Solar Panel Tilt Bracket Ideal Azimuth Formula Strike Angle

[Badore]

I have a W facing roof with no shade on a 35 Degree angle. My latitude is 43 Degrees. I have installed my panels on a 15 Degree angle facing S. What is the actual Azimuth to use in my calculations for output? Does anyone know the Formula? I need to know what to put into the Solar Pathfinder Assistant software.

This is what I have so far: the ideal tilt for max production based on weather data for my area is 34 Degrees. From what I can tell the panels are not really facing straight South. As the angle of the roof gets steeper the panels would start facing more West, right? But since the Sun is at a lower angle in the afternoon the ideal angle for a West facing roof would be steeper than 34 Degrees.

Suppose 34 is my ideal tilt on flat roof facing South. As I tilt the roof to the right the panel starts to face away from the ideal 34 Degrees x 180 Degrees, right?

I have attached a picture of another project. I install systems under the Ontario microFIT program and knowing this would allow me to boost production on my customers roofs.

[Mike90250]

Go to (RUN, dont' walk) the PV Watts site http://mapserve3.nrel.gov/PVWatts_Viewer/index.html and enter your location and PV size & angles. You can play around and see what gives you the most. I set my array for winter harvest, summer I have extra.

[SunnyTimes]

here another simple tool, for few regions around the world.. http://valentin-software.com/xcartgo...lculation.html

[Badore]

That's not Really what I was asking about

I attached a picture of what I am planning to do. The calculations can't be done with typical software. I will need to come up with a formula.

[albert436]

Reviving this thread just out of curiosity. The neighbors have cut down some trees so I don't need to worry about it.

But for the sake of academics--how do you figure these angles?

Anyone have any ideas?

I might have asked about this on another thread, I think I referred to it as a "sideways tilt".

[posplayr]

Reviving this thread just out of curiosity. The neighbors have cut down some trees so I don't need to worry about it.

But for the sake of academics--how do you figure these angles?

Anyone have any ideas?

I might have asked about this on another thread, I think I referred to it as a "sideways tilt".

Have you ever heard of Euler Angles?

The optimal solar angle is implicitly a two angle rotation AZ then, tilt (pitch). This is not a unique solution and can be arrived at with other rotations. See a review of Euler angles. If it was unique, then that would say your final roll angle has to be something specific but indeed the solar energy captured is only a function of incidence angle in two dimensions and not three. Landscape and Portrait orientation are but two of a continuum infinite set of optimal solutions defined by a full rotation about the normal.

https://www.google.com/webhp?sourceid=chrome-instant&ion=1&espv=2&es_th=1&ie=UTF-8#q=euler%20angles

There are other rotations that will give you the same panel normal angle to a local geographic frame.You can visualize this by orienting your panel ideally in az (180 due south) and tilt (or pitch) as optimally calculated and then spin the panel around it's center in the plane of the panel. That entire closed set, a full rotation in roll, represents a continuum of an infinite number of orientations that are all also optimal.

In summary:
The optimal panel orientation is one where the panel normal is oriented south at a tilt angle nominally equal to your Latitude. One way (the conventional one) to achieve this is to perform an AZ rotation of a flat panel so it is orientated N-S ( in either landscape of portrait) and then add a tilt equal to Latitude. Any rotation about that panel normal (imagine a vector coming out perpendicular from the panel) is also an optimal panel orientation. If you do a virtual spin of the panel, you will find an an azimuth angle (or approximately a projection to the ground) that matches your house orientation.


If you follow all of this we an work on how to do it analytically if you have trouble with that.

And yes there is something else called Quaternion which would do the same as Euler angles, but I'm trying to keep this simple.

PS: most all of the terminology I have used is in common usage in aerospace, aeronautical engineering or mechatronics. Apparently not so common in solar, I apologize for any confusion.

[sensij]

But for the sake of academics--how do you figure these angles?

I use 3D modelling software to figure out the equivalent Azimuth/Tilt plane for Poa irradiance calculations created when a panel is oriented with azimuth, tilt, and "sideways tilt", as you called it. More specialized software (incorporating the solar path) is necessary to determine how to lay out the panels to avoid external and self-shading in the compound orientation.

NREL's SAM (System Advisor Model) has a 3d shade modelling engine... it is free, but takes some time to learn.

[J.P.M.]

Albert: What are you actually looking for and for what purpose ?

If you are looking for the angle of incidence between the incoming direct (beam) solar irradiance and the normal to a flat surface of any random orientation, that information can be easily calculated by hand or machine. I'd suggest the hand calc a few times to understand the concepts, but doing that and suggesting such a thing seems to be anachronistic.

If you are looking for the optimum orientation to maximize the amount of solar radiation incident over a period of time for a clear sky, that calculation involves more than angles of incidence.

If you let me know what it is you are looking for and for what purpose, I believe I can point you in some correct directions. Euler angles, or direction cosines are one way to look at it. they are useful for many things. I've used them in engineering work unrelated to anything here. But, if I understand what I think you are asking, they won't get you where you want to go.

[albert436]

This is getting really interesting, thanks everyone for the comments.

It will take me a while to get my mind around some of it.

JPM, here's what I'm trying to figure out, or actually was trying to figure out would be more accurate, but now my curiosity has been piqued.

My mom's house had some shading issues which were resolved last week when several giant trees at a neighbors house were cut down.

Prior to that I was looking at the options for panel placement. The roof of her garage has a lot of room facing in an ENE direction, which is not particularly optimal. The roof has a fairly standard pitch for a California ranch. I was wondering whether tilting the panels "sideways" toward the south would give me a more favorable angle. If so, how much of a tilt would be optimal, and how much better would it be ? Does tilting them sideways change their direction to SSW ? And how much etc etc. For some reason this has got me slightly fascinated.

Hope that clarifies things.

[albert436]

There are other rotations that will give you the same panel normal angle to a local geographic frame.You can visualize this by orienting your panel ideally in az (180 due south) and tilt (or pitch) as optimally calculated and then spin the panel around it's center in the plane of the panel. That entire closed set, a full rotation in roll, represents a continuum of an infinite number of orientations that are all also optimal.

I totally get this.

More specialized software (incorporating the solar path) is necessary to determine how to lay out the panels to avoid external and self-shading in the compound orientation.

I'm was thinking that because the angle doesn't have to be very much, like in the range of 22 degrees (?), that the rows of panels might not have to be spaced too far apart. That's presuming they are placed in landscape orientation. But would we lose some sun in the afternoon sun just due to the angle of the roof itself, compared to if they were on a flat roof or a roof that actually was facing SSE ?

Inquiring minds want to know !

[posplayr]

I totally get this.



I'm was thinking that because the angle doesn't have to be very much, like in the range of 22 degrees (?), that the rows of panels might not have to be spaced too far apart. That's presuming they are placed in landscape orientation. But would we lose some sun in the afternoon sun just due to the angle of the roof itself, compared to if they were on a flat roof or a roof that actually was facing SSE ?

Inquiring minds want to know !

The issue becomes how high above the roof you can mount the panel or how strong your mounting structure is for additional leverage a high mounted array would have. If the panel is it is high enough above the roof there is no roof interference limitations.

With the method I described, you can translate any optimal program fixed angle solution by a roll orientation change to new panel orientation whose horizontal projection matches your roof orientation. You have to look at how much height you will need to mount that panel, but before imposing a height limit it will have the same inertial point angle as the unconstrained optimal panel angle.

Matching the roof horizontal projection will give you the largest contiguous dimensions for a side (of the array) and so there will be no shading of panels on that plane. If it is too tall you will have to start making compromises to limit the angles or break up the panel allowing for spacing (lower panel density). If you only have one space(between panels) it is not as good as no spaces but it is better than any other arrangement.

[albert436]

Headed off to the dentist in a few minutes and will try to get my mind around all of that later.

But basically, for aesthetic reasons I would think it best to keep one edge of the panels flush to the roof surface, as shown in the photo posted at the beginning of this thread:

(Don't you think this would look better, and would it be be better able to be securely fastened?)

[J.P.M.]

Albert: Sounds like the shading is no longer an issue. With that assumption:

1.) It seems what you are looking for, at least at this time, is en estimate of what the annual production enhancement might be between a panel orientation which is the same as the roof, and some other orientation(s) using some mounting/tilting scheme(s) that utilizes slightly different racking and mounting methods made for that purpose.

2.) On Posplayr's comments: Standard geometrical methods are the basis, in a strictly mathematical sense, for calculations dealing with a vector's orientation relative to a surface, or between two, or several vectors or between two surfaces, or other purposes.

For solar work, the position of the sun, and hence the angle of incidence (the direction) of the beam portion of the solar irradiance relative to a surface of arbitrary orientation, requires a form those calculations. But when used in the form suggested, Euler #'s or direction cosines or other methods are cumbersome, and in any case are not sufficient to get you what you need. No offense, but the tax credits will probably expire before you will be familiar enough in the use of such things, and you still won't have what you're looking for, because more is needed. That's the wrong way to go about it. Folks I know who are knowledgeable about such things and also about solar positioning use methods suited for the purpose at hand. Those methods are readily available. See Duffie & Beckman, chap. 1.

3.) Try this:

- Get a decent estimate of the orientation of the surface your proposed array will be mounted on. This means get the surface azimuth and the surface tilt. The accuracy is not SUPER critical, but as close as possible is best. GIGO. Google earth and a protractor on a computer screen will work to get an azimuth. A compass will also work, but don't forget to correct for magnetic declination. For elevation angle, projecting the roof slope to the ground and using a protractor set along that imaginary line is not a terrible 1st approx. for a tilt (elevation) angle.

- Find PVWatts on the net. READ THE HELP/INFO SCREENS. Try a few runs go get familiar with how it works and some of the flexibility/limits of model. If you are more than 10 miles from thje coast use Miramar #'s. if close to the coast use Lindbergh or Carlsberg. Remember the results are long term estimates of output, not predictions, and certainly not for short term guesses.

After you get your feet wet, for your starting orientation, use the roof azimuth and elevation angles you've measured. Also, set the system losses # at about 10%. Long story on that. For now, I suggest just set it at 10%. Consider that your "base case". Set the system size for now at 1kW. For a 1st SWAG, set the electricity cost at $0.25/kWh.

- Get the monthly output and "energy value" (savings) savings and save the results somewhere.

- Next, rerun PVWatts, this time changing the orientation ( the surface azimuth and tilt) to what you might think is a possible or wanted orientation using tilted racking or other means to change the position (orientation) of the panels.

- Get the monthly output and savings for that new orientation and save the results in the same place you saved the first set of results.

- Compare the two. Dividing one output by the other will give you a % enhancement. Subtract one energy value from the other for savings or additional cost.

- As a 1st approx., choose any system size (in kW) and multiply that system size (in kW) by the PVWatts output for an estimate of that system size output and savings.

4.) If you are considering a "sawtooth" or staggered row arrangement, PVWatts won't give reasonable results. Reason: PVWatts does not account for shading and rows of panels will shade the rows of panels to the north, often called self shading.

For self shading you'll need other software. SAM, which is sort of like PVWatts on steroids will do a fair job. However, without some fair amount of background on the subject of solar energy, that can cause confusion and lead to erroneous conclusions.

I encourage learning, but for right now, it seems you may have more immediate and pressing concerns.

5.) As a practical matter around here (San Diego), the additional distance between rows required to avoid self shading by rows in front (to the south in our case) that is required usually makes sawtooth or staggered row arrangements impractical and not as good a performer as you might think. Even if there is some improvement in annual production, the extra cost in engineering, materials and labor is usually not worth it when compared to the relatively small increase in electricity cost savings.

Take what you want of the above. Scrap the rest.

[posplayr]

2.) On Posplayr's comments: Standard geometrical methods are the basis, in a strictly mathematical sense, for calculations dealing with a vector's orientation relative to a surface, or between two, or several vectors or between two surfaces, or other purposes.

For solar work, the position of the sun, and hence the angle of incidence (the direction) of the beam portion of the solar irradiance relative to a surface of arbitrary orientation, requires a form those calculations. But when used in the form suggested, Euler #'s or direction cosines or other methods are cumbersome, and in any case are not sufficient to get you what you need. No offense, but the tax credits will probably expire before you will be familiar enough in the use of such things, and you still won't have what you're looking for, because more is needed. That's the wrong way to go about it. Folks I know who are knowledgeable about such things and also about solar positioning use methods suited for the purpose at hand. Those methods are readily available. See Duffie & Beckman, chap. 1.

.

JPM you clearly did not comprehend what I posted; I'll give you the benefit of the doubt you did not even read it.

[J.P.M.]

JPM you clearly did not comprehend what I posted; I'll give you the benefit of the doubt you did not even read it.

Assume what you want. I believe I understand what you posted, and in addition, and also in my professional opinion, I believe I know more about what you wrote as it pertains to solar energy resource assessment, and for that matter to engineering in general and mechanical engineering in particular, than you do.

Your description of what's required is, IMO, cryptic, technically clumsy to use for the application being discussed, and shows your ignorance of available and long standing technology that I've known about, studied and helped advance over the last 40+years to the extent your embarrassing yourself to not only me, but also to other folks who know something of the subject. Your written demeanor reminds me of the type of communication I'd get from someone who's in a subject way over their head.

Furthermore, your descriptions are, IMO, incomplete in describing what's required to the point of being misleading. I suspect however, that's done out of ignorance. Even so, that's unprofessional. Similar to the last dust up we (you and I) had over thermal time constants where you showed off your ignorance of heat transfer in particularly stunning ways, I'm of the opinion and suspicion you're in over your head on these currently being discussed aspects of solar technology and of engineering in general as it pertains to solar energy. Looks to me like the latest example of a little knowledge being a dangerous thing.

FWIW, I usually refrain from comments on subjects I'm ignorant about in the hope I'll not lead someone astray and embarrass myself in the process.

To the extent you continue making statements that are incorrect and misleading, and as a result perhaps causing readers to reach incorrect conclusions and act on them, I will comment on such statements.

Other than that, I have no intention of getting into an infantile like pissing match of the type this forum recently experienced as it's a waste of my time, hampers communication, and is harmful to the subject that's the sign over the door, which subject caused me to change careers and take up engineering as a profession about 40 or so years ago.

As the probable aggrieved party in this exchange, I'll leave the last word/comment to you.