This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PT100 Interface with MSP430F2410

Other Parts Discussed in Thread: MSP430F2410

Hi,

I want to use pt100 RTD temperature sensor with MSPF4302410. The sensor will be connected to ADC12.  The application is battery driven so the power budget is very short. The board size is also short its 24 mm X 34 mm.

I have seen following possible solution for PT100.

 RTD: Op-amp approach

RTD: with 4-20mA output: XTR approach: Using XTR105

RTD: AFE approach: Using LMP90100

http://www.ti.com/ww/en/industrial/sensors/rtd/design.html

I want to know which one is  most cost ,area and power effective.

Can you please send some Reference  schematic for .

Regards

Irfan

  • A PT100 can be operated in two modes.

    First is a constant current. It results in a linear voltage over the PT100. However, constant currents are difficult to do if one side of the PT100 is connected to GND. And if not, then you'll have to sample two voltages (or use an OpAmp) to get the voltage over the PT100.

    The otehr way is to use a constant voltage and a head resistor. Then the voltage output of the PT100 is not linear anymore. However, it is the simplest setup. Just have the PT100 in series with a head resistor connected to an MSP port pin. This way you can switch the sensoe ona nd off, saving power. And you only need to cample one voltage (the PT100 tied to GND on one side and the other side is checked).
    The result, because non-linear, then has to be converted using a lookup table. You can use Excel to create the table with the required precision, based on the know head resistor value, the known supply voltage and the also known PT100 resistance for each temperature.

    You'll have to carefully chose the head resistor, so you'll get the maximum resolution around the expected operating temperature.

    Using an XTR definitely isn't low power. And using an OpAmp usually isn't too. The LMP does the constant current approach. With more precision than a simple constant current source and some calibration. Still requires more power than the simple approach with a head resistor.

  • Hi Jens Michael,

    Thank you for your reply.  I will be using the following low power solution suggested by you.

    "The otehr way is to use a constant voltage and a head resistor. Then the voltage output of the PT100 is not linear anymore. However, it is the simplest setup. Just have the PT100 in series with a head resistor connected to an MSP port pin. This way you can switch the sensoe ona nd off, saving power. And you only need to cample one voltage (the PT100 tied to GND on one side and the other side is checked).
    The result, because non-linear, then has to be converted using a lookup table. You can use Excel to create the table with the required precision, based on the know head resistor value, the known supply voltage and the also known PT100 resistance for each temperature."

    From above suggestion my understanding is that I have to connect pt100 three terminal as follows.

    Terminal 1 : Should be connected to MSP430 through header resistor . Should it be connected to any  GPIO port or some specific such as Vref ? ( MSP430F2410)

    Terminal 2: Should be connected to  12 bit ADC of MSP430.

    Terminal 3: Should be grounded .

    Please correct me if I am wrong.

    Can you please refer to any reference circuit diagram for this.

    Best Regards

    Irfan Awan 

  • I'm not sure, as I don't know on which side of the PT100 resistive core the Terminal 2 is connected. Basically, you only need Terminal 1 (supply through resistor and directly measured by MSP) and Terminal 3 (grounded).

    About the supply, well, you can use VRef (requires some checks whether Vref can source enough current). This will give you the most precise reference. And oyu can switch the reference on and off if required. However, switching the reference on takes some time, so you'll have to wait. And this consumes power. A port pin will only output VCC (see port pin description for the voltage drop on the expected current), btu will do so instantly when you switch it to high and will stop drawing any current when you switch it to low.
    It's a current consumption/precision tradeoff.
    Also, higher current means higher voltage reading. And higher voltage reading means better resolution on the ADC (as long as the maximum voltage doe snot exceed the reference voltage). But then, higher current means higher self-heating and therefore less precision again. Again a tradeoff. The optimum point is to be calculated by you based on datasheets and applicaiton requirements. That's the real engineers work :)

  • Hi  Jens-Michae,

    We plan to use MOSFET as switch Inorder to on and off the reference voltage of bridge to pt100.

    The MOSFET base will be connected to MSP430  GPIO so it will get 3.3V when set high.

    Can you recommend any MOSFET suitble for This purpose. 

  • Why not using hte MSPs internal transistors? They are MOSFETs too - and you won't have the problems with the gate threshold voltage etc.

    When usign the MSPs itnernal reference, you can enable and disable the reference output in the ADC module.
    And if usign VCC, well, each port pin already is a MSOFET transistor between VCC and the port pin. :)

**Attention** This is a public forum