while testing ADE7868 on a custom board we are experiencing linearity issues on the current channels.
We have 2500:1 CTs connected to the current channels, these are the values obtained for various points of operation (ADE uncalibrated).
I1 3632 I2 3738 I3 3475 V1 3014804 V2 3011450 V3 3014501
I1r 0.007709251 I2r 0.007490637 I3r 0.008057554
I1 19807 I2 19786 I3 19506 V1 3002985 V2 2999628 V3 3002679
I1r 0.008178927 I2r 0.008187607 I3r 0.008305137
I1 35405 I2 35204 I3 35195 V1 3014990 V2 3011596 V3 3014684
I1r 0.008219178 I2r 0.008266106 I3r 0.00826822
I1 40660 I2 40387 I3 40288 V1 3011103 V2 3007706 V3 3010741
I1r 0.008214461 I2r 0.008269988 I3r 0.00829031
I1=AIRMS I2=BIRMS I3=CIRMS
As you can see, the response is not linear, and phase current 3 behaves strangely.
We already verified the CTs and their are linear, the behaviour is the same even swapping CTs.
Attenuation networks and burdens are the same on all the 3 lines.
What can we do?
thanks for your answer.
1. These currents are provided through a set of incandescent lamps. We have an accurate CT based meter reading the current flowing in the phase and we also verified with an ammeter in series. To achieve the higher currents (2.91A, 3.34A) we use a capacitor with the lamps.
2. I did not check the mV, just the mA flowing in the secondary of the CT and they scale linearly. But I can try.
3. I read the RMS registers upon the ZX and feed them in an exponential smoothing filter. The values I gave you are taken after the settling time of the filter. I also tried using the IxRMS_LRIP registers (I am using an ADE7868A) and the results are consistent with the data I posted.
4. At the moment I don't have it unfortunately.
5. Of course, but I'd prefer to do that in private.
The signal conditioning paths are identical for all the phases.