Led based continuity tester

Discussion in 'The Projects Forum' started by l33tfr34k, Feb 5, 2016.

  1. l33tfr34k

    Thread Starter New Member

    Feb 5, 2016
    So here's my project. I'm designing a test rack for the stage automation department at my University and part of the design is the ability to test the continuity of cables for encoder/limit signals (9 pins) and enable/reference signals (14 pins). For each of the 2 types of cables I want to be able to test male-male, male-female, and female-female cables. And I want to use leds for the verification lights, thats 69 total leds. I'd like to use a 24v power supply but that can be changed. My big issue is that I never know how many leds will be used at any one time. Multiple cables could be tested simultaneously and in those cables there maybe bad continuity across one or more pin(s).

    All the leds would need to be wired parallel to the source but I'm having trouble figuring out how to provide current control to the system. So far the best that I have come up with is to put the leds in groups of 3, each group would have a signal resistor. When all 3 leds are on, the current on each would be about half the rated amount (so a dimmer led), if only one in that group is one then the single led would have a higher current then the rated amount (but possible still within acceptable range depending on the led).

    I have very little experience using leds and would really appreciate some help. Would the use of led drivers help this circuit? And if so, what would be some good recommendations?
  2. wayneh


    Sep 9, 2010
    I don't understand your description. I get the part about one LED per conductor under test. How do you get to 3 LEDs in a group, and what defines a group?

    How many LEDs might be lit at once, max? A 5V supply with sufficient current to light the max number of LEDs would do the trick. You'll then need a current-limiting resistor for each and every LED. An old computer PSU would be an option, and you can find one for free in the recycle dump. But you may not need anywhere near that much current if you don't need to light more than, say, 20 LEDs at a time.

    Another option would be to use LEDs with built in resistors meant for a 12V supply.

    Note that for indication instead of illumination, 5mA is plenty versus the 20-30mA if you need to light something up.
  3. dl324

    Distinguished Member

    Mar 30, 2015
    Welcome to AAC!
    Don't overthink this. You just need to have appropriate connectors, loop back jigs, and a low voltage power source to illuminate the LEDs so you can visually verify connectivity. A cable tester should be portable (i.e. battery operated) and you don't need to be able to see the LEDs from any significant distance.

    Google for cable testers to get some ideas or even buy one.
  4. mcgyvr

    AAC Fanatic!

    Oct 15, 2009
    Do you need to detect improper wiring too? (aka pin 2 going to pin 3 on the other end instead of pin 2)
    If so that complicates it..
  5. dl324

    Distinguished Member

    Mar 30, 2015
    Something like this checks RJ11/12/45 cables for continuity, correct wiring, and shorts. It costs about $5. I bought one around 10 years ago and just checked it. The NiMH battery I had in it still has enough juice to power it.

    It applies voltage to the wires individually, so you have to wait for it to cycle through all conductors; though it does have fast and slow cycle speeds. The new ones don't seem to test BNC cables; mine does...

    You could use this as a starting point for your tester or look for something to buy.
  6. mcgyvr

    AAC Fanatic!

    Oct 15, 2009
    I'm using an Arduino mega for continuity and correct wiring verification..
    I set up 20 pins for output and 20 for input.. and the cable goes between them..
    I set a single output high and then scan all the input channels to make sure only that one is high..
    Then on to the next position.. over and over again..
  7. ebeowulf17

    Active Member

    Aug 12, 2014
    I'm helping a friend at work put together a tester which will use the same test methodology, except with a pro trinket (limited i/o) and three i/o expander chips. Rocky start though - either damaged a chip or we're having chip addressing problems. Only the first of the chips works as expected.

    Are you doing anything for current limiting so that in the case of a short you don't fry a gpio pin? I didn't think of the risk until after we had ordered parts. Adding current limiting resistors to every output nearly doubles the assembly work, but is probably worthwhile unless there's a better way. Any thoughts?