Sunn 2000s Pre

[D.M.N.]

My #1 amp for...jeez, 15 years? has been a Sunn 2000s. Absolutely love the things, still have 2 that trade duties. I saw a schem and build for a 200s in a box recently, and thought "hell, why not the 2000s". The 200s build was a bit in between the two, as it included a 'mid' control, but left out some other parts of the 2000s. So I did some reading on JFET adaptations, namely the runoffgroove article, and came up with this:


Runs at 18v, will have a DI out, includes the bright and normal inputs for the 2000s, the bass boost switch, and I tagged an output volume on to be able to drive the front end and tame levels. I have J201s in the schematic, but figure I'd give 2N5457s a go as well. But there's two things I still haven't managed to figure out...

1. Scaling values. Per the 200s pedal adaptation, the pots were scaled down, but it looks like they retained all the other parts values. In the ROG article they mention all the grid and cathode elements stay the same. Per this thread, it is mentioned to scale the caps and resistors to adjust for impedance. In doing so, that should help reduce noise, is that right? So I should be scaling throughout the circuit for the resistors and caps?
2. The spare leg on the Bass control. In the Sunn schematic below, it appears that the leg is left disconnected. It'd probably be better practice to tie that to the wiper, right?

Any suggestions would be greatly appreciated!

 

[temol]

1. Scaling values. Per the 200s pedal adaptation, the pots were scaled down, but it looks like they retained all the other parts values. In the ROG article they mention all the grid and cathode elements stay the same. Per this thread, it is mentioned to scale the caps and resistors to adjust for impedance. In doing so, that should help reduce noise, is that right? So I should be scaling throughout the circuit for the resistors and caps?

If you scale down tone stack resistors/potentiometers 10x than you have to scale up capacitor values the same amount.
So - R7 not 56k but 5k6. C9 not 22n but 220n. And so on. But I wouldn't do it blindly - I would start with a simulation in ltspice. I would draw a tube and JFET preamp nnd compare the frequency response.

 

[D.M.N.]

If you scale down tone stack resistors/potentiometers 10x than you have to scale up capacitor values the same amount.
So - R7 not 56k but 5k6. C9 not 22n but 220n. And so on. But I wouldn't do it blindly - I would start with a simulation in ltspice. I would draw a tube and JFET preamp nnd compare the frequency response.

Perfect! Thank you, that's what I thought, seemed odd that the 200s had the pots scaled but not the component values. The pot values in my drawing were based of some further discussion I found on the 200s plan, but I'll take the original values and set about calculating and modeling the changes then.

 

[temol]

20k may not be enough to bring J201 drain voltage down to 9V. Use 50k or even 100k for the trimmers.
I doubt that first stage of the 2000s clips the signal. With the j201 and C2 250u I would not expect clean signal after Q1.

 

[D.M.N.]

20k may not be enough to bring J201 drain voltage down to 9V. Use 50k or even 100k for the trimmers.
I doubt that first stage of the 2000s clips the signal. With the j201 and C2 250u I would not expect clean signal after Q1.

Ah, good catch, completely forgot to update those when I decided to run it at 18v instead of 9v.

I wasn't sure how much drive that would prove to have, most of the drive from the 2000s does come out of the power valves breaking up, so maybe it would be best to omit the master and socket C2 to futz with.

 

[Blackboarcult]

Cool design!!
This amp has an intereating topology. Kinda like the non-interactiveness of the treble control, although the mids don't seem to do heaps.
I wonder which stage clips first, 6AN8, cathodyne PI or power stage, and which one is responsible for the sound characteristics?