Post Go back to editing

AD8553 instrumental amplifier always outputting 0V?

Category: Hardware
Product Number: AD8553
Software Version: n/a

I am trying to amplify the difference between VQ5_N and VQ5_P. 
VQ5_N measures 2.125V
VQ5_P measures 2.435
I have set the Gain resistors R1 & R2 to produce a Gain = 10, so I would expect to see VQ5_FILT (the output) to be (2.435 - 2.125) * 10 or roughly 3.1V. However, Pin 4 (Vo) always measures 0.009V. This is true even if I remove the Anti-Aliasing Filter (R138 & C10). 
I have confirmed that Vref = Ground, ENABLE  4.25, and VCC = 4.25V

Do you have advice on my my output is always stuck at ~0V? 

Top Replies

  • Hi  

    Thank you for reaching out. We are validating your inquiries; we will need additional information from your application before anything else. Can you provide a scope shot on Pin2(VINP), Pin9(VINN), Pin7(VREF), and Pin4(VO). Then, what values are you using for R54 and your intended BW? and if this circuit was designed to drive an ADC? To check if the VOUT will stabilize at 3.1V, please try to isolate any circuit (ADC, etc..) connection to the VO, and remove R38 and C10. Please let us know if this help.

  • Hi,

    Thank you for your response. I will provide scope shot in the morning but I can answer the other questions. 
    R54  = 0 Ohms. I was hoping to set my Vref = 0 since I never expect my VINN - VINP to be negative. However, is this an issue because I'm using a single supply?

    This circuit is designed to drive an ADC, however the ADC does have a high impedance input. And, I did try isolating the circuit by removing R38 + C10 and still saw the same result. 


  • Hi  

     Thank you for your abrupt response. Your setup is fine on a single supply configuration; we just need tweaks to fine tune. I just missed your initial statement that you confirmed “Vref = GND.” To further isolate the setup, may I ask you to do the following: please set R54 =100K (similar to what you used in R53). For  VCC <5V and single supply configuration, the recommended  Vref value is VCC/2. After these changes and if the output is still stuck at 0V. You may try disconnecting the R48 = 10Kohm as this may hinder the intended Gain = 10. Gain accuracy depends on matching R1 and R2, and any mismatch of these resistor values results in a gain error.


    Please let me know if this recommendation helps.

  • Thank you for your feedback! 

    I added R53 = R54 = 100k to set the Vref to Vcc/2 and that seemed to solve the issue! Vout was Vinp - Vinn * Gain of 10.

    However, because I want to maximize the resolution & I never expect my Vinp - Vinn to be negative, I would like to set my Vref as close to 0V as possible. I saw in the datasheet that for a Vcc = 5V, the Vref range is 0.8 minimum. I am using a Vcc = 4.25V. Do you have advice on the minimum Vref I could use with a Vcc of 4.25V? 

  • Hi  

    Thank you for the update. It’s great to hear that your setup is now working. For a single supply configuration and VCC =4.25V, the suggested Vref =VCC/2. This would equate to 2.125V. The minimum Vref you can use would be -10% from the computed value, the Vref on this setup can go low as 1.9125V [2.125- (0.1*2.125)]. You can try 1.9V; still, I recommend using a voltage divider resistor network (100K||100K) from the VCC (Pin3).



  • Hi Argy,
    Thanks for the reply! Two follow-up questions: 
    1) If the minimum Vref is -10% from Vcc/2, how come in the datasheet page 3, for a Vcc = 5V, the Vref range is listed as 0.8 - 4V? 0.8V would be far more than -10% from 5/2 = 2.5V. 

    2) If my VinP - Vinn is always positive, can I get away with setting my Vref at 1V? 
    Best, Grace

  • Hi Grace,

     Thank you for your feedback.

    Regarding your follow-up questions; To guarantee specs (Table 1) on the Datasheet requires the condition of VCC=5V. For a Single supply configuration, I suggest modifying the value from 4.25V to 5V.The Vref range was taken for a Single operation of 1.8V to 5.5V.

    Basing values on your existing setup, you can use Vref of 0.8 ~1.1V  to adjust your offset and prevent going negative. But, I would recommend correcting your supply and using a Vcc=5V.


    Let me know if this helps

  • Hi Arsol,
    Thanks for getting back to me. I had a quick follow-up question: If I have a Vin (P - N) value of 0.15, Vcc = 4.25V, Vref = 1V, and the gain is set to 20, I would expect my output to be: Vout = Vref + 20 (0.15) or Vout = 1 + 20(0.15) = 4V. However, the measure of Vout is 4.25V (the Vcc value) which makes it seem like the Op Amp is getting railed. Why is it getting railed instead of outputting 4V? Thanks! 

  • Hi  

    Thank you for reaching out. Based on your questions, the values are different from before. Can you confirm if this is a different circuit design from what you used in your initial inquiries?

    As I suggested, kindly modify your Supply (VCC) for Single Supply configuration to 5V.

    To understand your inquiries and be able to answer them correctly, please share your design requirements and items needed below:   

    1. Schematic diagram of your Circuit design.
    2. Intended inputs and value for VINP and VINN, Input type: pure DC or Sinusoidal, where are these pins (VINN, VINP) connected to?
    3. Scope shots of Pin2(VINP), Pin9(VINN), Pin4(VO), Pin7 (Vref) and Pin3 (VCC).
    4. Vref Value:Is this +1V?
    5. Target Gain: 20?; what external resistors were placed between RGA to RGB and VFB to VO pins?