SolarEdge Metering help

[TheWorm3]

Hi all,

I am looking for some help in determining how to set up an import/export meter a large commercial site (480/277V) with (10) 100kW Solaredge inverters. The issue I am having is that the service has parallel 2000A panels coming from the utility owned transformer/meter, so we can't place CT's around the main feeders for the site, again, because they parallel directly from the transformer which is the utility's equipment. So my main question is where would we need to place the CT's, do we need multiple sets of CT's and/or meters in order to accurately capture import/export data for the entire site. The solar is tied into the main lugs of one of the 2000A panels. Any information would be greatly appreciated!

---

 

[pv_n00b]

Can you access the point where the parallel panels and the utility meter come together? That would be a good point for a single CT setup. If not then you will just have to put CTs in both panels and do some math. With modern multi-input meters, it should not be that hard to set up a system that will tell you if you are exporting of not.

 

[TheWorm3]

We can't access that point because the utility owns the transformer and will not allow it.

So we currently have the feeders from each panel CT'ed. The CT's are then paralleled (L1 from panel 1 paralleled with L1 of panel 2, etc) and wired to the Solaredge meter. The resulting graph shows overnight production being recorded, however the daytime consumption has spikes that mirrors the production curve and highest consumption is recorded at peak power production.

 

[tallgirl]

Are the CTs reversed? When consumption shows as production, that's usually a CT on backwards.

For bidirectional recording, direction matters. If a meter won't record both directions everything is just consumption. When you're logging both production and consumption the orientation of the CT matters. Turn the CTs around and see if that gets your production to be production, and your consumption to be consumption. I can't answer the paralleling question, because I'm just not THAT smart.

 

[TheWorm3]

Are the CTs reversed? When consumption shows as production, that's usually a CT on backwards.

For bidirectional recording, direction matters. If a meter won't record both directions everything is just consumption. When you're logging both production and consumption the orientation of the CT matters. Turn the CTs around and see if that gets your production to be production, and your consumption to be consumption. I can't answer the paralleling question, because I'm just not THAT smart.

I don't believe they are reversed, because overnight we see consumption recorded and it appears to be normal and as expected. It's when the solar production starts that the variations and spikes and mirroring occur. We are not currently using CT's for production data, as this data comes directly from the inverters themselves. Thank you for the the response though!

 

[BillK-AZ]

My experience with a SolarEdge 7.6 kW Energy Hub inverter on a residence with 400A service and a service entrance with 200A/200A/100A connections. To measure the total residence utility power required the use of six 200A SolarEdge CTs. These are connected in parallel for each line and the setup for the energy meter had to be for 600A CTs. If the setup was left at the default 200A, there were strange results. The Energy Hub system does not measure total load, but computes it as the difference between inverter power and utility power.

 

[jaggedben]

If you have multiple CTs paralleled to the same inputs on the meter, and most of them are correctly oriented but say one or two are not, then it's certainly possible that things would look plausible most of the time until something that's being measured by the wrongly oriented CT overwhelms the rest. That something being the relatively large peak PV output, that is.

In other words, I would second the advice to go back and check the orientation of all your CTs.

Also you confused me by saying 'overnight production' in post #3 and 'overnight consumption' in post #5.

 

[TheWorm3]

If you have multiple CTs paralleled to the same inputs on the meter, and most of them are correctly oriented but say one or two are not, then it's certainly possible that things would look plausible most of the time until something that's being measured by the wrongly oriented CT overwhelms the rest. That something being the relatively large peak PV output, that is.

In other words, I would second the advice to go back and check the orientation of all your CTs.

Also you confused me by saying 'overnight production' in post #3 and 'overnight consumption' in post #5.

My fault, I meant overnight consumption for both posts. Our electrician has verified the directions on the CT's, unless the direction on the CTs was labeled wrong, we should be good there.
I have been referencing the attached Solaredge document about metering and from our symptoms it seems we may have the CT's in the wrong location. Currently there is just one meter with the CT's paralleled into that meter.

 

[TheWorm3]

Here is a basic one line diagram showing the electrical.

 

[jaggedben]

My fault, I meant overnight consumption for both posts. Our electrician has verified the directions on the CT's, unless the direction on the CTs was labeled wrong, we should be good there.
I have been referencing the attached Solaredge document about metering and from our symptoms it seems we may have the CT's in the wrong location. Currently there is just one meter with the CT's paralleled into that meter.

I have seen instances of CTs labeled backwards by the factory among the ones Solaredge supplies. For what that's worth. Very very annoying.

Also note that the CTs can be accidentally reversed at any of three places:
- CT orientation around the conductor
- CT leads landed on the meter
- power wires to the meter which are used for voltage reference (which I guess with three-phase could be 'rotated' as well as reversed)

It would be helpful in your diagram to confirm the CT placement. But it would still make sense if one or more of the CTs on the feeder which has the PV are backwards. At night, the load measurements on that feeder may 'cancel out' and be close to zero, while the other feeder is reading correctly, such that the indicated consumption is postive but is lower than the real load. Then during the day when the PV starts cranking, its export to the grid is read as consumption and that is why your consumption readings partially mirror production.

 

[jaggedben]

Also yes, location could be screwing things up. I forget where and how it's set with Solaredge but you need to tell it whether the CTs include measuring the solar output or not. Depends which side of your tap you put the CTs on.