Power company not accepting production power

[jandrese]

I have a grid tie system using a SolarEdge Inverter and consumption meter. With the consumption meter I'm seeing values that don't make sense on most days. Even though my usage should be very low with a few spikes here and there (hot water, dish washer, etc...) what I see instead is the "self consumption" figure tracking closely my production and relatively little power being put on the grid. My assumption is that the consumption meter simply subtracts the "production" figure from the inverter by the total current going over the main line out to the grid. If the power company isn't accepting the power then it shows as consumption even though it's not actually being produced. This is a concern because we balanced our system on the assumption of building up kWh credits during the spring and fall to pay off during the summer and winter. If we can't build up credits it changes the payoff horizon considerably.

Is this expected? My contractors seems to be baffled by this, and nobody I've talked to claims to have seen it before. Some people have suggested it is due to improperly installed equipment, especially the consumption meter, but it seems to track what I got on my power bill, plus it does work as expected on a few days.

This is what I expect to see: solar_self_consumption_working.png

Instead I get this: solar_self_consumption.png

Note how the "self consumption" follows the production curve almost exactly, including the dip around 10 when a lonely cloud passed overhead. The little hitch around 8:15 is also shown on both graphs.

I'm really not sure how to verify this. Is there a particular question I can ask my solar installer or power company to verify what is happening? My power company is Dominion Virginia Power.

[oregon_phil]

What model inverter and consumption meter do you have?

What do you mean by it works for a few days? How many days working out vs not working?

[J.P.M.]

Another happy SolarEdge user.

[jandrese]

solar_self_consumption_partial.png


The Inverter is a SE11400H-US000BNU4
The Consumption Meter is a SE-RGMTR-3Y-277V-A

Over the past month I count:
15 days where the consumption traced the production closely as per the second graph.
7 days where the consumption traced production for only part of the day (almost always in the morning), but some production was shown. An example of such a day is shown above.
9 days where production seemed mostly normal.

I was not monitoring every appliance in my house constantly for the month, but the usage patterns are pretty regular so it's not hard to spot when the consumption figure is anomalously high.

[jandrese]

I do have one theory. In my neighborhood there are three other homes with solar panels. I'm guessing the substation can't feed back on the main grid, so if there is no other demand in the neighborhood the lines become fully saturated and the power has nowhere to go on mild days with lots of sun.

[scottbenson]

Hello Jandrese, I am a owner of a Solar Installation company in Costa Rica. We have installed hundreds of SE systems and was the first company to install the meters back in 2013 or so in the Northern Hemisphere.

First things first, confirm the CTS are placed correctly that is about 90% of the problems. With the new inverters it is getting simpler and simpler with more automation. The CTs of course measure the import and export the inverter shows the production. All the rest is calculated in the portal such as self consumption. You can see by this photo, this system is zero export. Using the battery and solar they are 88% self consumption. We will be getting that closer to the 99% by using soft starters on his well pump.


Screen shot ph.jpg

[oregon_phil]

↵jandrese , solaredge must have obsoleted the SE-RGMTR-3Y-277V-A because there isn't any documentation on their website at all. Not even in the support section. You didn't say how your system is wired up. Are you split phase, three phase?

scottbenson is suggesting that the installation of the revenue grade meter be verified. This means CT location/wiring/direction, power tap matches CT (CT1 to L1, etc), verify all extension wire splicing/wire colors/insulation/no shorts, verify cable shielding for extension wires if run in AC conduit, verify grounding, verify RS-485 communication wiring/termination settings/inverter settings. You could start by getting pictures of CT's.

I suspect your installer will have to verify the installation.

In the meantime, you can collect better data.

Over the past month I count:
15 days where the consumption traced the production closely as per the second graph.
7 days where the consumption traced production for only part of the day (almost always in the morning), but some production was shown. An example of such a day is shown above.
9 days where production seemed mostly normal.

I was not monitoring every appliance in my house constantly for the month, but the usage patterns are pretty regular so it's not hard to spot when the consumption figure is anomalously high.

These observations above are qualitative measurements and not be as useful in troubleshooting as quantitative measurements. Maybe the normal days aren't really normal. Sometime you have to force your system to a known condition. For example, if all breakers are shutoff, what happens? Do this for all of the different types of days stated above.

I do have one theory. In my neighborhood there are three other homes with solar panels. I'm guessing the substation can't feed back on the main grid, so if there is no other demand in the neighborhood the lines become fully saturated and the power has nowhere to go on mild days with lots of sun.

If other solar systems in your area are causing you grief, I think your inverter would momentarily shutdown due to line voltage rise issues. I don't see that in the graphs you show, but I may be wrong.

[jandrese]

I really hope the RGMTR isn't old and obsolete, it was installed in April.

[oregon_phil]

SE-RGMTR-3Y-208V-A is one of their current offerings. A call to SE or your installer would clear up any uncertainty.

[jandrese]

Hello Jandrese, I am a owner of a Solar Installation company in Costa Rica. We have installed hundreds of SE systems and was the first company to install the meters back in 2013 or so in the Northern Hemisphere.

First things first, confirm the CTS are placed correctly that is about 90% of the problems. With the new inverters it is getting simpler and simpler with more automation. The CTs of course measure the import and export the inverter shows the production. All the rest is calculated in the portal such as self consumption. You can see by this photo, this system is zero export. Using the battery and solar they are 88% self consumption. We will be getting that closer to the 99% by using soft starters on his well pump.


Screen shot ph.jpg

Is it possible for me to verify the CTS placement or do I need to call my contractor? I'm technical enough to trace wires, but I don't know what to look for.

[oregon_phil]

Since you don't know what you are looking for:

You will be working near live line voltage that cannot be turned off. If you do something wrong or in the wrong sequence you could cause injury or death to yourself. I am not an alarmist but electricity must be respected.

I don't know your skill set. I am not a sparky either so I am going to install my CT's in my two electrical panels and inverter after my automatic transfer switch is installed. That way I can shut off all power to all of my power panels.

So the one thing you can do is to located the CT's. There should be three of them. Two should be in your main power lines in your main power panel. The other one is probably in your main power panel on a wire going from your inverter to the breaker.

A photo of the CT is attached.

Take a picture of the CT's and the revenue grade meter.

[Ampster]

I do have one theory. In my neighborhood there are three other homes with solar panels. I'm guessing the substation can't feed back on the main grid, ......

Interesting theory but not one that is consistent with my understanding of how the grid works. Transformers are Bidirectional and solar generation just looks like a reduction of load to a substation. I have heard of issues in places like Hawaii but not likely in Virginia.

[PVAndy]

Interesting theory but not one that is consistent with my understanding of how the grid works. Transformers are Bidirectional and solar generation just looks like a reduction of load to a substation. I have heard of issues in places like Hawaii but not likely in Virginia.

Ampster

What you say is true for a radial grid. It may not be true if he's on a network grid where customers are fed from multiple transformers at the same time. Some of the protective devices in network grids don't allow backfeed. Network grids are found mostly in urban areas. See https://www.nrel.gov/docs/fy09osti/45061.pdf

Andy

[Ampster]

Ampster

What you say is true for a radial grid. It may not be true if he's on a network grid where customers are fed from multiple transformers at the same time. Some of the protective devices in network grids don't allow backfeed. Network grids are found mostly in urban areas. See https://www.nrel.gov/docs/fy09osti/45061.pdf

Andy

Andy,
Thanks for the clarification about the grid. What I didn't mention is that the hypothesis of jandrese is about what happens on the utility side of the meter and whatever happens on that side of the meter should not affect the meter reading. A more likely scenerio is the meter is not programmed correctly.

[Ampster]

It could just be a programming issue with the meter or power company configuration of the meter. Or the CTs may be configured wrong and no power is actually being exported and the CTs are misreporting.