Announcement

Collapse
No announcement yet.

Choosing a thermocouple vs rtd vs thermistor module

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts


  • Choosing a thermocouple vs rtd vs thermistor module

    Hello,

    I am planning to use Do-more BRX PLC to achieve PID temperature control of a plastic extrusion machine.

    The temperatures of the extruder rarely exceed 550C. I have a choice of using either an RTD (PT100-3 wire) or a thermocouple for this application. Thermistors won't be able to support this temperature range.

    I did a bit reading and found that RTDs have better accuracy and stability but slow response time compared to thermocouples. Thermocouples, on the other hand, have issues with EMI interference, the thickness of wires used and lower accuracy compared to RTDs.

    The dynamics of my extrusion system are slower and the response time of less than 5 to 8 seconds should be acceptable.

    Which sensor type should I go for? This also dictates which temperature module to purchase for my PLC (looking at the prices, both the modules are similarly priced in automation direct website)


  • #2
    Most plastic extrusion that I have been around have a standard on Type J thermocouples, which means there are spares around.
    I have used thermocouples with PLC PID control over the extrusion lines for ten years or more.
    Cant say that they are better, but are simple to use and personnel to troubleshoot.

    Usually try to use remote IO to keep thermocouple wire short.
    Have used Click plc with thermocouple modules on a Modbus connection, Pac Remote IO, and am currently putting in BRX remote IO on a line.

    Comment



    • #3
      Originally posted by RogerR View Post
      Usually, try to use remote IO to keep thermocouple wire short.
      Have used Click plc with thermocouple modules on a Modbus connection, Pac Remote IO, and am currently putting in BRX remote IO on a line.
      Thanks for the input. Do you have an idea of how long I can run the thermocouple wires without any negative effects? For my system, the length of the wire will be less than 6 ft.

      Comment



      • #4
        My short remote IO thermocouple runs are 8-20 ft.
        I have had some lines with 35-45 ft runs that worked OK.
        Always kept them in separate conduit and away from heater conductors.

        If there was a zone that was critical or had physical thermocouple damage possibility, I would put in three thermocouples.
        Take the average of the two closest values and disregard the third. Even with a total failure on one sensor, the PID loop would continue normally.
        When any one of the three sensors was out more than 10% from the other two, an alarm would set.


        Comment



        • #5
          Originally posted by RogerR View Post
          My short remote IO thermocouple runs are 8-20 ft.
          I have had some lines with 35-45 ft runs that worked OK.
          Always kept them in separate conduit and away from heater conductors.

          If there was a zone that was critical or had physical thermocouple damage possibility, I would put in three thermocouples.
          Take the average of the two closest values and disregard the third. Even with a total failure on one sensor, the PID loop would continue normally.
          When any one of the three sensors was out more than 10% from the other two, an alarm would set.

          thank you for the tip! From your experience how off are thermistor readings from actual temperature? I have seen cases where temperatures were off by 30 C.

          Comment



          • #6
            Someone else may have to weigh in here. Never used thermistors.
            It would seem that if the temperature was off that far and fluctuating, the sensor was faulty or there was some power/noise problems.
            If the reading slowly becomes more offset to a plateau, then maybe some self heating effect.
            If the reading was steady and just offset, then maybe a calibration/setup issue.

            Comment



            • #7
              Originally posted by RogerR View Post
              Someone else may have to weigh in here. Never used thermistors.
              It would seem that if the temperature was off that far and fluctuating, the sensor was faulty or there was some power/noise problems.
              If the reading slowly becomes more offset to a plateau, then maybe some self heating effect.
              If the reading was steady and just offset, then maybe a calibration/setup issue.
              Oops, I meant thermocouple. Not thermistor. Sorry for the confusion

              Comment



              • #8
                I think the feedback is within a few degrees or better at 400 deg F. Have not performed the thermocouple in ice water / boiling water routine for a long time.
                Usually when there is an obvious temperature measurement problem, there is physical damage, location of the sensor off the base metal, damaged/pinched cords which can cause additional junctions to be made in the middle, non thermocouple wire being used, terminal strips that are not thermocouple material or stainless steel, etc.
                Some thermocouple modules require that all channels be used or jumpered. This can make a channel drift also.

                Comment

                Working...
                X