diyAudio

diyAudio (https://www.diyaudio.com/forums/index.php)
-   Digital Line Level (https://www.diyaudio.com/forums/digital-line-level/)
-   -   XMOS-based Asynchronous USB to I2S interface (https://www.diyaudio.com/forums/digital-line-level/188902-xmos-based-asynchronous-usb-i2s-interface.html)

reo 3rd December 2011 02:24 AM

Quote:

Originally Posted by reo (https://www.diyaudio.com/forums/digital-line-level/188902-xmos-based-asynchronous-usb-i2s-interface-post2803903.html#post2803903)
For such a high price for that cchd-957 one would assume it was better than...
Having the NDK on this card it's one less thing to ponder on .

On another note . , i have installed a standalone 5V regulator replacing my PC'S usb2 output to WaveIO, it was not necessary after all, similar quality sound coming out either way. My PC's power supply is highly efficient and never drifted enough to substantiate a standalone regulator, least now i know, next thing i will try out is, isolating my usb2 cables from the pc with it's own standalone 5V, that's if it has noise, i wonder about that as well ,PH. regulators vs Trident might yield different results,i use them to control Mojo2 and that's another story . i'm thinking this card is already up to snuff, if one is in synchronous mode and use a separate regulator driving the card if incoming power supply is questionable, then both mods are a worthwhile move to reduce noise.
regards
rol


triode_al 3rd December 2011 10:32 AM

Quote:

Originally Posted by 4est (https://www.diyaudio.com/forums/digital-line-level/188902-xmos-based-asynchronous-usb-i2s-interface-post2803391.html#post2803391)
Upgrading clocks is certainly an option, but one needs to be careful on their selection. The specs on the clock may not tell the whole story. A case can be made for an isolated power supply and clock too.

Would a TENT VXCO clock (Guido Tent) of 11.28960 not be a perfect replacement? Extreme low jitter. The separate power supply can be bought too. You can use his special high frequency shunts too.
For playback of standard CD's (44.1 KHz and occasional special recordings of 88.2 KHz) this replacing the main clock is enough to start off.
Own recorded (i.e. sampled LP's) can stick with the other clock - that is used anyway to record if I understand correctly). This then is a later candidate.
  • Should the 12.0000 MHz clock have to be replaced too?

albert

Lorien 3rd December 2011 01:44 PM

11.28960 MHz would not be a replacement for any osc on that board. You'll need oscillators with two values only: 22.5792 MHz and 24.576 Mhz.
13 MHz (not 12 Mhz) crystal replacement: it's used by XMOS's PLL to lift the internal freq at 400 MHz. If you want, you can replace the whole circuit with any 13 MHz oscillator you like that works at 3.3V (to take advantage of on-board PSU) though, taking into consideration the internal XMOS architecture, all the signals are realigned using the corresponding Master clock signal which in WaveIO's case are: 22.5792 MHz and 22.576 MHz parts.
Cheers,
L

triode_al 3rd December 2011 01:51 PM

Quote:

Originally Posted by Lorien (https://www.diyaudio.com/forums/digital-line-level/188902-xmos-based-asynchronous-usb-i2s-interface-post2804706.html#post2804706)
11.28960 MHz would not be a replacement for any osc on that board. You'll need oscillators with two values only: 22.5792 MHz and 24.576 Mhz.
13 MHz (not 12 Mhz) crystal replacement: it's used by XMOS's PLL to lift the internal freq at 400 MHz. If you want, you can replace the whole circuit with any 13 MHz oscillator you like that works at 3.3V (to take advantage of on-board PSU) though, taking into consideration the internal XMOS architecture, all the signals are realigned using the corresponding Master clock signal which in WaveIO's case are: 22.5792 MHz and 22.576 MHz parts.
Cheers,
L

I see. Had it wrong in my head :headbash:
Then as TENT XO, there is indeed available 22.5792 MHz and 24.5760 MHz . Is that better then, can these be fed from a separate supply without disturbing the existing architecture?
Or is the advantage limited?
albert

Lorien 3rd December 2011 02:15 PM

PSUs can be improved but you have to desolder the main oscillators (22.5792 and 24.576) from WaveIO board using a hot air station (preferably). Also, I could leave them unmounted, at your request. This way, you can build your nice setup on top of WaveIO board using the shortest connections you're able to provide.
Limits: WaveIO has common ground with the source and this is something I will address on the revision. Even so, in my opinion it worths the try.
L

regal 3rd December 2011 03:46 PM

Quote:

Originally Posted by 4est (https://www.diyaudio.com/forums/digital-line-level/188902-xmos-based-asynchronous-usb-i2s-interface-post2803391.html#post2803391)
Upgrading clocks is certainly an option, but one needs to be careful on their selection. The specs on the clock may not tell the whole story. A case can be made for an isolated power supply and clock too.

-with hirez asynch usb where the usb ground(computer) is shared with the clocks power supply, I think there is a limit of return as far as clock quality. The power supply regulation is only as good as its reference which in case of hirez asynch usb isn't going to be spectacular even with a separate power supply it really isn't separate. Until there comes a reliable USB 2.0 480 isolation for asynch operation this will be an issue with all these devices.

Now RBCD asynch USB1.0 is a whole different ballgame where the isolation on the asynch side is reliable and proven, this opens up an entirely different realm of quality. For example IME a usb side isolated asynch USB1.0 to spdif device I bought sounds better with dithered/downsampled hdtracks than my other non usb side isolated asynch USB2.0 to galvanically isolated i2s device.

This is why I still think we are jumping the gun with USB2.0 "hi-rez" usb, should be designing and optimizing USB1.0 asynch USB RBCD, but unfortunately hirez marketing sells with these things (even if it doesn't as far as actual music.)

flocchini 4th December 2011 11:46 PM

Working Synchronously
 
Thanks to all your help I was able to muster the courage and hook up Lucian's XMOS board's clock to my Buffalo II. Works well.

The only unexpected outcome is that the Volumite stopped working. I think it gets a clock feed from the BII. Any thoughts would be appreciated.

Best

Bob

reo 5th December 2011 01:53 AM

strange , i re installed Volumite to see if it works, and it does ., maybe in your case it's a broken connection from BII to Volumite.
regards
rol

barrows 5th December 2011 02:28 AM

No problem
 
using the Volumite here with synchronous clocking. Maybe you should re-check all connections.

flocchini 5th December 2011 02:46 AM

Will do

I checked for 5V at the volumite - will check the other two connections


All times are GMT. The time now is 08:54 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio

Wiki