Hi Patrick,

There are only three rules for word clock cables from a masterclock:

1. Use good quality 75ohm impedence cable.

2. All cables should be of equal length.

3. All cables should be a short as possible.

Explanation: If you've got a word clock cable coming out of you Lucid that's 50cm long and then another that 3 meters long, the piece of equipment that's getting it's clock from the 50cm cable is going to get its clock signal fractionally earlier than the piece of equipment getting it's signal from the 3 meter cable. Obviously the amount of time the signal takes to travel the extra two and a half meters at the speed of light is tiny, probably only measurable in pico seconds. However, jitter itself is usually measured in pico seconds and as the whole point of adding a masterclock is to reduce jitter, it's definitely worth keeping the cable lengths the same.

The reason for rule 3 is that you want the signal to be as clean and strong as possible. But rule 3 is definitely secondary to rule 2. I've arranged my studio so that my clock cables are all 1 meter. 3 meters is a little long but I don't think you will have any problems.

Greg

Thanks very much, Greg.

I've been racking my head (now maybe that's one that I should put in for worst pun of the year!) trying to figure out a way to set things up to get shorter cable length, but even at best, it will have to be 2 metres. Damn, this place is so small! Everytime I buy something, I end up building or re-building shelves to get it in. (You know what they say about Japanese apartments? Well, it's true!) (If it seems funny that the lack of space prevents me from having shorter cables, it's because I have to spread the gear out too much, in order to have room to walk).

oh well, some day I'll have something I can call a "studio."

In the meantime, one more question. In the case of the machines that are placed very close together, what can I do with the extra length of word cable? Can I coil it or figure-8 it, or do I just have to send it around somewhere and back?

thanks

Oh yes, another question. For the gear that's connected to the master clock, will longer digital audio cables, or varying cable length, also have a negative effect on jitter?

I'm not questioning Greg's expertise, i'm just wondering if this is a unanimously held opinion, or a matter of debate. Can one meter really be measured or perceived at light speed? Does anyone else agree or disagree? I'm way too lazy to do the math, and wouldn't know what to do with the results! ;)

At the speed of light, one foot of wire equates to about 1 nanosecond of travel time. At 96KHz, audio samples are taken about every 10,000 nanoseconds.

Jitter is generally measured in picoseconds, so typical jitter specs are well under one nanosecond. Opinions vary on what level of jitter is audible. But jitter isn't the issue here.

The basis for using equal length wordclock cables is that it keeps your data synchronized as it moves through your studio. But in reality, while the delay of a few extra feet of cable is real, it is a nearly constant delay, which makes it pretty harmless.

The bottom line is that the two pieces of equipment have a tiny delay between them. But as long as all of your gear has a constant, jitter-free clock, the tiny delay between them isn't a factor.

Look at it this way: Electricity travels about a million times faster than sound. If you move a microphone back by one millimeter in your studio, you've added more signal delay than you would with a half a mile of wordclock cable.

The latency in most digital gear today is much, much larger than any delay that could be caused by varying your clock cable lengths.

There's certainly no harm in using equal cable lengths, assuming that they are reasonable lengths of quality cable. But there's no reason to worry about a few feet of difference either.

In general, I'd reverse the priority of Greg's "rules" 2 and 3.

Thanks Gilliland,

One less thing to agonize about! :p