I assume it is normal to have poor production up north during the winter?

[reader2580]

My system didn't get turned on until five or six weeks ago. Should I assume it is normal to have low production this time of year? I didn't generate enough power to offset even half of my electric usage. My 6.3KW system should over a year produce enough power to offset my entire electric bill.

[Mike90250]

Give us some specs, your zip code, array orientation, elevation angle, panel specs, inverter specs

[J.P.M.]

See PVWatts, read the help/info screens and do a few runs. Note Dec. modeled output vs. June.

The farther north you go , the greater the disparity between winter/summer.

[peakbagger]

PV Watts will give you real numbers. In general there are a couple of issues that aren't so good in winter and one issue that actually is better in winter.

Panel Angle - The sun is lower in the sky in winter, to get the optimal angle so that the panel is close to perpendicular it has to be set at an angle a lot steeper than a standard roof. At 45 latitude (far northern NH and upstate NY for example) the recommended angle is around 30 degrees off vertical. Rarely do you find a roof pitch much steeper than 12/12 (45 degrees). Therefore the panels rarely are at the optimal angle, this lowers the output.

Hours of sunlight - Days are shorter so less hours of sun

Panel Temps - This is better in the winter, the colder the panel the higher the production. On a sunny cold day the system could put a pretty impressive short term output, unfortunately this in no way offsets the first two.

The other two items are site specific. Shading can be more of an issue with low sun angles. A vent pipe can throw quite a shadow and trees to the south also can shade panels.

Snow coverage - Many installers lie about how quickly panels will clear after a storm. Unless you manually clear them, shallow panels may be covered for weeks.

Remember PV Watts or any design tool is based on past years averages, it will take a couple of years to really see how well the system works compared to the original design. If you happen to be in any area with good net metering that doesn't reset yearly, in northern areas, most people cut way down on their summer usage and build up a big surplus that carries them into some or part of the winter. I added an array to my prior system so I carry a surplus over the winter and burn it up with a mini split for heating.

[reader2580]

Give us some specs, your zip code, array orientation, elevation angle, panel specs, inverter specs

Zip is 55304
Array is 20 degrees west of true solar south
Angle is 26 degrees
Panels are 300 watts and there are 21 of them
Inverter is Solaredge SE6000H
No shading

[reader2580]

I had not looked at PV Watts since I did the initial design. According to PV Watts I should be getting around 400 KWh even in the worst month. I have had snow on the panels at times so that doesn't help. We just got 12 inches of snow this weekend, but it is supposed to be near 40 degrees this week so the snow should be gone off panels pretty quickly.

I also have 3,120 watts of solar that isn't in production yet on a ground array. I have to wait until ground thaws to put in underground conduit to array.

[emartin00]

I'm not sure what the weather has been like in your area, but I know we have had quite a few overcast and snowy days this winter. If there is snow on the panels, it's obviously going to decrease your output a lot.
Typically cold weather is good for panels, because it actually increases the efficiency. But at the same time, the days are much shorter this time of year.

[J.P.M.]

PV Watts will give you real numbers. In general there are a couple of issues that aren't so good in winter and one issue that actually is better in winter.

Panel Angle - The sun is lower in the sky in winter, to get the optimal angle so that the panel is close to perpendicular it has to be set at an angle a lot steeper than a standard roof. At 45 latitude (far northern NH and upstate NY for example) the recommended angle is around 30 degrees off vertical. Rarely do you find a roof pitch much steeper than 12/12 (45 degrees). Therefore the panels rarely are at the optimal angle, this lowers the output.

Hours of sunlight - Days are shorter so less hours of sun

Panel Temps - This is better in the winter, the colder the panel the higher the production. On a sunny cold day the system could put a pretty impressive short term output, unfortunately this in no way offsets the first two.

The other two items are site specific. Shading can be more of an issue with low sun angles. A vent pipe can throw quite a shadow and trees to the south also can shade panels.

Snow coverage - Many installers lie about how quickly panels will clear after a storm. Unless you manually clear them, shallow panels may be covered for weeks.

Remember PV Watts or any design tool is based on past years averages, it will take a couple of years to really see how well the system works compared to the original design. If you happen to be in any area with good net metering that doesn't reset yearly, in northern areas, most people cut way down on their summer usage and build up a big surplus that carries them into some or part of the winter. I added an array to my prior system so I carry a surplus over the winter and burn it up with a mini split for heating.

Mostly agree but with a couple comments: PVWatts will give "typical" output probably within +/- 30 % or a bit less for any month and maybe +/- 5% to 10% or so for any 12 month period.

Also, you state PVWatts is based on many years' averages. That is incorrect. Assuming you are referring to the weather inputs to PVWatts, those inputs are from the TMY (Typical Meteorological Year) data base. That database is not an average of weather (or other data).

As an aside, and separate from the issue of "average": A dirty secret hiding in plain sight: The TMY models, either TMY2 or TMY3 are themselves mostly a model of "typical" weather for over 1,000 U.S. and additional other worldwide locations. Almost none of the data is from actual measurements or observations or instruments. It's mostly synthetic data.

In either or any case, synthetic or actual data, as for why the data is not, or cannot be called average, each month data for a location's TMY file is chosen by examining what is mostly or entirely modeled data for prior years and then choosing the month of all appropriate months (say, all Jan., Feb., March, etc.) that is the "most typical" (note - not average) of all such months examined, with heavier weighting factors applied to irradiance data. The months are then strung together to make what's called a "typical year" of 12 months. So, a TMY for a location may consist of data for Jan. of, say, 2002 because that year's Jan. came closest or was the "best" representation of what "typical" Jan. weather might be, considering the weighting factors. In the same way, Feb. for that location might be best represented by, say, Feb. 1995 for example, and so on. As for why an average of the last, say, 30 years' data is not as useful: An average of each day's weather (or each hour's weather for that matter) will give what's probably a similar but not identical set of daily numbers, but that's not typical of how weather works most of the time. This year's Feb. 26, for example, is different from every other prior Feb. 26 and not necessarily or even probably like Feb. 25 or Feb. 27. As an average, those daily numbers will probably be close to the adjacent days' numbers, but that will mask weather patterns and local anomalies. A couple of other ways averages are not as good as the TMY logic of using "typical" data: 2 same date days for different years may have identical irradiance, but one will be clear until solar noon and heavily overcast the rest of the day while a other may be partly cloudy all day with scattered clouds. The different irradiance patterns will affect solar device output. Also, sunny days tend to correlate to some degree with higher temperatures. Average data will bury that. The TMY methodology is an attempt to account for such variation in ways that using average values cannot.

It's also not perfect, but given that most of the data is modeled, it's probably adequate. Seeing how close models that use TMY data can often and commonly get to actual system performance may confirm it's adequacy.

See the TMY manual or the open literature for a better explanation.

Respectfully,

[reader2580]

It seems like what I am seeing is probably normal. If I continue to only produce enough power to cover half my usage I will likely never break even on my solar system. From what I have seen to date I'm looking at 20 to 30 years for break even without any equipment failures. I can assume I'll have to replace the inverter and all of the optimizers at least once in 20 to 30 years which would mean no break even ever if production continues to be low.

I am planning that production will be much higher come spring/summer.

[J.P.M.]

It seems like what I am seeing is probably normal. If I continue to only produce enough power to cover half my usage I will likely never break even on my solar system. From what I have seen to date I'm looking at 20 to 30 years for break even without any equipment failures. I can assume I'll have to replace the inverter and all of the optimizers at least once in 20 to 30 years which would mean no break even ever if production continues to be low.

I am planning that production will be much higher come spring/summer.

Most quality components professionally installed and reasonably maintained (which mostly means hosed off/cleaned 1X/awhile) produce close to what the models' output suggests. I'd SWAG on what PVWatts models for each month, +/- 30 % on a monthly basis, and +/- 10% for a yearly total.

Most arrays simply produce less in the winter, more in summer. If you model on winter's output, you will underestimate your system's output. Your summer output will probably be ~ 2X winter output or something like that. Think annual, not seasonal or monthly.

[bcroe]

I also have 3,120 watts of solar that isn't in production yet on a
ground array. I have to wait until ground thaws to put in underground conduit to array.

Here at 42 deg latitude 60504 my June production is about 3 times Dec, that is just the way it works.
Just about now I am seeing some serious production when the sun shines, maybe 2/3 of a June day.

I won't tell you what happens to wires here before I can bury them. Bruce Roe

[reader2580]

The ground mount array has no wiring at all yet. It won't until I get the conduit run.

[bcroe]

The ground mount array has no wiring at all yet. It won't until I get the conduit run.

I do hope, you don't have panels ground mounted with the wires just hanging the
way they come off the skid. Those MC4s should never be left unplugged in any
weather, or even for long storage, as the contacts would deteriorate rapidly. I
have had to replace some of mine. Bruce Roe

[reader2580]

I do hope, you don't have panels ground mounted with the wires just hanging the
way they come off the skid. Those MC4s should never be left unplugged in any
weather, or even for long storage, as the contacts would deteriorate rapidly. I
have had to replace some of mine. Bruce Roe

I have rubber plugs for MC4 connectors. Will those work for another two to three months until the panels are hooked up? The panels have Solaredge optimizers built-in if that matters.

[specialgreen]

I'm in the Twin Cities. As a data point, during February 2017, I produced 395 kw-hrs; but produced only 49 kw-hrs in February of 2018 (88% less). Over one-third of this month's generation came-in just today. It's pretty meaningless to look at 6 weeks of winter and try to guess what the system will produce over years.