• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 47 subscribers
  • Views 516 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

INA592: Designing with INA592IDT

Shibin Varghese
Intellectual 835 points
Part Number: INA592
Other Parts Discussed in Thread: DAC80502,

Hello Team,

We are designing with INA592IDT for a Howland current source as per the technical documentation Analysis of Improved Howland Current Pump Configurations.
Attached is the circuit for the same.


The signal DAC_CH_A is from a DAC (DAC80502DRXT) and the voltage varies from 0V to 5V.
The current thus generated is used as the excitation current of some RTD sensors like PT100, PT1000, and diode.

We are currently developing the firmware with a decade box as the load resistor (Sensor).
As the resistance is changed from 1K to 3K gradually, the current seems to be increasing (the current is initially set to 1mA, using the DAC voltage and resistance using mux).

When we measure the voltage at the DAC, the DAC voltage seems to be stable at the voltage we set for 1 mA.
But the voltage at the resistor R150 (node between R150 and C127), the voltage seems to be increasing as we increase the load resistance using the decade box.

What can be the reason for this?







  • 0
    TI__Mastermind 26278 points

    Shibin,

    Here are some things to consider for proper operation of this circuit.

    1. The feedback and input resistor on both the inverting and non-inverting input need to be balanced.  In the Holland Current Pump document you reference it shows this on page 4.  R1 = R3, and R2=R4.  In your case you added R150.  This will unbalance the circuit.  You should directly drive the input with a low impedance source.  Can you drive directly with the DAC?
    2. The current at the output of the Howland will need to flow through an impedance to ground.  Also, the voltage developed by the current needs to be in a range that the INA592 can drive.  Once it is outside of the INA drive capability it is called the "compliance limit".  In your circuit I do not see the full current path.  Current would flow through R76 to ground.  Is this where your decade box is connected?
    3. I simulated the circuit (see below).  I used a -10V to +10V DAC signal.  The "sense" resistor is set to 400ohms.  This will generate a current from -14.06mA to +10.938mA.  The calculations are also shown below.  The calculations and simulations match.  If you look at the simulation results I parameter step the load.  That means that I test with different loads and see how the response changes.  Notice that the outputs all track regardless of the load until the INA592 hits either it's positive or negative output swing limitation.  When a large resistor like 5k is used this happens sooner.  When a smaller resistor like 500 ohms is used, the amplifier can swing across the entire input range without limitation.  
    4. One possible issue with your circuit is that you are hitting a compliance limit.  Another issue is the usage of the external resistor. R150.

    holland.TSC

    I hope this is helpful to you.

    Best regards,

    Art

  • 0 in reply to Art Kay
    Intellectual 835 points

    Hello AK,
    Thank you for your reply.
    In our case, INA592IDT is powered from a +-15V supply.
    The DAC can swing only between 0 and 5V.
    We are using the Howland current source to generate the excitation current for PT100, PT1000, and some other RTDs.
    The resistance of the PT1000 can be as high as 2.6K ohms while the excitation current is remaining at 1mA throughout the range of PT1000 resistance change.
    We can drive the Howland directly from the DAC.

    In our design, the current output of the Howland current source (through the R76 resistor) is passing through the sensor (PT1000, PT100) and then to a sense resistor. The sense resistor is a 2K Ohms resistor used to ensure the desired value of current (Set by the DAC) is passing through the sensor.
    The sense resistor is connected to the inputs of a current sense amplifier (INA190A1).






    Hope our design is clear to you.
    Looking for your reply.


  • 0 in reply to Shibin Varghese
    TI__Mastermind 26278 points

    Shibin,

    I selected components so that you get a range of 0mA to 2mA.  You can use the equations below to adjust your range as needed.  The voltage reference on the inverting input if the INA592 is not needed if you only want positive current.  I tested the output swing for a 100ohm, 1k and 3k load.  There is no issue with compliance in this range as the 3k x 2mA = 6V and the INA592 can output close to 15V (lots of margin).  The precision of this design is determined by the DAC output accuracy, the sense resistor, and gain / offset errors of the INA592.  The current shunt circuit is designed to output 4V for a 2mA signal.  It's accuracy is determined by the shunt resistor tolerance, and gain / offset errors of the INA190A1.  Below is a picture of the simulation results as well as the simulation file.  The circuit should work well as a current output with the load ranges specified.  Here is a good document with a lot of details on RTD (A Basic Guide to RTD Measurements ).  In summary, the circuit works as expected.  Let me know if you have additional questoins.

    holland2.TSC

    Best regards,

    Art