Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter
Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 18th May 2012, 03:15 AM   #551
glt is offline glt  United States
diyAudio Member
 
Join Date: Oct 2004
Quote:
Originally Posted by qusp View Post
he didnt say that, yes the demo board comes with 40Mhz but the dac was specified to run MAX 192 and async mode. the maybe yes, maybe not 1db loss he mentions will only be because of increased thermal, voltage and current noiseI would think. seems its a bit of a different sitruation we are now in, using the dac beyond its initial spec in sync mode with 384+ sources that didnt exists at the time of the dacs inception

...

The message is that Crystek's best clocks are available in this range...
__________________
www.hifiduino.wordpress.com
  Reply With Quote
Old 18th May 2012, 03:21 AM   #552
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
but we arent talking about crystek clocks =) for the higher speeds not a chance would we get the MOQ together for a special order and anyway I think there are other bounties to be found than crystek

Last edited by qusp; 18th May 2012 at 03:28 AM.
  Reply With Quote
Old 18th May 2012, 03:38 AM   #553
iancanada is offline iancanada  Canada
diyAudio Member
 
iancanada's Avatar
 
Join Date: Dec 2009
Location: Toronto
Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter
Quote:
Originally Posted by qusp View Post
he didnt say that, yes the demo board comes with 40Mhz but the dac was specified to run MAX 192 and async mode. the maybe yes, maybe not 1db loss he mentions will only be because of increased thermal, voltage and current noiseI would think. seems its a bit of a different sitruation we are now in, using the dac beyond its initial spec in sync mode with 384+ sources that didnt exists at the time of the dacs inception


yup, i'll be removing mine once the testing phase is over ; I probably should have actually just sent my dac boards down to acko when I sent the MCUs for update given I cant use them while they are gone anyway. but while testing different stuff out its handy to be able to use either mode and connect sources without HAVING to have the fifo inline, particularly while its still not updated to accept higher speeds
I'm thinking about come back to the 3 pin dip connector for the reserved position, so that I can switch between sync and async mode quickly for compairing the sound between... .
  Reply With Quote
Old 18th May 2012, 04:13 AM   #554
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
I think the only way to do such a thing accurately is with a switch tbh. i'm more comparing titan with/without fifo and i'm still stuck really, much as I like the idea of the fifo in my main rig; it cant work with anything but my headphone amps until a multichannel fifo comes about. is there any way I can use 2 fifo main boards, for 4 channel with a single dual clock board? in my previous experiments I have found sync mode to be a bit more cohesive, soundstage is more solid, similar to your impressions. how much of this is because of being able to use lowest setting on PLL i'm not sure

the problem with setting up the fifo clock boatrd as the master clock without this functionality means I would have to devise a way to switch master clocks and have a second master clock for multichannel operation, because the fifo clock board will not work standalone without the control signals from the fifo main board.

complicated...

its doubtful performance is better in async mode given the quality of clock and i2s source is allowing you to use lowest, but I suppose it might be subjectively preferred. if it were me in your position, I think I would just turn my gaze back towards clocks and forget about the distraction of sync vs async

Last edited by qusp; 18th May 2012 at 04:24 AM.
  Reply With Quote
Old 18th May 2012, 04:40 AM   #555
hochopeper is offline hochopeper  Australia
diyAudio Member
 
Join Date: Feb 2009
Location: Brisbane, Australia
Quote:
Originally Posted by qusp View Post
is there any way I can use 2 fifo main boards, for 4 channel with a single dual clock board?
I'm certainly not the expert here but ... isn't the final reclocked i2s output leaving from the dual xo board? I agree you're in a bit of a bind here for multichannel.

Are you talking 2 x ES9012 or 4 ch output from 1 x ES9018? I think I know which but wanted to ask.

Quote:
Originally Posted by qusp View Post
in my previous experiments I have found sync mode to be a bit more cohesive, soundstage is more solid, similar to your impressions. how much of this is because of being able to use lowest setting on PLL i'm not sure


Quote:
Originally Posted by qusp View Post
if it were me in your position, I think I would just turn my gaze back towards clocks and forget about the distraction of sync vs async
I concur.

Last edited by hochopeper; 18th May 2012 at 04:44 AM.
  Reply With Quote
Old 18th May 2012, 05:46 AM   #556
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
ahh yes of course youre right, wasnt thinking that through, not something I had considered before, so its impossible, cant use this in my rig for speakers. perhaps I could clock another set of flipflops on a little daughterboard, but sounds like a pita.

Chris, you know what dacs ive got..dont you? 2 x 9012
  Reply With Quote
Old 18th May 2012, 06:18 AM   #557
coolhead is offline coolhead  Malaysia
diyAudio Member
 
Join Date: Mar 2010
Quote:
Originally Posted by iancanada View Post
I hooked up the FIFO KIT and B III into synchronized configuration yesterday. It runs great without any problem. The setup is easier than I thought. Please see the attached pictures for details(find the names for the meaning of the pictures).

Step1: Remove the CCHD950 OEM 100Mhz local oscillator
After the CCHD950 was removed, I found TP has already left a backdoor for the sync mode. There is a 3 pin socket underneath, GND, VDDXO and CLOCK
Tips: protect the surrounding component with thermal tape and heat the back of the PCB first to avoid any possible damage.
Ian,
That is awesome job. Thank you for sharing.

I have 1 question: if I have a system with BIII duo mono, in theory, can the FIFO clock feed to 2 boards? Or i should use another clock board for this purpose? Thank you.

-coolhead

Step2: Assemble a U.FL socket
I still think U.FL socket is better for the clock signals, so I just soldered it on the BIII board. Signal (MCLK input) to CLOCK and shield to GND. However, using the 3 pin dip connector will make it easier switching between sync and async mode, so later on, maybe I have to give up the U.FL socket and switch back to 3 pin dip connector.

Step3: set up the Dual XO Clock board into double speed mode with 45.1584 MHz and 49.1520 MHz oscillators.
Please see the attached PDF guidance for details. Please be notice, jumpers have to be set exactly as what directed in the doc. Actually I finished this document 6 weeks ago, but I have to confirm it by myself before I public it.

Step4: connecting the cables
I2S output to I2S input, MCLK output to MCLK input.

Step5: Enjoy the music.

It runs right away without any hesitating. The LOCK LED keeps lighting all the way indicating it synchronized with the input I2S stream unconditional even I set the bandwidth switches to 'lowest' and run music at 192KHz from an USB.

According to my understanding on the ESS9018 sync mode:

1, The DAC is running by the MCLK directly no matter what mode it is, sync or async, DPLL is only used for re-sampling job according to the ASRC algorithm , it doesn't generate any real clock;

2, In the async mode, DPLL re-samples the input I2S asynchronizely, the bandwidth of the DPLL could not go infinite to lock with input I2S due to the phase difference between clocks. With the limited bandwidth, all jitter below will be path through and converted into re-sampling jitter and all above will be removed and determined by the MCLK.

3, In the sync mode, DPLL locks to input I2S un-conditionally with infinite (even better than the 'lowest') bandwidth because the I2S is generated by same MCLK with fixed relationship of 256*,512*,1024* and so on. In this case, the DAC jitter is 100% determined by MCLK itself, so the finial sound will be highly depend on the FIFO clock.

In my configuration, ESS9018 sounds great in both of the modes, sync and async. I even couldn't tell which one is better. But there is slightly difference on the sound style. Async is more like ESS9018 J while sync is more like classical high-end DACs. I usually don't talk much about the sound, because too many psychological factors and personal feelings mixed. I'll left this part to others, trust your ears and try to find out the best parameter settings of 9018 for the sync mode.

90.3168MHz and 98.3040MHz would be more optimized frequency for ESS9018, but itís very hard to find out the really nice XOs with those frequencies so far, except the si570 based real time programmable low jitter XO I could think about.

I added an additional piece of heat sink and make it facing up. The stable working temp on the surface reduced from 60C to 53C J.

Have a nice weekend.

Ian
  Reply With Quote
Old 18th May 2012, 07:46 AM   #558
coolhead is offline coolhead  Malaysia
diyAudio Member
 
Join Date: Mar 2010
Sorry guys, I realized that I screw-up the reply, this is my message.
________________________________________
Ian,
That is awesome job. Thank you for sharing.

I have 1 question: if I have a system with BIII duo mono, in theory, can the FIFO clock feed to 2 boards? Or i should use another clock board for this purpose? Thank you.

-coolhead
  Reply With Quote
Old 18th May 2012, 03:35 PM   #559
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
it can, but you wouldnt be able to use the spdif board, or if you did you would have to work out some other way of sending Mclk back to it. I was meaning to mention that before you go to press on the next run mate, you should really duplicate one of those clock outputs, probably doesnt need another driver if thats a big deal
  Reply With Quote
Old 18th May 2012, 11:24 PM   #560
iancanada is offline iancanada  Canada
diyAudio Member
 
iancanada's Avatar
 
Join Date: Dec 2009
Location: Toronto
Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter
Quote:
Originally Posted by coolhead View Post
Sorry guys, I realized that I screw-up the reply, this is my message.
________________________________________
Ian,
That is awesome job. Thank you for sharing.

I have 1 question: if I have a system with BIII duo mono, in theory, can the FIFO clock feed to 2 boards? Or i should use another clock board for this purpose? Thank you.

-coolhead
There are 2 U.FL MCLK output sockets on the Dual XO Clock Board, so I don't think you have any problem feed them into two dacs. You don't need feed the MCLK to S/PDIF Interface Board if you don't use the DIT function, or you may connect it directly to the U.FL socket on the FIFO board. However, your have to make the I2S cable with branch by yourself.

Last edited by iancanada; 18th May 2012 at 11:35 PM.
  Reply With Quote

Reply


Asynchronous I2S FIFO project, an ultimate weapon to fight the jitterHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
exaU2I - Multi-Channel Asynchronous USB to I2S Interface exa065 exaDevices 1406 4th June 2018 01:47 AM
XMOS-based Asynchronous USB to I2S interface Lorien Digital Line Level 3175 1st April 2018 04:24 PM
DAC chip selection + I2S jitter questions drwho9437 Digital Line Level 2 26th July 2010 12:50 PM
Simple FIFO to I2S CPLD, for MCU players / reclocking KOON3876 Digital Line Level 21 19th September 2008 04:00 PM
asynchronous reclocking and low jitter clocks ash_dac Digital Source 3 8th February 2005 09:22 AM


New To Site? Need Help?

All times are GMT. The time now is 02:54 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki