On grid zero export system???

[HX_Guy]

People that don't have a generation and a consumption meter...are you able to see what your house's monthly electric usage is?

I know the that consumption meter only shows net usage from the grid, and the over production of the solar is also a net number to the POCO...anything the house uses directly they don't show on the bill.

With two meters, you can actually calculate the actual usage of the house, which I really like. For example my generation meter shows 535kWh of of 12am last night. The consumption meter shows 175kWh from the grid and also 393kWh fed back to the grid. So taking the generation of 535kWh, subtracting the 393kWh that was fed back to the grid, I get 142kWh. Add the 175kWh that the house consumed from the grid and I get a total of 317kWh, which is what the house actually used this month to date.

If my bill was to be cut today, it would show "Delivered from APS: 175kWh - Delivered to APS: 393kWh - Net Usage - 218kWH" but none of that would actually tell me what the house used.

[sensij]

Devices exist that can download live net consumption data from smart meters; you may see the Rainforest Automation Eagle discussed in other threads. It is compatible with SDG&E, some SCE, and some PG&E meters. With it, net usage can be recorded, along with total generation from the inverter... PVOuput.org makes it really easy. The difference between inverter generation and net consumption from the meter reveals household consumption. Alternatively, a device like the TED5000, the Locus LGate, or some other 3rd party system can be used. The +/- 5% calibration of most inverters adds some uncertainty to it, but for energy auditing purposes, it is probably good enough.

[HX_Guy]

True, with the TED5000 you can actually see the the numbers in real time (house usage, solar generation, net usage) and I compared the TED numbers to the meter and they are spot on, like to the digit.

Was just wondering how people without something like TED go about it, if there is a way for them to figure out how much electricity they are using if they only have one meter.

[bcroe]

People that don't have a generation and a consumption meter...are you able to see what your house's monthly electric usage is?

I know the that consumption meter only shows net usage from the grid, and the over production of the solar is also a net number to the POCO...anything the house uses directly they don't show on the bill.

With two meters, you can actually calculate the actual usage of the house, which I really like. For example my generation meter shows 535kWh of of 12am last night. The consumption meter shows 175kWh from the grid and also 393kWh fed back to the grid. So taking the generation of 535kWh, subtracting the 393kWh that was fed back to the grid, I get 142kWh. Add the 175kWh that the house consumed from the grid and I get a total of 317kWh, which is what the house actually used this month to date.

If my bill was to be cut today, it would show "Delivered from APS: 175kWh - Delivered to APS: 393kWh - Net Usage - 218kWH" but none of that would actually tell me what the house used.

All I need to do is check the cumulative KWH output displayed by the inverters. If I log that
monthly it can compared to the electric bill to see how much I consumed. If the billing date
doesn't coincide with my log, a little interpolation will work, or just do KWH per day which
compensates for different length months. Actually I log it much more often. Bruce Roe

[HX_Guy]

All I need to do is check the cumulative KWH output displayed by the inverters. If I log that
monthly it can compared to the electric bill to see how much I consumed. If the billing date
doesn't coincide with my log, a little interpolation will work, or just do KWH per day which
compensates for different length months. Actually I log it much more often. Bruce Roe

True, though in my experience the inverter kWh output is a bit off, at least for SolarEdge, which shows slightly higher than actual.
I called tech support about it and they said it has a +/-5% accuracy. Mine seems to be almost exactly 3.5% on the plus side compared to actual.

[sensij]

True, though in my experience the inverter kWh output is a bit off, at least for SolarEdge, which shows slightly higher than actual.
I called tech support about it and they said it has a +/-5% accuracy. Mine seem stop be almost exactly 3.5% on the plus side compared to actual.

I looked at the Solaredge public page of three forum members (thejq, sunpowered, bikerscum), and found that in every case, the difference on a clear day between a quarter of the sum of individual 15 min power readings (from the weekly view), and the kWh total reported at the end of the day (from the monthly view) was 3.0%. I would guess that if you sum up your 15 min power readings for a clear day and divide by 4, you will get an energy value closer to what the TED5000 and the meter show. If I had my tin foil hat on, I'd say SolarEdge is intentionally calibrating their daily energy to be +3%, to make their systems look more effective than the neighbors, while still staying safely in the 5% error range they publish. The 15 min data (which is sent to PVOutput.org, if that is set up), looks more accurate.

[HX_Guy]

I looked at the Solaredge public page of three forum members (thejq, sunpowered, bikerscum), and found that in every case, the clear day difference between the 25% of the sum of individual 15 min power readings (in kW, from the weekly view), and the kWh total reported at the end of the day (from the monthly view) was 3.0%. I would guess that if you sum up your 15 min power readings for a clear day and divide by 4, you will get an energy value closer to what the TED5000 and the meter show. If I had my tin foil hat on, I'd say SolarEdge is intentionally calibrating their daily energy to be +3%, to make their systems look more effective than the neighbors, while still staying safely in the 5% error range they publish. The 15 min data (which is sent to PVOutput.org, if that is set up), looks more accurate.

Interesting theory on that sensij. So are you saying that the reading are accurate in real time, but the end of the day number is actually bumped up by 3%?

That actually makes sense and it's something I noticed myself. During the day, if I look at the real time production number via TED, it's practically a dead on match with the real time production of the inverter...but at the end of the day they are quite different, and I couldn't figure why the difference. I thought maybe in real time the difference would be too mall but cumulative over the day it would add up.

[sensij]

I don't really know what I'm saying, I try not to wear that hat too often. I'm only trying to come up with an explanation for what we can observe. I called SolarEdge tech support today to ask about it, and pretty much got a non-answer. They say different sensors are involved in the power readings (which post every 15 min) and the energy readings (which update hourly and create the daily total), so there might be a calibration difference between them. To my eye, that difference does not look random. I haven't watched an inverter (or the portal) in real-time to see how the error develops during the day, but will, once I have my own.

[bcroe]

I looked at the Solaredge public page of three forum members (thejq, sunpowered, bikerscum), and found that in every case, the difference on a clear day between a quarter of the sum of individual 15 min power readings (from the weekly view), and the kWh total reported at the end of the day (from the monthly view) was 3.0%. I would guess that if you sum up your 15 min power readings for a clear day and divide by 4, you will get an energy value closer to what the TED5000 and the meter show. If I had my tin foil hat on, I'd say SolarEdge is intentionally calibrating their daily energy to be +3%, to make their systems look more effective than the neighbors, while still staying safely in the 5% error range they publish. The 15 min data (which is sent to PVOutput.org, if that is set up), looks more accurate.

That's interesting, so the software is wrong again? As I have said, the only meter I REALLY
trust is my rotating disk. I periodically sample things for performance, but don't really
appreciate the value of 15 minute energy readings. Putting all those short readings
together sounds like an opportunity for much roundoff error.

There are losses everywhere in my system, primarily 1 % DC & 3 % AC transmission, 4%
conversion. But the values are known and can be compensated in calculations. If inverter
results are consistently cooked, a disk meter could be inserted by the inverter and run for
a month to compare answers. Then a consistent error could be taken into account. Come
to think of it, I might have to do that. The inverter readout is only the first 3 decimal places,
which means today I can only read monthly production to the last 100 KWH (out of several
thousand). Once the inverters hit 100,000 KWH, I will only be able to read to 1000 KWH.

I wouldn't bet TED is 1% or less error, till comparing it to an even better standard, not done
yet. And don't forget repeatability of opening & closing those current clamps. I did make
a check down to around 3% on true & reactive currents, and it did well. Bruce Roe

[HX_Guy]

Like I said, in my experience, at least with my TED unit, it's spot on. Not sure what more testing I could do aside from comparing the total kWh reported to the house meter. I've done this on three separate days and every time it was right where it should be...TED would say 38.4kWH, meter says 38. TED shows 36.8, meter shows 37.

As for the inverter...it makes reading much more often than 15 minutes. Those are just what it uses to give you a visual graph, but with SolarEdge, you can click on the Layout tab and then click on the inverter (or each individual panel) and get a near real time readout...it even says the last time it was updated and it's usually within a minute or two of the current time....for example right now it says "Last Measured 9:02am".

[gregvet]

Like I said, in my experience, at least with my TED unit, it's spot on. Not sure what more testing I could do aside from comparing the total kWh reported to the house meter. I've done this on three separate days and every time it was right where it should be...TED would say 38.4kWH, meter says 38. TED shows 36.8, meter shows 37.

As for the inverter...it makes reading much more often than 15 minutes. Those are just what it uses to give you a visual graph, but with SolarEdge, you can click on the Layout tab and then click on the inverter (or each individual panel) and get a near real time readout...it even says the last time it was updated and it's usually within a minute or two of the current time....for example right now it says "Last Measured 9:02am".

I don't use the TED5000 much for the minute to minute readings other than looking at the dashboard for instantaneous readings. I do use it every day to read daily energy consumption/production from the various CTs. I enter these daily totals on a spreadsheet that automatically calculate monthly totals. I assume the POCO numbers on meters are correct and I adjust the TED5000 numbers to match those on my bill. So far in January TED was long by 1.96% compared the POCO's production meter and short by 4.77% compared to POCO net meter but in February, TED5000 was short 4.79% compared to POCO production meter and long 5.19% compared to POCO net meter. The net accuracy for the first 2 months of 2015: TED5000 is 1.67% off compared to POCO production meter and 1.03% compared to the Net meter

[sensij]

Sorry for hijacking the thread. I will start a new thread on the topic of SolarEdge accuracy soon, once I have my own live data to include. To Enphases's credit, they do their calculations in a way that are at least internally consistent. The calibration may be off, but at least you don't have to worry about which number to calibrate, because they all report the same.

[bcroe]

Sorry for hijacking the thread. I will start a new thread on the topic of SolarEdge accuracy soon, once I have my own live data to include. To Enphases's credit, they do their calculations in a way that are at least internally consistent. The calibration may be off, but at least you don't have to worry about which number to calibrate, because they all report the same.

I would be happy to make adjustments so that everything agrees with the PoCo meter. They
have the best odds of being right, and a couple percent in this adventure aren't worth a dispute.
If you want to challenge them, you need something traceable back to the bureau of standards.
We used to pay big bucks every year to do that at work.

Bruce Roe

[azdave]

To the point of why SRP only charges a "demand" fee to solar users and not all users...

SRP does charge a demand fee to all commercial/business customers but this is the first time applying it to homeowners with solar (for now).

I'm certain their goal was to intentionally pit non-solar users against solar users in order to get demand charges accepted for the first time into residential bills. They needed to make the solar owners out as bad guys to get their foot in that door and they did that wihout any problem. I'm sure they will apply demand fees to all homeowners in the near future so at least we will all be back in the same boat when that happens. Meanwhile, they may say they support alternative energy but that's only so long as long as it come from their solar farm or wind generators and not yours and mine.

In an interview with SRP General Manager Mark Bonsall, a question was aked why SRP didn't apply the demand charges to all residential users. He said it was "too complicated" for non-solar customers but would be easy for solar users. Is it easy to re-alig panels and purchase load-shifting/monitoring devices?