For hundred+ Volt work, diode is diode. You are unlikely to over-current, and the voltage drop is always "small". (Gets different with high current and low voltage.)
Let me forget the sim and work it by hand. Start with 1uFd per mA. That's 4uFd, which in today's market is hardly worth buying. (There was a time when 4uFd was big.) Round-up to 40uFd.
A commercially economic goal at the first cap is 5% ripple. 5% of 250V is 13V ripple. We have gone-up 1X on the first cap so expect 1/10th of that, 1.3V ripple.
To get low ripple for preamp use multiple R-C stages. Use more of the same C, 40uFd.
Figure there is no problem with 10% voltage loss in the filter (it won't hurt audio performance much, and buzz is a bigger hurt). 10% of 250V is 25V. 25V at 4mA is 6kOhms. A preamp will need 3 or 4 caps, say 4, so 3 resistors. 6k/3 is 2k each resistor.
A 2k with a 40uFd is 2Hz. 120Hz ripple will be reduced 120/2 or 60:1 in each stage. Three stages is 1/(60*60*60) reduction. This is about 1/200,000. Times 1.3V ripple at the 1st cap is 6/1,000,000 or 6 microvolts.
Power supply rejection ratio of a simple triode gain stage is 1/2 to 1/3rd. So 3uV ripple at the plate.
What do we need? The hiss at the first grid is about 1uV. Say stage gain is 10. Then hiss at the plate is 10uV. Numerically, hiss overwhelms buzz. However hiss is natural and buzz is artificial. I'd like buzz a little lower.
NFB is effective at reducing buzz but it is against some religions; and needs a non-minimal amplifier.
Power caps are likely to be small-change in total cost with tubes case and knobs. (Even if you love high-buck "audio" caps, surely they only matter in the last stage of the filter?) At today's prices we can use bigger caps, or add another R-C stage.
Thank you PRR!, that's very detailed and informative info, so my 470uf caps are pointless for the CRC and even the 100uf I purchased are overkill! as I already had them ive used 3 100uf then one 470uf for left/right breakout so 5 in total. The resister size has been set by the circuit designer, I dont know if this affects the noise reduction? It is meant to be 800-1K2 so I went for 3x 330R. Does this impact hum reduction?
Ok so I have finished the cdp, I ended up with a CRCRCRC circuit on the Anode, 100uf caps where what I had ordered before PRR posted so that's what i used for the first 3 then I split the supply for left and right and used a 470uf as they where already on my PCB. I can honestly say this is the quietest SRPP I've ever built, I don't know if it's an optimal circuit but it's certainly excellent. Thanks for all the info and replies guys, it does seem I needed the extra filtering even though all I've changed is the transformer.
All the best
Alan.
All the best
Alan.
Last edited:
flux coupling
Hey DawgNo1
I wonder if the hum is caused by your 2 transformers' magnetic flux feeding each other and impinging on the small signal circuits. Quick experiment is to move 1 trafo. Especially try making the 2nd trafo flux field perpendicular to the 1st.
I would also investigate the earth plane and how the trafo static shields (if any) are tied to earth.
If flux coupling is the culprit for your hum, you may consider moving the main pcb away a bit, giving more space between the trafo etc. One idea is to move the small pcb of lytics next to the tubes to give more space between the main pcb and the trafos.
Hey DawgNo1
I wonder if the hum is caused by your 2 transformers' magnetic flux feeding each other and impinging on the small signal circuits. Quick experiment is to move 1 trafo. Especially try making the 2nd trafo flux field perpendicular to the 1st.
I would also investigate the earth plane and how the trafo static shields (if any) are tied to earth.
If flux coupling is the culprit for your hum, you may consider moving the main pcb away a bit, giving more space between the trafo etc. One idea is to move the small pcb of lytics next to the tubes to give more space between the main pcb and the trafos.
Thanks for all those great tips, I have stopped looking for the cause though because as I said the unit is now deadly quiet after building a 4 stage filter, my next one will be 5, the capacitors are half the cost nearly for the same range at 68uf they are miles cheaper than the 470uf's I was using too.
Alan
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.