Power Supply Design, Big Reservoir Cap, Small Preamp Caps

[jhaas]

I am in the planning stages of a multi-channel build that borrows some ques from the Soldano SLO100. I've decided to design the power supply after the SLO's and I have a few questions about it.

Omitting the dropping resistors, the SLO100 power supply looks like this:
200uF Reservoir Cap feeding the plates
100uF Cap on the node feeding the screens
40uF for the PI and Cathode Follower
Preamp fed by two parallel nodes:
10uF feeding V2
10uF feeding V1

There's also a big 5H choke between the plates and screens

So in summary, the SLO PS has five nodes, real big capacitance up front, and small value caps in parallel RC networks to the preamp.

I understand that the big values for the power tube nodes, and the choke, will make for a very stiff power supply, but I'm curious about the design goal behind the Soldano's choice of small caps and parallel nodes to the preamp.

My amp will have a lot more going on, three preamp channels (Clean/Drive/Lead) that will require a total of nine power supply nodes. I'm considering implementing parallel R/C strings with 10uF caps for each channel to match the SLO100 design, but I'd like to know the effect this will have first.

-John

[Firestorm]

The parallel nodes allow each preamp to see the same VDC (no series dropping resistors) and may reduce interaction. The small value caps are plenty considering the brute force filtering done earlier and control frequency response: the preamp plates are RC coupled. SLO is a preamp distortion topology, so I expect the filter values are part of the overall frequency response design.

[roberto]

SLO poweramp sounds really good when you lowers the NFB, just because the power supply keeps the response constant even at mid-hi volumes (even with the depth on 10 and detuned guitars).

The 10uF is enough when you have 15k decoupling, so basically you keep the 1Hz RC rule for the power supply.

Do not forget that the 40uF feeds six gain stages ( gain+CF for the loop, gain+CF for the tonestack, PI), and a decoupling could be a good idea (or removing a tube).

As for the parallel configuration, it's just because you can obtain higher voltages and higher rejection between the stages.

[jhaas]

Thanks for the info, that all makes sense to me. I'm glad to hear the SLO power amp continues to sound good with lower amounts of NFB. My amp will have a PPIMV, and I'm not likely to be able to dime it at most the venues I play, so reduced NFB will likely be the norm.

[Milkmansound]

slightly OT:

I lived in Burlington for 8 years. Miss it!

Must be getting cold up there about now

[jhaas]

slightly OT:

I lived in Burlington for 8 years. Miss it!

Must be getting cold up there about now

"Slightly"?

I've been here 10. Live about 16 miles east of Burlington, but work in Winooski. Yeah, first snow to stick fell last night. Only an inch, but a co-worker of mine put his car in the ditch this morning!