0

I have a quick question regarding the ACS71020 chip. I am using SPI to communicate with this chip, and the iRMS value is represented as a "15-bit fixed point number with 14 fractional bits". My guess is that this means that the MSB equals 0 or 1, and the rest of the bits all represent a fraction of "1". So if you have 1 as your MSB and 630 (in decimal) for your 14 bits, your value is 1.0 + (630/2^14). Unfortunately, the datasheet explains that we need to multiply this floating point number by the max current that the chip can detect which is 30A. That makes no sense because why would you allow for values up to 60A to be reported since this 16-bit floating point with 15 fractional bits can equal 1.9999 at the highest end? Am I not understanding something correctly?

Thanks for the help

This can be found on page 27 of the datasheet.

Link to datasheet: https://media.digikey.com/pdf/Data%20Sheets/Allegro%20PDFs/ACS71020.pdf

John
  • 1
  • 2
  • It's polite to include a hyperlink to the datasheet for the device in question. There's a button on the editor toolbar. Welcome to EE.SE. – Transistor Jul 26 '20 at 21:33
  • https://media.digikey.com/pdf/Data%20Sheets/Allegro%20PDFs/ACS71020.pdf page 27. Sorry about that – John Jul 26 '20 at 21:33
  • Hit the [edit] link below your question to fix it. Don't leave information buried in the comments. – Transistor Jul 26 '20 at 21:36

2 Answers2

0

You always have unsigned 15-bit fixed point number with 14 fractional bits, not 16 with 15 fractional bits.

Why would you report 60A with a device that measures 30A: Because the power meter is subject to inrush currents, motor starting currents, ... that last for short time, but you want to record them even if they are outside the precision tolerance or clamped. So this device is capable of measuring 60A, but not as precise as the nominal measuring range and not for continuous duty.

Marko Buršič
  • 23,562
  • 2
  • 20
  • 33
0

If the chip can measure up to 30A RMS current, it means that the peak current for sinusoidal current is 1.414 times 30A. And the overcurrent detect threshold can be set up to 1.75 times the 30A RMS current.

Therefore it might be that the ADC instantaneous measuring range must be larger than 30A, and it must be at least 1.414 times 30A to measure peaks of sinusoidal current, and it must be at least 1.75 times 30A to detect that threshold.

The only logical explanation is that the ADC can measure 60A full scale peaks to get 2.0 as result, so that 30A RMS result is 1.0.

Justme
  • 127,425
  • 3
  • 97
  • 261