S/PDIF termination; practical guide? - Page 3 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Line Level

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

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 25th August 2011, 07:53 PM   #21
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
Question still is, how does this affect the operation of the D/A? The Hiface had a terrible-looking square wave (probably due to the transformer choice), but the analog output of the D/A looked fine.
__________________
You might be screaming "No, no, no" and all they hear is "Who wants cake?" Let me tell you something: They all do. They all want cake.- Wilford Brimley
  Reply With Quote
Old 26th August 2011, 01:15 AM   #22
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Quote:
Originally Posted by marce View Post
Again I am collating some links and references as well to go with my answer, as I want it to be as compherehensive as possible, as I quite often get shot down in flames for some of my views, also the simulations will show the effects of cable length and terminations.

Remember, the 1st thing after the resistive termination is a CMOS Schmitt trigger.
Think about applying signal gain and limiter at the average data level ( AC coupled). This is good because the decision point isn't fixed, it has hysteresis. which means there is significantly more noise and glitch immunity than standard logic gates. see Fig. 4 for the visually inclined
http://www.fairchildsemi.com/an/AN/AN-140.pdf
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust

Last edited by infinia; 26th August 2011 at 01:36 AM.
  Reply With Quote
Old 26th August 2011, 09:44 AM   #23
Zoran is offline Zoran  Serbia
diyAudio Member
 
Join Date: Jan 2004
Location: Belgrade
Sy,
I used HiFace interface with I2S bus as well as for spdif.
That fine product is working good even without mods, comparing to others from same fashion.
The software was installed without any problem for win (standard PC and mac bootcamp win) and mac of course.
I didn't expect this will happened from first try and I was "impressed" a bit, because it is not the usual case. Somehow from that moment I knew that it is very good little device.
.
I opened the unit and find that clock generators are not the same, booth, like it was at the beginning of serial production, for my opp. those was not good as it was? Why they accept that policy I don't know. they put smaller and not accurate like it was when the device launched?
.
I am not sure, I will check it again,
BUT
with win, I2S bus looks like it should be, Latch L/R comes prior one cycle of Sys CLK to the MSB...
At mac osx that is not the case, and Latch is starting the same time, not prior but in sys clk time. That is could results from losing MS bit at the L/R ?
I saw "1" at the place of data bus on the logic analyzer scope, at the place that should be "0", MSB
anyway I am not 100% sure and that should be checked and if so reported.
.
I add own power supply for the module, without using computer PS.
What I found is that spdif output is TTL 5V and insulated with trafo.
I put "P" or "T" network, not sure at the moment, to att. at the right level.
google for
.
Unfortunately device NOT accepting, usb digital isolation at the input, so I add at the output, following by re-clock cir. and after that diff. driver for the cables... (in between I split digital bus from standard L/R into L/L and R/R.
So I made mono I2S diff. outputs as wel as standard.
Judging by the cope there is much, much better signal, when transferred differential line against coaxial.
.
About the I2S bus I tried with coax cables, at first, but I switch to diff. driver
with a adding a resistor in path of I2S bus, I spot much better signal integrity
at the scope. That should be done wit each line because values are not the same... best signal integrity shape i trimmed and replaced with R.
.
cheers
__________________
###
  Reply With Quote
Old 26th August 2011, 09:58 AM   #24
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
And after all that, what was the improvement in the output of the DAC?
__________________
You might be screaming "No, no, no" and all they hear is "Who wants cake?" Let me tell you something: They all do. They all want cake.- Wilford Brimley
  Reply With Quote
Old 26th August 2011, 10:11 AM   #25
marce is online now marce  United Kingdom
diyAudio Member
 
Join Date: Jun 2007
Location: Blackburn, Lancs
The art of signal integrity is preserving the switching points of the downstream logic on the rising and falling edges of the square wave, and keeping the time interval the same between each change of state, especially for clocks. The rest in not as critical but still important. The ringing stresses the silicon with the extra energy, causing more heat generation, amongst other things. But as long as you keep the switching points clean the wave will be recovered.
The link to the Tektronix app note shows some DDR memory waveforms, and the attached picture shows a snapshot of a real one taken on a 13GHz scope, you can clearly see the high frequency noise from simultaneous switching and SMPS etc causing ground bounce, superimposed on the wave, this is from a 14layer-5 ground plane layout. The right hand image shows some more nice and clean square waves. DDR interfaces tend to be one of the fastest these days and are becoming the memory standard, hence using it as an example. It is laying out these sort of high speed interfaces and seeing them work, that makes me a firm believer in "bits is bits", for the slower frequency interfaces it is even harder to ruin them provides good design practice is followed. Some interfaces have 100-200+MHz clocks, using good engineering practices, layout etc we should be able to sort the lower frequency interfaces used in Audio.
http://www.isotest.es/web/Soporte/Fo...20de%20DDR.pdf
Qusp,
software a bargain at 16K on top of the base software 20K, luckily work buys me these toys, and pays for the training, which is ongoing as Signal Integrity and high speed logic is back to school and back to basics, physics, and thinking in waves traversing the PCB instead of electrons. Its fun and adding a lot to PCB design, though not making it any easier. Still got a lot of training to do, at the moment I'd say I'm 25% there, but even that gives me a tool and the confidence to release a layout and be pretty certain the signals are going to get from a to b, whereas before it was follow best practice and hope.
Attached Images
File Type: png a0_ddr2.png (7.6 KB, 198 views)
File Type: png ddr-d0-d1-a1-a2-clko0p.png (40.7 KB, 204 views)
  Reply With Quote
Old 26th August 2011, 10:35 PM   #26
Zoran is offline Zoran  Serbia
diyAudio Member
 
Join Date: Jan 2004
Location: Belgrade
this was significant impact on the sonic result
in the better way, for my opp..
__________________
###
  Reply With Quote
Old 27th August 2011, 12:53 AM   #27
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
excuse me, Marce in case you missed it in my post, the Schmitt trigger part is already inside the Rx data interface input/s, read up on the data sheet about S/PDIF data input you will see it's not single ended logic levels.
Attached Images
File Type: png CS8416.PNG (104.5 KB, 181 views)
File Type: png SPdif.PNG (55.5 KB, 181 views)
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 27th August 2011, 07:54 AM   #28
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
hey marce, oh indeed, if you are involved in any area related to particle and subparticle states then indeed the research is ongoing. the Bose-Einstein condensate matter-wave experiments and research throws a real curveball.

sounds like you have a pretty fun job. btw most sabre dacs run with pretty fast clock speeds too for mclk, my ackodac and buffalo 2 both run at 100mhz
  Reply With Quote
Old 27th August 2011, 08:19 AM   #29
marce is online now marce  United Kingdom
diyAudio Member
 
Join Date: Jun 2007
Location: Blackburn, Lancs
Hi, Infinia, I think we may be at crossed purposes, I started my post ages before I posted it. The example showns a real square wave (that isn't very square, between the two dots the time interval is 1.284ns), yet the DDR2 interface still works. I was pondering last night whether to do a discourse on changes with measurment and control that I have seen over the years, 4-20mA current loops, the increase in electyrical noise in the envoironment and how most electronics is moveing away from analogue, to mainly digital and digital transmission, with small areas of analogue. I am refering to digital in general, I dont think digital in audio is anything extra special, its digital, just like analogue is analogue to me, I go for 100% perfection in any design I do (I dont achieve it, but still strive for it). I sometimes wonder whetehr sometimes in Audio peiople loos some focus on design and reality, we are not creating an instrument, we are creating some electrical equipement to provide the best possible fidelity in playback of music, speach etc within a tightly defined frequency range.
One of the biggest changes in digital signal transmission has been the change from parallel links, to serial and now because of the advantages low voltage differental signaling, I did VME and G64 motherboards in the late 80s, where all busses were parallel and serial data transmission was the good old RS 232 interface, now once a signal is going off board it will be some form of serial transmission, no more great reals of ribbon cable all over the equipement.
But when it comes to magical effects in digital or analogue I tend to get skeptical, especially when good engineerinig practice (ie I hear its better but have no measurements to show whats happening, head in the clouds brigade) I tend to get cynical...
What I am saying is I do high speed digital, analogue, RF layout every day, whether its audio, communications, video, control or some other aspect of electronics, to me its electronics and should be layed out to the best of both my and the tools I use ability.
Electronic Engineers, well now thats a different kettle of fish
  Reply With Quote
Old 27th August 2011, 08:32 AM   #30
marce is online now marce  United Kingdom
diyAudio Member
 
Join Date: Jun 2007
Location: Blackburn, Lancs
Qusp,
basicly I'm a PCB designer, there are a few of us out there, usualy found in the dark recesses of some office, gibbering away in a corner. !00-200MHz clocks are common these days, look at every PC motherboard. RF is common place and not the dark art it use to be, look at the prolifaration of rf based communication. When you grow up doing such a job, you dont notice wizzy bang developments as you are in the middle of it doing your job, so it dosn't seem so fantastic. The nearest I've done to particles is some work for Daresbury, a firm I worked for did some electronics to control the magnets, and some PCB's for electron microscopes. Again sorry to sound cynical (I am this morning, got a weeks holiday that I thought was gonna be peaceful, and I wake up to a house full of grandkids whove come to stay for a week, with respective mothers etc).
I have been lucky to work on a lot of different projects and products over the years and now do some real fun communications stuff.
I am just getting back to physics with Ralph Morrisons book the field of electronics, I'm an old dog having to re-learn old tricks
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
WTT TP S/PDIF transceiver for S/PDIF 4:1 MUX/Receiver Module merlin el mago Swap Meet 0 28th June 2011 10:48 AM
Load termination cathode_ray Instruments and Amps 5 23rd August 2007 10:57 PM
S/PDIF Optical to S/PDIF RCA Replacement PrecisionAudio Digital Source 2 11th February 2007 09:35 PM
Michael Fremer's Practical Guide to Turntable Set-Up fcel Analogue Source 0 2nd September 2006 05:53 AM
Interconnect termination hbarki Parts 0 10th April 2004 01:51 PM


New To Site? Need Help?

All times are GMT. The time now is 10:44 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2