We are using an ADE9000 chip to obtain the total active, reactive and apparent power of a 4-wire wye setup (using current transformers, so the digital integrator is disabled). With the recollected data, if we compute the apparent power from the active and reactive one (S = sqrt(P²+Q²), we have noticed that the apparent power provided by the chip is not the same one than the computed one for certain scenarios. The differences are the following ones:
-The apparent power provided by the chip is delayed with respect the computed one.
-A variation of the apparent power (produced by a peak of consumption) is softer in the power provided by the chip than the computed one
The user guide indicates that the active and reactive powers are computed by the chip through the instantaneous current and voltage and the Low-Pass Filter to obtain the DC value, and the apparent power is computed through the RMS of the current and the voltage.
https://www.analog.com/media/en/technical-documentation/data-sheets/ADE9000.pdf (pag 28 and 29)
Also, in the user guide, in page 15, it is indicated in tables 11 and 12 that the integrator of the active and reactive powers can be on and off, but in table 13, the only configuration that can be selected for the apparent power, the integrator is on. Does this mean that the digital integrator is always affecting to the apparent power provided by the ADE9000 chip?
May this integrator the one that is producing the difference between the provided apparent power and the computed one? And if that is the case, can the chip provide an apparent power that is synchronized with the active and reactive powers?
Thank you in advance.
I add an example of the lectures we are obtaining:
This is a reading of a transitory state. As you can see, the apparent power computed by the formula S=sqrt(P²+Q²) differs greatly from the obtained from the ADE9000. Nevertheless, when the peak of consumption disappears, and we reach a stationary state, both apparent powers get a similar value.
Can the ADE9000 provide a correct apparent power in a transitory state that satisfies the theoretical relation with the active and reactive power?
The algorithm that the chip uses for Active Energy is S = Vrms * Irms and the equation S = sqrt (Active^2 + Reactive^2) will give you a different result in non-stationary state becase the calculation methods are different.
1. There is a filtering for Active and Reactive datapath which is different from the RMS calculations filtering that might cause the difference during transition state.
2. Apparent Power is multiplication of signal by itself so any harmonics, correlated noise ends up in the apparent power. Using the Active and Reactive energy, it is multiplying V and I so there is much less effect from the harmonics and noise.
3. The IEEE 1459 defines this algorithm for apparent calculation in metering.
Thank you very much.