Post Go back to editing

AD8497 Setpoint controller has three states?

Hello as part of my bachelorthesis i am working on a setpoint controller for the temperature. therefore i used the ad8497. But the amplifier is not working properly. Because the sensor has 3 states. One state for off which means 0V and one state for on 5V. But after it turns off and then again it wants to turn on it doesnt go to 5V. Instead it remains in 2.5V. Is this some kind of undocumented feature or do someone know how it can be resolved? I have two potentiometers for setting the setpoint voltage and for setting the hysteresis. so the setpoint voltage is generated by a voltage divider with a poti. I know the setpoint voltage should be generated trough a buffer. Is this the problem causing this weird behauvior? Please i need help.

Parents
  • Hi,

    The weird behavior could be a result of not using a buffer on the reference pin of the AD8497. The datasheet recommends that this pin must be driven by a low output impedance source. This could be done by connecting your resistor voltage divider network to a buffer.

    Essentially the AD8497 is an instrumentation amplifier as you can see on the block diagram. The addition of the resistor divider network directly to the reference pin would cause an imbalance to the internal substractor circuit or impedance mismatch. This would result to a reduced CMRR of the AD8497 and possibly introduce some gain errors. 

    Let me know if this works out for you.

    Thanks,

    Donnie

Reply
  • Hi,

    The weird behavior could be a result of not using a buffer on the reference pin of the AD8497. The datasheet recommends that this pin must be driven by a low output impedance source. This could be done by connecting your resistor voltage divider network to a buffer.

    Essentially the AD8497 is an instrumentation amplifier as you can see on the block diagram. The addition of the resistor divider network directly to the reference pin would cause an imbalance to the internal substractor circuit or impedance mismatch. This would result to a reduced CMRR of the AD8497 and possibly introduce some gain errors. 

    Let me know if this works out for you.

    Thanks,

    Donnie

Children
No Data