steveg@ritcsh.UUCP (in stereo where available) (05/17/88)
Here at RIT we have had the THICK Ethernet installed for some time. We had also been using (but are slowly phasing out) the IBM 327x type equipment tied to controllers with RG 62 A/U cable. I had always thought that this cable was 50 Ohms, but after looking in a Belden catalog, the specs show that the cable is in fact 48 Ohms or was it 48.? Ohms. Well either way I know that the Thin Ethernet cable is supposed to be 50 Ohms as is it's bigger fatter brother. Just about every office in our computer department has the RG62A/U running from it to a patch panel in the CPU room. I will probably try this out anyway, but am looking to see if anyone has done this before. That is using the RG62A/U cable for Thin Ethernet? Will the slight difference in OHMS have a detrimental affect on the Ethernet? When looking for THIN ETHERNET supplies in catalogs, I have only come across cable with stranded center conductor. I am wondering if there is any THIN ETHERNET around with a SOLID COPPER center conductor as then I could use BNC "Fast Fit"/"Twist On" connectors. -- ___ ___ ___ Steve Good ...uunet!ccicpg!cci632!ritcsh!steveg |@/o o\@| @/- -\@ @/o o\@ Network Administrator BITNET%"SNGDCO@RITVAXD" | \ o / | / o \ \/-\/ Rochester Institute of Technology (716)475-2702 \ --- / | --- | /---\ "I hear nothing, I see nothing, I say nothing" Shultz
eshop@saturn.ucsc.edu (Jim Warner) (05/18/88)
In article <2471@ritcsh.UUCP> steveg@ritcsh.UUCP (Steve Good -- ISC) writes: >Here at RIT we have had the THICK Ethernet installed for some time. We had >also been using (but are slowly phasing out) the IBM 327x type equipment >tied to controllers with RG 62 A/U cable. I had always thought that this >cable was 50 Ohms, but after looking in a Belden catalog, the specs show >that the cable is in fact 48 Ohms... Look at your catalog again. RG62 is 93 ohm cable. It won't work. jim warner
ron@topaz.rutgers.edu (Ron Natalie) (05/25/88)
RG 62A/U Cable is not 48 Ohms, its 93 Ohms. I am skeptical that this would be any good for thin Ethernet. I know it doesn't work the other way around (I tried once to use some of these 50' RG58 cables Sun keeps sending us on a 3179) and that does not work. -Ron
brian@ucsd.EDU (Brian Kantor) (05/26/88)
Actual experience: one of the outlying buildings here was mistakenly wired with RG62/U cable (yes, 93 ohm) instead of RG58/U, and the thin Ethernet did NOT work dependably for a few of the taps, and not at all for many of the rest. Brian Kantor UCSD Office of Academic Computing Academic Network Operations Group UCSD B-028, La Jolla, CA 92093 USA brian@ucsd.edu ucsd!brian BRIAN@UCSD
steveg@ritcsh.UUCP (in stereo where available) (05/27/88)
Thank you for all the response via Mail and via "comp.dcom.lans". I double checked my cable catalogs as was suggested and indeed RG 62 A/U is not 48 ohms but ninety something ohms, so I will go out and buy some real "thin ethernet" cable spected at 50 ohms. I did not even want to try the RG 62 A/U after some of the warnings that were sent to me. -- ___ ___ ___ Steve Good ...uunet!ccicpg!cci632!ritcsh!steveg |@/o o\@| @/- -\@ @/o o\@ Network Administrator BITNET%"SNGDCO@RITVAXD" | \ o / | / o \ \/-\/ Rochester Institute of Technology (716)475-2702 \ --- / | --- | /---\ "I hear nothing, I see nothing, I say nothing" Shultz
bae@ati.tis.llnl.gov (Hwa Jin Bae) (05/28/88)
In article <2631@ritcsh.UUCP> steveg@ritcsh.UUCP (Steve Good -- ISC) writes: >I did not even want to try the RG 62 A/U after some of the warnings that >were sent to me. How about RG 59? (as opposed to RG 58 or RG 11). I am using RG 59 in one of the segments of our thin Ethernet part of LAN, and it seems to work just fine... Can anyone think of any reason not to use RG 59? It seems to be a better cable than 58 and a nice compromise between 58 and 11. Maybe I don't know what I'm talking about... Hwa Jin Bae | The Devil made me do it...yeah, that's right! Control Data Corp. | (415) 463 - 6865 4234 Hacienda Drive | bae@tis.llnl.gov (Internet) Pleasanton, CA 94566 | {ames,ihnp4,lll-crg}!lll-tis!bae (UUCP)
brian@ucsd.EDU (Brian Kantor) (05/28/88)
In article <22226@tis.llnl.gov> bae@ati.tis.llnl.gov (Hwa Jin Bae) writes: >How about RG 59? (as opposed to RG 58 or RG 11). >Can anyone think of any reason not to >use RG 59? It seems to be a better cable than 58 and a nice compromise >between 58 and 11. Maybe I don't know what I'm talking about... RG59/U is a nominal 75 ohm cable; the standard for Ethernet (both thick and thin) is a nominal 50 ohms. Thus you have an impedance mismatch which can cause standing waves (i.e., reflections) which may cause or mask collisions on the cable. Note also that much of the RG59/U cable sold nowadays is foam dialectric rather than solid, since that is what is popular in the cable-TV industry. Foam-dialectric cable (the center insulator is made of polyethylene foam instead of solid polyethylene) has a velocity of propagation that is around .79C instead of .66C; thus the speed at which signals travel is different. Minimum spacing requirements for taps into the cable have to be recalculated; they would have to be .79/.66 = 1.2 times further apart in order to assure the same minimum time delay between taps. Further, foamed cable is much less resistant to damage. RG6/U cable is much the same as RG59; it's the slightly-thicker stuff that most cable-TV companies are using nowadays for house drops (and often for house wiring too); it has lower loss and is physically stronger, since it has a CopperWeld (copper over steel) center conductor, but it's still 75 ohm and it's also usually foam. RG11/U is very thick 75 ohm cable; its primary use used to be long video cable runs or special antenna feeds. RG55/U is double-shielded RG58; it should work well for thin Ethernet if you really need the extra shielding (perhaps if your Ethernet runs past a transmitting antenna) but the outside diameter is a bit larger and you'll have to find special connectors for it. Finally, a note to you ham radio operators: thick Ethernet cable (the yellow stuff) works pretty well at up to 450MHz, so that's a pretty nifty way to use up those little scraps that were left over. At current prices, it might be cheaper than RG8 at the local hamstore, although Belden 9913 seems to be the cable of choice above 30 MHz these days. Brian Kantor UCSD Office of Academic Computing Academic Network Operations Group UCSD B-028, La Jolla, CA 92093 USA brian@ucsd.edu ucsd!brian BRIAN@UCSD
eshop@saturn.ucsc.edu (Jim Warner) (05/29/88)
Brian is right. What is important is the impedance of the cable.
The only thing that is magic about 50 ohms is that's what all the
transceivers in the world expect to see. If you are willing to
make your own transceivers you can use any strange impedance cable
that you happen to have installed in your buildings. To be sure,
what you'll have won't be Ethernet, but it will work. The details
of how to do such a foolish thing are in National Semi's publication
550083-001 "Advanced Peripherals IEEE 802.3 Local Area Network
Guide."
We have successfully used 110 ohm balanced video cable for an
underground Ethernet link. This cable is sometimes called twinax
because it has two center conductors surrounded by a shield. We
found that by attaching the two center conductors together at both
ends and treating it as coax, the actual impedance was 47 ohms. The
slight shortfall in impedance is OK as long the cable isn't too short.
(The DC loop resistance adds to the terminator resistance so that
collision detection can still be guaranteed.)
jim warner
Sr Dev Engineer
Divison of Natural Sciences
University of California, Santa Cruzdave@stcns3.stc.oz (Dave Horsfall) (05/30/88)
In article <22226@tis.llnl.gov> bae@ati.tis.llnl.gov (Hwa Jin Bae) writes: >How about RG 59? (as opposed to RG 58 or RG 11). >I am using RG 59 in one of the segments of our thin Ethernet part of LAN, >and it seems to work just fine... Can anyone think of any reason not to >use RG 59? It seems to be a better cable than 58 and a nice compromise >between 58 and 11. Maybe I don't know what I'm talking about... Nope - it's 75 ohm. It's used for CATV, broad-band LANS etc. Certainly not for base-band Ethernet, which wants 50 ohm co-ax. You'll still get impedance mis-matches, and although you may not _see_ any degradation, it'll be there in the form of retries etc. Stick to 50 ohm cable. And what do you mean by "a nice compromise between 58 and 11"? -- Dave Horsfall (VK2KFU), Alcatel-STC Australia, dave@stcns3.stc.oz dave%stcns3.stc.OZ.AU@uunet.UU.NET, ...munnari!stcns3.stc.OZ.AU!dave "ADA - From the people who brought you COBOL"
lobo@tub.UUCP (Alexander Lobodzinski) (06/02/88)
In article <875@ucsd.EDU> brian@ucsd.edu (Brian Kantor) writes: >RG59/U is a nominal 75 ohm cable; the standard for Ethernet (both thick >and thin) is a nominal 50 ohms. Thus you have an impedance mismatch >which can cause standing waves (i.e., reflections) which may cause or >mask collisions on the cable. Beware that RG58/U cable has 53 ohms impedance; RG58C/U gives you the 50 ohms you want. >Foam-dialectric cable [...] has a velocity of propagation that is >around .79C instead of .66C. [...] >Minimum spacing requirements for taps into the cable have to be >recalculated; they would have to be .79/.66 = 1.2 times further >apart in order to assure the same minimum time delay between taps. I doubt that; as far as I know, the Cheapernet specs (max 185m, max 30 taps, min 2m between taps) are calculated based on .79C; not accidentally, this velocity matches that of TEthernet. The difference between RG58C/U and Belden 9907 (*the* Cheapernet cable) is just that relative velocity (and the pretty yellow insulation, not to forget :-) Ciao, l. -- Alexander Lobodzinski, <lobo@tub.UUCP>, <lobo@tub.BITNET> UUCP: ...!pyramid!tub!lobo (Overseas) ...!unido!tub!lobo (Europe only) Drink positive! --RKL