ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Gain stage placement relating to noise

:

12/8/2003 9:20 PM

Gain stage placement relating to noise

How does putting a gain stage closer to the power amp as opposed to the input stage effect noise/hiss/hum, etc.? A long time ago I asked kg why he ditched the Williamson power stage on his BAGA. He mentioned all the gain it had and why he wanted to move it to the preamp stage, but I can't for the life of me remember what the reasons were.

I would think if you're not looking to distort the gain stages themselves, it would make more sense to put them toward the power amp rather than the input stage which would amplify noises more down the chain, no?

What are the down sides to the Williamson design in guitar amps? There is an extra coupling cap between the split load and the diffy amp. So what? Compared to the number of gain stages in high gain amps it's inconsequential, I would think.

I'm thinking of resurrecting an old Williamson amp I had into a newly made, much larger chassis. This will be no NFB with UL hookup. If I go Williamson I'll use 12AU7 tubes for the splitter/diffy amp section. If I go LTP I'll use a 12AT7.

Any thoughts or suggestions? Thanks.

John

12/9/2003 3:14 PM

quote:
"How does putting a gain stage closer to the power amp as opposed to the input stage effect noise/hiss/hum, etc.? A long time ago I asked kg why he ditched the Williamson power stage on his BAGA. He mentioned all the gain it had and why he wanted to move it to the preamp stage, but I can't for the life of me remember what the reasons were."

in my situation, the pre and power are separate chassis. that means a signal wire jumps between the two. moving a gain stage from the beginning of the power amp chassis to the end of the preamp means that there is less gain after the signal wire, which means less noise. functionally the combination of the two still results in the same overall gain for the signal.

note that if you've got a pre and power amp in one chassis, this is kind of a moot point.

the williamson is a very elegant circuit. the downside is that it uses an "extra tube" vs. a current-sharing splitter like the typical schmitt. you DO get more gain though, since you've got a common cathode voltage amp, and then a differential driver after that.

you might be surprised if you try it out if you disconnect one of the diffy driver's grids from the concertina. essentially half-driving the circuit, turning it into a self-split. this is a guitar amp, remember!

another issue: if you like to run open loop, with no NFB then there's probably too much gain with medium or high mu tubes. your 12au7 isn't a bad choice. also consider the sv6n1p which is a tube i really like, and has about the same mu (33, iirc).

hth
ken

12/10/2003 4:09 PM

[QUOTE]in my situation, the pre and power are separate chassis. that means a signal wire jumps between the two. moving a gain stage from the beginning of the power amp chassis to the end of the preamp means that there is less gain after the signal wire, which means less noise. functionally the combination of the two still results in the same overall gain for the signal.

note that if you've got a pre and power amp in one chassis, this is kind of a moot point.[/QUOTE]

Thank you so much. That is extremely helpful. I don't quite understand why the separate chassis make a difference in this way, but I don't have to since this will be pre and power sharing the same chassis and power supply.

I'm going to make a nice big chassis so I have plenty of room to try out a few things, but I want to keep the Williamson power stage intact. I may be wrong, but I can't help but think it's responsible for a good part of what I liked so much about that amp.

What I'm thinking of trying right off is a 6AU6 or EF86 direct coupled to a mosfet set up like a cathode follower ---> Baxandall tone ---> volume ---> Williamson power stage. Should be plenty of gain for my purposes.

John

12/11/2003 2:09 PM
kg	john, the issue with the separate chassis isn't so much power supply related--it's the matter of the signal venturing outside of the relatively protected chassis and into the "noisy" free-air environment, where it encounters all kinds of EMI. that's why if you're all in the same chassis, it doesn't matter much. ken

12/11/2003 6:15 PM
anonymous	I have a tri-amp setup (actually six mono blocks). I have set up these amps to be removed from the crossover and the preamp by as much as eight feet!!!! Absolutely free of noise. Hmmmmmmmm - makes ya think don't it?

12/12/2003 5:51 PM
kg	it makes me wonder what your signal driving impedances are, what voltage levels are being passed, if your interconnects are shielded or not, and whether they are balanced or single ended. to that end, yes, it makes me think. ken

12/12/2003 6:16 PM
anonymous	Note, this was a bit of silliness really; I wanted the amps co-located with the speakers. I don't have the amp(s) data with me so I'd have to check on I/O impedance(s) and levels. The amps were nothing special - old Stromberg-Carlson and Heath stuff. SAE crossover. Standard 2 conductor sheilded interconnects. Very quiet.

Page 1 of 2