Something limiting my PV production all of a sudden

[doronazl]

IMHO, Your installer is an ass. No professional should have configured it that way. On the red string, the 4 South roof array panels connected will be crippled by the performance of the flat roof panels.

They may be limited by the series voltage on each string (8S). If shading on the South roof isn't an issue, I would insist that they move those 4 panels from the flat roof to the South roof and add them in, so you would have 2X 8S strings on the roof.

If the MPPT can accommodate the 10S, then that's the easy way out.

There are some other consideration regarding the placement, i actually asked him to fill up the flat roof first.
Regarding the strings, 100% , i already feeded it to him.
My question is can we estimate how big of an improvement if they do just the rewiring, making it
10+10
6+6
?

And how do i know if my inverter can handle the 10s?

 

[TomC4306]

And how do i know if my inverter can handle the 10s

Easy. The open circuit voltage of 10 panels added together must not exceed the open circuit voltage rating of your inverter. And I would lower that inverter spec by 15% for temperatures.

 

[sunshine_eggo]

There are some other consideration regarding the placement, i actually asked him to fill up the flat roof first.
Regarding the strings, 100% , i already feeded it to him.
My question is can we estimate how big of an improvement if they do just the rewiring,

Can you pull individual MPPT performance (power)?

 

[doronazl]

Can you pull individual MPPT performance (power)?

Usage chart? Or its capability?
Will the spec data of the inverter help?

 

[doronazl]

This is what i have in solar assistant

if assume its showing the MPPT1 of both inverter on the left graph and MPPT of both inverter on the right one.
and some information from the inverterr label

 

[sunshine_eggo]

Can you extract the data into a CSV file or Excel spreadsheet?

Can you identify which string is connected to which MPPT?

Your MPPT can do 800V input, so I expect 10S will be no issue provided your panels are "typical" 60 or 72S panels. Looking back at your voltage plots, the 8Ss strings are never hitting 400V, so yep.

 

[32 volt boater]

 

[doronazl]

Can you extract the data into a CSV file or Excel spreadsheet?

Can you identify which string is connected to which MPPT?

Your MPPT can do 800V input, so I expect 10S will be no issue provided your panels are "typical" 60 or 72S panels. Looking back at your voltage plots, the 8Ss strings are never hitting 400V, so yep.

Great
Il have them fix this.
Would you guesstimate how much of a difference would it make? Once no mixed directional strings?

 

[TomC4306]

Would you guesstimate how much of a difference would it make? Once no mixed directional strings?

Nowhere near your dreamed of 15.3k.
Marginal improvement. Quantifying that is near impossible. Especially from a distance.

 

[doronazl]

Nowhere near your dreamed of 15.3k.
Marginal improvement. Quantifying that is near impossible. Especially from a distance.

i think we gathered that it wont reach 15k with this array setup.
i just want to know what to expect and if its so important to have them facing the same direction its gotta be meaningful right ?

 

[TomC4306]

i think we gathered that it wont reach 15k with this array setup.
i just want to know what to expect and if its so important to have them facing the same direction its gotta be meaningful right ?

Yeah, prolly, but without crawling on your roof and getting a little supercomputer to crunch the numbers on one panel 12° different than four other panels...
Meaningful? sure in theory. Measurable estimatable beforehand probably not.

 

[TomC4306]

Bottom line is it should not be connected the way it is connected. Very basic solar panel 101.

 

[Doggydogworld]

Only the 8 panel string with panels at different angles will improve. Right now current is limited by the 2 or 4 panels at the "worst" angle. The penalty varies with time of day and day of the year. There are even times, close to noon near the equinoxes, when all 8 panels will coincidentally be off-axis by the same amount and you won't see any penalty at all!

Imagine if the panels on the flat roof were pitched 45 degrees instead of 10 or so. Before 9am the west-facing panels would be shaded and after 3pm the east facing ones would be shaded. The shaded panels would pull the whole string down to near-zero output. But your angles aren't that dramatic. Your 4 or 6 "good" panels are probably only losing 10-20% of output on average.

Assuming each string has its own MPPT, 20% of your panels (6 of 32) losing 20% of output on average would be a 4% overall hit. I'll guess 2-5% improvement overall, and probably closer to 2 than 5.

 

[TomC4306]

I'll guess 2-5% improvement

Right, guess.

 

[sunshine_eggo]

My thinking is we can determine that the mixed string will perform the same as the flat roof array since that's the limiter.

The 2X 6S strings will perform like the 1 8S string.

Knowing the current performance of the existing strings, we can reasonably estimate the reconfigured array.

 

[Doggydogworld]

Knowing the current performance of the existing strings, we can reasonably estimate the reconfigured array.

Sure, if he has has string-level data. The pic showed two full strings on the flat roof in green and the string on the south-facing roof in cyan. The mixed string was red. The reconfigured array should output:

1.25 * (Green1 + Green2) + 1.5 * Cyan

Just compare that to existing output (Green1 + Green2 + Red + Cyan) to predict the percent improvement.

 

[doronazl]

IMHO, Your installer is an ass. No professional should have configured it that way. On the red string, the 4 South roof array panels connected will be crippled by the performance of the flat roof panels.

They may be limited by the series voltage on each string (8S). If shading on the South roof isn't an issue, I would insist that they move those 4 panels from the flat roof to the South roof and add them in, so you would have 2X 8S strings on the roof.

If the MPPT can accommodate the 10S, then that's the easy way out.

Had them fix it today.
It is now divided 10+6 and 10+6 to each inverter
So everything balanced now and no multi direction facing on same string.

Only thing limiting now would be heat right?
When its cloudy i saw it peaks to 14kw pv power but my question is if my system is 15.3kw, what should i expect when its constantly sunny and hot, how much approx in percentage the heat affecting the solar output capability?

 

[sunshine_eggo]

Had them fix it today.
It is now divided 10+6 and 10+6 to each inverter
So everything balanced now and no multi direction facing on same string.

Only thing limiting now would be heat right?
When its cloudy i saw it peaks to 14kw pv power but my question is if my system is 15.3kw, what should i expect when its constantly sunny and hot, how much approx in percentage the heat affecting the solar output capability?

We already established that your array orientation is suboptimal for this time of year. When it's cloudy, you're getting the cloud edge effect - clouds allow the panel to cool, then they are hit with full sun while cool. You then get a power spike that tapers off as the cells heat up.

Find the NOCT rating for your panel, and that's a reasonable expectation as it factors in hot cells and a 20% loss of incoming energy compared to STC.

I refer you back to this post:

Something limiting my PV production all of a sudden

Not really. Clipping is a hard limit where you essentially show truly flat production. You have what seems to be a taper and a mid-day spike. This seems more like a safety limiter. Maybe something is getting hot, and production is reduced. Clipping looks more like this (top flat line...

diysolarforum.com

Where I include a resource that will allow you to estimate your system's performance based on your location and weather patterns.

It won't be able to tell you your momentary power output, but it will be able to tell you typical daily and/or monthly kWh performance.

 

[doronazl]

We already established that your array orientation is suboptimal for this time of year. When it's cloudy, you're getting the cloud edge effect - clouds allow the panel to cool, then they are hit with full sun while cool. You then get a power spike that tapers off as the cells heat up.

Find the NOCT rating for your panel, and that's a reasonable expectation as it factors in hot cells and a 20% loss of incoming energy compared to STC.

I refer you back to this post:

Something limiting my PV production all of a sudden

Not really. Clipping is a hard limit where you essentially show truly flat production. You have what seems to be a taper and a mid-day spike. This seems more like a safety limiter. Maybe something is getting hot, and production is reduced. Clipping looks more like this (top flat line...

diysolarforum.com

Where I include a resource that will allow you to estimate your system's performance based on your location and weather patterns.

It won't be able to tell you your momentary power output, but it will be able to tell you typical daily and/or monthly kWh performance.

OK but am asking theoretically , say you have an array facing the optimal direction, simplest form, but bloody hot like where i am, in that situation , how bad is the degrading due to heat ?

 

[sunshine_eggo]

10-15% due to heat alone. Again, check the NOCT rating on your panels.