What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Twisted pairs??

szicree

Well Known Member
I'm hoping somebody can give me a quick explain on when and why to use a twisted pair of wires. Please don't tell me to read Bob's book, cuz I have and I still don't quite get it. Thanks in advance.
 
Twisted pairs are typically used when you're trying to relay a bunch of differential signals. You twist the pairs of wires together that go with each other (the + and - of one particular signal, for example). By doing this, you reduce what's called "crosstalk" between pairs of wires because the field keeps reversing at every twist and things tend to cancel out over the length of the cable. Also, it brings the wires very close to each other and reduces the capacitive effects in the first place. Between this, the twisting and being differential, you can propogate twisted pair signals for very long distances with very good SNR at the other end (CAT5 for ethernet and RS422 are good examples of this.....they can go VERY long distances and still be reliable).

Offhand, I can't think of anywhere I'm planning to use twisted pair other than maybe some of the interconnects on the back of the instrument panel. For example, it wouldn't really do any good to do twisted pair on any of the RS232 stuff since that's not a differential signal and you loose the benifit of twisting in the first place. I'll defer this part of the question to someone who's actually wired an airplane before :D
 
Last edited:
John... a twisted pair is still effective on non-differential signals if one of the wires is the sole signal return for the other wire (i.e. a dedicated ground)

A microphone input comes to mind.....

Glideslope and VOR deviation signals to an indicator are an example of true differential signals we deal with.

gil in Tucson

PS only works if it's a dedicated return line....
 
Last edited:
Steve,

The simplest way to KNOW whether or not twisted pairs are required for a particular signal is to check the installation manual for the specific piece of equipment you are installing. If you're doing some "pre-wiring" before you get all your boxes (I did a lot!), just download the manuals for what you expect to have from the appropriate web sites, and they will be pretty specific about what you need. You could use rules of thumb, but with avionics, I find it better to use what they tell you - then if there are problems, they can't tell you to re-wire and call them back... :rolleyes:

Paul
 
Guys I agree with everything previously said, but just as a precaution, don't want anyone to get the idea they can use single strand wire in a plane, always use multi strand wire in airplanes.
JMO, Larry
 
Years ago (I won't say how many, but a lot) I worked for a power company and when installing phase 3 lines, we always "twisted" them. If you ever notice power lines, you will occasionally see a pole without cross ties and the insulators are all in a row up the pole. This allows each of the lines to be turned to a new position on the next pole, thus introducing a twist in the lines. Kind of like twisting the pair of wires in a telephone or computer line. Anyway, this was supposed to reduce the electromagnetic field around the wires so that it didn't induce a current onto adjacent wires, like telephone lines. Induction current will create a humming or static on a telephone line.
One thing some farmers would do, was run a long wire from one pole to the next, on insulators, along our power poles, and the induced current would actually electrify the line, like an electric fence wire, to keep the cattle, or whatever, inside the fence. If it was long enough.
B&C calls out putting a twist in the main wires coming from their back up alternator, probably for this reason.
I'll let the real electronic wizards jump in on this one, but those are some of the things I remembered about that.
 
Here is another 2 cints worth.... somehwere in the distant past I remember an article that stated that the wires feeding the panel lights can be twisted... particularly those in the vicinity of the compass.. thus reducing the "pesky" stray magnetic fields in that area..... ??
 
I chant this to my kids in school almost daily:

Around every current carrying conductor, there is a magnetic field. This magnetic field increases with current flow.

Now with that said, an AC circuit behaves differently than a DC one. An AC circuit (due to it's alternating nature) has a magnetic field that builds up, collapses and then sets back up again in the reverse polarity. By simply pairing the "feed" and "return", you cancel the magnetic affects of each wire out. Twisting them keeps them in proximity of each other and aids to the canceling affect.

A DC circuit is different and this magnetic field sets up in one direction and stays there. It will grow and shrink depending on current flow. Due to the fact that it DOES NOT change direction, induction is not a huge concern.

Running twisted pairs to everything will not harm a thing.

Run twisted pairs in AC circuits, mikes, speakers, and pretty much anything where noise would cause a problem.

Does this help?

:cool: CJ
 
Oh, and Yes...

Wires run in the vicinity of the compass will cause a deviation. Twisting "PAIRS" will mitigate the deviation.

They must be pairs containing feed and return, which you don't always have.

:) CJ
 
Back
Top