Solar install turning out really bad - looking for some help/advice please.

[HX_Guy]

Today *should* be the last day on the roof for this project. Buttoned up everything, even did a couple things I wasn't planning on but figured what the hell while I was up there.

First started out by painting the conduit on the side of the house, came out looking pretty good...


This was the before...


Next I redid all the MC4 connectors on the roof. As expected, every single one of them was bad in varying degrees. This was probably the worse, cable pulled right out with no effort.


After that I moved the top row of panels over a few inches so they lined up nicely with the bottom.
Mainly for aesthetics but also to cover up the wiring at the combiner box better.

Before...




After...



Leaves me with the perfect amount of space for a future extra panel on the upper left side too.




Painted all the visible parts of the aluminum mount covers. Mainly to stop the refection from the sun...it was like a mirror up there and the heat reflected back was crazy (heat is not good for solar panels, it actually lowers their power output).


Repositioned the conduit on the roof so it sits higher off the roof, and painted it.


And finally, bought a nice big dust mop to clean off the panels. They got pretty dusty in the past month (as you can see on the left side of the panel being cleaned) and I want them super clean for tomorrow's production "test".


Here's to hoping inspection passes on Monday!!!

[HX_Guy]

Had a good production day today, just shy of 55kWh. Graph is a bit strange, normally the curve is supposed to be very smooth but mine isn't...I'm guessing it's due to the partial shading of some panels? About a month left of shading and them I'll be shade free until October.

It almost looks like clipping but I don't believe that is what's happening, or at least shouldn't be happening.



Here is a normal, non-shades curve of a considerably bigger system for comparison (16.0kW).

[Alisobob]

solar53.JPG

My output is so different from the "normal" graph you posted... Mine basically plateaus from 11am to 1:15 pm , kinda' like yours. But I have zero shade issues.

Perhaps its the way your style graphs are condensed into a shorter timeline.

...weird.

I'm glad you got everything recrimped.

I was worried about that.....

[foo1bar]

It almost looks like clipping but I don't believe that is what's happening, or at least shouldn't be happening.

To me it looks like clippping.
It looks like it reaches it's max power output from ~11AM to ~1:30PM.

I'd look at each panel's power output - if you have a number of them that have similar flat-topped graphs, maybe that string is reaching it's maximum power.
Or maybe you've reached the max on your inverter. (I don't recall what size inverter you have.)

I wouldn't think shading would cause a plateau like that - I would expect shading to cause the line to be less smooth in the evening or morning. I wouldn't expect a bunch of shading at around noon. My guess is the line from ~9 to ~9:30 is showing some shading impacting your production.

[bcroe]

Had a good production day today, just shy of 55kWh. Graph is a bit strange, normally the curve is supposed to be very smooth but mine isn't...I'm guessing it's due to the partial shading of some panels? About a month left of shading and them I'll be shade free until October.

It almost looks like clipping but I don't believe that is what's happening, or at least shouldn't be happening.



Here is a normal, non-shades curve of a considerably bigger system for comparison (16.0kW).

Don't forget, when you have panels facing different directions, there will be time that
one group in increasing but another is decreasing. That might relate to the curve.

Big clouds today, you made a lot more power than me. Bruce Roe

[HX_Guy]

To me it looks like clippping.
It looks like it reaches it's max power output from ~11AM to ~1:30PM.

I'd look at each panel's power output - if you have a number of them that have similar flat-topped graphs, maybe that string is reaching it's maximum power.
Or maybe you've reached the max on your inverter. (I don't recall what size inverter you have.)

I wouldn't think shading would cause a plateau like that - I would expect shading to cause the line to be less smooth in the evening or morning. I wouldn't expect a bunch of shading at around noon. My guess is the line from ~9 to ~9:30 is showing some shading impacting your production.

I understood clipping to mean when the power output of the panels is greater than that of the optimizers/micro-inverters/inverter. None of those are (or should be) and issue in my system. My panels are 310W, the optimizers are rated to 400W and my inverter is rated to 12,000W (though only 5250W per string). I have one string of 16 panels which at 310W per panel is 4960W, but the panels were producing at most 255W at peak and 3 of the 16 where shaded. There shouldn't be any clipping.

I think as bcroe pointed out, it's the transition of not only different facing panels, but also of the shade across the panels as the sun moved across the sky.

As you can see in the day's summary of each panel, the 4 along the bottom in the middle were really affected by the shade, but so were the ones to the left and right of them. Plus you have the 11 panels facing east so as those transition out as the sun moves from east to west, but more panels that face south come "online", it creates that plateau I'm thinking?



Another reason why I don't think it's clipping is because my chart from 3 weeks ago looks pretty much the same, except the day was shorter but it still plateaus in the same fashion at a lower power output...so it couldn't have been clipping at ~7.6kW then and at 8.2kW now.

[foo1bar]

but the panels were producing at most 255W at peak and 3 of the 16 where shaded. There shouldn't be any clipping.

3 of the panels were shaded at noon?

That just seems odd to me that you'd have shading at noon, even in winter.

I'd still look at each panel's graph for the day (or isn't that an option?)

[HX_Guy]

Theres no option for a panel level graph, but what I can see is the power output of each panel at 15 minute intervals.

Here is what it looks like at noon...4 panels shaded.




1pm is actually the time of day that the least amount of panels are shaded, 3 panels. I visually confirmed this up on the roof yesterday.






Reason being is the neighbor's 2 story house that is literally 10' wall to wall, and 8' roof edge to roof edge.

[HX_Guy]

Just called the city inspector and she said she would be here today between 10:00 - 12:00 for re-inspection.


I was thinking there isn't anything major I should be worried about but now I'm not so sure.
The permit spec'd out two AC disconnects, one on each side of the gate which is required by code. My installer of course said that it's fine and we only need one. The inspector didn't say anything about it the first time she was out, so she either missed it (pretty hard to miss though, no?) or she's fine with just one.

I did ask the permit designer about it previously and I guess the code says that the disconnect has to be in line of sight at all times, otherwise two are required. When the gate is open, the one already installed is right there, probably 6 ft away at most so maybe it's ok? Guess we'll see what the inspector says when she gets here.

Permit specs:



As installed:

[Willy T]

Just called the city inspector and she said she would be here today between 10:00 - 12:00 for rein spection.

If she doesn't see it, I am sure you'll point out to her. It was another one of your mistakes to insert yourself between the installer and the city. Once you have bureaucrats playing CYA you end up paying the price.

[HX_Guy]

If she doesn't see it, I am sure you'll point out to her. It was another one of your mistakes to insert yourself between the installer and the city. Once you have bureaucrats playing CYA you end up paying the price.

What? Why would I point it out to her? And what does it have to do with my inserting myself between the installer and the city?

[SunEagle]

What? Why would I point it out to her? And what does it have to do with my inserting myself between the installer and the city?

That wiring diagram shows a two pole "switch" on the AC side of the inverter that opens the circuit for the L1 and L2 connections. If that can be "locked out" it might be considered to the the first of the 2 required AC disconnects.

[Ian S]

Keeping my fingers crossed that all goes well today!

[HX_Guy]

That wiring diagram shows a two pole "switch" on the AC side of the inverter that opens the circuit for the L1 and L2 connections. If that can be "locked out" it might be considered to the the first of the 2 required AC disconnects.

Yeah except that's not accurate. The Inverter only has a built in DC disconnect switch, no AC...even though the permit shows it. SolarEdge changed their systems somewhat recently, they used to have both AC and DC built in and the data sheets you find on Google still seems to show that.

[Amy@altE]

Most inverters stopped doing the on-board AC disco since most locations were requiring an external one. Made theirs irrelevant, so it made sense to remove it and save the $.