Rate my quote

What is the annual expectation? I have a 7.9 kW array and my estimate is 11,400 kWh or 11.4 mWh in Northern California. For me that is an annual factor in kWh of 1443 times the capacity in kWs of the array.

You're most welcome.

Thank you for the info.

How long can you roll the excess ?

What's your zip ?

With respect to modeling annual output, it's all an approximation of running 365 day output anyway. Many folks around here have found that using 10 % vs. 14 % for that loss parameter results in a better match to model vs. actual output. Your results will vary. FWIW, my running 365 day outputs have varied by about 4.5% after the first year's panel burn ins. That running output will generally vary inversely as the annual local clearness index which makes sense if I think about it. The sunnier the area, the higher the probability of less variation in annual output if only because more days will have less day/day variation in irradiance.

You can always add batteries. Depending on the rate of any battery price reductions it may even make economic sense to wait, particularly if/when the price reductions in batteries over a time period are greater than the savings over the same time period.

The solar company quote estimates at 11.5kWh which I can retrieve with PVWatts. We are higher in altitude and the solar radiation is very high.

From the Xcel brochure : "Any excess generation from your net-metered PV system will be rolled over month-to-month, year to year, for the Time of Use period in which it was generated. The credits will never expire, and will be used whenever your consumption from the grid exceeds your generation on the net meter."

I'm aware that Xcel could modify these terms anytime, at that time I'll shop for a battery

80108

Yeah that's fine, as long as on the long term my generation exceeds or equals my consumption

Thank you.

See the note on the PVWatts output page about annual variation and probabilities.

Also, FWIW, check out something called "PVoutput.org".
There seems to be a few sites close to zip 80108.
Actual real-time and historical outputs for working arrays.
Nothing like real data over rules of thumb and models. Just be careful with interpretation of what you're looking at and like all the info you get here, take it as less than gospel.

I suppose that could be seen as 3.95 sun hours per day. The array here
does about 2.4 sun hours under the clouds, actually not bad considering
so many are pointed E or W, predictions are for 3 sun hours. But taken in
terms of the relatively small inverter plant, it does 5.5 sun hours, which
was the original idea. Bruce Roe

Well no, it couldn't.

"sun-hours" is a rather imprecise term that's easy to throw around but can easily cause confusion precisely because it is imprecise and also because it masks units. To make matters worse, there seems to be several definitions of the term "sun-hour", all of which are not useful for any type of design purposes and some of which are just sloppy in definition. Google "sun-hours" and see what you get and then try to size an array using any of the definitions.

Anyway, and regardless of how one feels about the worth of the term "sun-hours", it has nothing to do with array output.

One "sun-hour" = 1 kWh/m^2/day (or per 24 hours) of solar radiation received on a surface of any orientation - not array output.

What I believe Ampster is referring to is array specific (energy) output. In usual units and using his numbers: [(11,400 kWh/yr.)/(7.9 STC kW)]/365 days/yr. = (3.95 kWh/day)/(installed STC kW). But that's not sun-hours.

So there it is, I have not completely abandoned sun hours as it keeps appearing
places. A quite unscientific term. Whatever it is, I am not using it to determine
an array output, that is alreadfy known. Rather, to determine relative component
utilization of different panel alignment approaches, a unitless number. Bruce Roe

Do as you see fit, NOMB, but your statements were incorrect as you wrote them and may have led others to incorrect conclusions.

We've already discussed and beat to death why, under low/no shade (and most all other) applications a single optimum array orientation will always have greater specific annual output than multiple array orientations.

That point has not been in doubt, though you keep saying it.

But that is not the objective here. The objective is to get as many
KWh as possible out of my 15KW inverter plant. That is about the
max size possible here for a lot of reasons. The numbers above
show, that this can be achieved with less than optimum orientation,
of a lot more now quite cheap solar panels. In this very poor solar
situation, I can not only match a 15KW plant optimized in the desert,
I can out produce it.

Alternately, a 30KW inverter plant with the same panels turned south,
under these clouds, would improve output some. Cost would include
twice as expensive an inverter plant, with wiring upgrades all the way
to my pole transformer, and I could no longer economize with double
sided panel mounts, which tend to cost more than the panels on them.
Today I do not need the extra energy, it would just be more I give to
the PoCo every April true up.

And I would be back to where I started, with the very unsatisfying feeling
of watching a huge inverter plant doing almost nothing a large part of
the time. Instead, it is very pleasant to watch my plant in clipping
8 hours straight, and even getting over 4KW in a rain storm. Bruce Roe

Bruce: I believe I understand your objectives. And it seems clear to me that whatever those objectives are, a cost effective way to generate electricity is not one of them.
But, bottom line, NOMB - and hale the freedom to make choices. Seriously.

But, purposely attempting to keep 15 kW of inverter in clipping 100% of the time by oversizing arrays and putting a relatively high proportion of the arrays in less than optimal orientations is not what a lot of folks consider an optimum design from at least a cost-effective standpoint.

IMO only, you explain yourself very well when you write "...it is very pleasant to watch my plant in clipping 8 hours straight ...".
If you're happy, I'm happy.
Not my arrays, life, choice, But I'd not choose to waste financial resources in that way.

All of which has naught to do with the error I called out in your statement of 07/13/2022 @ 1225 hrs: "I suppose that could be seen as 3.95 sun hours per day."

Now, can we stop this thread hijack and get back to Petitout's application?
Respectfully,

J.P.M.