Post Go back to editing

AD8220

Hello!

In a circuit I use AD8220 for measuring the sensor resistance between 100k ohm to 200k ohm. The problem is that I only get a good linearity from 100k to 110k at the output. Over 120k the output saturates.

How can I solve the circuit so a can get an linear output for 100k to 200k?

PDF

Thread Notes

Parents
  • Hi  

    Thank you for reaching out and sharing your inquiry.

    To be able to recommend the correct recommendation, I would like to know the different design requirements of your application.

    Isolating other components that may affect the AD8220 performance, Please cut the connection from Pin7  to (10uF cap and 10Kohm res) and confirm if this is correct.

    • What is your Sensor (s1) and intended applications?
    • Basing from your application circuit, AD8220 is using a single operation; +Vs =5V and -Vs =0V?
      • Vref (pin6) = 2.5V if variable resistor is 0ohms? If var res is 100K then, Vref = 2.62V?
    • Intended Gain =1??
    • If S1 (Sens) = 100Kohms, what is the value for +In (Pin4), is this 2.5V?
    • If S1 (Sens) = 200Kohms, what is the value for +In (Pin4), is this 3.33?

     Thanks and Best Regards,

  • Hello!

    Thank you for helping mePray

    I have disconnected from Pin 7.

    The sensor is a gas sensor which changes the impedance depending on the concentration of the gas. For no gas it is 100k ohm for maximum gas concentration it is 200k ohm.

    Yes I use single operation 5V. +Vs = 5V and -Vs = 0V

    Vref (pin6) is used for offset trimming. 2.37V for 0 ohm and 2.62V for 100k. The offset shall be trimmed so between Earth ground and Vout (Pin7) shall be less then +/-1mV.

    There is no resistor between Pin2 and Pin3. Gain = 1.

    If S1(Sens) = 100k ohms +In (Pin4) is 2,503V  and output Pin7 is 2,470V

    If S1(Sens) = 200k ohms +In (Pin4) is 3,27V  and output Pin7 is 2.632V

    Why is the output so low for 200k? it should be at least 3,2V

    What have I done wrong?

Reply
  • Hello!

    Thank you for helping mePray

    I have disconnected from Pin 7.

    The sensor is a gas sensor which changes the impedance depending on the concentration of the gas. For no gas it is 100k ohm for maximum gas concentration it is 200k ohm.

    Yes I use single operation 5V. +Vs = 5V and -Vs = 0V

    Vref (pin6) is used for offset trimming. 2.37V for 0 ohm and 2.62V for 100k. The offset shall be trimmed so between Earth ground and Vout (Pin7) shall be less then +/-1mV.

    There is no resistor between Pin2 and Pin3. Gain = 1.

    If S1(Sens) = 100k ohms +In (Pin4) is 2,503V  and output Pin7 is 2,470V

    If S1(Sens) = 200k ohms +In (Pin4) is 3,27V  and output Pin7 is 2.632V

    Why is the output so low for 200k? it should be at least 3,2V

    What have I done wrong?

Children
  • 0
    •  Analog Employees 
    •  Super User 
    in reply to SMENA

    Hi  

    Thank you for providing these design requirements for your intended gas-sensing applications.

    I have additional questions on my end to confirm several items; please see below:

    • You mentioned that the applications will be used to measure the gas level/ concentration. Will it be for all gases, or is it only for specific types of gas?
    • For the sensor, is there a chance you can check what the typical tolerance (resistance equivalent) is? +/-5, +/-10%, etc..
    • Can you please double-check the values on your -In (Pin1) for both conditions? If S1 =100K and S1=200K.

     For the inquiries on your applications,  

    • Vref (pin6) is used for offset trimming. 2.37V for 0 ohm and 2.62V for 100k. The offset shall be trimmed so between Earth ground and Vout (Pin7) shall be less then +/-1mV.

                 ArSol: The Vref should be half of 5V when the 100K is set to 0. Voltage divider = 5V * (1M/2M). There might be line parasitics that affect your Vref value and another reason is the                           resistor tolerances of the 1Mohm that you used.

    • If S1(Sens) = 100k ohms +In (Pin4) is 2,503V  and output Pin7 is 2.470V and If S1(Sens) = 200k ohms +In (Pin4) is 3,27V  and output Pin7 is 2.632V

                ArSol: If S1 =100K. I’m seeing 30mV offset on the output. Considering both inputs should be equal (2.5V), and your Vref set is 2.5V. Again, this might be the effect of the tolerances used in 1Mohm.

                 ArSol: If S1 =200K, the amplifier starts to saturate, at ambient temp, the maximum allowable Input voltage range on both pins (1, 4) is +Vs -2. In this case, it would be 3V only.

                            

      • My recommendations
        • To satisfy the input voltage range  (IVR) of  AD8220, If you can modify the divider ratio on the -In (pin1) side; suggest using 1/3 ratio ( 1K||2K)

                               

      • Better to use a precision resistor (find for 1% value) for 1M and resistors on the right-left side of the bridge.
      • If any chances there are precision varistor for your 100K, it will helps

      I hope these suggestions will suit your requirements.

      Regards

    • Hi!

      Thank you so much for trying to help.
      All resistors I use are +/-1%
      The Multimeter I use is HP 3478A
      The Power supply is TTi CPX400D
      The two resistors to the left, each of 1k ohms, are connected to earth ground in order to minimise the noise level.
      REF Pin6 on AD8220 is used for trimming offset with 100k ohms trim potentiometer so that the output Pin7 has the same Voltage value as earth ground. The operation amp is OPA2277.
      I now measured carefully the following:
      If S1(Sens) = 100k ohms +In (Pin4) is 2.497V and output Pin7 is 2.497V (perfectly trimmed).
      If now S1(Sens) = 200k ohms +In (Pin4) is 3.307V and output Pin7 is only 2.684V. 
      The output Pin7 saturates when S1(Sens) reaches around 120k ohms. Why can't the output get higher to about 3,3V?
      I tried to use your suggestion to change the upper left resistor to 2k ohms. Ground earth was then 1.645V.
      I trimmed the REF Pin6 with the second output from the power supply with the voltage at REF 0.79V
      If S1(Sens) = 100k ohms +In (Pin4) is 2.504V and -IN (Pin1) 1.645V output Pin7 is 1.645V (again perfectly trimmed).
      If now S1(Sens) = 200k ohms +In (Pin4) is 3.270V and -IN (Pin) 1.645V and output Pin7 is only 1.792V. 
      Again the output saturates at around 120k ohms.
      How can that be?
      I'm very confused about this.
    • +1
      •  Analog Employees 
      •  Super User 
      in reply to SMENA

      Hi  

       Thank you for sharing the instruments that you are using.

       I forgot to mention that we should keep the node voltage on the right side of the bridge (+IN; pin4) < (+Vs -2V) to prevent the amplifier from going into saturation. Hence, I propose two (2) conditions: varying the value of R1 and R3. I reattached your design applications schematics here and added several designators to ensure everything is clear. Please see below:

      For clarity, I made a table correlating the expected vs the measured values. Please retake the measurements using the conditions below:

      Condition 1

      EXPECTED

      MEASURED

      R1 = 1K, R2 =1K , R3 = 200K, VR1 =0 Ohms

       +IN

       -IN

      REF

      OUT

       +IN

       -IN

      REF

      OUT

      S1(min) = 100K

      1.666667

      2.5

      2.5

      1.666667

       

       

       

       

      S2(max) = 200K

      2.5

      2.5

      2.5

      2.5

       

       

       

       

      Condition 2

      EXPECTED

      MEASURED

      R1 = 2K, R2 =1K , R3 = 200K, VR1 =0 Ohms

       +IN

       -IN

      REF

      OUT

       +IN

       -IN

      REF

      OUT

      S1(min) = 100K

      1.666667

      1.666667

      2.5

      2.5

       

       

       

       

      S2(max) = 200K

      2.5

      1.666667

      2.5

      3.333333

       

       

       

       

      I hope this will help you

      Regards

    • Hi!

      This was of great helpPray

      Thank you so much. The output dose not saturate now for both conditionsHugging

      Condition 1

      EXPECTED

      MEASURED

      R1 = 1K, R2 =1K , R3 = 200K, VR1 =0 Ohms

       +IN

       -IN

      REF

      OUT

       +IN

       -IN

      REF

      OUT

      S1(min) = 100K

      1.666667

      2.5

      2.5

      1.666667

       1.669

       2.471

       2.471

       1.669

      S2(max) = 200K

      2.5

      2.5

      2.5

      2.5

       2.455

       2.471

       2.471

       2.455

      Condition 2

      EXPECTED

      MEASURED

      R1 = 2K, R2 =1K , R3 = 200K, VR1 =0 Ohms

       +IN

       -IN

      REF

      OUT

       +IN

       -IN

      REF

      OUT

      S1(min) = 100K

      1.666667

      1.666667

      2.5

      2.5

      1.672 

       1.649

       2.478

       2.500

      S2(max) = 200K

      2.5

      1.666667

      2.5

      3.333333

       2.458

       1.649

       2.478

       3.287

      The calibration works well also.