Difference between inverter kWh output # and meter output #?

[HX_Guy]

Is this pretty typical or something that needs to be looked into?

Now that I have an actual meter from the POCO, I can check it against my inverter and again the TED5000 (which has always shows a little less than the inverter). Well, the meter itself actually shows lower than the inverter as well.

SolarEdge Inverter: 53.38kWh
POCO Meter: 51kWh
TED5000: 51.5kWh

I thought maybe the inverter is showing DC kWh and not AC, but they are too close, if that was the case the difference between them would be bigger.

[Sunking]

Very first thing every electric student learns in his first day of lab his numbers are a bit off. That leads to the student asking the instructor what is going on and the instructor replies:

"Stupid we measure our 2 x 4 lumber with a micrometer, mark the spot with chalk, cut it with a broad ax, and make it fit with a sledge hammer"

The only correct reading and the one that counts is the POCO meter.

[gregvet]

Yes, I assume the POCO meter is correct. I do use the TED5000 number to correct PVOutput numbers. The spreadsheet I created using TED5000 number for PV production as well as consumption are entered in this spreadsheet. From TED5000, I calculate net production/usage. I then adjust the TED5000 numbers to match POCO numbers monthly. I then sum these monthly Net and PV production adjustments separately(some months are +, others are -). So far, the End of Year adjustment sums of net kWh used as well as the PV production compared to the POCO net meter and PV meter have been almost negligible. At the end of 2014, POCO PV production meter was about 1% less than TED5000 and POCO Net meter was 2% less than TED5000 readings.

[bcroe]

First, I would hope the inverter reading is higher than the POCO, because it is
pushing out current and the voltage is highest there. I lose 400W in my 600'
loop. The inverter reading should be the AC output; compare it to in/out
voltage X current.

I would not expect a clamp on ammeter to be as precise as a fixed, calibrated
commercial meter. I trust my old mechanical meter the most; hope the latest
electronic ones are as good.

And a TED5000 might not deal with power factor as well, though my tests
indicate its close.

I would not be terribly concerned with the variations listed; possibly some
improvement is possible. Listen to Sunking. Bruce Roe

[Alisobob]

Is this pretty typical or something that needs to be looked into?

Why is it that when I bring something to your attention, it falls on deaf ears??

[JFinch57]

I think it has to do with the accuracies of the meters. The SolarEdge is most likely (guessing, similar to my Enphase) +/-5% accurate and a revenue grade meter must meet the ANSI C12.1-2008 accuracy requirements of +/- 2%. The difference can be as much as 7%!

[HX_Guy]

Why is it that when I bring something to your attention, it falls on deaf ears??

What did you bring to my attention?

[Alisobob]

That theres something funky in your KWH reporting software....

[sensij]

That theres something funky in your KWH reporting software....

What are you taking about? A difference of a few percent doesn't seem like a big deal, and the TED agrees well with the meter. I think that most of the kWh numbers he had reported have been from the TED, and although there is reason to think the inverters output might be slightly high, nothing that has been posted is inconsistent with what the system should be capable of. I don't think the numbers are any more or less meaningful than what you've posted on your own system.

Edit: since no one has really said it... given the history of this system, it wouldn't hurt to check for hot spots in the connections between the inverter and the meter, just in case some power is being lost. A fun use of that camera the op just bought.

[silversaver]

The inverter will shows the highest number since that's where the energy produced. POCO meter is the one it actually count.

I have Locus monitoring + TED 5000 as well, but the report aren't that far off. Locus monitoring is compliance with ANSI 12.20 class revenue-grade power-meter. The actual reports are less than 2% off comparing with inverter's output (by looking at inverter's output LED screen). TED 5000 report are like about 5% off.

OP mention the solaredge inverter output at 53.38kWh, I'm not sure if he is looking at the output from inverter's LED screen or his solaredge monitoring software.

[HX_Guy]

That theres something funky in your KWH reporting software....

As was mentioned by sensij, I put up numbers that the TED5000 was reporting, which seems very accurate to the POCO meter. You seem to like to bust my balls for some reason.

[HX_Guy]

Edit: since no one has really said it... given the history of this system, it wouldn't hurt to check for hot spots in the connections between the inverter and the meter, just in case some power is being lost. A fun use of that camera the op just bought.

I gave the thermal camera a look but honestly not even sure what I'd be looking for. I can't open up the meter box or the disconnect anymore (POCO put their lock on it) so everything look "cool" through the camera, I guess that's normal.

I kind of wanted it to check the house inside for any air leaks to make sure the insulation is good, maybe it'll be more useful for that.

[HX_Guy]

The inverter will shows the highest number since that's where the energy produced. POCO meter is the one it actually count.

I have Locus monitoring + TED 5000 as well, but the report aren't that far off. Locus monitoring is compliance with ANSI 12.20 class revenue-grade power-meter. The actual reports are less than 2% off comparing with inverter's output (by looking at inverter's output LED screen). TED 5000 report are like about 5% off.

OP mention the solaredge inverter output at 53.38kWh, I'm not sure if he is looking at the output from inverter's LED screen or his solaredge monitoring software.

I was looking at the online figures, not the LCD on the inverter itself. I would assume they should be the same, no?

BTW the SolarEdge inverter is reporting an AC voltage of 245.16V while the TED5000, which is inside the main service panel, is showing 121.3V (242.6?). So that's about a 1.1% loss? The run is only about 15 ft, is that normal?

I can take a multimeter to it later and check actual voltages, but that's what I was able to gather quickly.

[bcroe]

I was looking at the online figures, not the LCD on the inverter itself. I would assume they should be the same, no?

BTW the SolarEdge inverter is reporting an AC voltage of 245.16V while the TED5000, which is inside the main service panel, is showing 121.3V (242.6?). So that's about a 1.1% loss? The run is only about 15 ft, is that normal?

I can take a multimeter to it later and check actual voltages, but that's what I was able to
gather quickly.

That is an interesting point; inverter running at 240 VAC but TED plugging into 120 VAC. So
the TED knows to double the voltage for correct power. But any imbalance between the 2
120V halves (caused by imbalanced loading in the house and other things) will shift the
Neutral voltage between the halves, introducing an error.

Guess bridging a 240 variac across the inverter output leads and connecting the TED to a
center point could take the N imbalance out of the picture, and allow a chance to fine tune
accuracy. But unless the variac was connected to the inverter terminals, the adjustment
would only be precise at a particular output level. And this assuming the variac doesn't
introduce a power factor error.

Is the TED in that box the POCO padlocked?
Bruce Roe

[Sunking]

That is an interesting point; inverter running at 240 VAC but TED plugging into 120 VAC.

If that is in fact how TED measures the voltage, then it is invalid. While in theory the halves should be equal in practice they are not.