CCS/TIDM-THDREADING: TIDM-THDREADING harmonic measurement not working with MSP430F67791A

Jose,

I am looking into this and will provide some feedback.

Eddie,

Tanks for looking at this issue. I have a meeting tomorrow at 10:30am CT to go over the test results. Would you be able to attend? If so, could you send me your email address so I can send you the invite through Microsoft Teams?

Regards,

-Jose

Eddie,

Per system design theory on the THD manual, there are 3 diff methods for THD calculation.

Then later on the manual, it says there are eight different conditions for THD testing. (see Test setup step 6 of page 29).

Aside from 6.1 (no harmonics), which one is the most suitable to test harmonics for our needed range of 60%-600% THVD)??

Hi Jose, 

Unfortunately, I will need to get some feedback from our energy measurement experts, but will not be able to get this feedback until next week.  I will get this feedback to you by Mon. or Tues.  Sorry for the inconvenience.

Hi Eddie,

Thanks for the replying back. I also noticed on section 3.1.1.1 "....To properly sense the RMS voltage when there is a lot of harmonic content, the front-end design should be designed with an even greater safety margin."

I believe this might be another reason the meter can not read high THD content in our unit under test, both THD and RMS Voltage drop to zero after load is applied.

Hi Eddie,

Any updates on this urgent request? I've been waiting for feedback.

Hi Jose,

Yes, I apologize for the delay.  First, I would like to confirm.  Are you using i20xx code on the F67xx device?  

 One thing that was observed is that youj are applying large amounts of harmonics over 100%.  Perhaps the code isn’t able to handle such large amounts of harmonics. 

Initially, I thought it was the AFE circuitry filtering out the applied harmonics, but the LPFs on the AFE have a cut-off frequency of around 3.3MHz. Most customers are measuring harmonics up to the 51^st harmonic which equals 60 Hz x 51^st = 3060 Hz which is much less than 3.3MHz, so that’s not a concern. It will be good to understand the highest harmonic that you want to detect and make sure the sampling rate is at least Nyquist. If the sampling rate is 4kHz, then the max harmonic that can be accurately measured is the 40^th harmonic for 50 Hz fundamental or 33^rd for 60 Hz fundamental.

It would also be good to make sure you're not exceeding the FSR. That’s what’s discussed in Section 3.1.1.1. I’m not sure if that would make the results zero or not, but the ADC performance really falls off if FSR is exceeded.

For the harmonic content, also make sure that the voltage provided to the input is still within the recommended analog input voltage range of the ADC, which is somewhere around ±900 mV.

Hi Eddie,

Yes, I'm  using i20xx code on the F67xx device.

Yes, I agree with you, the code isn't able to handle high harmonic content.

I have attached harmonic spectrum up to 5kHz, the main issue is the 70-80% increase on the 5th and 7th harmonic when load is applied, I believe that is what contributes to the overall high THVD. I believe measuring up to the 51st harmonic is ideal, as you can see on the graphs the most content of these high harmonics are below 2kHz.

Regarding to exceeding the FSR, the max voltage we are putting (attached target) is 112V which is below the 120/240V allowed for the board to handle. So I believe we are good in this aspect.

Is there any possibility the code can be change to handle the high harmonic content we need for our application?Board Measurments 2.2.docx

Eddie,

Any updates on this urgent request?

Thanks,

Jose,

I am working to get some feedback.  I hope to be able to provide some later today.  

Is there a reason you are using the i20xx code instead of F67xx?

Jose,

Please see the feedback below.

• In one of the screenshots, the GUI communication button was red instead of green.  I wonder if this was caused by trying to debug or if the power supply stop working for some reason.
• Can you provide the frequency readings associated with each phase for both the no harmonics and harmonic conditions?  I saw a value of 59.92 Hz on phase A in one of the screenshots, but I would not expect that inaccurate of a reading from the MSP430F67xxA.  I would expect that type of reading from a MSP430i20xx though.
• It would be a good idea to calibrate the meter, in case you haven’t done so already.
• Which THD calculation option are you using?  For the THD design, there were options for IEEE, IEC_R, and IEC_F, as is mentioned in the TIDM-THDREADING design guide.

Eddie,

the GUI communication button was red instead of green = this was caused by trying to debug, remember the fundamental voltage is still reading. I have other screenshots showing this now green.

Readings were taking only at phase A (only phase A was used for our testing). The 59.92Hz is the reading from the fundamental. You can see frequencies from the spectrum, the 300Hz and 420Hz (5th and 7th harm) where the more predominant from no load to full load.

Meter was calibrated 4 times.

Also all 3 diff methods for THD calculation were used (previously mentioned), the issue is still there, no readings regardless of the THD method.