Article View: comp.dcom.lans.fddi
Article #29Re: Query on overdriving short fiber optic links
From: Paul Koning
Date: Mon, 03 Aug 1992 14:42
Date: Mon, 03 Aug 1992 14:42
49 lines
2943 bytes
2943 bytes
In article <1992Jul29.161817.2757@ncar.ucar.edu>, hyder@niwot.scd.ucar.edu (Paul Hyder) writes: |>After three months tracking down a problem with a fiber based Ethernet |>that ended up being overdriven fiber connections I'd like to get some |>feedback from anyone out there that has or is planning configurations |>similar to ours. |> |>This new network used fiber as the major interconnect in preparation for |>full scale FDDI and beyond implementation. Details on request (were |>posted in comp.dcom.lans.ethernet), in short we used short fiber runs |>(3 to 300ft) to connect between close buildings, between floors, and to |>in-room hardware. We have been seriously impacted by the lack of |>published minimum loss/length requriements. (Overdriven fiber is truly |>a nasty debugging chore.) |> |>QUESTIONS: |> 1. Has anyone seen overdriven FDDI? Is there a minimum |> loss/length requirement? |> |> 2. If you have short fiber optic runs that had overdrive noise |> how did you solve the problems? Where did you get |> attenuators? What did your solution cost? |> |>As always TIA. Email to me please. |> Paul Hyder |> NCAR |> Boulder, CO The transmitter, receiver, and fiber specs are spelled out in detail in the two FDDI PMD standards (multimode and singlemode). In the multimode case, the max transmit power is within the receiver specs, so there is no minimum loss when using standard (62.5 micrometer) fiber. I've been told that there MAY be problems (depending on transceiver) if you use 100 micrometer fiber. See the appendix on alternate fiber sizes in the multimode PMD standard for more detail, but you might want to use a power meter to check such configurations. Better yet, use conventional size fiber. In the singlemode case, there are two classes of transceivers. One is intended for moderate distances, the other for long distances. The long distance version definitely DOES have a minimum loss requirement (on the order of 10 dB, but check the individual product since there may be some small variations among transceivers from different suppliers). If you plan single mode installations with significant cable lengths, you should do the appropriate planning and measurement to get signal levels right. Given the cost of such installations, the additional cost of doing a thorough job is trivial. paul
Technical Details
Message-ID:
<1992Aug3.144239.25230@nntpd.lkg.dec.com>Path:
arc1-ca.pugleaf.net!sparky!uunet!olivea!decwrl!deccrl!news.crl.dec.com!news!nntpd.lkg.dec.com!usenetReferences:
<1992Jul29.161817.2757@ncar.ucar.edu>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!olivea!decwrl!deccrl!news.crl.dec.com!news!nntpd.lkg.dec.com!usenet From: koning@koning.enet.dec.com (Paul Koning) Newsgroups: comp.dcom.lans.fddi Subject: Re: Query on overdriving short fiber optic links Message-ID: <1992Aug3.144239.25230@nntpd.lkg.dec.com> Date: 3 Aug 92 14:42:39 GMT References: <1992Jul29.161817.2757@ncar.ucar.edu> Sender: usenet@nntpd.lkg.dec.com (USENET News System) Reply-To: koning@koning.enet.dec.com Organization: Digital Equipment Co., distributed systems architecture Lines: 49