Solaredge and clipping - anyway to use the clipped power?

[jbuszkie]

I just got my system installed and up and running! It's nice! My system is big but not big enough as we use a lot of electricity. But it's as big as I can make it. We are inverter limited. We had to keep the inverter to 10kW to stay under the net metering limit in MA. But the inverter allowed oversizing the panels to 155%. So I have about 15kW of panels. I'll get a lot of clipping. I hate to throw away all that potential energy. I know it's only for a couple hours a day in full sun, but I guess it could really add up.

So hypothetically.... if I could design a DC-DC converter that can draw just enough current from the PV array to keep the inverter from clipping, could that work?

Will the solaredge optomizers see a current that's more than the inverter is saying it's getting and then the inverter will then try to adjust the optomizer and cancel out my attempt to draw the extra current?

Now what would I do with the extra current? Not sure. Since this is hypothetical.. I, ideally, would use it to charge a battery bank and then when the sun is lower power, it would
feed back to the system... Either right to the DC of the inverter... Which I don't know if that would work or not. What does the inverter do with a DC voltage that it didn't get from the optomizers?
Or get another, smaller, grid tie inverter to dump back into my panel. I also have a electric water heater that's used as a pre-heat tank from my geo-thermal heat pumps (Which doesn't do much btw) I could feed the extra current to those elements and get "free" hot water. hmm...


Now this is probably all academic as I probably don't want to do anything to void my warranty..

But I'd be interested to see if anyone knows how the HD wave inverter and optomizers would behave in my hypothetical situation...

Or if someone would have a different idea to hypothetically harvest the extra DC power..


Jim

[J.P.M.]

Before I did anything, I'd get a better handle on an estimate of just how many kWh/yr. might be lost due to clipping. Once folks understand what's going on and do a little mind work, they often come to the conclusion that the lost production is a lot less than they thought. That lower likely loss often changes the economics of things.

Details on request.

[Ampster]

If SolarEdge comes out with a conversion kit like it has in other parts of the world you would be able to have the features of a StoreEdge inverter. Then you could attach an LG Chem DC coupled battery that would harvest that energy. Your installer should be able to tell you how much energy is lost to clipping. I have a 5.7kW DC system with a 3.8kW AC inverter and my installer ran a simulation and the annual lost production was a couple hundred kWrs. Your mileage may vary.

To answer your question about where does that lost energy go, the answer is that it doesn't go anywhere because it is not harvested.

[bcroe]

I just got my system installed and up and running! It's nice! My system is big but not big enough as we use a lot of electricity. But it's as big as I can make it. We are inverter limited. We had to keep the inverter to 10kW to stay under the net metering limit in MA. But the inverter allowed oversizing the panels to 155%. So I have about 15kW of panels. I'll get a lot of clipping. I hate to throw away all that potential energy. I know it's only for a couple hours a day in full sun, but I guess it could really add up.

Or if someone would have a different idea to hypothetically harvest the extra DC power.. Jim

Jim, its not just hypothetical, many others have such a problem involving high DC:AC ratio.
Here the motivation is to boost output under clouds. Having set my DC:AC even higher,
how can sunny output not be wasted?

Here the solution is to tilt panels away from the solar noon sun, to favor rising and setting
sun angles. Do this to increase that output, until clipping at noon is minimum or none.
Here is a curve from some test panels demonstrating this.

PVm17Jn16.jpg

And here is an actual output curve here, 15KW inverters, hardly a bell curve.

NScurJn17.jpg


Forget batteries, they are a dead loss if you do not have to have them. Bruce Roe

[Ampster]

My next installation is going to be ground mount so I can do that. I am also going to talk my wife into a induction cooktop so I can eliminate natural gas in the home.

[jbuszkie]

Jim, its not just hypothetical, many others have such a problem involving high DC:AC ratio.
Here the motivation is to boost output under clouds. Having set my DC:AC even higher,
how can sunny output not be wasted?

Here the solution is to tilt panels away from the solar noon sun, to favor rising and setting
sun angles. Do this to increase that output, until clipping at noon is minimum or none.

Interesting idea. I like it. But it won't work for me. My morning and evening sun has shadows. And my array is fixed. So
no changing it now!

[Paul Land]

Find a cheap EV

[Ampster]

Find a cheap EV

If you are answering the OP's question, the issue is on the DC side of the inverter.

[Paul Land]

If you are answering the OP's question, the issue is on the DC side of the inverter.

I know direct to ev thru how ever he wants to tap excess DC. Just like having dummy(dump) load on wind turbine.

dumploadfin.jpg

[Ampster]

Easy enough for Don Quixote, but not as simple as it seems.

[Paul Land]

Easy enough for Don Quixote, but not as simple as it seems.

"Truly I was born to be an example of misfortune, and a target at which the arrows of adversary are aimed."D.Q

[Ampster]

I was thinking more of the wind turbine comment as symbolic of DQ's fighting windmills. LOL

[neweclipse]

Jim, its not just hypothetical, many others have such a problem involving high DC:AC ratio.
Here the motivation is to boost output under clouds. Having set my DC:AC even higher,
how can sunny output not be wasted?

Here the solution is to tilt panels away from the solar noon sun, to favor rising and setting
sun angles. Do this to increase that output, until clipping at noon is minimum or none.
Here is a curve from some test panels demonstrating this.

PVm17Jn16.jpg

And here is an actual output curve here, 15KW inverters, hardly a bell curve.

NScurJn17.jpg


Forget batteries, they are a dead loss if you do not have to have them. Bruce Roe

Paper graphs are wonderfull, now may see the arrays with the mounts?

[bcroe]

Paper graphs are wonderfull, now may see the arrays with the mounts?

Paper graphs and slide rules were all we had when I started, still providing backup here
that will never be hacked. Here are 3 single panels facing west (1), south (2), and east
(3). In 2016 they were equipped with dummy loads to measure output on a sunny day.
The elevation angle was varied to try and find a combination of angles and directions to
produce the longest and flattest overall daily power curve. This is one of my better curves.


Test3dir.jpg


The second curve was produced by this array. Bruce Roe

NSview.jpg

[bcroe]

now may see the arrays with the mounts?

All the above mounts are now considered obsolete here. This 2018 mount could replace all.

18ArrayR.JPG


Bruce Roe