Firestone BRAVO 24/96 USB transport w/o drivers + reclocker? - Page 30

I use this one: http://aim-eco.com/?pid=10601619

It outputs 12.2V(w/o load) and initially came w/ a 19" LCD screen I think...it's overkill as the Tenor chip needs <100mA, but hey I got it for free and it sounds good

I could, but then

I was promised to be notified when it'd become available, patience my friend.

the TeraX2 PSU might actually be really sweet to power the Olimex dongle: http://cgi.ebay.com/Teralink-X2-X1-Linear-Low-noise-Power-Supply-black-/250754604110

The Olimex can take 8-15V and this PSU can output up to 10V :)

And this thing to provide a clean USB power would render the USB3 board + 5V batteries kludge useless: http://www.aqvox.de/usb-power_en.html

It might even provide all those self-powered dongles w/ a second youth.

Soon I will have more PSU's than audio units(as tomorrow I will receive three 24VDC LM317T based linear regulated PSUs), but we're literally listening to them...so I guess that makes sense

Good point, but it'll be a challenge to make it look good...and not like the nasty kludge that it is.

BTW, the Olimex+laptop brick PSU really makes a drastic change on this cheapo DAC(PCM2704): http://cgi.ebay.com/SMSL-Hifi-Sound-Card-DAC-w-USB-OPTICAL-COAX-SD1955-/270696777603

The cheaper the design, the more isolation matters I think. The Bravo already has a rather complicated design, and the HA-160D USB input has a pulse transformer embedded...but on a cheapo ebay DAC, the SQ improvement is pretty amazing!

OK, I've ordered the TeraX2 PSU...someone on ebay claims that it sounds great on the EVO, and I'm sure it'll do wonders on the Olimex dongle. Michael from Teradak said he'll send me the ripple measurements he made.

Now to decide whether I'll keep it on 8V and go through the voltage regulator of the Olimex, or lower it to 5V and tap it directly onto the USB output of the Olimex

230V, and I think I'd rather avoid kludges so I'll go safe and simply use it at the default 8V. If there's no audible diff. over the laptop brick whatsoever, I'll reconsider.

This one looks yummy too: http://www.teradak.com/ProductsView.Asp?id=24

Originally Posted by ahodge 

  I am also looking at usb cables.

Everything that's been said about USB cables sounds like major bs, but the thing is that I've got half a dozen USB cables here...and they all sound very much different, even though they're all USB 2.0 certified, shielded, braided, got ferrites/gold plated plugs, Fanny by the gaslight and so. When you read what happens in a coax cable, it might very well be that USB cables matter a lot more than they seem: http://en.wikipedia.org/wiki/Coaxial_cable

signal leakage, induction, impedance, skin effect, signal reflections, yada yada...the differences I'm hearing are definitely not placebo

my fav cable is still a fiver, it's only got one ferrite and it sounds amazing...I should get a hold of some pricey cables soon, time will tell.

Ppl keep asking what jitter sounds like, well look no further than playing 44.1kHz through PCM2704

48kHz sounds ok'ish, but 44.1 sounds sibilating to death, like a kid singing through the space between his middle teeth. That's what I was told about the STX ages ago on the official Asus forum because it only has a 24.576MHz(48kHz*512) clock slaved to its DSP, and that seems to be exactly what jitter sounds like. I thought PCM2704 sounded colored, but that's just harmonic distortion due to jitter. Could it be due to its 12Mhz clock? That's a possibility.

Upsampling to 48kHz minimizes the sibilating problem, but going 44.1>48kHz is also very lossy...so there's no free lunch to be had.

Anyway, my TeraX2 PSU has been shipped...I hope the Chinese new year won't delay it too much

OK, change of plans!  I've run more tests, if I tape the 5V and GND of the USB cable the Bravo isn't recognized whatsoever...if I tape 5V only, the data connection works on both sides, but no audio! and if I tape GND only, the "USB" LED on the Bravo blinks.

I believe this proves that -just like I thought- the Tenor chip -in the Bravo- uses the USB cable 5V...that would explain the SQ difference when using an external PSU.

This thread shows that it's the computer ground that's dirty, and it's been confirmed here: 

So we want to lower the USB 5V ripple, keep it independent from the computer and provide a clean ground. There really is no need to mess with the data lines, it's uncalled for and will only color the sound(mostly harming the trebles).

I also had a few emails w/ the manufacturer of this thing, and he doesn't think that galvanic isolation is useful for computer audio at all: http://www.odysseyaudiohk.com/eng/tod2.htm

I've just put an order on this thing: http://www.aqvox.de/usb-power_en.html

It could also provide the Hiface w/ a second youth, but quite frankly I'm not going back to proprietary drivers...besides, I was told that the problem w/ the Tenor chip isn't the crystal, it's the isochronous drivers that provide poor jitter. I like to believe that WM8804 overcomes this issue in the Bravo.

I'm also awaiting a WW cable, and I'll be lent a premium killer audiophile USB cable...time will tell. I'm skeptic, but my real world experience has shown me that all coax and USB cables sound different...why? It's rather easy to answer for coax, as it's all explained here: http://en.wikipedia.org/wiki/Coaxial_cable#Signal_leakage

and USB in isochronous mode is merciless, nothing to do w/ USB storage solutions that work in block mode and include EEC lowering the speed if the cable quality is poor. Everyone raves about the WW USB cable, time to hear what it's all about.

Another interesting thing is that AQVOX in their FAQ advise to disable Spread Spectrum...I've had Gigabyte mobos for a while, and these didn't provide any option to disable SS. I even whined to their tech.support(hoping to fix dropped video frames when using Reclock) who once gave me a beta BIOS to disable it, but it didn't allow any overclocking...stalling at POST. They told me that SS helped to stabilize o/c and that it was company policy to not allow disabling it.

I've had troubles stabilizing my 2.66Ghz Q9450@3.5Ghz overclock on this high end Asus mobo, so I left the CPU/PCI-E SS on...together w/ the NB/CPU "clock skew" on "auto". But now that it's dead stable, I just disabled them all...the mouse is sluggish, but I'd swear that the SQ has improved. It certainly hasn't gotten worse