Shop OBEX P1 Docs P2 Docs Learn Events
PT100 Signal Conversion IC — Parallax Forums

PT100 Signal Conversion IC

hcwonghcwong Posts: 38
edited 2005-11-15 22:11 in General Discussion
Dear all,

Just wondering if any of you know of an IC which can connect a PT100 on the input and on the output it send out digital data of temperature.

I am currently working on a project to measure temperature of liquids·with a·0.001 deg C resolution, 0.005 deg C accuracy and·for the range of 0 - 300 deg C. Using DS18S20 is out of the question as the chip cannot be immersed into liquid and its resolution is not high enough. I have made my own circuit as below but the reading fluctuate up and down by +/- 0.04 deg C. I am lost and do not know how to proceed further.

Two·precision constant current source of 100uA is connected to a PT100 and a 100 ohm 0.1% resistor. The differential voltage of the PT100 and 100ohm resistor is fed to an instrument amplifier, which is amplified by 50x. The data is then fed to a 24-bit ADC and the ADC is connected to a BS2. The ADC by itself is giving very stable reading of upto 1uV·and I can get an accurate reading of 0.1 deg C if I connect the PT100 / 100 0hms resistor directly to the ADC. In order for me to get a reading of·0.001 deg C, I need to amplify the signal by 50x. But adding the instrument amplifier creates the fluctuation problem and the reading is never stable. I even use moving average with·50 data points to·stabilise it but it is still fluctuating by +/- 0.4 deg C.

Is there·such an IC which has a constant current source, an instrument amplifier and a 24-bit ADC it it, all in one package?

Thanks.

HC Wong···

Comments

  • pjvpjv Posts: 1,903
    edited 2005-11-15 05:12
    Hello Mr. Wong;

    I believe you will have an impossible time getting to 0.005 C repeatability over a range of 300 degrees with anyting but the most exotic techniques. That is almost one part in 65,535; in an analog world, that's REALLY tough as references of that stability are no cinch. Good luck!

    Cheers,

    Peter (pjv)
  • hcwonghcwong Posts: 38
    edited 2005-11-15 21:17
    I have overstated by requirements. I can live with the following specs :

    Resolution·: 0.01 deg C
    Accuracy : 0.05 degC
    Range : 0 - 300 deg C

    Anyone can help? Please....

    HC Wong
  • pjvpjv Posts: 1,903
    edited 2005-11-15 22:11
    Hello Mr. Wong;

    I believe you have already met your spec of 0.05 C as your cicuit gives 0.04 C; or am I missing something?

    Cheers,

    Peter (pjv)
Sign In or Register to comment.