LTC2984 and 4-wire thermistor measurement?

I would like to use mux chips for measuring many (over 100) thermistors.  The mux chips have too much Ron resistance to provide accurate measurements for each thermistor.  Can I use the 4-wire measurement method (like RTD) but still configure as a thermistor for Steinhart-Hart coefficient calculations?  What would my configuration look like for this?

  • 0
    •  Analog Employees 
    on Apr 13, 2021 11:53 PM

    Hi ,

         Sure, I show an example of doing something very similar on page 9 of my LT Journal article here: 
    https://www.analog.com/media/en/technical-documentation/lt-journal-article/LTJournal-V27N1-00-df-LTC298x-LoganCummings.pdf

          Essentially the channel config is the same as for a directly connected thermistor, but because of the desire to separate the current and voltage paths (Kelvin connect), sharing and rotation must be disabled. 

    • Some things to keep in mind:
      • Auto-ranging is not allowed for custom thermistors. If you're using custom Steinhart-Hart coefficients you'll need to be smart about the excitation current you choose, possibly even adjusting it in use if the thermistor temp changes a lot. 
      • Be mindful of how filter capacitance, etc. is arranged in the circuit. Settling time for the excitation current may be a factor in your measurement.
      • At high values of thermistor resistance leakage currents could cause significant errors. 

    Hope it helps,
    -Logan
     

  • Logan,

    Thank you for the reply.  I have a few comments and a few more questions.

    My IT group has somehow blocked access to Analog.com... would you please send the journal article that you mentioned via email?  I have submitted a ticket... but their LC filter is Low Pass.  :-)

    The thermistor that I am using is a 10kΩ@25°C.  The manufacturer has suggested no more than 10uA source current. 

    For the MUX chip, what level of leakage currents would be acceptable (I am an Mech Eng. living in an Elect. Eng. world).  The MUX chip that I am considering is a ADG732.  Will this MUX IC provide enough channel to channel isolation to not significantly affect the source/measurements?

    Thank you.