Post Go back to editing

Fast RMS over one line Cycle

Thread Summary

The user is experiencing inconsistent RMS current measurements using the ADE9000's fast RMS (RMSOne) register, which should update every half cycle. The support engineer suggests using the Vpeak method and synchronizing with the zero-cross (ZX) status bit or IRQ for more accurate results. The RMSONERDY bit indicates when new RMSOne values are ready, but these values can vary due to the random start point of the capture.
AI Generated Content
Category: Datasheet/Specs
Product Number: ADE9000

Hi!

I'm dealing with a load controlled by a PWM controller. Sometimes the output of the controller can go to very small values, conducting only one linecycle. I tried to measure the RMS values using two strategies, one with fast RMS and the other one with Vpeak. Vpeak works perfectly but it requires me to calculate the values on Firmare to get the RMS value

Just found it weird that fast RMS register (AIRMSONE - 0x219) gives me about 4ARMS. The wave has a peak of about 7,59Amps so I was hoping to get  5,3ARMS or similar.

Sometimes  I can get the right value if I time the measurements using the IRQ0 interrupt by masking the RMSONERDY bit . Still, the ADE9000 gives me wrong values as well like 3,0 - 2.04 -4.05 etc. It never outputs clean 5.36ARMS

Can I be doing something wrong?

Currently I set RMS_SRC_SEL to one on the CONFIG0 reg(0x060) and thats just it I just write 1 to RUN register and start reading.

If somehow I can get the right measurment will other parameters follow? Like THDI, PF, THDV etc. Checking the datasheet I got confused, not sure if those registers can be updated so fast as well.

Parents
  • RMSOne is updated every half cycle and the rms of the last 2 half cycles. 

    The status 0 bit  RMSONERDY  is 1 when values are ready. 

    NOTE: The RMSONE is over a number of samples needed to capture one cycle. This does not mean from zero cross to zero cross. The start point of the capture is random.

    In this example I am showing the rmsone ready at the peak. this is an example it can start anywhere in the waveform. You can see the rms will change from reading to reading.

    I think your vpeak method is the best and you can sync to zx status bit or IRQ.

    Dave

  • Thank you, Dave.

    Will proceed with the derived measurements. Its the only option with those fast cycles.

Reply Children
No Data

Before You Switch


Switching languages will make ADI Explorer unavailable. Resume your session by switching back to English and reopening ADI Explorer.