Building power supply for master clock questions

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello Peter,

Thanks for the info, I checked DigiKey and the regs are in stock.

Can you help me out and make a suggestion for a transformer from Digikey?

I was thinking of using either 1000uf Nichicon Muse KZ or 3300uf Elna for Audio caps as filter caps and 100uf Blackgate or 100uf Elna Silmic as cap after reg. What do you think would be better?

One final question, for the bridge I have Fairchild fast switching soft recovery diodes, will this be OK in a digital supply or should I use something else?

Thanks

KevinLee:)
 
I found that MSR860 (OnSemi) work much better than MUR860 in digital supplies.

All my observations here are with regards to CS8412, but I think it should work similar with a clock.

The best bypass on the chip was BG N. I tried 33u, but it seemed too much; 10u or 4.7 was much better choice.

Those small Tamura transformers (from DK) should be more than enough 12VA (MT11xx).
 
I install 470u Cerafines or 1000u BG before regualtors (and after battery). After regulators capacitance should be rather small: 4,7 or 10u BG N for digital chips and around 100u BG N for TDA1543 DAC. I found that those values work the best.

But I would suggest you install some sockets and play with capacitirs yourself and see what you like most;)
 
Question:

Iv'e always wondered why such low levels of filtering after a regulator. Is this due to possible oscillation problems rearing their ugly head..and if so, why take such crap from a regulator? What the heck is it's problem anyway?

All I know, is I don't need microdynamic destroying 'micro correction' from a regulator ruining my sonics. I keep thinking that this is exactly what a limited amount of uf after a regulator will do.. even if it sounds like it is more detailed... that may be why....like some dog chasing it's tail....is transient noise being mis-diangnosed as detail these days?

tell me I'm wrong. tell me exactly why I'm wrong. thanks!

What am I saying? Every time I have gone to excessive filtering after regulators.. I've always realized gains in sonic quality.... not harm. Every single stinking time.

So where is my thinking wrong?
 
They say a capacitor that is too big can be wrong. "Bigger is better" rule simply doesn't work here. Note that a big capacitor loads also slower, so its more like a question of balance.
On the other hand regulators these days have very good noise suppression against low frequency noises (eg 50 Hz). What you should care about instead is the high frequency noise (yes, that can come from the supply also, causing more jitter!), so it is recommended to use some good quality drop-tantalum capacitors of, say, 10uF after the reg instead of some big electrolits. At least this is what was told me... :)
 
Oops..yes. after regulation.. for DIGITAL.. I tend to use 100-330uf Sprauge-Vishay SP series OS-CON caps. Works best. A bit larger than is reccomended here.. but works fine. Analog? I tend to use 1000uf or so (line level circuit), but a single low imp, wide band one. Beside the load, of course. Local-to-Regulator buffer(100uf and sometimes MUCH larger (10x..if the PS rail lines are long)....and then the local (to load) large cap. Sorry for the thread jacking, but is at least relevant....
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.