tierces

22:06 Jul 9, 2010 

Thanks for keeping me right, Tony. I have an elevated sense of etymology and sometimes it sets me off.

multiple
1640s, from Fr. multiple, from L.L. multiplus "manifold," from L. multi- "many, much" + -plus "fold," from base of plicare "to fold, twist;" see ply (v.)). Multiple exposure first recorded 1923.

That said, "triad" is rather more common than "triplet" as characteristic of signal wiring.

21:55 Jul 9, 2010 

Sorry, Gene, but that particular argument doesn't hold water: there is no notion of 'ply' in triplet. The term is often used in all sorts of technical contexts, simply to mean 'something that is combined in threes'.

Thnaks, Gene! I too am passionate about etymology, and you can see on closer analysis that the idea of 'ply' comes from the '-fold' idea, but has long lost any direct connotation; and we do often talk about 'doublets' in electrical contexts.

However, the mere fact that there are 3 wires instead of 2 means that the interference signal cancellation achieved by twisting a pair of conductors inevitably won't work in the same way, since the symmetry is of course lost.

I take your point about 'triad' being slightly more common than 'triplet' in relation to signal wiring; however, you do have to be extremely careful, since there is a lot of 'noise' out there, and once you have eliminated the red-herrings and fortuitous collocations, the figures appear to be much closer — too close, in fact, for any difference to be statistically meaningful.

18:23 Jul 9, 2010 

If the purpose is to protect the signal from the E/M interference (as it is in a twisted pair), the direction and the amount of twist must be maintained along the entire length of the bunch.

http://www.tewire.com/noisred.php

And "triad" is used in that case.

I guess the choice is due to the fact that "triplet" has a "ply" in it, which is a wrong idea in case of a signal wire.