Musings on snow on panels and production estimating

[electrofelon]

Do the values in PV watts take into account snow covering the modules for significant periods during the winter? Anyone heard of an analysis of different panel tilts and its effect on snow shedding?

Ideal fixed tilt here near Albany NY is 34 degrees. 22 degrees only results in 2% less annual production - in theory. But how much quicker will they shed snow? Dec Jan and Feb together are 17% of annual production. This gets back to the first question, but I assume that assumes no snow cover. WAG, 1/4 of the time covered with snow at 34 degrees, 3/4 of the time at 22 degrees?

 

[electrofelon]

Very old, but interesting. In particular, see page 11:

https://www.osti.gov/scitech/servlets/purl/5232456


A brief primer on a model in SAM:

https://sam.nrel.gov/sites/default/files/04-Severin_Ryberg-Snow_Model.pdf

 

[jaggedben]

Keep in mind that snow reflectance from the ground can increase irradiance and thus increase winter output beyond the expected. A higher tilt is better for that, too. I can't point to any kind of robust analysis but my intuition says to go with the higher tilt, all else being equal.

 

[gadfly56]

Do the values in PV watts take into account snow covering the modules for significant periods during the winter? Anyone heard of an analysis of different panel tilts and its effect on snow shedding?

Ideal fixed tilt here near Albany NY is 34 degrees. 22 degrees only results in 2% less annual production - in theory. But how much quicker will they shed snow? Dec Jan and Feb together are 17% of annual production. This gets back to the first question, but I assume that assumes no snow cover. WAG, 1/4 of the time covered with snow at 34 degrees, 3/4 of the time at 22 degrees?

When you measure tilt that way, are you talking about raising the top from horizontal, or raising the bottom from vertical?

 

[jaggedben]

From horizontal.

 

[gadfly56]

From horizontal.

That didn't seem to make sense, so I went here and confirmed my hunch that it is from vertical, as in pin the top of the panel like it's the rear door of a hatch-back and lift from the bottom. 90 degrees would be parallel to the ground, 0 degrees is perpendicular to the ground. According to the calculator the ideal sun angle for Albany, NY varies from 24 in December to 70 in June.

 

[electrofelon]

Keep in mind that snow reflectance from the ground can increase irradiance and thus increase winter output beyond the expected. A higher tilt is better for that, too. I can't point to any kind of robust analysis but my intuition says to go with the higher tilt, all else being equal.

Yeah good point.

 

[electrofelon]

That didn't seem to make sense, so I went here and confirmed my hunch that it is from vertical, as in pin the top of the panel like it's the rear door of a hatch-back and lift from the bottom. 90 degrees would be parallel to the ground, 0 degrees is perpendicular to the ground. According to the calculator the ideal sun angle for Albany, NY varies from 24 in December to 70 in June.

This site has the equations:

http://www.solarpaneltilt.com/

The common theory of "set them to your latitude" is incorrect.

 

[wwhitney]

That didn't seem to make sense, so I went here and confirmed my hunch that it is from vertical.

You found a site using that convention. This thread has been using the other convention, as does the site that electrofelon mentioned, as does the (non-optimal) rule of thumb "tilt angle = latitude".

Cheers, Wayne

 

[jaggedben]

Industry standard, at least in this country, is tilt is from horizontal. In the decade I've been doing this I've never seen or heard anyone measuring from vertical. See Solmetric tools, for example. Or PVwatts. Or CSI.

 

[gadfly56]

This site has the equations:

http://www.solarpaneltilt.com/

The common theory of "set them to your latitude" is incorrect.

If you look at the site I had gone to, for Albany NY the tilt angle is 70 degrees. Per Google, Albany NY is at 42.6526 degrees North. Using your web site's equation for 4-season adjustment, the summer angle, converting for my site's convention, is 71.4 degrees. Not very different. Optimization asks two questions; when do you need the most electricity vs the sun angle at that time.

If you're looking for the maximum utility offset for a single fixed angle, the sun angle will be biased toward the summer angle as that's when you have the longest collection period. In concept it's a pretty simple max/min problem, but I'm so far removed from high school senior year calculus I can't even set it up now.

 

[electrofelon]

If you look at the site I had gone to, for Albany NY the tilt angle is 70 degrees. Per Google, Albany NY is at 42.6526 degrees North. Using your web site's equation for 4-season adjustment, the summer angle, converting for my site's convention, is 71.4 degrees. Not very different. Optimization asks two questions; when do you need the most electricity vs the sun angle at that time.

If you're looking for the maximum utility offset for a single fixed angle, the sun angle will be biased toward the summer angle as that's when you have the longest collection period. In concept it's a pretty simple max/min problem, but I'm so far removed from high school senior year calculus I can't even set it up now.

Yeah to be clear I was thinking fixed angle for max annual production in my previous posts. Yes that site has other situations that an off gridder might want such as adjusting the angle several times a year or setting at the ideal winter angle. The ideal winter angle would def shed the snow nicely!

 

[pv_n00b]

Do the values in PV watts take into account snow covering the modules for significant periods during the winter? Anyone heard of an analysis of different panel tilts and its effect on snow shedding?

Ideal fixed tilt here near Albany NY is 34 degrees. 22 degrees only results in 2% less annual production - in theory. But how much quicker will they shed snow? Dec Jan and Feb together are 17% of annual production. This gets back to the first question, but I assume that assumes no snow cover. WAG, 1/4 of the time covered with snow at 34 degrees, 3/4 of the time at 22 degrees?

PV Watts is pretty primitive. If you go into the loss calculator you can enter an annual loss percentage for snow coverage and that's it. If you want to do more advanced simulations then you need to use SAM or PVSyst.

Simulating loss from snow is actually pretty complex. There are a number of papers on it available on the internet. I found it particularly interesting to find out the huge effect having a place for the snow to slide off too has on losses. If the array is not high enough off the ground for the sliding snow to drop clear of the modules the increase in annual losses is quite large.