KAMBVL@BLEKUL13.BITNET (Guido Dilles) (06/02/88)
We have several spare pairs of 0.8 square millimeter solid copper cable between 2 buildings (a 250 meter stretch). Our technicians call it 'signalisation cable'. Can we use them as a PhoneNet LocalTalk 'backbone' or do we need to put in new standard twisted pair cabling :-( ? How many pairs do we need ? And how would you connect this cable to the Faralon supplied cable that we intend to use on both sides ? We're just in the planning stage and technical advice is hard to get by over here. Thanks, Guido Dilles <KAMBVL@BLEKUL13.BITNET> U.P.C. St.-Kamillus - Catholic University Leuven - Belgium Acknowledge-To: <KAMBVL@BLEKUL13>
denbeste@bgsuvax.UUCP (William C. DenBesten) (06/03/88)
From article <Added.UWdJzOy00Ui3Q52U9b@andrew.cmu.edu>, by KAMBVL@BLEKUL13.BITNET (Guido Dilles): > > We have several spare pairs of 0.8 square millimeter solid copper > cable between 2 buildings (a 250 meter stretch). > > Can we use them as a PhoneNet LocalTalk 'backbone' or do we need to > put in new standard twisted pair cabling :-( ? No problem using it. To connect it in, get two modular jacks, and hook one wire up to the black cable and the other wire to the yellow cable. Phone net is very forgiving in what you use for cable. -- bill denbesten denbeste@bgsu.edu
eshop@saturn.ucsc.edu (Jim Warner) (06/05/88)
In article <614@mtxinu.UUCP> ed@mtxinu.UUCP (Ed Gould) writes: > >As I think more, though, I remember that the cable that PhoneNet uses >is flat, untwisted telephone cable. So it might work better to use >two pairs, selecting one conductor from each for the PhoneNet, and >grounding the other. > No. Apple's LocalTalk is RS-422 balanced differential signalling. The two wires should be twisted together for the maximum noise immunity. You can read about it in EIA-422-A, available from the Electronic Industries Association (Dec 1978). Furthermore, "grounding the other [wire]" has little meaning in an interbuilding connection. It is not safe to assume that ground will be the same in different buildings. My reading of RS-422 does not lead to the conclusion that Farallon's distance limits are conservative. But whatever works works. The rules appear to be empirical rather than based on engineering analysis. I have peeled open a Farallon transformer and looked at the measures they have taken to guard against transients. They use the same transient suppression components that are used to protect 120 VAC circuits. I honestly can't imagine how these are at all appropriate to protecting the attached equipment. Jim Warner Univ of Cal Santa Cruz
dorourke@polyslo.UUCP (David O'Rourke) (06/06/88)
In article <3611@saturn.ucsc.edu> eshop@saturn.ucsc.edu (Jim Warner) writes: >No. Apple's LocalTalk is RS-422 balanced differential signalling. I was under the impression that differential signalling requires two wires for each signal. So that would require Apple's LocalTalk to use 5 wires, 2 for Transmitt, 2 for Recieve, and 1 for ground. Well LocalTalk only has 3 wires, 1 transmit, 1 Recieve, 1 ground. Besides I was told by an EE person who looked at Apple's Transformer specs in inside AppleTalk that their wiring scheme reaks more of RS-423 (RS-432??). In any after looking at the specs for the LocalTalk box, and the number of wires in the LocalTalk cable, the 20 or so people I've talked to seem to think apple isn't using the RS-422 at all. And after looking at how they wired the 8530 these people were even move convinced that Apple isn't using the RS-422 standard at all, they seem to be using a tweaked version somewhere between RS-232 & RS-422. In any case I might be wrong, but I don't think LocalTalk is run on top of RS-422 as you stated. -- David M. O'Rourke Disclaimer: I don't represent the school. All opinions are mine!
eshop@saturn.ucsc.edu (Jim Warner) (06/07/88)
In article <3090@polyslo.UUCP> dorourke@polyslo.UUCP (David O'Rourke) writes: >In article <3611@saturn.ucsc.edu> eshop@saturn.ucsc.edu (Jim Warner) writes: >>>No. Apple's LocalTalk is RS-422 balanced differential signalling. > > I was under the impression that differential signalling requires two wires >for each signal. So that would require Apple's LocalTalk to use 5 wires, 2 >for Transmitt, 2 for Recieve, and 1 for ground. The five wires you list exist in the cable leading from the Mac to the transformer. They are combined in the transformer box. There are two active wires in LocalTalk, plus the shield that is connected to ground through 1 Kohm in parallel with 0.1 uF. PhoneNet dispenses with the shield. >Well LocalTalk only has 3 wires, 1 transmit, 1 Recieve, 1 ground. LocalTalk is a party line. Any station can send to any other station. The sender transmits on both wires and everyone else listens on the same wires. The full duplex scheme you propose (separate wires for send and receive) won't work for more than two stations. LocalTalk is half duplex -- one data path is used for both sending and receiving. >Besides I was told by an EE person >who looked at Apple's Transformer specs in inside AppleTalk that their >wiring scheme reaks more of RS-423 (RS-432??). >In any after looking at the specs for the LocalTalk box, and the number of >wires in the LocalTalk cable, the 20 or so people I've talked to seem to think >apple isn't using the RS-422 at all. In "Inside AppleTalk" in the Electrical Specifications section it says: "The use of the EIA RS-422 signalling standard for transmission and reception over AppleTalk provides significantly higher data rates over longer distances than that of the EIA RS-232C standard." >And after looking at how they wired >the 8530 these people were even move convinced that Apple isn't using the >RS-422 standard at all, they seem to be using a tweaked version somewhere >between RS-232 & RS-422. "Inside AppleTalk" says, "The recommended driver is the 26LS30 used with both +5V and -5V as power supplies, and the mode control connected to give differential outputs." Perhaps this is the source of confusion of your 20 friends. Dual power supplies are not required for RS-422. I suppose that Apple engineers made that choice because they wanted a larger signal. What Apple did isn't "tweaked". It is a straight forward application of the info in the Advanced Micro Devices Linear and Interface Data Book. > In any case I might be wrong, but I don't think LocalTalk is run on top of >RS-422 as you stated. > >-- jim warner