Usb power

IMO the whole "usb power is 'dirty'" thing is completely overblown, and a legacy of the "all switchmode power supplies are bad" mentality that hasn't been true for a decade or more. (Or a product of people with fancy 'scopes who think that any noise above -98dB between DC and daylight is completely unacceptable.)

If you're really so concerned about USB power from the computer... just use a powered hub.

There's probably already some filtering in the uDAC (at the very least, some rail caps). I'm sure the designers took into account where they were getting the power from and how much it needed to be filtered (if at all) to produce good sound.

Of course, you're certainly welcome to try making your own supply with a split cable. Use any regular 5V supply (linear 7-9V wallwart -> 7805, TREAD, etc). Cut up the cable and terminate the +5V coming from the computer side. Connect the +5V from your supply to the DAC end of the cable. Leave the ground intact, and hook the ground from your power supply into it. Leave the data lines intact. Not too complicated.

In my experience how "dirty" the USB power is highly system dependent. I can not hear the difference between ac power and USB power on my Gamma2 from my desktop but there is audible noise on my older laptop.

Try it and see when your uDac comes in.

I would bet that the uDAC is already regulating the USB power supply. I haven't found anything specifically stating that, but USB is 5V and the uDAC says they're providing 2V at the RCA outputs. That's at least 1/2 volt burned by regulation, most likely. Even the Alien and Bantam DACs do it with only 1/4 Volt. (Regulation means the dirty supply is cleaned up, for most intents and purposes.)

but this link seems to say that no regulation whatsoever could ever "clean up" a dirty 5V USB supply

http://www.head-fi.org/forums/6078622-post123.html

I'd re-read that link. the clean supply is the USB and the bad one is the external.

Hehehe, owned!

hehe, we're playing w/ words here

he said that the interface w/ the external PSU was worse than the plain USB power....when you see the internal cheapo SMPS within the Musiland 02US, it's pretty obvious that it'll be much worse than a quality ATX PSU and that it will pollute the case w/ massive EMI.

but saying that the USB power is clean is just not true, and you know it

you get all the 5V ripple from the ATX SMPS(120 mV maximum by the ATX standard on the 5V rail), and the very dirty ground that's polluted to death by all the other computer components...and when some components(CPU/GPU/HDD) are under heavy load, the ground is even more dirty as this is where they flush...reason why many ppl get ground loops reacting to the graphic card's activity.

all the audioholics will say that a SMPS is just a terrible idea for audio, and even worse if it's shared w/ tons of noisy computer parts.

a random link discussing the issue: Do SMPS have a home at high end audio? - diyAudio

and a random link w/ ppl(including the main head-fi admin) discussing how much of a difference a clean PSU makes: http://www.head-fi.org/forums/f21/fi...pplier-410199/

using a computer 5V supply to feed a DAC stage is totally fool to me...there, I said it. now prove me wrong, by all means.

and a guy going nuts w/ ground loops over firewire: FireWire 4-wire cable to break "ground loop" - mLAN Forums

Building a PSU for an external firewire soundcard - diyAudio
sharing the audio ground w/ the computer will automatically lead to SQ degradation, why bother w/ USB when we have S/PDIF...that offers full galvanic isolation right off the bat?

USB power brings high ripple/harmonics and dirty grounding...how is that good to anything for audio? digitis? yes, that's the main source

It is interesting that you say that when I get into a couple of those links and some of the comments are that a good SMPS can be very good.
Assuming that they've transformer coupled it. We won't even start to talk about the impedance matching issues.

As I sit here and listen to my USB powered dac, which include a 400kHz charge pump generating a negative rail, where the noise floor is below that of the output opamp, and its quiet, I'm not worrying about it.

yes, I agree that you could easily use a pulse transformer on S/PDIF coax...but IRL, I have yet to find a USB/firewire device that does. And the OP was asking about the uDAC, do you think they added one?

Can you name some USB/Firewire devices that do? not even the pro-audio manufacturers such as TC Electronic seem to care...and no DIY please

and you can't quite cut the ground wiring in a USB/firewire cable because most devices seem to use it to do hotplug and cable length detection.

yes, I know that some ppl on the diyaudio.com link said that SMPS can be good for audio...but it's pretty obvious to me that they haven't heard a discrete linear regulated PSU on high end gear.

I can provide many links about the "Firestone Supplier" where ppl were blown away by the SQ difference between a cheapo wallwart and a linear regulated PSU...yes this is old school, but the old school knows how to make good sounding gear.

Amazon.com: Customer Reviews: Audiophile Products Supplier - Silver Power Supply Unit

CD Reviews by C60Crew: Firestone Audio Supplier: Delightful Sidekick

What I find interesting is that there are people who have stated clearly that they have heard excellent SMPS and you discount them as saying they have never heard good linear gear. However some of those same people are experts on power supplies and are used as references on measuring how good linear gear is.

I'll bow out now. You've obviously made up your mind.

no, I haven't...I am well aware that your are far more competent than me on those matters ...but what you can do in DIY like ground galvanic isolation over USB/firewire doesn't seem to EVER happen in commercial gear.

and some ppl even said that running a computer S/PDIF interface on a clean linear PSU had yielded a worthy improvement.

thoppa measured that his Corsair PSU 12V ripple was going up the roof when his HDD was under stress...so I don't see how someone can put a discrete clean linear regulated PSU at the same level as an unfiltered 5V USB supply, as I'm quite confident that most <$1K USB/firewire DAC's don't bother filtering the nasty computer ground.

another thing I've noticed is that if I hotplug/unplug my headphones on gear that uses a clean ground, everything's fine...but on anything that's sharing ground w/ my computer, I get loud pops when the jack goes in and out...a ground is supposed to be neutral in audio, but a computer ground is just very filthy IRL.

and I don't think that a $75/150 computer ATX PSU could be referred to as a "high end SMPS"..

I don't claim any competence on anything. I'm always learning.

What I would say is this though: I agree that the cleaner the power supply the better. Whether that comes from USB or SMPS or a Linear or Shunt, it doesn't matter (at least to me). They are all tools to be used to create the desired result.

Your mileage on anything will vary. I've only used laptops for at least 10 years so my power dynamic may not be the same as yours. Even plugged in it will not be the same.

All the power supplies are building blocks. You can make them poorly and well made. That first link of walwart versus USB is a great example of this - a cheap linear reg that makes more noise than USB. Your comments on the "Firestone supplier" are another point to support this. They provide a cheap wal-wart and then up-sell you into a better linear. Great marketing or deliberate choice?