calculating north tilt vs afternoon tree shading

[S0larAlex]

first time posting on this website, thank you all for the stickies and such it was very helpful in my research.

My house has low pitch 10% degree roof in both south and north orientation. South orientation does not have enough space to fit all the panels and part of it gets tree shading in the afternoon (probably after 2pm, in the winter and probably none in the summer month since the tree is 35' south, 30 feet west' and about 30-35 feet high)
The design I got from the installer had 6 panels (out of total of 21) on the north facing slope, and part of the south roof not filled out completely.
I'm questioning their design because they do it remotely and obviously don't see the real conditions as google earth is outdated.

So I did a design using http://pvwatts.nrel.gov/ for both north and south facing slopes and would like to do an excel calculation sheet to determine whether filling up the south roof completely with some shading is more beneficial than putting 6 panels on the roof facing north.

My main question would be, how can I accurately account for tree shading in the afternoon.
In the excel I was planning to take south facing design and north facing design and mold each one to the two options I have. I planned to take off some percentage points for the tree shading in the winter - for example 20% off total December energy production.

Any idea what percentage of energy an average tree without any leafs might block?

Thanks

[ButchDeal]

Any idea what percentage of energy an average tree without any leafs might block?

a leaf off tree is going to block nearly all production for the modules it shadows. With optimizers you will get a little but I would for planning consider it zero for the effected modules.

[bcroe]

Shade can easily cost 90% loss of output from the shaded cells, and any directly feeding through them. To maintain reasonable efficiency
under partial shade, your system needs to have means to bypass the shaded cells, such as optimizers or use of micro inverters. Accurate
simulation of shading is going to be difficult, except assuming zero output for the interval. Bruce Roe

[S0larAlex]

is there a way to calculate where the shadow will be will based on the tree size (height width etc), how far it is from the house for each month of the year and time of the day?

[ButchDeal]

most design systems will do this for you.
Aurora, SolarDesigntool, Helioscop, etc.

[J.P.M.]

is there a way to calculate where the shadow will be will based on the tree size (height width etc), how far it is from the house for each month of the year and time of the day?

Yes there is. I did calculate some such stuff a long time ago and later using a spreadsheet for the very simple cases of a small panel area when the shadow line of the top of a tree would begin to shade the panel. I use something similar for my array and the tree that is to the west of the array and it seems to produce reasonable results that match measurements to a fair degree.

I can go into it, but it's a bit complicated and take more space than available here. Basically it involves finding, with some tolerance, the angle of incidence, relative to some reference system, that an obstruction such as a tree, building or structure (think chimney) makes with a solar device, or some point on a solar device, and then comparing that to the solar incidence angle for that point for any time series over a year. That's pretty much how most such mathematical algorithms work in the simple sense. Unless your doing this as a learning experience, you're probably better off buying canned stuff. If you're doing it for the educational value, you'll learn a lot about solar geometry. Like a lot of this solar energy business, it's far from an exact science. For all my work, I got about as close with back of the envelope calcs and a protractor.

[bcroe]

A Solar Pathfinder or similar tool will allow you to see what time the sun is shadowed, at different times of the year, for a particular spot.
Move it around to see how other spots are affected. Bruce Roe

[J.P.M.]

A Solar Pathfinder or similar tool will allow you to see what time the sun is shadowed, at different times of the year, for a particular spot.
Move it around to see how other spots are affected. Bruce Roe

FWIW, I've used this tool. It seems easy to use and IMO, it produces useful information. I borrowed one from a vendor to check it against my estimates. However, it seems a bit pricey to me for a single or a few uses such as a homeowner might need.

As an aside, shading analysis might be a decent/interesting sideline for someone in popular solar markets. On 2d thought, maybe not so much of a demand now that residential PV looks to be headed for a possible downturn, at least in the U.S.

[bcroe]

FWIW, I've used this tool. It seems easy to use and IMO, it produces useful information. I borrowed one from a vendor to check it against my estimates. However, it seems a bit pricey to me for a single or a few uses such as a homeowner might need.

I got mine A1 cond used on Eb+y for a substantial discount; spare parts were readily available. Had to order some scales
for my latitude; could probably get much of my money back the same way if I no longer needed it. I would say it has been
quite useful first in dealing with trees, and then for planning future layout changes (ground mount). Bruce Roe

[J.P.M.]

I got mine A1 cond used on Eb+y for a substantial discount; spare parts were readily available. Had to order some scales
for my latitude; could probably get much of my money back the same way if I no longer needed it. I would say it has been
quite useful first in dealing with trees, and then for planning future layout changes (ground mount). Bruce Roe

Understood. Thank you for the info.

Looking back, there was a prior thread that discussed the solar path finder with some talk of several homeowners sharing the cost. Also, it seems there are several places that will rent the device for what may be reasonable rates.

[peakbagger]

If you have access to a Cad program its not that hard to setup up a shading simulation using US Naval Observatory Sun angle data and measurements you take around the house. I expect someone good at it could do it all in 3D but I just do hourly section elevations and bracket the dates where the suns angle doesn't line up with typical generation hours. I wish it was as easy to factor in sticky snow covering the panels (like I have currently on my roof mounts

[ButchDeal]

If you have access to a Cad program its not that hard to setup up a shading simulation using US Naval Observatory Sun angle data and measurements you take around the house. I expect someone good at it could do it all in 3D but I just do hourly section elevations and bracket the dates where the suns angle doesn't line up with typical generation hours.

why not use the more accurate and simpler to set up tools that already exist, like Aurora, SolarDesigntool, Helioscop?

[wwu123]

I was nearly fully shaded by my neighbor's trees from in December from 11 am onwards because of the low sun in the sky, until they topped off the trees more than a year ago. I'm still shaded after 2 pm, but gaining full sun from 11 am-2 pm only increased my December production from 150 kwh to 230 kwh or so, a 50% percentage jump, but 80 kwh increase is only a tiny fraction of my annual 5,000 kwh production.

The reason it had so little difference is that December is already the lowest production month, but also there's more fog, morning cloud cover, storms that month. So shading doesn't matter if it's cloudy or rainy. So optimizing layout for December sunlight after 2 pm sounds unproductive, vs optimizing for peak summer production. On top of that, if your POCO pricing is TOU and seasonal, a kwh of off-peak winter production is worth far less than a kwh of peak summer production.

[DC1]

Just a DYI guy with my 2 bits. Solar Shading app for $16.00 has helped me to lay out my 8.4 system. And you can go to Univ. of Oregon SRML and make a Elevation and Azimuth angle gauge to map out the suns path for the full year. Probably not as accurate as a Solar Path finder but a lot cheaper for a one time use.

[DC1]

Forgot to add. I tried to buy a used S. P. F. or rent one but could not find either.