Energy Metering - Mains Frequency Issue

Hello Abhay,

Have you seen the following resources ? The application note and design guide have some valuable information on how to setup and calibrate the i2040 for this application type.

www.ti.com/.../TIDM-TWOPHASEMETER-I2040
www.ti.com/.../evm430-i2040s

Regards,
JH

Abhay,

There can be many cause of this, there is not enough information for me to analyse the cause. Which code and hardware are you using ? Did you modify something in the reference code ? Please provide me more information to proceed.

mars

Hi Mars,

I'm using MSP430i2041 based hardware. IT is similar to the that of EVM based on  the same processor. Also I'm using energy metering code provided on TI webpage of EVM.

I will provide some background - Initially we were not getting proper readings for Irms (values were all over the place), but I was getting correct mains frequency. We corrected our hardware and then we started getting correct values of Irms but at the expense of mains frequency (which is zero always).

If I try the same code with EVM, it gives proper reading for mains frequency. So is there some problem in our hardware which is causing this behavior with mains frequency. But then I should not get the Irms or Vrms  as mains frequency is calculated from these signals.

Thanks,

Abhay

You got Irms correct, Frequency = 0, what else ? If you see frequency = 0 it might mean it detect itself in a DC measuring condition, this might due to a DC offset for voltage which is too large and lift the AC voltage reading to a point that the firmware is not able to see a zero crossing. I suggest you to check the voltage DC offset value in the calibration data (check out the application note that I refer to you in the other question you asked in the e2e how to read that data) and let me know the value.

No. Please use the debugger and check the variable: nv-->seg_a->s->cal_data->phases->[0]->initial_v_dc_estimate->[0]

Hi Mars,
I checked the value of nv_parms-->seg_a->s->cal_data->phases->[0]->initial_v_dc_estimate->[0] and it is zero. The values specified in metrology-calibration-template.h file are getting populated inside NVM.

Thanks,
Abhay

Ahbay,

That sound strange. Can you share your schematic you are using ? Thank you.

mars