Any experts in choke loaded tube rectified power supplies?

[phrarod]

I've had a few threads on this circuit already based on different aspects. One contributor suggested I make a new one solely about the choke loaded supply.

I'm including the schematic. The PSU output is 205VDC 12mA. This goes to the tube by way of output transformer. The only R/C is the tube bias.

Previously the prototype had this psu circuit with a regulated supply in front of it. The ripple was 45mV (.045vac) Now the ripple is 310mV (.31vac).

Someone modeled this in another thread and said it should be 37mV of ripple.

Both supplies show identical ripple. Each has its own separate ground plane.

This is a two-box pre-amp, dual mono. Star grounded. Ground lift is a 100 Ohm/.22uF R/C which had the prototype dead quiet although that was on a single board with the rectified psu.

No scope here and while I try to find someone with a scope is there anything that looks off that I might explore? Sonically it sounds fine so fixing the ripple which will hopefully help expose a slight 60 cycle hum solution is all that's left.

[DougL]

I am not an expert.
However, there are a couple of things I would tweak in PSUD2.
First, a 12mA Current Sync is usually more useful than a resistor as a load.
Second, you are using the default 2 ohms ESR on the caps. I would look it up on the spec sheet, or use .1 R. 2 R ESR caps are floor sweepings today.
Third, I would delay 10 sec to let the supply settle, then you can read the ripple off of PSUD2 directly.

HTH

Doug

[Rundmaus]

310mV of ripple sounds much to me.

A supply I'm building right now shows 11mV(p-p) ripple in PSUD, using a B+ of 410V DC and a load current of 110mA. Filtering is LCLC using 5H, 120µF, 5H, 120µF.

With a 29H choke and 12mA load current I would expect less ripple...

Greetings,
Andreas

EDIT: As I am no tube expert, "much" is very much relative^^. Morgan Jones notes somewhere in one of his books that an upper limit of 5% of B+ would be a good design choice concerning ripple - that would be about 10V with your B+ of 200V....

[phrarod]

Yes its far too much ripple. That's what I'm trying to reduce. Any suggestions are appreciated. This is all p-p.

[phrarod]

Here's the file with the suggested changes. I'm new to this program. Real world its about 7vdc higher than the program shows.

Is it saying the ripple should be 12mV?

[Rundmaus]

Hm,

running your .psu file still gives about 170mV pp ripple...

Greetings,
Andreas

[phrarod]

Andreas. Where are those numbers in PSU panel? I would like to understand it better. So I'm getting almost twice that. 320mV. Even 170mV is too much.

[Rundmaus]

Hi,

for the ripple, mark the checkbox for U(I1). This will show a graph of the voltage over current source I1, which is your load.

Zoom in (or out) as much as needed to see the minima and maxima of the (sinusoidal) voltage curve. Read out minimum and maximum B+ on the voltage scale --> ripple.

Greetings,
Andreas

[phrarod]

Thanks Andreas! I'll play around with it and see if I can find ways to lower it.

[phrarod]

Did you mean V(I1)? There isn't any U checkbox.

How's this modeling look?