Big difference in power displayed between Solar Inverter and emontx

I have a SunnyBoy inverter that has its own data publisher to Sunny Portal. However the realtime power displayed there is a lot lower than what emontx is reading. The difference is higher the higher the Solar power being generated.

Do I need to calibrate the input or how do I figure out who is reporting this correctly. I guess I could let it run for a few days and compare both kwh/d readings with the actual generation meter and see who is more correct, because after all that is the one truth that we all need to be concerned about.

Robert Wall's picture

Re: Big difference in power displayed between Solar Inverter and emontx

How much is "a lot", i.e. what is each reading? Can you list corresponding values over a range of powers? Hopefully a stupid question: are you measuring the same quantity?

You really should calibrate your emonTx if you're making comparisons, because like everything else, the emonTx is subject to component tolerances. Whilst it is unlikely, your V3 (if that's what you have) could be in error by more than 10% if all the errors are worst-case and they all add up the wrong way.

bruce_miranda's picture

Re: Big difference in power displayed between Solar Inverter and emontx

I'll first let this run for a few days and compare the daily values to the meter, then I'll know who is 'more' correct. I have v2.2

Robert Wall's picture

Re: Big difference in power displayed between Solar Inverter and emontx

Ah, the V2 is liable to even worse uncertainty, without calibration the worst-case error could be +30% or -25%. I suggest the quickest route to accurate values is to calibrate your emonTx first against the inverter, then check both against the revenue meter.

The reason for the difference is that the V2 uses the internal voltage reference in the ATMega 328. Whilst this has very good stability, the actual value is subject to an uncertainty of 10%, which is why it is included in the calibration procedure.

bruce_miranda's picture

Re: Big difference in power displayed between Solar Inverter and emontx

I definitely need to get this calibrated. But my differences don't seem like they are a fixed percentage off, they are more like exponentially off, because higher the power more the gap between what the inverter is reporting v/s what the emontx is picking up. 

I've noted my solar meter reading now in the night, will do so each day for a week and then compare the kwh/d reported by the inverter and the emontx. 

The one thing I noticed is that in the night the solar CT on the emontx is still picking up a +2/3W passing through. Is this the error expected or the power drawn by the inverter electronics? I thought I should get a -2/3W if it was the latter. 

Robert Wall's picture

Re: Big difference in power displayed between Solar Inverter and emontx

If you have the standard CT and burden resistor, the max current is 100 A which is presumably 24 kW (as you don't say what your system voltage is). So please don't expect the emonTx to be good below about 1%. It is a 10-bit ADC that's recording the shape of the wave and calculating the rms value, so when the amplitude of the wave is only a few ADC steps, you cannot expect high accuracy.

bruce_miranda's picture

Re: Big difference in power displayed between Solar Inverter and emontx

Yes I have using the standard 100A CT with the std burden. My Vrms is showing between 245-250.

 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.