKS!

https://tubeamparchive.com/viewtopic.ph ... +capacitor

Huh....interesting. I guess I wouldn't think of a Marshall and a Brown Deluxe as needing stability....but what do I know? I can see the high end cut being needed, though...but my question is now, I suppose, wouldn't a Cut control (a la AC30, etc..) serve the same (or similar enough) purpose as putting a 47-100pf cap across the plates of the PI?

[paulster]

Often known as the 'Fizz' cap when applied to Marshalls, which pretty well describes what it's there to get rid of!

[gearhead]

Very cool!
2) When is a small value (47-100pf) cap necessary across the PI plates, and what purpose does that serve?
THANKS!

https://tubeamparchive.com/viewtopic.ph ... +capacitor

Huh....interesting. I guess I wouldn't think of a Marshall and a Brown Deluxe as needing stability....but what do I know? I can see the high end cut being needed, though...but my question is now, I suppose, wouldn't a Cut control (a la AC30, etc..) serve the same (or similar enough) purpose as putting a 47-100pf cap across the plates of the PI?

Believe so. haven't tried myself, but as per Jackie Treehorn in that thread, "The Komet has a cut control across the plates, so I don't think Ken would have a problem with that approach. "

[dehughes]

Very cool. Thanks gearhead.

One other question, regarding screen resistors:

In the 6G3 schematic I see that there are no resistors on the screen, but rather they're connected to the B+ right after a 1k resistor in the B+ line. As it is now, that's how I have the 6V6s wired. What would the difference be in terms of feel and tone should I put 1k resistors on each 6V6 grid as well? Basically, the 6G3 vs. AB763 way of wiring up 6V6 screens...