[time-nuts] Sound cards

Magnus Danielson magnus at rubidium.dyndns.org
Tue Jan 13 18:01:13 EST 2009


Lux, James P skrev:
>>> optical media, the latter being attractive in further isolating PC
>>> noise from any measurement setup. And of course, a manufacturer's
>>> evaluation board is much better documented and more suited to
>>> measurement-specific mods than a random sound card.
>> The optical link commonly being used for S/P-DIF is TosLink
>> and it seems like it can be the cause of many problems. It
>> seems like some care in doing the optical link setup is
>> needed. I have never digged into why the optical links have
>> that problem. I can only guess, but bad optical coupling
>> seems reasonable. The multimode "fiber" seems to be leaving
>> one or two things to ask for.
>>
> 
> Consumer audio optical links (TOSlink) are 1000 micron (yes, it sounds better than 1mm) plastic fiber, which is fairly high loss (1000 dB/km), which is fine for 1-2 m cables.  TOSlink is actually a trademark of Toshiba
> http://www.toshiba.com/taec/components/ProdLineGuide/toslink.pdf describes all.
> 
> TOSlink has worked fine for me, except in one case, where the cable was flexed a lot, and it eventually developed cracks and the loss shot up.

Trouble is, many want some 10 m runs. The unbalanced one into a good 
cable should be able to do much longer runs. The Phono (aka RCA) 
connector isn't really suited to the task, but there is pretty good 
limits on the slopes, so it should be fine. Propper BNC with good cable 
handles 90m of HD-SDI signal so I don't think I am stretching my 
knowledge beyond unreasnoble limits here... :)

Using 4,5 GHz BW cable for AES/EBU testing is kind of interesting... :)

> It should be relatively trouble free. The connector is a positive mate and keyed, and as long as you don't get debris in the hole in the chassis side, it should work fine.
> 
> It's not a bad interface, although, I think not well suited to many mate/demate cycles (I don't know for sure, but the trusty 1/4" phone plug is a pretty rugged design). One data sheet I have says 500 mate/demate cycles.
> 
> 
> I was amused when the guy at the stereo store tried to sell me on RF shielded TOSlink cables, claiming it would provide more clarity and definition in the sound.  Uh-huh..   Sort of like the green marking pen for the edges of your CDs to reduce internal reflections, etc. (I, of course, would only use the finest brush made from selected hairs of Tibetan mountain goats to apply a dye made from chlorophyll molecules selected using an electron microscope by trained technicians, etc..  A "marking pen"? My $14000 speaker cables supported on carefully oriented pure fused silica supports would wither in shame.)

You must have gold-plated connectros for your TOSlink, you KNOW that. :)
(Yes, those cables exists!)

Cheers,
Magnus




More information about the time-nuts mailing list