wb8foz@mthvax.cs.miami.edu (David Lesher,,255RTFM,255rtfm) (11/21/90)
>I'm not clear on what you meant by "all twisted", so here is a >little more: >Each pair is twisted. Other than that the only placement or >arrangement done is to make it easy to identify which is which. {} >Take a look at some of that non-twisted quad and see if it doesn't >really have a twist about every 1-2 feet. Well, it's like this.... Twisted pair is just that. Now if you get (for example) 25 pair, each pair is twisted around itself, then the whole group is twisted at a slower rate. If you have 25 pair key cable, you can see this easily. Now if you have 100 pair, each pair, each group of 25, and then all 100 are twisted at different rates. And so forth up the line, to the BIG stuff of thousands pairs of light ;-}. All this Chubby Checker work is to cancel out the influence of external magnetic fields. I can remember seeing the proof in fields class, but my memory blissfully shields {booo} me from the gruesome details. BUT! Old IW {interior wire} came in lots of varieties. Most of it was 'quad' with the four conductors being Red, Green, Black & Yellow. This includes old JTK, SK, and other stuff I cannot remember the name for. This was NOT 2 twisted pairs. [I guess that Bell Labs never envisioned us running Trailblazers on the Yellow & Black ;-}] IW likely has a slight {total} twist but to tell you the truth, I never stripped back a bunch to be sure. More recently, I've seen installations with real 2 pair: typically colored BL-W, and OR-W. I have NO idea what the standards are -- does the installer now just assume we will have three lines, and thus pull 3 pair or what? [Of course, it is now YOUR wire.] There is an old joke about the twists. In the "Good Old Daze" the cable crew dragged the stuff down the road with a mule, then hauled it up to each crossarm. They had to add the final {total} twist of each cable. One of the things the foreman would do to the new guy was convince him that the mule had been at it so long that she would stop every 1/4 mile, and roll over to add the needed twist, and roll the other way the next time! Now this is not a lot of help for the guy who asked the question to begin with. I agree that he should try using the Y+B pair, being sure that no phones are still connected to them, and see if he has a problem. If he does, THEN worry! -- A host is a host from coast to coast.....wb8foz@mthvax.cs.miami.edu & no one will talk to a host that's close............(305) 255-RTFM Unless the host (that isn't close)......................pob 570-335 is busy, hung or dead....................................33257-0335
root@zswamp.fidonet.org (Geoffrey Welsh) (11/23/90)
David Lesher,,255RTFM,255rtfm (wb8foz@mthvax.cs.miami.edu ) wrote: >All this Chubby Checker work is to cancel out the influence of external >magnetic fields. I can remember seeing the proof in fields class, but >my memory blissfully shields {booo} me from the gruesome details. Short & simple version: picture the pair, connected at each end to some kind of device, as a loop. The interference picked up is proportional to the electromagnetic flux passing through the loop. Twisting the loop reduces the area to negligible. That's why signal wires should be twisted with their returns (read "ground" where appropriate). -- UUCP: watmath!xenitec!zswamp!root | 602-66 Mooregate Crescent Internet: root@zswamp.fidonet.org | Kitchener, Ontario FidoNet: SYSOP, 1:221/171 | N2M 5E6 CANADA Data: (519) 742-8939 | (519) 741-9553 MC Hammer, n. Device used to ensure firm seating of MicroChannel boards Try our new Bud 'C' compiler... it specializes in 'case' statements!
floyd@hayes.ims.alaska.edu (Floyd Davidson) (11/24/90)
In article <5072.274CAD24@zswamp.fidonet.org> root@zswamp.fidonet.org (Geoffrey Welsh) writes: >David Lesher,,255RTFM,255rtfm (wb8foz@mthvax.cs.miami.edu ) wrote: > > >All this Chubby Checker work is to cancel out the influence of external > >magnetic fields. I can remember seeing the proof in fields class, but > >my memory blissfully shields {booo} me from the gruesome details. > > Short & simple version: picture the pair, connected at each end to some >kind of device, as a loop. The interference picked up is proportional to the >electromagnetic flux passing through the loop. Twisting the loop reduces the >area to negligible. > > That's why signal wires should be twisted with their returns (read >"ground" where appropriate). What you are saying is true; however, the last statement is going to lead people out to left field. The terms "signal wires" and "returns" may not mean the same thing to everyone. Twisted pair works for *balanced* circuits. It will make things worse if the circuit is single ended. RS-232 is single ended. NEVER use twisted pair for RS-232. Telephone lines are balanced circuits. ALWAYS use twisted pair. The terms "signal" and "return" generally refer to single ended (unbalanced) circuits, where twisting the wires is just going to increase the capacitive coupling between the two wires. Any noise picked up on one will be transfered to the other. Balanced circuits cancel induced noise because it is common mode, and the signal is differential mode. Twisting the wires tends to equalize the coupling of external fields to both of the wires by providing equal exposure. That maintains a higher ratio between the common mode and the differential mode. There is a lot more to this than that simple answer. Anyone who is really interested should read up on transmission lines. Floyd -- Floyd L. Davidson floyd@hayes.ims.alaska.edu Salcha, AK 99714 paycheck connection to Alascom, Inc. When I speak for them, one of us will be *out* of business in a hurry.