ADE7878 calibration

We are currently developing a series of energy monitoring units with either CT or Rogowski Coil (RC)current sensing. It is for 3 phase 4 wire grid systems, based on ADE7878. As the units are delivered without matched CT / RC current sensors, we need to calibrate the units current sensing inputs. The CT inputs could be calibrated with a good sinus source. However RC is more difficult. We have read AN-1076
1. Could we calibrate a RC input, using a good voltage sinus source, with the ADE7878 set to CT mode, or do we need a physical calibrated RC.
2. Are there any RC emulators available, like e.g. thermocouple / PT100 calibrators
3. What EMC hardware protection of the current sense input do you recommend.



  • 0
    •  Analog Employees 
    on Nov 17, 2020 6:59 PM 4 months ago

    Let me make sure I understand your question, you want to calibrate a system without the sensor (CT or Rogowski) present? Or rather the question is about calibrating after the sensor is connected?

    We always recommend calibrating the system with the sensors that will be used as sensors will have variation from sensor to sensor

    1. For a rogowski coil, there is an integrator inside the ADE7878 that must be enabled in order to correctly calibrate the channel.

    2. You can simulate the output of a rogowski with pure sine waves shifted by 90 degrees but that doesn't necessarily mean that it will be calibrated when the sensor is attached. Sensors have errors which are corrected for in the calibration step

    3. We don't have any specific recommendation for protection, diodes can be used to protect the inputs if necessary but just be aware that at high temperatures the diodes can start turning on which causes increased THD on the inputs.

  • Well, It is to be used as a voltage, current , power monitor ( not for billing ), in our RTU system, where the customer connect CT or Rogowski coils of his choice ( and calibration class ), depending of application. The customer is able to configure the monitor with Rogowski coil data ( mv pr KA ). So it is not shipped as a match set of monitor unit / CT or Rogowski coils.
    We want to do a basic calibration of the monitor unit, and voltage / CT inputs we can calibrate with good sources, but Rogowski coils inputs calibration is unclear. How do you suggest we do this ?.

  • 0
    •  Analog Employees 
    on Nov 20, 2020 8:37 PM 4 months ago in reply to Geertp

    In that case, you should be able to perform something similar to what you do with the CT by assuming a Rogowski coil is the sensor and essentially the rogowski inputs a 90 degree phase shifted waveform.

    First off, instead of a turn ration like a CT, the Rogowski will have a mV/A constant associated with it. Then if you already have an accurate source, all you have to do is input an AC waveform that is 90 degrees out of phase with the voltage and then after the internal digital integrator the current waveform will be in phase with the voltage waveform.

    Then you can perform the same procedure as you would with the CT. Just instead of doing it based on an assumed turns ratio and burden value, it would be based on a assumed mV/A constant.

  • Thanks, we will continue with this approach.

Reply Children
No Data