Power Scaling the Liverpool

[withnail_jet]

I am planning a L'pool build with only 2 power tubes, and I would like the ability to get the thing even quieter if need be. Power Scaling seems like a good option, and I was looking for suggestions regarding two things:

1) How much of the amp I should I scale? Output stage only? The whole thing? (I am leaning towards output stage only)

2) Should I implement a Post-PI Master Volume in conjunction with the Scaling (as "Drive Compensation")?

Anyone scaled one of these things, and if so, what's the opinion?

[gtrcollectr]

I am also currently building a Liverpool and would love to have some more info on scaling this amp specifically. I have thumbed through all of the TUT books and have not found any clues as to the better method for this style amp or exactly how to implement it.

Has anyone ever scaled the liverpool? I have not found an example that has been scaled.......

[dartanion]

There has been a lot of power scaling discussions going on over at the 18watt forum, so you may want to take a gander over there.

Other wise, look into an attenuator if that meets your requirements.

[withnail_jet]

Yeah, I've seen some of the 18Watt discussion of this topic. What I am hoping to hear is someone's firsthand experience in scaling a Liverpool (or clone), and suggestions. If I followed it correctly, the 18Watt crowd suggests scaling the whole amp, but I've heard from at least one person that this method tends to kill the touch responsiveness of the Liverpool. So this is why I was thinking of scaling the output section and adding a PPIMV. Still not sure whether this would help, though...

[fierce_carrot]

If you are going to install a PPIMV, then why add power scaling?

[bueller]

I have not made or scaled a Liverpool, but I have scaled a similar SE type TW style amp. I used TUT vol 3, "The Standard" schematic as a guide for the scaling and built my own circuit. I made it with and without the master.

The master volume is necessary for reducing the drive level to the power section. If you scale down the power amp but not the preamp and don't have a drive compensation then you lose all articulation and it's too much distortion. In that the smallest signal will distort the power section and not in a good (useful) way. If you scale the pre-amp you may not need the master, but I didn't like the sound/feel of the 12AX7s on low voltage.

I have to say that power scaling is better than attenuators, e.g. Powerbrake, Hotplate, and DrZ Airbrake. The Airbrake is the best of all of them, but scaling was more natural sounding to me than attenuation. Although a lot of the feel is lost when the volume is not there, maybe speaker compression/reaction is key.

After trying scaling for two months, I tore it all out. There is no substitute.

[bueller]

My bad, it's The Ultimate Tone (TUT) volume 5 that has the Standard schematic in it not vol. 3

[withnail_jet]

If you are going to install a PPIMV, then why add power scaling?

From KOC:

"As you make the output stage look smaller, if you do not reduce the drive signal from the splitter, then the output stage will distort more than when it looked large. The 'Drive Compensation' control is essentially a post-phase-inverter master volume. However, unlike the usual post-PI MV, we're using this control to reduce distortion instead of increase it. This is a refinement that is not needed in every situation, but does allow true power scaling of the output stage and allows the use of your "usual" settings for all of the other amp controls. "

[Fischerman]

I built a power scaled Express but not a Liverpool. The real issue IMO is whether or not to scale the PI. You're definitely going to scale the power tubes and I would definitely not scale the preamp...which only leaves the PI for consideration.

If you scale the PI then you'll need a pre-PI MV...if you don't scale the PI you'll need a post-PI MV. In any case, you need a MV right before the first thing you scale. Since EL84s overdrive so easily and that EL84 overdrive is such an integral part of the tone...I would think scaling the PI would be the better choice but that's just a guess. I did not scale the PI in my Express but EL34s don't overdrive as easily as EL84s (but EL34s are still relatively easy to overdrive compared to most power tubes).

I can honestly say that power scaling doesn't make a huge difference in my Express...it seems much of the distortion occurs in the PI and before. So just using the post-PI MV gets me most of the way there. But PS does add to the tone...and changes the feel of the amp too (in a good way).

[withnail_jet]

My bad, it's The Ultimate Tone (TUT) volume 5 that has the Standard schematic in it not vol. 3

I have that volume of the series and am using it as a reference for the build. Still trying to follow some of the notes in the Standard build though!

If you scale the PI then you'll need a pre-PI MV...if you don't scale the PI you'll need a post-PI MV.

So, if you look at the attached "artistic modification" of Mark Abbots L'pool schematic, does putting a 1 Meg pot where I've added the red "MV" box get me a pre-PI MV?

I think I will try your suggestion to scale both the PI and the output. Concurrent with implementing power scaling, my goal is to retain the dead stock tone of the amp when the scaling and MV controls are maxed...

[Fischerman]

You'd actually prefer to put the pre-PI MV where there isn't high voltage on it...you'd have the full plate voltage on the pot doing it like that schematic. But you can't just move it to the other side of the coupling cap because then it mucks with the PI grid (and possibly the bias voltage on that PI grid). The Express/'Pool schematic is a bit 'diifferent' in that there is only one cap between last preamp plate (or cathode) and PI. You'd have to add a cap, then the MV, then another cap after it. The stock .1uF cap going into the PI is extremely large so even if you used another .1uF cap I don't think you'd lose any noticeable low end.

[gtrcollectr]

[img:800:822]http://home.roadrunner.com/~toddrjackso ... ivermv.jpg[/img]

Would this be the correct implementation as you stated above?

I added a .1uF coupling cap in front of the pot to isolate the MV before the PI.....

Thanks,
Todd

[Fischerman]

Not quite. You want the 'top' lug connected to the first .1uF cap, the wiper (center lug) to the other (second) .1uF cap, and the third lug (left lug when looking at back of pot) to ground. Typically a 1M-A pot is used. It's just like the MV in a 2203/2204 (JCM800) Marshall except the Marshall has a tone stack instead of that 'first' .1uF cap.

[gtrcollectr]

Not quite. You want the 'top' lug connected to the first .1uF cap, the wiper (center lug) to the other (second) .1uF cap, and the third lug (left lug when looking at back of pot) to ground. Typically a 1M-A pot is used. It's just like the MV in a 2203/2204 (JCM800) Marshall except the Marshall has a tone stack instead of that 'first' .1uF cap.

[img:800:926]http://home.roadrunner.com/~toddrjackso ... poolmv.jpg[/img]

I corrected the pot wiring per your suggestions. Let me know if corrections are needed.

Thanks for the help,
Todd

[withnail_jet]

So, when scaling the power and PI sections and using a pre-PI MV as "Drive Compensation", would it be correct to assume that one should keep the MV pot well away from the scale control b/c of the voltage going through there? It would be nice if these two controls were next to each other on the panel, but I don't want to induce any noise...

thanks!