Why can't I convert 10mV voltage from Thermocouple to current and measure it?

Discussion in 'General Electronics Chat' started by booboo, May 13, 2015.

  1. booboo

    Thread Starter Member

    Apr 25, 2015
    165
    2
    Hey guys
    I have a thermocouple and trying to convert its 10mV(at 450degree) to current by a resistor(1.5ohm) and measur it by a digital multimeter (it's a VC9805). this is the circuit:

    [​IMG]

    I'm trying to convert the output of this thermocouple to current and measure it. the multimeter shows nothing. to adjust the multimeter, I connected the black prob to the "COM" terminal and the negative of the thermocouple. and the red prob to the "mA" terminal and the positive of the thermocouple.

    please note that I don't want to do it with Op-Amp.
     
  2. Reloadron

    Active Member

    Jan 15, 2015
    963
    232
    First you may want to give this a read. The link is an informative read from AAC (these forums).

    The current from the thermocouple is a function (is limited by) circuit resistance. Now figure the voltage you have to start with is way down in the mV region, therefore with the very low resistance the current will be negligible. Also your drawing will have additional junctions where your + Probe and - Probe connect to your TC alloy, that needs considered for CJC (Cold Junction Compensation).

    Also keep in mind the thermocouple is not a current source. If you measure the voltage across your 1.5 Ohm resistor I doubt you will see 10 mV. :(

    Ron
     
    Last edited: May 13, 2015
    booboo likes this.
  3. booboo

    Thread Starter Member

    Apr 25, 2015
    165
    2
    Thanks Ron
    My goal is to convert the voltage from the thermocouple to current and use a CA3130E as a Transimpedance amplifier and measure the current. finally, calculate temperature. that's what I think. CA3130 has low Input Bias and Offset current.
     
  4. Reloadron

    Active Member

    Jan 15, 2015
    963
    232
    Well I will share with you how I have done it in industry. The first problem we face with a thermocouple is that CJC which I mentioned. Then with CJC resolved the best method is to amplify the small mV TC output using an IA (Instrumentation Amplifier) configuration. There is also a matter of when amplifying the TC output the parts placement on a PCB is critical, especially if you have any hope of accuracy and repeatability. All of this makes for a pain in the neck. Even using amplifiers like the ones made for TCs which include CJC.

    I like using "Temperature Transmitters" and as you can see by the link everybody makes the things. The newer ones are even programmable for TC Type and Range. For example you would program a temp transmitter for Type K (it does the linearization curve) with a range of maybe 0 to 500 Degrees F. Then you program the output for example for 4 to 20 mA. So now you have 0 to 500F. = 4 to 20 mA. Want volts? Run the 4 to 20 mA current loop through a 250 or 500 Ohm resistor and get 1 to 5 volts or 2 to 10 volts.

    Here are a few images of a small project I did years ago regarding computer case temperatures. I used J Thermocouples and programmed the temp trans for 0 to 400 F.

    Depending on your specific application and budget something like this might be a viable solution.

    Ron
     
    booboo likes this.
  5. crutschow

    Expert

    Mar 14, 2008
    13,028
    3,236
    Why do you want to measure current rather than voltage?
    The accuracy of the current measurement will be degraded by any resistances in the circuit including that of the thermocouple junction and wire.
    Thermocouple wire can have significant resistance as shown by this table.
     
    booboo likes this.
Loading...