Hi there,
[Premise: I'm not expert in electricity stuff]
I'm using OEM since 3-4 months for monitoring my house consumption and pv production. All is fine.
Now, that there is a lot of sun, I'm seeing a "strange" behavior. My pv is 6000 Wp, but sometimes OEM records 6500 - 6900 watt of the production. Is it possible? Or there is some trouble in the measurements?
Thanks!
Re: How to check the correctness of the measurements?
What "OEM" do you have?
Everything out of the shop comes with the calibration set to be correct provided that every component in the calibration chain comes at the nominal value. Normal manufacturing tolerances will ensure that this is rarely if ever the case, so some variation is expected. However, statistics say that many of the variations in value will cancel each other out and the overall accuracy is normally within a few percent, but if luck is not on your side (statistics again), the measurement could be wrong by 10%. Your 15% error is not what we expect. What does your inverter tell you? Can you count pulses on your generation meter over a timed period and cross-check that way? (Ask if you don't know how to do that.)
Re: How to check the correctness of the measurements?
Hi Robert,
unfortunately now I'm away form my pv, so I cannot read the Inverter value. Therefore, I cannot read pulses because I have not the hardware. I am using this Firmware
https://github.com/openenergymonitor/emonTxFirmware/blob/master/emonTxV3/RFM/emonTxV3.2/emonTxV3_2_continous_kwhtotals/emonTxV3_2_continous_kwhtotals.ino
Re: How to check the correctness of the measurements?
The only hardware you need to check against your generation meter is something to measure time and your eyes! You stand in front of the meter and count the number of pulses over a period of time, say one or two minutes, depending on how frequent the pulses are. Then you look at the meter, where it should tell you what each pulse represents. For the meter here (a), it is "1000 imp / kWh", or 1 pulse per 1 Wh. That is a common value, but yours might be different. If you were generating 5.5 kW, that meter would flash at the rate of 5500 pulses per hour, or 91.66 per minute. Then you look to see what your emonTx reads.
If you have an emonTx V3.2, you should calibrate it, because the accuracy depends on the accuracy of the ac adapter, the CT and several components inside the emonTx. You need to determine the error against your (certified?) generation meter and adjust the constants in the sketch, as described in the comments starting around line 101.
If (say) your emonTx is reading 12% high (× 1.12), then because the present calibration constant is
powerCal_CT2 = (276.9*(3.3/1023))*(90.9*(3.3/1023)) = 0.2619 ADC-steps2/W
you must multiply that by 1.12 to reduce your reading by the correct proportion.
(It might be easier to write
const float powerCal_CT2 = 0.293;
but change the number to the one you calculate).
Reload the sketch, then check again against you generation meter.