blais@ut-emx.UUCP (Donald Blais) (06/05/90)
Vendors of fiber optic cabling and fiber optic transceivers often provide the option of using SMA or ST connectors. In selecting a standard for a site where no standard is yet established, why should one choose ST over SMA or vice versa. The application, in this case, would be a large campus Ethernet data distribution system that would later be upgraded for FDDI. -- Donald E. Blais Internet: blais@emx.utexas.edu Computation Center BITNET: blais@UTAIVC University of Texas UUCP: uunet!cs.utexas.edu!ut-emx!blais Austin, TX 78712 Phone: +1-512-471-3241
haas@cs.utah.edu (Walt Haas) (06/05/90)
In article <30971@ut-emx.UUCP> blais@ut-emx.UUCP (Donald E. Blais) writes: >Vendors of fiber optic cabling and fiber optic transceivers often >provide the option of using SMA or ST connectors. In selecting a >standard for a site where no standard is yet established, why should >one choose ST over SMA or vice versa. We have both on campus but like the ST better because it goes on more easily thus keeping costs low, and connects in a more reproducible manner. The ST is like a miniature BNC with a key that causes it to align a certain way every time you connect. The SMA on the other hand allows the connector to rotate while it is tightened. The skill with which the connector is installed will have a major effect on the performance of the connector. -- Walt Haas haas@cs.utah.edu
kwe@bu-it.bu.edu (Kent England) (06/06/90)
In article <1990Jun5.082908.26719@hellgate.utah.edu>, haas@cs.utah.edu (Walt Haas) writes: > > > We have both on campus but like the ST better because it goes on more easily > thus keeping costs low, and connects in a more reproducible manner. The ST > is like a miniature BNC with a key that causes it to align a certain way > every time you connect. The SMA on the other hand allows the connector to > rotate while it is tightened. The skill with which the connector is installed > will have a major effect on the performance of the connector. > > -- Walt Haas haas@cs.utah.edu I second Walt's analysis. STs align much better than SMAs. We started out with SMAs on Proteon's Pronet-80 and switched to STs for all but the f/o interfaces. You can make or buy a custom patch cord for end-node termination to anything that is not ST. I'd like to find a good non-epoxy, non-polish ST or FSD connector. That would really save some money in the field. We have been playing around with AMP's FSD (Full Shroud Duplex) connectors. We are finding them very tricky to build and so are considering staying with STs for field connectorization in the FDDI environment. We would buy the FSD patch cords. Anyone found a field connector for FSD that works well? Kent England Boston University
vjs@rhyolite.wpd.sgi.com (Vernon Schryver) (06/06/90)
In article <30971@ut-emx.UUCP>, blais@ut-emx.UUCP (Donald Blais) writes: > Vendors of fiber optic cabling and fiber optic transceivers often > provide the option of using SMA or ST connectors. In selecting a > standard for a site where no standard is yet established, why should > one choose ST over SMA or vice versa. The application, in this case, > would be a large campus Ethernet data distribution system that would > later be upgraded for FDDI. > -- > Donald E. Blais Internet: blais@emx.utexas.edu The FDDI standard requires neither STs nor SMAs, but MICs. Much of the FDDI hardware now being shipped uses STs, often hidden behind the shells that turn a pair of ST plugs into a MIC recepticle. I do not know if such shells are available for anything except STs. If you use STs now, you can later use such shells and some short MIC jumper cables to convert your cable plant. It would cost the insertion loss of an extra connector, but would be easier and could be cheaper than reterminating. (Actually, unless you are good with the polishing pads, you might have less insertion loss with two machine terminations than with one field termination, and so might come out ahead in loss budget by using such jumpers.) Also, if you use STs, you might be able to connect them directly to your FDDI equipment, after discarding a shell or jumper cable. Vernon Schryver Silicon Graphics vjs@sgi.com
kwe@bu-it.bu.edu (Kent England) (06/12/90)
In article <61655@sgi.sgi.com>, vjs@rhyolite.wpd.sgi.com (Vernon Schryver) writes: > > The FDDI standard requires neither STs nor SMAs, but MICs. That is true, but if you are building a campus backbone fiber optic cable plant, you may wish to build your patch panels with STs and not FSDs (Fixed Shroud Duplex) or MICs or whatever we call the FDDI connector. > > Much of the FDDI hardware now being shipped uses STs, often hidden behind > the shells that turn a pair of ST plugs into a MIC recepticle. I do not > know if such shells are available for anything except STs. We are piloting some desktop fiber using the AMP FSDs. AMP (and others) let you make FSD wall outlets and patch panel outlets using a pair of STs and a jig to create a FSD. You can field terminate a pair of STs, plug them into a mechanical connector, and, presto!, you have a FSD. But we are talking "horizontal fiber" or "desktop fiber" and not "vertical fiber" or "riser". There is a big difference between inter- building and intra-building distribution. The desktop fiber needs to take account of the FSDs of the workstations/routers, but the intra- building cable plant does not (unless you decide that the duplex connector is a great idea). > > If you use STs now, you can later use such shells and some short MIC jumper > cables to convert your cable plant. It would cost the insertion loss of an > extra connector, but would be easier and could be cheaper than > reterminating. (Actually, unless you are good with the polishing pads, you > might have less insertion loss with two machine terminations than with one > field termination, and so might come out ahead in loss budget by using such > jumpers.) Also, if you use STs, you might be able to connect them directly > to your FDDI equipment, after discarding a shell or jumper cable. > > We are finding the AMP FSD kits to be quite difficult to do in the field. The mechanical tolerances are quite tight, and you are forced to polish both ferrules simultaneously. If one connector is ruined, you are forced to redo both. Makes us wonder whether we should use FSDs any more than we have to. But the technology is still young and we are hopeful that we will find better field termination strategies for the FSDs. But we are planning to use patch cords to a wall outlet with the fiber pair terminated on a patch panel, cross-connected to the f/o equipment. This standard approach requires connector matings at the: workstation wall outlet patch panel (2) concentrator I would not think it wise to try to eliminate any of these connections, or one would lose the flexibility needed in the cable plant. This methodology is independent of medium; we use this approach on unshielded twisted pair, thin coax and everything else. Thanks, Vernon. I'm not disagreeing with what you said, just pointing out some other considerations in deploying fiber. --Kent England, Boston University
vjs@rhyolite.wpd.sgi.com (Vernon Schryver) (06/12/90)
> In article <61655@sgi.sgi.com>, I wrote... > > > > If you use STs now, you can later use such shells and some short MIC jumper > > cables to convert your cable plant. It would cost the insertion loss of an > > extra connector, but would be easier and could be cheaper than > > reterminating. .... In article <58608@bu.edu.bu.edu>, kwe@bu-it.bu.edu (Kent England) writes: > We are finding the AMP FSD kits to be quite difficult to do in the field. > The mechanical tolerances are quite tight, and you are forced to polish both > ferrules simultaneously.... I think we are talking about two different duplex thingies. I think you are talking about the official, genuine, fearsome MIC. I meant a piece of plastic that makes two ST plugs into a MIC recepticle. The ones I've used are intended for p.c.board mounting, but work fine dangling in air. They have the semblence of a pair of ST recepticles on the back. Vernon Schryver vjs@sgi.com