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.

ADS8598S: Is it possible to use ADS8598 in energy meter class 0.1

Part Number: ADS8598S
Other Parts Discussed in Thread: ADS8588S, TIDA-00834, REF6225, OPA2188, INA188

Hello,

i'm trying to figure out is there possibility to use ADS8598S in energy meter project with class 0.1.

After looking some of reference projects I've only found ADS8588S reference with 0.3% accuracy (TIDA-00834), but from calculating possible accuracy I should get more from this project. I'm thinking that my calculations are wrong for class 0.1 meter accuracy.

I'm planning to use ADS8598S ADC, OPA2188 as voltage buffers, INA188 as current transducer measure and reference voltage REF6225. Also i'm planning to do one time offset and gain calibration. My ADC device must be SAR ADC with simultaneous sampling because I need zero-latency readings.

Is there any chance to use this devices to get 0.1 class metering? I don't want to use 24-bit DeltaSigma, because Sinc5 gets 5 cycles delay. Or maybe use ADS8589H and do some oversampling or feed ADC inputs with prepared PWM signal and then do oversampling to increase resolution?

  • Hi Lukas,
    Thanks for looking at TI precision ADC. The final design accuracy depends on many factors.

    Firstly, based on your current ratings, you can calculate the requirement to ADC, here just an example for you, if your current rating for your class 0.5 meter is 5(80A), the current range required is from 0.1I to Imax if the power factor is 1, so the dynamic range is 160:1. If 0.2% accuracy is desired which is below required 0.5% accuracy, the required accuracy in bits will be ln(0.001/160)/ln2= 16.28 bits. Regarding ADC, total unadjusted error(TUE) can be calculated as per this link:
    e2e.ti.com/.../adc-accuracy-part-2-total-unadjusted-error-explained.
    Finally, as 0.1 class accuracy is critical, a coherent sampling solution may be required to achieve this level of performance and accuracy, please refer to this app note:      www.ti.com/.../getliterature.tsp

    Thanks&Best regards
    Dale

  • Hello and thank you for your answer.

    I don't understand only one thing - how you do calculate dynamic range from this parameters: 5(80A), 0.1I to Imax - how you convert this to dynamic range of 160:1.

    If I have transducer 80A->5A my input is also 5Amps. 0.1I of 5Amps is 0.5A, so by a IEC i need 0.1% accuracy from 0.5A to 5A input current. This current is in RMS, so Ipk is from 0.707A to 7.07A and this current is alternating, so I need 0.1% accuracy from -7.07A...-0.707A and +0.707A...+7.07A. Or I'm calculating this in wrong way?

    Also "ln(0.001/160)/ln2" is 17.29bits. Your answer is for "ln(0.002/160)/ln(2)".

    Coherent sampling isn't problem for me.

  • Hello Lukas,

    0.1I to Imax for 5A(80A) means that the minimum current is 0.1*5A=0.5A, the maximum current is 80A, so the dynamic range is 80/0.5=160:1. this is just an example calculation and didn't consider RMS or PK value.

    ln(0.001/160)/ln2 should be ln(0.002/160)/ln2 because we supposed that 0.2% accuracy is desired. I apologized to make you confused.

    Thanks.

    Best regards

    Dale