querubin@uhccux.uhcc.Hawaii.Edu (Antonio Querubin) (07/13/90)
Not sure if this is the appropriate news group, but if it's not, let me know and I'll re-post: Don't know if anyone has run into this situation before but we have a requirement to setup a point-to-point link between two buildings using a previously installed 75-ohm video cable. For various reasons, we can't just replace the cable with a 50-ohm thin-net cable to run an ethernet link. So the simplest and cheapest way to get the link is to use the existing cable. Now it seems that one should be able to solve the impedance mismatch and associated reflection problems by using a 75-to-50 ohm impedance- matching transformer at each end of the video cable so that the ethernet transceivers on either side effectively sees a 50-ohm cable. I'm considering building two of these transformers as a small afternoon project and giving it a try. Question: has anyone done this before and what kind of problems did you run into? I know the turns ratio has to be the square root of 1.5 but I haven't the foggiest idea of how many turns to wind onto a toroid. Too many turns and I might have a saturation problem, too little and I might get a lot of signal loss. Anyone got a good recommendation here? Does anyone sell an inexpensive ready-made transformer for this kind of situation? Antonio Querubin, Jr. Internet: antonio_querubin-manoa@uhplato.uhcc.hawaii.edu BITnet: antonio_querubin-manoa@uhplato
oberman@rogue.llnl.gov (07/13/90)
In article <8592@uhccux.uhcc.Hawaii.Edu>, querubin@uhccux.uhcc.Hawaii.Edu (Antonio Querubin) writes: > > Don't know if anyone has run into this situation before but we have a > requirement to setup a point-to-point link between two buildings using a > previously installed 75-ohm video cable. For various reasons, we can't > just replace the cable with a 50-ohm thin-net cable to run an ethernet > link. So the simplest and cheapest way to get the link is to use the > existing cable. Now it seems that one should be able to solve the impedance > mismatch and associated reflection problems by using a 75-to-50 ohm impedance- > matching transformer at each end of the video cable so that the ethernet > transceivers on either side effectively sees a 50-ohm cable. TCL, the granddaddy of Ethernet hardware companies, used to make Ethernet eqipment already set up for 75 ohm cable. We have quite a bit of this stuff because of a similar problem, but it's all been here for a while and my not still be available. R. Kevin Oberman Lawrence Livermore National Laboratory Internet: oberman@icdc.llnl.gov (415) 422-6955 Disclaimer: Don't take this too seriously. I just like to improve my typing and probably don't really know anything useful about anything.
morrison@thucydides.cs.uiuc.edu (Vance Morrison) (07/14/90)
Hello, If you are interested, National Semiconductor puts out a LAN chip book that describes the 8350 (I think) ethernet controler. In this book it describes the changes that need to be made for use on a 75ohm cable. Note that for a point to point link you should be able to get away with simply connecting up the (Thin) cards with a T connector and placing a 75ohm resistor at both ends. There will be no reflections, because a ethernet card looks like an open (after all, if it didn't when you connect a card to the middle of a net, there would be reflections at the card). Practically speaking I have seen ethernet set up normally (except for the 75ohm cable), and seen it work. This undoubtedly has some effect on length, and node count restrictions, but with a point to point setup like yours, you have a lot of slop to spare. Vance --------------------------------------------------------- P. S. The transformer idea in its simple form may not work. The problem is that Ethernet uses a DC signal to detect collisions. (When a card is transmitting, it transmits a AC signal imposed on a DC bias if two cards try to transmit, then twice the DC voltage will be seen, and a colision can be detected. Now in reality you still might get away with it, since you could design the transformer to pass signals effectively down to say 1000Hhz. Since collisions last quite a bit less than 1ms, the transformer will pass the DC signal 'long enough' for the colision to be detected. It depends on exactly how the tranceiver detects colisions. To answer you direct question, the number of turns of a transformer is not a critical design factor. It is only lower bounded by your desire to have a lot of coupling per winding resistance at frequencies above 1000Hhz. At such high frequecies, even 100 turns should be sufficent (on a good, high frequency toroid core). All of the above is simply for academic interest. I really think you can get away with treating the 75ohm cable as if it was a 50 ohm cable (execpt for the different resistance values in the terminating resistors). Vance
querubin@uhccux.uhcc.Hawaii.Edu (Antonio Querubin) (07/14/90)
Thanks to all those who responded to my base note. Almost everyone pointed
out a problem which I hadn't considered - the DC collision detection process
would be affected by a normal transformer. A normal transformer would pass
the AC signal but not the DC. After pondering this new twist for a few minutes
I realized that an 'auto-transformer' would pass the DC as well as AC.
For those of you who don't know what an auto-transformer is, in it's simplest
form it is a single coil with the secondary as just a tap on the primary (or
vice versa):
(---------------- secondary out
(
-------------)
primary )
)
)
)
-------------)---------------- common
Eg. An equivalent turns ratio of 3:4 would mean the primary is attached
3/4 up the coil while a 2:1 ratio would mean the secondary is a tap at
the half-way point of the coil. Theoretically this *should* work :-).
But someone also suggested using the 75 ohm cable directly but with 75 ohm
terminators instead. This makes more sense since all that's needed is a
point-to-point link. Connection to a 50 ohm segment would require the
impedance matcher above (or a repeater) between the two cable types.
Further comments on the above two schemes are welcome!
Antonio Querubin, Jr.
querubin@uhccux.uhcc.hawaii.eduBOBW@cc.usu.edu (BOB WOOD WA7MXZ) (07/16/90)
In article <8601@uhccux.uhcc.Hawaii.Edu>, querubin@uhccux.uhcc.Hawaii.Edu (Antonio Querubin) writes: > Thanks to all those who responded to my base note. Almost everyone pointed > out a problem which I hadn't considered - the DC collision detection process > would be affected by a normal transformer. A normal transformer would pass > the AC signal but not the DC. After pondering this new twist for a few minutes > I realized that an 'auto-transformer' would pass the DC as well as AC. Look at your autotransformer design again. Not only does an autotransformer pass DC but it also shorts it out, directly to ground. The larger the wire used in the transformer to improve the Q the lower the D.C. resistence to ground. In short, the matching transformer plan isn't going to work. > > But someone also suggested using the 75 ohm cable directly but with 75 ohm > terminators instead. This makes more sense since all that's needed is a > point-to-point link. Connection to a 50 ohm segment would require the > impedance matcher above (or a repeater) between the two cable types. This is going to be your only game plan. Use only 2 nodes on the cable and use good 75 ohm 1% terminating resistors. The cable length will have to be no more than 1200 feet, (1640 is the maximum THICKnet length) IF the video cable is RG-11 (1/2 inch diameter). If the cable is the smaller 1/4 inch RG59 stuff then the maximum distance will need to be 500 feet or less. These figures do not guarantee that your idea to use 75 ohm cable will work. The Ethernet spec is timed for 50 ohm cable and timing will be different on the 75 ohm cable. If you want to use additional onde on either end you will need bridges and/or repeaters to keep the number of nodes of the cable down to 2, one at each building. -- =============================================================================== Bob Wood WA7MXZ bobw@cc.usu.edu Utah State University bobw@usu.bitnet Chemistry and Biochemistry tel. (801) 750-1614 UMC 0300 Logan, Utah 84322-0300 ===============================================================================
querubin@uhccux.uhcc.Hawaii.Edu (Antonio Querubin) (07/16/90)
In article <28221@cc.usu.edu> BOBW@cc.usu.edu (BOB WOOD WA7MXZ) writes: >Look at your autotransformer design again. Not only does an autotransformer >pass DC but it also shorts it out, directly to ground. The larger the wire >used in the transformer to improve the Q the lower the D.C. resistence to >ground. In short, the matching transformer plan isn't going to work. Yes, you're quite right. But considering that the DC signal is a relatively short pulse on the order of milliseconds at the most the autotransformer may work anyway. The trick is to make the thing have wide enough bandpass to accept both the low frequency pulse and the high frequency component. I have a few ideas on how to do this. My curiosity about the thing is high enough for me to pursue the matter time permitting. However, as you said: >This is going to be your only game plan. Use only 2 nodes on the cable and >use good 75 ohm 1% terminating resistors. The cable length will have to be >no more than 1200 feet, (1640 is the maximum THICKnet length) IF the video >cable is RG-11 (1/2 inch diameter). If the cable is the smaller 1/4 inch RG59 >stuff then the maximum distance will need to be 500 feet or less. These >figures do not guarantee that your idea to use 75 ohm cable will work. >The Ethernet spec is timed for 50 ohm cable and timing will be different >on the 75 ohm cable. If you want to use additional onde on either end you >will need bridges and/or repeaters to keep the number of nodes of the cable >down to 2, one at each building. I think we'll play with that idea first. And it is RG-59. Thanks for the response! Antonio Querubin, Jr. querubin@uhccux.uhcc.hawaii.edu
rpw3@rigden.wpd.sgi.com (Rob Warnock) (07/17/90)
This is NOT going to work, guys... In article <8611@uhccux.uhcc.Hawaii.Edu> querubin@uhccux.UUCP (Antonio Querubin) writes: +--------------- | In article <28221@cc.usu.edu> BOBW@cc.usu.edu (BOB WOOD WA7MXZ) writes: | >Look at your autotransformer design again. Not only does an autotransformer | >pass DC but it also shorts it out, directly to ground. The larger the wire | >used in the transformer to improve the Q the lower the D.C. resistence to | >ground. In short, the matching transformer plan isn't going to work. | Yes, you're quite right. But considering that the DC signal is a relatively | short pulse on the order of milliseconds at the most the autotransformer may | work anyway. The trick is to make the thing have wide enough bandpass to | accept both the low frequency pulse and the high frequency component. +--------------- Nope, the D.C. signal is *not* necessarily "short". Consider a long stream of back-to-back long packets, and note that the gaps between the packets do not allow time for the D.C. balance to "relax". There is no limit to the maximum length of the D.C. "pulse" as seen by the autotransformer. Eventually you will get enough current in the autotransformer that the voltage will (1) not get high enough *during* packets, and (2) start kicking the "wrong way" *between* packets, which will mess everything up. By putting a (carefully chosen) capacitor between the bottom of the auto- transformer and cable shield, you *may* be able to avoid that problem. However, see below... +--------------- | >This is going to be your only game plan. Use only 2 nodes on the cable and | >use good 75 ohm 1% terminating resistors. +--------------- Nope, this doesn't cut it either. Ethernet transceivers are *current* drivers. The collision-detection logic measures *voltage*, which is the sum of the currents into the cable times the impedance of the cable. Raising the cable impedance (and the terminating impedance) by 50% (50 -> 75 ohm) will raise the voltage by 50%, and you'll constantly be on (or *over*) the ragged edge of a (false) collision detect. Notice that the Ethernet spec says that transceivers must detect collisions within two bits times, so no matter what D.C. termination you put on the cable, as little as 40 meters of 75 ohm cable will cause constant "collisions". The only thing you can do is use the autotransformers, *with* the blocking capacitor I mentioned above in the ground leg, and 50-ohm cable and 50-ohm terminators at the ends. All transceivers must be on the 50-ohm sections. And even then, no guarantees it's going to work... Good luck! -Rob ----- Rob Warnock, MS-9U/510 rpw3@sgi.com rpw3@pei.com Silicon Graphics, Inc. (415)335-1673 Protocol Engines, Inc. 2011 N. Shoreline Blvd. Mountain View, CA 94039-7311
wdr@wang.com (William Ricker) (07/20/90)
I have seen here around Wang Towers some devices which balance
ThinNet to run over pairs of old cables already installed.
(The old Wang cabling I'm told is 75ohm, although I successfully used
an extender of that cable as a splice-extension in the middle of a
point-to-point ThinNet link, which should have created horrible
reflections -- maybe ignored because the reflections were within 3m
of both transceivers? Who knows.)
The magic coverter would presumably also work with a single run
of 75 ohm cable, with some sort of terminator.
(Wang minicomputer customers cabled 75ohm pairs between workstation
and computer room; now that they're replacing that net with PC's running
802.3, they want to run on the same wire. Since a thin-net is usually
a loop, this works fine: do a star-shaped loop, patching each
office's return to the next's feed at the patch-panel. The hard
part is matching the impedenance and DC components, as previously
pointed out in this thread. Allegedly, which cute little coverter
cable with a WANG logo does this.)
I've heard conflicting rumors whether we're selling these to paying
customers or not; I can't find it in the catalogs, but then, I don't
know where/how they'd index it if it was.
To find out if you can order one, you can call either 800-225-0234
or 800-TEL-WANG, which is Wang Express, our direct-sales group, and
ask about adapters to run Ethernet or 802.3 over "928 cables" or
"old wang workstation cables".
If they can't find it, and you want it, call me & I'll see if I can
find they guy who told me what it was and get a number for you.
Bill Ricker
Applications Software
(508)967-2663
No, Wang doesn't guarantee this information -- if we aren't selling
'em, I probably shouldn't suggest we could! So don't fink on me!-)
--
/bill ricker/
wdr@wang.com a/k/a wricker@northeastern.edu
*** Warning: This account not authorized to express opinions ***karn@envy.bellcore.com (Phil R. Karn) (07/21/90)
In article <64296@sgi.sgi.com>, rpw3@rigden.wpd.sgi.com (Rob Warnock) writes: |> Nope, the D.C. signal is *not* necessarily "short". Consider a long stream |> of back-to-back long packets, and note that the gaps between the packets do |> not allow time for the D.C. balance to "relax". There is no limit to the |> maximum length of the D.C. "pulse" as seen by the autotransformer. Quite true. In fact, I found a cute use for this DC component. Because the Ethernet modulation is Manchester, and because the signal swings between 0 and -2V, by putting a 0-1V DC analog voltmeter across the cable you get a very simple and effective real time network loading monitor! 0 volts corresponds to an idle network, and 1 volt corresponds to a saturated network. Of course, the accuracy may suffer across a long cable, but hey... Those who used to watch the analog CPU utilization meters on the old IBM 370/168 consoles should appreciate this little hack. Phil