I am working on a university project using the ADE7753 IC (with Rogowski Coil current sensor). I have successfully calibrated de active energy measurement, but I'm having problems reading the IRMS and VRMS registers. The voltage RMS value particularly oscillates a lot, and there seems to be some kind of weird overflow on the current RMS channel.
I am using three incandescent light bulbs as a test load. When I turn on one, and then two, I can see the current RMS value rising, although with a lot of offset (didn't compensate the offset yet), but when I turn on the third light, the value decreases , although the 24-bit value is still far form overflow.
Also, during debugging, I have tried changing the PGA gains for both the channels, and got even stranger results. I have an application on the microcontroller which allows me to read and write registers on the ADE via serial interface with a computer. When I change the PGA configuration for the voltage (CH2) channel (3 MSBs of GAIN registers), the VRMS channel increases (although still with a lot of noise), but the current channel is also affected!
At first I thought it could be some problem with the SPI communication, but the routines where exhaustively tested, even comparing waveforms on the oscilloscope, and all reads are tested against the checksum register. I have found nothing on the datasheet that hinted this kind of interaction between the PGAs, is there something I am missing?