Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!mp.cs.niu.edu!ux1.cso.uiuc.edu!uxh.cso.uiuc.edu!german From: german@uxh.cso.uiuc.edu (Gregory German) Newsgroups: comp.dcom.lans Subject: Re: another 10BaseT wiring question Message-ID: <1991Apr10.172500.24529@ux1.cso.uiuc.edu> Date: 10 Apr 91 17:25:00 GMT References: <9736@mentor.cc.purdue.edu> <1991Apr10.035307.29375@netcom.COM> Sender: usenet@ux1.cso.uiuc.edu (News) Organization: University of Illinois at Urbana Lines: 85 cmilono@netcom.COM (Carlo Milono) writes: >In article <9736@mentor.cc.purdue.edu> lairdkb@mentor.cc.purdue.edu (Kyler Laird) writes: >>I saw the front of a Gateway (?) concentrator and ordered network cabling >>that terminates in modular (RJ-45) connectors. I want this because I'll >>have ~14 connections and only 12 ports on the concentrator. (I don't intend >>to ever use all at once.) >> >>Unfortunately, we got a David System's concentrator which uses a 50-pin telco >>connector. >> >>I'm looking for something like a harmonica adapter that would allow me to >>plug in 12 RJ-45 connectors and come out with a 50-pin telco connector. >Well, the standard jack for 10BASE-T uses 8-wires...every 10BASE-T card >*I* have ever seen is keyed for 8-wires...so I would assume that you >want an adapter that is an amphenol (50-pin) on the Equipment side and >(ahem!) 96 wires on the (ahem!) jack side with 12x8 wires splayed in >the correct order - if that is what you want: NO, I haven't seen it. >However, there are makers of 12x4 (skipping the Violet/Slate pair), >at which point you would have four-conductor mounting/patch cords and >rather unstandard wiring at the station jack, at which point you would >have to re-wire when you use the jack for something else - IMHO, a poor >choice for wire management. >A suggestion: >Run the 50-pin Amphenol to 110-like hardware and cross-connect from there >to your station field. Leave your four-pair alone from the closet to the >jack and sleep well into the future. >-- >+--------------------------------------------------------------------------+ >| Carlo Milono: cmilono@netcom.apple.com or apple!netcom!cmilono | >|"When a true genius appears in the world, you may know him by this sign, | >|that the dunces are all in confederacy against him." - Jonathan Swift | >+--------------------------------------------------------------------------+ Well I have started to avoid any of the products with the 50pin connectors. It is much easier to trouble shoot a problem or to switch a station from on network to another when each port is connected with its own jumper to a patch pannel rather than an Aphenol connector. I think you will find that how you wire everything up will depend on how you are setup to begin with. We have 8 pairs going from the wire closet to each station. There is an RJ-45 on top for voice and a second (B-Jack) for data. In general we wire only 2 pair of a connection back to the distribution location and terminate in an RJ-11 patch pannel. Our jacks are wired as inside pairs and outside pairs such that an RJ-11 plugged into an RJ-45 jack at the end user's location will hit the inside pairs (5&6 here). When we wish to use pairs 7&8 we use a custom splitter that provides 7&8 as the middle pair on the B-side of the splitter and just passes 5&6 through to the middle on the A-side. We would wire 7&8 as a separate port on the RJ-11 patch pannel. When I am done with a building each station is wired to the patch pannel and what we connect them to determines what kind of jumper we used. For AppleTalk it would be a straight through cable from the patch pannel to a StarController (or first to a 4:1 patch pannel like MacLAN then to the SC). For 10baseT we use a cable that is 1234 on the RJ-11 end and 1236 on the RJ-45 end, one from the station to the 10baseT card or transceiver and one from the patch pannel to hub. For Token Ring we also use straight through jumpers. This system requires us to use custom jumper cables and drop cables for 10baseT but allows us to use our distribution system in a more flexible fashion. We have had very little trouble getting cables made to our specs and make custom length drop cables when we need to. Sometimes it seems that the people who installed all this phone cable had no idea that I would want to use it for data! :-) It is hard to work out a system that is both flexible and 100% efficient in use of cable. AppleTalk really only requires 1 pair, but in order to be able to switch to ethernet or 10baseT you need to wire for 2 pair. When I started out I knew that not every building was going to be easy, but I had hoped that a few of them would be! -- Greg German (german@sonne.CSO.UIUC.EDU) (217-333-8293) US Mail: Univ of Illinois, CSO, 1304 W Springfield Ave, Urbana, IL 61801 Office: 129 Digital Computer Lab., Network Design Office