Skip to main content

Is a distribution box preferrable in a situation where I chain my word clock this way :

(AD-16x WC out) > (DA-16x WC in) > (DA-16x WC out) > (MOTU 2408 WC in)

...where I import the digital output of the AD-16x via the 2408's ADAT inputs and likewise take the ADAT outputs of the 2408 and send them to the inputs of the DA-16x?

Is there a better way to chain the WC or is chaining a bad idea in any event ? Until I invest in a few firewire cards I need to have the 2408 in the chain and use it as an input/output bridge for conversion...

Should I be concerned?

Is there a better way to do this?

Topic Tags


johnwy Fri, 01/07/2005 - 23:18

Long daisy chains of word clock can get sloppy because each piece of gear has an inherent delay, and it is cumulative. Long wordclock cables can cause problems too. Basically, the longer the chain (and the longer the cable), the more delay of the wordclock signal and increase of clock jitter. Delay + Jitter = high frequency distortion artifacts.

As for your current set up, you should be fine for the moment. Just keep your word clock cables as short as possible.

ghellquist Sat, 01/08/2005 - 14:59

The designed for way to do a wordclock is to chain them using T-connectors, and have a terminator at the last card. Never chain them in -- through the box -- out.

Here is what you should have

Master clock box sending word clock out
T-connector -- in to first box receiving word clock
T-connector -- in to next box
T-Connector -- in to next box

If you do it that way, you should turn off any switchable terminators in the the boxes receiving word-clock (some boxes has that). Keep the cables reasonably short, each less than a meter if possible.

The terminator looks like a small "cap" whith a BNC contact. The cable should be 75 Ohm coaxial (one of several very common coaxial cable types). This is not expensive stuff, but you might need to go to a "radio amateur" kind of place to find it, given my experience. Or find a general electronics supply firm. They can supply the terminator, cable, connectors and the special tool used to place them on the cable. In that way you can make the cable to exactly the length to fit perfectly back in the rack.

There should be no problem at all with 5 or more receiving boxes. The delay throught the cable is basically the speed of light so that should not influence you.

Funny thing is, I visited a pro sound shop a few days ago. The persons there had never even seen a T-connector, even less a terminator. And they do sell quite a bit of ProTools TDM systems, and proud themselves on beeing experts on installing them.

Basically, what you have been doing, and the advice given to you in a previous post, is wrong. It might work anyway, given that wordclock really is a very robust thing.

Good luck with your recordings.

Gunnar Hellquist

anonymous Sat, 01/08/2005 - 15:56

Gunnar....thanks very much !!! Great advice and I will hit radio shack and grab up a few BNC "T" type connectors. I’ve seen them but never knew what they were for...certainly something for video but I didn't know they could be used for chaining clock info. Thanks again for the very informative post...cheers

ghellquist Sun, 01/09/2005 - 02:44

Rereading my post, I see one thing that might be unclear.

The terminator should be 75 Ohm type, same as the coax cable (they come in different types). Basically it is just shorting the central conductor to the screen using a 75 Ohm resistor.

From my vague recollections of the theory, it is about the signal progating as a wave through the coaxial cable. The combination of resistances and capacitancs in the cable is what becomes the 75 Ohm. The wave then happily moves along the cable and is then happily "eaten" by the terminating resistor (they didn´t use quite those words in school). Unless the cable is terminated, the wave is mirrored back from the end of cable and interferes with the next signal. This would probably result in jitter in the word clock.