Using an LM334 as a constant current source to measure CRV on a pot

Discussion in 'The Projects Forum' started by dannomassey, Mar 15, 2014.

  1. dannomassey

    Thread Starter New Member

    Jul 28, 2010
    Hi guy's, first of all, thanks in advance for any help its much appreciated.
    I have a very simple test circuit for measuring contact resistance variation on potentiometers and i'm using a LM334 as a 1mA constant current source. The cap is just being used as a DC filter, I'm only interested in the transient voltage. I have put a 1ohm series resistor on the LM334 output to monitor the current and am finding it is ramping up exponentially as the load is reduced on the pot. So the question is, should this cct work and if so what have I overlooked. Cheers all.
    Last edited by a moderator: Mar 15, 2014
  2. #12


    Nov 30, 2010
    Bad design. First, that drawing does not show how to make a constant current circuit. Second, the 317 chip has a minimum load of 10 ma. Third, the capacitor coupling to the scope will show all the dirty spots on the pot AND the wiper contact. I'd show you a constant current circuit, but the basic concept is flawed. I don't know how to separate the contact resistance variations from the resistance path variations.
  3. dannomassey

    Thread Starter New Member

    Jul 28, 2010
    Hi, thanks for the reply, i've been away this weekend so apologies for not getting back.
    The device used is an LM334 not a 317 and so has no minimum load issues, also I do want to see the dirty spots on the pot and the wiper contact as I need to see the full variation over the length of the contact. The LM334 datasheets provide a basic aplication note for a constat current source as per attachment CCT5 where Rset is given by ISET = 67.7 mV/RSET.
    If I am looking at this far to simplisticly then please point me in the right direction.
    • cct5.png
      File size:
      60.7 KB
  4. atferrari

    AAC Fanatic!

    Jan 6, 2004
    In the same datasheet there is a simple circuit that would allow you to keep it stable vis a vis temperature. I used it and works OK. Few components to be added.
    Last edited: Mar 17, 2014
  5. AnalogKid

    Distinguished Member

    Aug 1, 2013
    What is the value of the capacitor in your circuit, and what are typical and max values for the pot under test?

    I think you are seeing the effects of your coupling cap and the pot under test forming an RC time constant circuit. Try this: First, add a high value resistor from the junction of the pot and the cap to ground (the low side of your 10 V source). The resistor should be 10 times the max value of the pot. Second, eliminate the capacitor and put your scope in AC mode. The added resistor will decrease the amplitude of the signal you want by about 10%, but it also will lower the output impedance of the signal. If you are testing pots in the megohm range, the input impedance of thw scope will complicate things.

  6. dannomassey

    Thread Starter New Member

    Jul 28, 2010
    I'm suing a 22nF cap which was an arbitrary value that was at hand and the pot under test is 5K max, I have added a 50K resistior as suggested but this has made no difference, I'm still seeing up to 130mA as the pot is turned right down. Putting the scope in to AC coupling would remove the need for the cap maybe but I will still have this excess current pull from the pot. I was hoping that this simple LM334 circuit would work but clearly its not up to it. atfarrari, which circuit are you refering too, most I could see are for thermal control apart or voltage regulating?
    I really am just after a very simple 1 to 2mA constant current source
    Thanks again,