|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
101.161.33.69
In Reply to: RE: Spice is better, but definitely not easier posted by Chris O on March 24, 2012 at 11:31:46
You need one of the Spice solutions, maybe LTspice, which is free and gives unlimited size models - subject to practical limits, of course. Spice can be used to simulate not just the PS bit but the whole amplifier.
PSUD2 is OK but is quite crude in comparison, being designed for the beginner who wants a quick and easy answer. But if you find even PSUD2 too tricky to use, forget Spice because it requires much more effort.
Follow Ups:
Not tricky, just at times slow taking a minute to calculate on a very fast newer computer.
I perform a lot of computer software functions at work as management & presentation tools. Complications is not really an issue.
Bizarre little values in chokes or caps drives PSUII into fits. If you stick with "normal" stuff, it is very fast.
No micro chokes for me unless I am working in the RF spectrum.
I have been analyizing the entire PS including the next section after the power tubes. I look for little nuances that can appear on the power tube side of the PS.
My power supplies as designed for a while are low DCR at least at the power transformer & tube rectifier. I use a hybrid bridge to keep the PIV lower. The foremntioned already makes a big difference. I also add .2uF input cap to a LCLC as a pre-suppressor before the LCLC.
I find my LCLC filter design has less recovery overshoot with higher than the as recommended at times 30 ohm maximum chokes. The last PS I modeled to the power tubes drops 8 volts for 16ms for a load change of 100 to 160ma, yet is a low ripple design.
4 volts better regulation (2.8% vs 1.4% regulation) & 4 ms faster time recovery (16ms vs 12ms) in a LESS design with high ripple is not a good tradeoff IMO.
Of course I can blowoff the whole discussion and build a shunt regulated PS, but I do not hear any difference.
When you are using a hybrid bridge is the final B+ close to the rated secondary voltage of the transformer you are using? I had read something to that effect. I was wondering how you had modeled it in PSUD, or if you had simply ran one model with a SS bridge, and one model with a tube rectifier, and averaged the two.
About .9x secondary voltage using LCLC with measurement at first cap.
.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: