• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 1092 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ADS1220EVM: is it possible to test acquiring +/- 5V input to the analog input with external reference input?

Sridhar M S
Prodigy 70 points
Part Number: ADS1220EVM
Other Parts Discussed in Thread: ADS1220,

In my application, I want to measure +/-5V analog signal. We have designed and developed the board with ADS1220 for analog siganl measurment. We are facing problem when we are measuring the negative -5V analog signal. ADC measures well up to -0.3VDC then measurement drops.

Register was set to external reference and ADC Channel configured to AI0, Single Ended.

RefP0 was connected +2.5V and RefN0 was connected to -2.5VDC.

So we have a doubt in circuit design. So we are planning to buy the EVM. is it possible to test the above said condition in ADS1220EVM? 

  • 0
    TI__Guru** 112895 points

    Hi Sridhar,

    You cannot measure below analog ground unless you are using bipolar (+/- 2.5V) analog supplies.  If you cannot get good results past -0.3V input, then this sounds like AVSS is connected to ground instead of -2.5V.  If you apply an input voltage (analog input or reference input) that is outside of the analog supplies by more than 300mV, you have exceeded the absolute maximum input range of the ADS1220 and risk damage to the device.  Also, to measure all the way to the supply rails, you must disable and bypass the PGA in the configuration settings for Register 0.

    Something else to consider is the +/-5 V input range is measured AINP relative to AINN.  For example when using a unipolar supply, if AINP is at 5V and AINN is at ground, the output code results is output code for +5V.  But if AINN is at +5V and AINP is at ground the output code is at -5V.  If AINN is fixed at analog ground (AGND), you can only measure up to AVDD which will be 1/2 of the FSR using a unipolar supply (+5V AVDD, AVSS at AGND).  If you want to measure the input AINN fixed at AGND using bipolar supplies (+/- 2.5V) then the input range of AINP is limited to +/- 2.5V as you cannot apply an input beyond the supply rails. 

    If your sensor output is +/-5V (above and below AGND) you can use a voltage divider to divide the voltage in half so that the input range is maintained within the supply range.  Also, it may be better to use 2.5V reference instead of 5V to maximize the dynamic range.  Within your code, you would multiply the result by 2 to get the proper result.

    The ADS1220EVM is not designed to have a bipolar supply connected to the EVM, so the EVM will not help you in this case as you cannot apply a negative voltage below AGND to the ADS1220EVM.

    Best regards,

    BobB

  • 0 in reply to Bob Benjamin
    Prodigy 70 points

    As advised by you, I have made the corrections in my circuit. my setup details and read response are given below

    ADC Response

    input signal = 0.2V,0.5V,0,-0.2,-0.5V response of ADC was 2.0V

    Setup

    DVDD - 5V, DGND- GND, As given in the ADS1220 datasheet, page 60. instead of 3.3V. we have given 5V.

    AVDD - 2.5V, AVSS- -ve2.5

    RefP - 2.5V, RefN - -VE 2.5 (with/without reference input tried and the only differences observed in the ADC response was it saturates at 1.0 or 2.0V)

    Ain - +/-2.0V signal, Signal GND connected to DGND.

    The register is set to ch0, single ended, external reference, normal mode, 10 SPS, PGA disabled, continuous mode

    Could you please help me what is wrong with my setup?

  • 0 in reply to Sridhar M S
    TI__Guru** 112895 points
    Hi Sridhar,

    Can you send me your schematic instead of descriptions? Can you send me the exact register settings for the ADS1220? Can you give me your results as raw codes from the ADC instead of computed values for the input voltages you are using? Do you have scope or logic analyzer shots of the communication?

    When you state the ADC saturates at 1 or 2V, does this happen for both positive and negative input voltages? To me it appears that there is some gain being applied. The ADS1220 can have gain up to 4 using PGA disabled.

    Best regards,
    Bob B
  • 0 in reply to Bob Benjamin
    Prodigy 70 points

    Please see attached the circuit schematic and ADC raw values.

    First thing the bipolar measurement was resolved. I do not know exactly how it was corrected. But we observed different issue now which I have explained in the attached document.

    probably after finding the solution this thread could be appropriately categorized.

    ADC Technical Bipolar issue.docx

  • 0 in reply to Sridhar M S
    TI__Guru** 112895 points
    Hi Sridhar,

    After reviewing your schematic, if you attempt to apply a voltage below GND to analog input while the analog supply to the ADS1220 is unipolar (AVDD= +5V and AVSS = GND), you will see your original complaint as you will be exceeding the absolute maximum ratings for the analog input. This issue is corrected by using bipolar supplies (+/- 2.5V) which allows for an input to be applied below GND.

    For the case discussed in the document your provided, the measurement mux selection you are making is single-ended (AINP = AIN0 and AINN = AVSS). This means the measurement is referred to AVSS. With a reference of 5V (REFP = 2.5V and REFN = -2.5V) the full-scale range is 2*Vref or +/-5V. But because the measurement is AIN0 to AVSS, where AIN0 is always greater than AVSS you will only see the positive codes.

    The ADC measurement is always differential even though you have selected the measurement as single-ended. What this means is AINP is always compared to AINN. If AINP is greater than AINN then the code result is positive. If AINN is greater than AINP then result is negative. As AINN can never be greater than AINP you will always have positive code results.

    Your measurement range is from -2.5V to +2.5V (total of 5V). If AIN0 is -2.5V your output code will be 0. If AIN0 is +2.5V your output code will be positive full-scale (0x7FFFFF). If you want to determine if the input voltage is above or below GND, then you would need to subtract the equivalent codes (or voltage) representing 0V from the result.

    If you set AIN0 to GND, you can determine the code and use this code as a factor for all remaining measurements. As an example, let's say you set AIN0 to GND and you see a value returned of 0x3FFFFF (4194303). You would then subtract this code from all other measurements. This will give a result as +/-2.5V for the +/-2.5V input range. The resulting codes will range from -4194303 to 4194304. This is a corrected value but still only uses 1/2 of the full-scale range as you cannot use the full-scale range for a single-ended measurement. Is this a big issue? Most likely not as noise will subtract at least one bit anyway.

    Best regards,
    Bob B
  • 0 in reply to Bob Benjamin
    Prodigy 70 points
    Thank you for your patient reply.
    Yes, it is the matter of losing one bit, and there is no significant change in the final resolution.