A1T@PSUECL.BITNET (Andy Tefft) (08/30/88)
- - The original note follows - -
Can anyone out there tell me how to make a null modem cable? I tried just
connecting corresponding pins of my 2 serial connectors on 2 computers
together, with the serial cards set the same way. No dice.
Please reply via e-mail unless you think others will be interested.
Andy Tefft
a1t@ecl.psu.edu
a1t@psuecl.bitnet
a1t%psuecl.bitnet@psuvax1.uucp (maybe)
...psuvax1!psuhcx!a1tjohn@wa3wbu.UUCP (John Gayman) (09/04/88)
In article <8809030943.AA02215@ucbvax.Berkeley.EDU>, A1T@PSUECL.BITNET (Andy Tefft) writes: > > - - The original note follows - - > > Can anyone out there tell me how to make a null modem cable? I tried just > connecting corresponding pins of my 2 serial connectors on 2 computers > together, with the serial cards set the same way. No dice. > [ My mailer failed to get this through so I'm posting it] Andy, There are many configurations for null-modem cables. It seems that each manufacturer wants to see different signals. All the null-modem cable is doing is crossing over the appropriate signals to make each computer "think" he's tied to a modem, when in fact their connected directly together. I have used the following null-modem on dozens of systems, it uses a couple more wires but it should satisfy anything you hook it to. On the other hand, the worlds simplest null modem only uses 3 wires, and works sometimes with terminals. Here's what always works for me: Side A Side B Pin no Pin no ------ ------ 1 ---------------------------- 1 2 ---------------------------- 3 3 ---------------------------- 2 4 --+------------------------- 8 | 5 --+ 8 -------------------------+-- 4 | +-- 5 6 ---------------------------- 20 7 ---------------------------- 7 20 ---------------------------- 6 I think you see the theory behind this arrangement. Depending on the equipment, you can get away without certain signals. Im sure you'll get a lot of different suggestions from your query. Good luck. John --- John Gayman, WA3WBU | UUCP: uunet!wa3wbu!john 1869 Valley Rd. | ARPA: john@wa3wbu.uu.net Marysville, PA 17053 | Packet: WA3WBU @ AK3P
animal@ernie.NECAM.COM (Alan R. Silverman) (09/07/88)
In article <8809030943.AA02215@ucbvax.Berkeley.EDU>, A1T@PSUECL.BITNET (Andy Tefft) writes: > > - - The original note follows - - > > Can anyone out there tell me how to make a null modem cable? I tried just > connecting corresponding pins of my 2 serial connectors on 2 computers > together, with the serial cards set the same way. No dice. > > Please reply via e-mail unless you think others will be interested. > > Andy Tefft 1 -> 1 Straight thru 2 -> 3 2 crosses to 3, and visa versa 3 -> 2 4 -> 5 4 crosses to 5, and visa versa 5 -> 4 6 -| 6 shorts to 8 on the same side, then crosses to 20 on the 8 -|->20 other side 7 -> 7 Straight thru 20 --> 8 |->6 Note: On VAX Modems, pin 22 is needed for ring indicator. For more info, read RS232 Made Easy. Good simple book.
aad@stpstn.UUCP (Anthony A. Datri) (09/08/88)
I was wary of trying to respond to that path via mail, so here goes: For many things, a null-modem cable merely means a cable wired thus: 2----3 3----2 7----7 20--20 A full-blown, general-purpose null-modem cable would be thus: 1---------1 (always friendly to include the frame ground) 2---------3 3---------2 4---------5 5---------4 6--------20 7---------7 8---------8 (often varies, often not needed) 20--------6 In short, pins 1,7,8 wired straight, 2x3, 4x5, 6x20. The 4-wire form works for most things (terminals, printers), the full-blown is for things that want hardware handshaking. And, of course, if you've got something like an MTI mux or (shudder) a DMF-32, everything's screwy. -- @disclaimer(Any concepts or opinions above are entirely mine, not those of my employer, my GIGI, or my 11/34) beak is beak is not Anthony A. Datri,SysAdmin,StepstoneCorporation,stpstn!aad
syap@cc.rochester.edu (James Fitzwilliam) (09/08/88)
In article <2066@stpstn.UUCP> aad@stpstn.UUCP (Anthony A. Datri) writes: [... two good diagrams omitted...] *In short, pins 1,7,8 wired straight, 2x3, 4x5, 6x20. *The 4-wire form works for most things (terminals, printers), *the full-blown is for things that want hardware handshaking. -- see below *And, of course, if you've got something like an MTI mux or *(shudder) a DMF-32, everything's screwy. I had a different unusual problem when making a null adaptor to connect my laptop and my Macintosh's modem cable (a commercial Din-8 to RS-232). The laptop wanted a handshake and wasn't getting it from the cable, so I wired the adaptor thus: 2 --------- 3 3 --------- 2 7 --------- 7 4 -- -- 4 | | 5 -- -- 5 Each computer handshakes itself, and is constantly in a ready-to-transmit state. Loss of the handshake's function (to make sure both ends are ready for data transfer) is irrelevant, since you have both machines under your control and can see for yourself whether they're both turned on. Probably you won't encounter this situation, but just in case... I was scratching my noggin for quite awhile on it! James domain: syap@vera.cc.rochester.edu path: rochester!vera!syap (I think...) "Piano is my forte" (-: GEnie: FITZWILLIAM ===========================================================================
mrm@sceard.UUCP (M.R.Murphy) (09/09/88)
In article <2066@stpstn.UUCP> aad@stpstn.UUCP (Anthony A. Datri) writes:
[...simple null modem diagram deleted]
..>
..>A full-blown, general-purpose null-modem cable would be thus:
..>
..>1---------1 (always friendly to include the frame ground)
[...rest of null modem diagram deleted]
..>
..>In short, pins 1,7,8 wired straight, 2x3, 4x5, 6x20.
..>The 4-wire form works for most things (terminals, printers),
..>the full-blown is for things that want hardware handshaking.
..>And, of course, if you've got something like an MTI mux or
..>(shudder) a DMF-32, everything's screwy.
..>
..>
..>--
..>@disclaimer(Any concepts or opinions above are entirely mine, not those of my
..> employer, my GIGI, or my 11/34)
..>beak is beak is not
..>Anthony A. Datri,SysAdmin,StepstoneCorporation,stpstn!aad
Including the 1-1 connection can really be exciting if someone (electrician,
homeowner, facilities group, ...:-) has miswired the AC and reversed
hot and neutral somewhere. The 1-1 connection can then become a fuse.
It's not always a good thing to connect frame-frame.
--
Mike Murphy Sceard Systems, Inc. 544 South Pacific St. San Marcos, CA 92069
ARPA: sceard!mrm@nosc.MIL BITNET: MURPHY@UCLACH
UUCP: ucsd!sceard!mrm INTERNET: mrm%sceard.UUCP@ucsd.ucsd.edupoage@sunny.UUCP (Tom Poage) (09/10/88)
In article <829@sceard.UUCP> mrm@sceard.UUCP (0040-M.R.Murphy) writes: ... >Including the 1-1 connection can really be exciting if someone (electrician, >homeowner, facilities group, ...:-) has miswired the AC and reversed >hot and neutral somewhere. The 1-1 connection can then become a fuse. >It's not always a good thing to connect frame-frame. The result of miswiring the power to any properly manufactured computer by reversing hot and neutral will be an increase in 60 Hz leakage to ground through the third wire (chassis ground) by capacitive coupling. If per chance hot and CHASSIS ground (the third prong on your computer or peripheral) were swapped, the computer would blow a fuse at the switch box before anything ever got turned on. For those devices that have only two prongs, they are normally isolated. It makes no difference if the prongs on this type of device are switched. The risk to computers and peripheral equipment is relatively low. In addition, all surges across pin 1 of, e.g., an RS-232 connection go directly to ground through the chassis of either device -- ISOLATED FROM THE REST OF THE DEVICE AND IT'S POWER SUPPLY. Now for the advantages of pin-1. Since the lines which carry signals, including signal ground, are isolated from chassis ground, a potential problem arises. Ground wires in the walls have a finite resistance, and as a result two pieces of equipment may be operating at different DC potentials from each other relative to chassis ground. If this DC offset is great enough, there is a risk of blowing the circuitry that drives the signals. The purpose of pin-1 is to reduce or eliminate this potentially catastrophic DC potential. In some cases, pin-1 could cause problems: lightning storms for one; but in this case you should shut the stuff down anyhow. Tom Poage UCDMC ucbvax!ucdavis!sunny!poage
friedl@vsi.UUCP (Stephen J. Friedl) (09/10/88)
In article <2066@stpstn.UUCP> aad@stpstn.UUCP (Anthony A. Datri) writes: < < [building a null-modem cable] < < 1---------1 (always friendly to include the frame ground) < In article <829@sceard.UUCP>, mrm@sceard.UUCP (M.R.Murphy) writes: < < Including the 1-1 connection can really be exciting if someone (electrician, < homeowner, facilities group, ...:-) has miswired the AC and reversed < hot and neutral somewhere. The 1-1 connection can then become a fuse. < It's not always a good thing to connect frame-frame. I believe at one time the National Electrical Code said that wiring 1 <---> 1 is a violation of code for this reason. Logic ground is fine with pin seven, but a 32-gauge wire is not sufficient to provide a safe frame ground. Anybody know if this is still the case? -- Steve Friedl V-Systems, Inc. +1 714 545 6442 3B2-kind-of-guy friedl@vsi.com {backbones}!vsi.com!friedl attmail!vsi!friedl ------------Nancy Reagan on conductance: "Just say mho"------------
kaufman@polya.Stanford.EDU (Marc T. Kaufman) (09/10/88)
In article <829@sceard.UUCP> mrm@sceard.UUCP (0040-M.R.Murphy) writes: >In article <2066@stpstn.UUCP> aad@stpstn.UUCP (Anthony A. Datri) writes: ..>A full-blown, general-purpose null-modem cable would be thus: ..>1---------1 (always friendly to include the frame ground) >Including the 1-1 connection can really be exciting if someone (electrician, >homeowner, facilities group, ...:-) has miswired the AC and reversed >hot and neutral somewhere. The 1-1 connection can then become a fuse. >It's not always a good thing to connect frame-frame. Generally, if this happens (and I HAVE seen it happen), pin 7 (logic ground) will also be at 115 volts, since most equipment connects logic ground to frame ground somewhere. Note that, even if the electrician has miswired the socket, most likely nothing will happen since in either linear or switching supplies there is a transformer between the AC and secondary voltage sides. Where I found problems was in a (TVI 950) terminal, where the supposed "neutral" side of the line was connected to the frame through a resistor, to keep the frame from floating above ground (did you ever rub your fingers lightly on a chassis and get a slight tingling? That is from the capacitive coupling between the hot side of the line and the chassis... sometimes due to deliberately placed capacitors which are intended to keep RFI out of the device). The correct use of frame ground in RS-232 environments is to provide electrostatic shielding for the signal lines. If you are using Belden cable with an overall shield, connect pin 1 to the shield at ONE END ONLY of the cable. Then make sure that the equipment is correctly grounded. The connection at one end only prevents current flow on the shield which can couple into the signal lines. Marc Kaufman (kaufman@polya.stanford.edu)
jgp@moscom.UUCP (Jim Prescott) (09/14/88)
In article <2066@stpstn.UUCP> aad@stpstn.UUCP (Anthony A. Datri) writes: >For many things, a null-modem cable merely means a cable wired thus: > >2----3 >3----2 >7----7 >20--20 I think this is bad. Since null modem connects two DTE devices it isn't a good idea to tie lines asserted by DTE's together; in the above cable both sides will be asserting 20 (DTR). I've heard that having both sides try to assert the same line results in a greatly shortened lifespan of one or both sides; anybody know if this is actually true? Just dropping 20 is probably fine for a minimal null-modem cable. >A full-blown, general-purpose null-modem cable would be thus: >1---------1 (always friendly to include the frame ground) >2---------3 >3---------2 >4---------5 >5---------4 >6--------20 >7---------7 >8---------8 (often varies, often not needed) >20--------6 Tying pin 8 through isn't as bad since it is DCE but it probably won't have any effect (both sides are reading the pin, nobody writes it). I usually find 6,8 -> 20 in both directions to be best. This assumes that the devices at least follow the distinction between DTE and DCE lines. Are there many devices that don't (ie assert some DTE and some DCE lines)? -- Jim Prescott moscom!jgp@cs.rochester.edu {rutgers,ames,harvard}!rochester!moscom!jgp