|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
74.215.192.112
In Reply to: RE: Dc Regulator Comaprisons posted by fmak on June 24, 2015 at 00:18:17
Fred,
How much stuff are you using that has requirements that are stepped (i.e. change in current). Probably nothing...
I have been making regulators for the last 2 weeks. I seem to do this every couple of years. Shunt, series, combo's, etc...
Really everything we do is pretty much steady current, maybe a few mills here and there, but big deal caps take care of that.
I have to agree with Charlie Hansen that feedback regulators seem to choke the sound and loose space. But maybe not in all places.
~~~~
If you looking for a good low noise 5V to power some USB dac, via VBUS then really the LT1806 is not a bad choice. Sure you could make a Sulzer or some of the modified series or shunt things. But really the work after the regulator (i.e. in the device) is going to make much more of a difference than LT1086 vs something else.
JMO!
Gordon
J. Gordon Rankin
Follow Ups:
"I have to agree with Charlie Hansen that feedback regulators seem to choke the sound and loose space."
Seems entirely reasonable. The low power Bellesons pretty much start rolling off response around a couple hundred hertz. The high power model almost get up to a kHz. Even the LT part rolls of below 1kHz.
I'd guess the high performance ticket is go some other way for audio band work.
CPUs for starters.What are you complaining about; it is a comparison as stated?.
You say the wrong test. The wrong test for what?
Edits: 06/24/15
"You say the wrong test. The wrong test for what?"
Most $$ regulator consumers are likely putting them in analog circuitry I'd bet. I doubt many have been put on digital circuit power rails. But here you are, so maybe there are more of you.
and Gordon has repeatedly said here that he designed motherboards.
I think there was one (by Fit) that used 78xx ???????
I like to convert my computer files to analog before I listen to them.
Might as well save your money and go with a cheapo regulator since you'll end up needing to heavily decouple your digital loads with bypass caps anyway. That is, if you want it to work. Even your ultradollar regulator has no response at anywhere near the rate a modern cpu can change it's demand for power. Same with USB and its extremely fast transient requirements.
You are absolutely correct about CPU power demand varying. It varies with the program loops being run, even varies with the data being processed.
This variation has been used to extract secret cryptographic keys from satellite TV decoder chips and from "chip and pin" smart cards.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Anyone with a good scope and has monitored the power rails of a music playing PC can see the instantaneous changes in current demand.
Those with knowledge of measurement systems will know that step changes are used for gauging transient response, otherwise there will be no need to measure the square wave response of analog and digital circuits.
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: