Shop OBEX P1 Docs P2 Docs Learn Events
CAT-5 Cable Tester — Parallax Forums

CAT-5 Cable Tester

Basic JimBasic Jim Posts: 106
edited 2008-07-29 15:22 in General Discussion
Hello all,

I am currently working setting up temopary offices and wiring them with Cat-5 cabling for networking. Most of the techs I work with have a tester to test the cables they make. I thought since I can program an SX48 Proto board I could make one myself and save a few bucks.

I was thinking that where some runs are quite long , I should use a drive·transistor connected to 9 volts to send a signal. And a loopback plug on the other end that would connect two of the wires and the Proto board would sense the signal and light the appropriate LEDs.

One question is, depending on the length of the cable the voltage returned to the input of the SX could be higher than 5 volts. So, I intend to insert 330 ohm resitors to the input lines. Should I be safe with that?

For now I'll be happy with just an led on or off that wiil tell me that my cables are good. In the future I think that I might build one that uses an LCD display that would give more Information like read resistance in the wire or even tell me the length of the wire by measuring resistance or measuring delay time. or tell me the style cable being tested.

Second question, has anyone already done this?

Thanks,
Jim W.

Comments

  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2008-07-23 19:11
    Jim,

    There will be a little more to it than that. You don't have to work with voltages over 5V, but you will want to buffer the output in case there is a short in the wiring somewhere. Back in the late 90's my business was contracted to build several pieces of test equipment. Of those we built a cable tester for Serial, Printer, PS/2, etc. The design was fairly simple but you need to be able to detect shorts, open circuits, crossed/miswired cables, etc. On our tester both ends of the cable plugged into the same box. The end was a module that could be sawpped for different cables. Essentially the device allowed up to 50 conductor cable assemblies to be tested. As I recall the device used several 8255 PIA chips, so latches and some buffers. Kind of makes me want to dig back into it and redo the whole thing on a newer controller.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
  • Basic JimBasic Jim Posts: 106
    edited 2008-07-23 20:10
    Hello Chris,

    Wouldn't a transistor with a small ohmage resistor on the base that is driven by an SX output pin with a grounded emiter and the colector having a resistor to +Vin provide enough protection? If the wire under test is shorted to ground you still have the resistor going to +vin for protection. The other end would be a simple loopback plug where say the white/ orange wire is connected to the orange/white wire. So, I wouldn't see a problem unless someone pulled it off and pluged it into a router. Even then I don't think it would cause a catastrofic failure given that even the input side would have a 330 ohm resistor. Do you think I would be better off to use an open colector buffer chip like a 7406?

    Thanks,
    Jim W.
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2008-07-23 21:26
    Jim,

    Unless I am misunderstanding the transistor configuration you mean, it sounds like your outputs into the cable would be high by default until the microcontroller applied a high to the base of the transistor to turn them off. In our system I remember we wanted the power-up state to be all lines off/low. One unit had them set to inputs. I know I still have the Z80 source around somewhere. I can always check to see what exactly we did back then (it was awhile ago). I know we didn’t use transistors for the outputs though.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
  • pwillardpwillard Posts: 321
    edited 2008-07-23 21:47
    Eons ago I had to make a 25-pair telco cable tester using just TTL logic (we received a shipment of unsplashed/unmarked cable and had to act fast to be sure were were keeping the right pairs together) We used open collector logic (SN7407) and used a 4.7K pullup on each pin of a sequential counter. We were able to get the signals to travel 4000ft with no problem winking the return indicators. A little signal loss due to cable length is OK in a Go/No-Go test.

    Our test would send a level change on a PIN (pull low) and expect a related response in indicator pin. Cable was BAD if response/sequence was wrong.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


    There's nothing like a new idea and a warm soldering iron.
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2008-07-23 22:38
    Here is a link to build your own CAT-5 cable tester.

    http://www.netadmintools.com/art145.html

    and some more...

    http://www.bikearound.co.uk/index.php?pr=Cable_Tester

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support


    Post Edited (Chris Savage (Parallax)) : 7/24/2008 2:33:56 PM GMT
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-07-24 06:21
    Um, Chris, your link is self-referential. I think you may have pasted the wrong URL...

    -Phil

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    'Still some PropSTICK Kit bare PCBs left!
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2008-07-24 14:32
    Phil,

    Thanks for letting me know…I have no idea how that happened since I had to edit the link prior to pasting it. Anyway, I added a second link in which the output drives look very similar to the design we used (buffer/driver with resistors).

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
  • Basic JimBasic Jim Posts: 106
    edited 2008-07-29 15:22
    Thank you guys for the great advise, I'm sorry it took so long to get back. I have been on the road since. I will start on my project when I get back.



    Thanks,

    Jim W.jumpin.gif
Sign In or Register to comment.