MSP430I2041 Calibration Issue

Varaprasad,

Have you looked at the TLV structure  that contains the peripheral registers for calibration on the i2040?

What are your values for these? Have you adjusted these for your specific application? More information on these registers can be found in the family user guide slau335.pdf: 

Hi Evan,

     We don't have much idea about the TLV calibration values . We will try to understand them.

     In the mean time, we would like to integrate the energy library and test the values. We believe that the variance in the values could be due to the noise .Applying digital filtering may make the values more stable, which are already available in the Energy Library.

 In that aspect , is there a instruction guide for integrating the energy library with this micro controller using CCS6. Any direction in this regard would be of great help to us.

   

Varaprasad,

I will have to loop in an expert in our metering group to help you. However, due to the holidays, a response might be more delayed than usual. Thanks for your patience.

Hi Evan,

    That will help. Thank You.

Hi Evan,

It would be very helpful if you could refer us to an expert.
Thank you.

Varaprasad,

You should hear very soon from our experts. My apologies for the long delay. Thank you for your patience.

Hi Evan,

We know that you are working on our problem. But, we would be really grateful if you can refer someone as soon as possible.

Thank you.

Varaprasad ,

With your description I do not exactly know how do you do your determine the 10 AC cycles. With the observation you mentioned I suspect you are determine the AC cycle by number of samples. If you are doing so that would probably have an alignment problem. As the sampling clock is not exactly the same multiple of the AC frequency you are measuring thus the AC zero crossing would fall into a fraction of the last sample and thus causing a periodic error pattern. To be more exact you can detect a sign change in the sample and do an interpolation to determine how much fraction of the last sample should be included in the calculation. Please refer to the application note in the link below and see if that could give you some idea to solve your problem.

mars

Varaprasad,

In case you can't open that file here is the link: www.ti.com/.../slaa638.pdf

Hi Mars and Evan,

First of all, thank you very much for the inputs. Now we are getting more stable values.

We are not very happy with this approach since it seems not an efficient way. we are sampling at 150 and taking only 80 samples in that discarding the remaining. Is there a better way to detect zero crossing.

And also is there a app note on energy metering library integration for CCS?

Regards,
Varaprasad.

Varaprasad,

I don't quite get what do you mean your problem right now can you explain more ?

For converting to CCS of the library code. There is no application note on that but I did successfully make it work. I am on business trip now. I will jot you some note of how to do that when I am back.

mars

Hi Mars,

That helps, we will wait for your note.

On the issue we mentioned, we implemented the Zero Crossing Detection as suggested by you in Software and it solved the problem of getting the Voltage and Current values independently. We implemented it by sampling about two cycles and doing software detecting to get the samples for one cycle. The approach worked well but we are not very convinced with the efficiency of our implementation. This approach also does not help to calculate Power , Power factor .....and hence the request for guidance on integrating the Energy library.

Varaprasad ,

I think you are not using the library provided and you are doing everything from the lowest level by yourself. Our metrology library is already having the zero crossing alignment that it will automatically wake the CPU from sleep mode and set a flag to indicate that and then in the main loop when this is signaled call to a calculation routine to perform the necessary calculation. In short you do not have to align the zero crossing yourself nor write your own meterology calculation. Our reference design has the example code of how to use the metrology library please check for TIDM-SERVER-PWR-MON in ti's web site and the application slaa638.pdf tells you how the software is structured into detail.

mars

Varaprasad,

Has this solved your inquiry? Any updates?

Hi,

We are working on the problem. We should be able to reply you about the results in this week.

Regards,
Varaprasad.

Varaprasad,

Thanks, looking forward to hearing from you.

Hi Evan,

The energy library is a binary file and is available as .r43 file. All the reference integration use IAR. We could not find any note for using CCS. It will be helpful if you can share the notes for CCS.

Regards,
Varaprasad.

I am sorry I still not have the note done due to other priority. As the process is not that simple and I have to record every step from the original code it will take a while

mars