My Roof Faces Southeast, Now What?

[eric@psmnv]

My roof faces southeast. Therefore, I will not get the best use of sunlight. Generally speaking, how much additional panel surface area is required to compensate for panels that do not face south? Or would it be better to rig up mounts that force the panels into a south-facing orientation?

[SunEagle]

I would say the angle of the panels is more important then the different between South and SE. So the question is how far North do you live and at what angle are those panels.

[organic farmer]

I have seen some who advocate for a South-facing array and an East-facing array. In either case, you get the same number of hours of sunlight in a day, but power production is displaced in terms of at what time of day you produce the power.

[eric@psmnv]

I would say the angle of the panels is more important then the different between South and SE. So the question is how far North do you live and at what angle are those panels.

I live about 34 degrees north and my roof angle is 45 degrees, best I can tell.

[eric@psmnv]

I have seen some who advocate for a South-facing array and an East-facing array. In either case, you get the same number of hours of sunlight in a day, but power production is displaced in terms of at what time of day you produce the power.

Our cabin is surrounded by trees and steep, rocky terrain. The only place an array can comfortably live is on my roof, which faces SE. It's not an ideal situation, so I'm doing my best to optimize things with a single array. (I'm aware that some of the nearest trees will have to be capped.)

[eric@psmnv]

Actually, there's one other place an array could live. We have a 24' high 8x8 three-story enclosed wooden tower next to our house. The top story houses a 500 gal. water tank. We could put an array on top of the tower, but we would need at least 5 panels, so the array would hang over the edges of the roof on both sides by several feet, essentially turning the array into a giant wind sail. I'm not sure how much force the wind would exert if it blows hard. I'd hate for the whole array to get torn off the top the tower on a gusty day.

[J.P.M.]

The answer to your question lies in a user friendly model from NREL called PVWatts.

Read all the help screens a couple of times, get your array orientations and other asked for parameters as close to those of your array(s) as close as possible, use a 10 % system loss parameter rather than the 14 % default the model uses and make a few runs. Then, change the orientations of the array tilt and azimuth in the model and see the differences in the model's output.

Comparing the differences in output as f(array orientation) will give you the answer to your question about additional panel surface.

Usually, and since the game is mostly about cost effectiveness, when different orientations require more/fewer changes in fixations (mountings), practical, code, economic and other considerations come into play.

An array that's optimally oriented for production per m^2 of array that requires a lot of mounting and other costs, is not the best choice in terms of what it costs to construct that array.

[eric@psmnv]

The answer to your question lies in a user friendly model from NREL called PVWatts.

Read all the help screens a couple of times, get your array orientations and other asked for parameters as close to those of your array(s) as close as possible, use a 10 % system loss parameter rather than the 14 % default the model uses and make a few runs. Then, change the orientations of the array tilt and azimuth in the model and see the differences in the model's output.

Comparing the differences in output as f(array orientation) will give you the answer to your question about additional panel surface.

Usually, and since the game is mostly about cost effectiveness, when different orientations require more/fewer changes in fixations (mountings), practical, code, economic and other considerations come into play.

An array that's optimally oriented for production per m^2 of array that requires a lot of mounting and other costs, is not the best choice in terms of what it costs to construct that array.

Thanks, JPM, that's good information and I will check out the PVWatts resource. Thankfully I have no code issues to worry about as we are located in a mountainous area far from any city limits, and state/county officials never come up here. You can build pretty much whatever you want and nobody cares. The main issues for me are efficiency and ease of installation. If I can lay the panels flat on the roof and get enough energy, that's preferable even if I have to add a few panels. On the other hand, if I have to build custom mounts to get the angle right, then I might as well raise one side to also make the panels point south.

[J.P.M.]

Thanks, JPM, that's good information and I will check out the PVWatts resource. Thankfully I have no code issues to worry about as we are located in a mountainous area far from any city limits, and state/county officials never come up here. You can build pretty much whatever you want and nobody cares. The main issues for me are efficiency and ease of installation. If I can lay the panels flat on the roof and get enough energy, that's preferable even if I have to add a few panels. On the other hand, if I have to build custom mounts to get the angle right, then I might as well raise one side to also make the panels point south.

You're welcome.

I'd make sure you are not under some jurisdiction. Cheap insurance. Besides not being the safest way to go, not following building codes and/or ignoring the AHJ for something that is as obvious as an array is a lot chancier (but no less dangerous) than skunking, say, an electrical upgrade that falls under the jurisdiction of codes and the AHJ.

Building codes or not, if you mount panels nonparallel to their support structure (including the ground in the case of a ground mount) or with a lot of clearance between the support structure and the array, if you don't take wind loads into consideration, you most likely will not have the array in the same county if/when you get a wind event. It will happen.

[eric@psmnv]

Building codes or not, if you mount panels nonparallel to their support structure (including the ground in the case of a ground mount) or with a lot of clearance between the support structure and the array, if you don't take wind loads into consideration, you most likely will not have the array in the same county if/when you get a wind event. It will happen.

I did think of that. My son-in-law is a mechanical engineer. He's helping to ensure that whatever we install is not at risk of encountering wind forces that are too great for it.

[J.P.M.]

I did think of that. My son-in-law is a mechanical engineer. He's helping to ensure that whatever we install is not at risk of encountering wind forces that are too great for it.

As an M.E., I'm sure he's familiar with ASCE 7-16. If not, Google "ASCE 7-16 + solar arrays".

[eric@psmnv]

As an M.E., I'm sure he's familiar with ASCE 7-16. If not, Google "ASCE 7-16 + solar arrays".

Thanks, I'll ask him about it.