• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 80 subscribers
  • Views 1097 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

emeter 6779 - CT phase calibration

Skip Rye1
Prodigy 140 points
Other Parts Discussed in Thread: EVM430-F6779
When we use EVM CT on our hardware, all is good PF = 1 V and I in phase. 
When we switch  to off the shelf Veris CT's and have our calibration source in
phase for V and I we read around 0.7 for PF. I can tune PF to about .82 using
I_HISTORY_STEPS and DEFAULT_BASE_PHASE_C_CORRECTION_LOW high
and low byte numbers. If I change I_HISTORY_STEPS to 7 and BASE_PHASE_C_CORRECTION_LOW to 2032 I get maximum phase shift
with a PF of 0.82. If I change I_HISTORY_STEPS any higher and continue to use
 the higher bits of DEFAULT_BASE_PHASE_C_CORRECTION_LOW PF starts to
get worse. Example ;
I_HISTORY_STEPS 6, DEFAULT_BASE_PHASE_C_CORRECTION_LOW 1276, PF = .76
I_HISTORY_STEPS 7, DEFAULT_BASE_PHASE_C_CORRECTION_LOW 2032, PF = .82
I_HISTORY_STEPS 8, DEFAULT_BASE_PHASE_C_CORRECTION_LOW 2176, PF = ..80
I_HISTORY_STEPS 10, DEFAULT_BASE_PHASE_C_CORRECTION_LOW 2800, PF = Negative -0.8181
I_HISTORY_STEPS 20, DEFAULT_BASE_PHASE_C_CORRECTION_LOW 3584, PF = Negative -0.7263
Any idea why I cannot change phase further using higher I_HISTORY_STEPS values and how to fix this?
See pictures below;
  • 0
    TI__Guru**** 202256 points
    Have you tried using the calibrator GUI from SLAA409?

    www.ti.com/.../slaa409a.pdf
    www.ti.com/.../slaa409a

    And here are some other threads involving EVM430-F6779 calibration:

    e2e.ti.com/.../1894288
    e2e.ti.com/.../419263

    Let me know if these resources do not solve your problem and I will loop in a metering expert to further assist you.

    Regards,
    Ryan
  • 0 in reply to Ryan Brown1
    Prodigy 140 points

    Hi Ryan,

    We are not using load banks to calibrate. Instead we purchased a high accuracy calibration source. We can and do calibrate to the slaa577f (More recent version of slaa409a) document using the EVM's custom CT's on our hardware at PF 1, 0.5 and -0.5. The problem is when we use an off the shelf CT the phase is shifting further that what section 4.3.2 (4.3.3 in slaa409a) can handle. At the end of that section they state;

    "When using CTs that provide a larger phase shift than this maximum, sample delays along with fraction delay must be provided"

    That is where I_HISTORY_STEPS and DEFAULT_BASE_PHASE_C_CORRECTION_LOW come into play. With our cal source at unity (PF = 1) we can only adjust the phasing to get to around PF = 0.82. We need more phase adjust using additional sample delays to get to PF = 1 with off the shelf CT's.

    We need sample delays or maybe I just don't know the proper usage of I_HISTORY_STEPS or some other way of phase shifting.See the data in the original submission. 

    The question : Is background.c code fully functional in this area for large phase shifts using sample delays of I_HISTORY_STEPS? 

  • 0 in reply to Skip Rye1
    TI__Guru**** 202256 points
    Hi Skip,

    Further comments from the metering expert I've spoken with:

    •We recommend that first gain calibration is done with 0° phase shift between voltage and current. After gain calibration is done, phase calibration should be done using 60° phase shift between voltage and current. When performing phase calibration, we recommend using the active power % error value to determine when phase calibration is completed instead of using the power factor reading. The SLAA577 application note should have more details on the recommended procedure.
    •During calibration, I_HISTORY_STEPS normally does not need to be modified. The phase compensation factors that need to be modified are the DEFAULT_BASE_PHASE_x_CORRECTION_LOW factors and in some cases DEFAULT_V_PRESCALE_FACTOR. We recommend restoring I_HISTORY_STEPS back to 2 and only modifying the DEFAULT_BASE_PHASE_x_CORRECTION_LOW value.
    •What is the typical phase shift of the new CTs?
    •To simplify calibrating the EVM, the EVM’s GUI can be used instead of modifying the DEFAULT_BASE_PHASE_x_CORRECTION macro.

    Regards,
    Ryan
  • 0 in reply to Ryan Brown1
    Prodigy 140 points

    1) As mentioned in the original post "V and I are in phase" therefore phase shift is 0 on the calibration source.

    2) OK I changed I_HISTORY_STEPS back to 2. According to SLAA577f section 4.3.2 the limits of DEFAULT_BASE_PHASE_x_CORRECTION_LOW are +/- 2.62 Degrees. 

    3) As mentioned in the original post "off the shelf Veris CT"  caused the PF to go from 1 to 0.7. This is roughly a 45 Degree Phase shift. According to  SLAA577 DEFAULT_BASE_PHASE_x_CORRECTION_LOW is only good for 5 Degrees. 

    4) Yes I know this. We are actually using Modbus and FRAM for calibration but for this thread I downloaded EVM code and the .h file works fine for  now.

    Below is my data adjusting DEFAULT_BASE_PHASE_x_CORRECTION_LOW and DEFAULT_V_PRESCALE_FACTOR.

                                                                                                      | DEFAULT_V_PRESCALE_FACTOR  |                 

    DEFAULT_BASE_PHASE_x_CORRECTION_LOW       |                                                                    |          Power Factor

    --------------------------------------------------------------------------|--------------------------------------------------|-------------------------------------------------------

                                       0                                                             |                         1023                                 |           0.659

                                                                                                       |                             0                                   |            0.659

    --------------------------------------------------------------------------|--------------------------------------------------|-------------------------------------------------------

                                  128                                                              |                         1023                                |            0.693

                                                                                                       |                               0                                 |             0.693

    --------------------------------------------------------------------------|--------------------------------------------------|-------------------------------------------------------

                                  255                                                             |                           1023                               |             0.725

                                                                                                       |                                0                                 |            0.725

    Again Calibration source V and I are in phase and using an off the shelf Veris CT instead of the EVM CT.

    Now what should I do to get more phase adjustment for off the shelf CT's?

                                                           

  • 0 in reply to Skip Rye1
    TI__Guru**** 202256 points
    Skip,

    •We recommend that the phase compensation is done at 60° phase shift between voltage and current. For phase compensation, the best results would be obtained by trying to minimize the active power % error (assuming that gain calibration has already been done) instead of performing compensation using the power factor.
    •The +/- 2.62 degrees limit mentioned in section 4.3.2 is referring to the limits of using the preload register for providing fractional sample delays. The DEFAULT_BASE_PHASE_x_CORRECTION_LOW macros correspond to both the fractional sample delays provided by the preload register as well as whole sample delays. As a result, the limits mentioned in section 4.3.2 do not apply to these macros.
    •Are you using standard e-meter CTs? If so, can you verify in the CT datasheet what is the expected phase shift? A 45 degree phase shift is not common for a standard e-meter CT. From my side, I have typically observed phase shifts from e-meter CTs that are less than 6°.

    Regards,
    Ryan
  • 0 in reply to Ryan Brown1
    Prodigy 140 points

    I found out that by using a different CT there is much less phase shift. In my calibration I get a skew of the
    power factor around 1. See results. How would I compensate for this?

    Calibration
    Source emeter
    0 -1.0
    60 -0.5588
    -60 0.4284

**Attention** This is a public forum