No announcement yet.

Please help with a PID issue

  • Filter
  • Time
  • Show
Clear All
new posts

  • Please help with a PID issue

    Hi all,

    I am trying to connect another thermometre sensor to my roaster which as a PID. The external one is able to connect to my computer.

    The problem i have when i conenct them is that i am getting slightly different about 1 - 5 degrees different readings. The delta movements are also not the same either between the reading (Roaster PID and the external sensor).

    I have ensured that i have connected the correct +/- to the right wire. What else could be causing this issue?

    Many thanks in advance,

  • #2
    Hi Mike,

    Not quite sure of what the situation is - a picture is often worth a thousand words.....

    In any case - I firstly assume that you are using thermocouples. There are different types of thermocouples and not all are created equal. But they are all designed to measure temperature. Most common from what I understand that you will see in the coffee industry will be K type thermocouples, sometimes J. Personally I have used K types when I pimped a La Marzocco a few years ago due to the PID controller requiring a K type, and today I am just changing over the thermocuples on my roaster from exposed bead type K to probe type K's (my datalogger uses K type thermocouples).

    If they are a different type, you will get a slightly different reading. likewise, if they are different style (grounded, ungrounded, probe, bead etc) you will get different response rates.

    Often there is a function for an offset in the PID controller - that might be of assistance, but again i do not know what you have or the programming of it.......and I am just an interested amateur...

    In any case - first if it was me I would be wanting to compare apples and apples............get some crushed ice with a smiggin of water (you want the crushed ice and a little water - so you can get the 0 degree C point). Insert the probes and check the difference in the 2. That will give you a good start of knowing how far they are apart from each other. The response rate I am assuming that is possibly because of different manufacturers or types of probe.

    For what is worth, the thermocouple on my roaster that came with it connected to the digital temp indicator is a type J, the probe is big and bulky and has hugely slow response rate. Not sure why it came with it, but it is worth #$%^t. A bead type in the same proximity gives me up to 60 C difference depending on rate of change of temperature......

    Hope it helps a little.


    • #3
      Thank you so much for your response an3!

      I tried to add attachments but i couldnt since i might be too new a member.

      I will check with the manufactuer of the roaster as to which probe and cable they are using. My thermometer and probe is a K type.

      I do have a dumb question. If i cut the probe and simply wire the probe going to the thermometer with the wire going to the PID in the roaster, in theory if the cables are the same type then should it give the same reading?

      Many thanks,


      • #4
        I would not be in such a rush to cut the cable.

        When you cut the leads and apply a different lead, it must be of the correct metal......otherwise it will effect the reading of the thermocouple. Each type of thermocouple has a specific requirement for cabling type. Best to leave it alone until at least you know what you have.

        Still not quite sure what you are trying to do? Is it have dual temp output from the roaster so that one will drive the roaster PID and another for a datalogger / computer datalogger?


        • #5
          Hi an3,

          I am trying to connect my roaster, which has a PID to a datalogger. I wish i could upload pictures to show it.

          Thanks for letting me know about the load of the wires and the reasons why it looks like i am getting different results. I think the best way would be to remove the thermocouple from the roaster and replace it will another that can go straight into the datalogger, right?


          • #6
            Not sure what you have or how it works.

            For mine - it is gas powered and completely manual. I have 2 x k thermocouples into my Center 309 datalogger which in turn is connected via RS232 USB converter into the Macbook. 1 thermocouple I have positioned in the bean mass and the other to measure the environmental temperature.

            I am imagining you may still need that thermocouple connected to your pid - as obviously the pid output is driving your heating? and it needs a source input (thermcouple) to run its algorithm ie no thermocouple input to your PID = no ability to drive the PID/heating.

            Sorry - love to help some use of PID's has been with single loop Yamatake and Fuji to drive constant temps and not exactly what I have a sneeky suspicion of what yours could be controlling a ramp/soak any case if you work out your picture function I am sure someone passing through will be able to assist.