Solar Panel Tilt Bracket Ideal Azimuth Formula Strike Angle

[posplayr]

Questions on your analysis
1. At SW orientation you are saying one half production is during peak 2-8pm hours. How would one theoretically orient to get most possible production within 2-8pm? I get that the sun moves around during the year and that different climes have differing seasonal demands, but here in SD the peak is for summertime AC use I believe. Let's forget about unique shading issues for now as well.

The only way to optimize is to have a full up model that will compute hourly production that a rate schedule can be applied to. Senseij has a model he is using to post results. I used a simplified back of an envelop analysis to guess where the optimum orientation would be without the aid of the model. This is useful to understand how to steer the full up model as I have not see anybody produce a 3D visualization of $ or KW-h production (spread across azimuth and tilt angles).

2. Following out your calls, I couldn't tell how much the savings would be if we took an example power bill with an example of typical use pattern, in terms of $$$. Just because you mentioned being surprised at the magnitude.

The models described are production models. If you want to model Net Metering you would have to estimate seasonal hourly demand and subtract it from the equivalent production model. It is basic arithmetic, but you need a tool to manage the data.

I think I already described the issue of small differences. A fixed panel is never correct except when it is as best aligned which itself is a compromise. So it is hard to make it any better unless you track. Even seasonal adjustments can only improve harvest by about 4-5% depending upon strategy. In that regard the TOU optimization is one of the most productive things you can do as you are improving your afternoon production every afternoon when rates are higher.

[albert436]

Ok yeah had a look at that other post and though I didn't read the entire thread, and so I don't know the entire context, yes.

Good question, especially for those. In lower latitudes where not much sawtooth in is needed.

Ie. Changing the panels on a SE facing roof by means of tilt, to a SW orientation in order to take advantage of TOU options.

So another application, probably a more worthwhile one, than my original question. Or for that matter would changing front NE to SE Impact TOU?

Questions on your analysis
1. At SW orientation you are saying one half production is during peak 2-8pm hours. How would one theoretically orient to get most possible production within 2-8pm? I get that the sun moves around during the year and that different climes have differing seasonal demands, but here in SD the peak is for summertime AC use I believe. Let's forget about unique shading issues for now as well.
2. Following out your calls, I couldn't tell how much the savings would be if we took an example power bill with an example of typical use pattern, in terms of $$$. Just because you mentioned being surprised at the magnitude.

Posting from iPad at the moment, may change rooms tomorrow and don't want to setup laptop yet so this is all off the top of my head.

[albert436]

When selecting an optimality criteria, TOU is often relevant for GT systems. It can shock even some of the experts.

http://www.solarpaneltalk.com/showth...l=1#post166332

Total lay a good point if you've got those orientation options available.

Another good reason this might be good for a contractor to have this in their armamentarium.

Arrived London this morning, still jet lagged, will check that other thread but yes very much aware. You've got to be able to produce during that time of day though, or not use any electricity, right? (Minimize use)

[albert436]

Right that's what I figured.

Oh, I got a big lol from that impetus-impudence thing ! ! !

I figure that pic must be from way up in the wild northland, what do you think ?

And thanks for bearing with my impudent ramblings !

[posplayr]

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?)

My comment was in reference to this.

[posplayr]

OK this I have no idea what you are talking about unless it is some special engineering meaning of the word.

Impudence ? ? ?



Please remember to use smileys when you are joking ! ! !

Sorry, spelling-checker snafo

im·pe·tus
ˈimpədəs/
noun
the force or energy with which a body moves.
"hit the booster coil before the flywheel loses all its impetus"
synonyms: momentum, propulsion, impulsion, motive force, driving force, drive, thrust; More

Of course you know engieners can't speel.

[albert436]

Nah, I just thought maybe one could collect more solar energy goodness by tilting the panels.

On further consideration however, it occurs to me that depending on how much "sawtoothing" (love that word!) is required, x amount of roof space which could be used for additional panels might be lost. Depending on the orientation and how much solar energy collection that represented as opposed to the gain from tilting the panels, etc etc.

But I'd have thought in the lower latitudes that not much tilting would be needed, and therefore not much sawtoothing woud be needed. At 32 degrees latitude aren't we looking at something like 22 degrees tilt ?

And I still don't see how tilting that little amount should be such a big expense -- small standoff on one side, longer one on the other. I haven't heard much so far to explain why the expense would be so much more.

But Mr. Posplayer, you are the first to bring up the subject of getting it signed off or certified. I think you are talking about the city/county building departments and so forth. That's an interesting point that had not occurred to me, and I know they can be stickly at times, sometimes justifiably and sometimes not. So thanks for pointing that out.

[albert436]

Bear in mind that your original impudence seemed to be aesthetics and so in none of the analysis you have see to date has used that value to justify the associated costs.

OK this I have no idea what you are talking about unless it is some special engineering meaning of the word.

Impudence ? ? ?

impudent
adjective im·pu·dent \ˈim-pyə-dənt\
: failing to show proper respect and courtesy : very rude

Please remember to use smileys when you are joking ! ! !

[posplayr]

In fact, after letting this subject percolate and rest, let me take a stab at a quick and dirty solution, minus only 1.5 key factors.

Let's just say we work out the staggering to our satisfaction, which can't be that hard to do.

Then let's pretend that we are mounting them on a roof that is oriented in just that SSE direction.

Then we'll check out how much solar energy they "should" collect at that optimal angle at which we will tilt them. I don't know if the diffusion, irradiance, etc. is built into PVWatts or not, or if the think it just has to do with the sun shining directly at the panels, but let's forget that and let's see where we stand.

So . . .

The only things left to worry about are

1 The ridge of the roof cutting off sun/shading the panels in the afternoon -- to what degree does that affect panel output. On one hand, it will probably shade them more than a roof that is actually oriented toward the SSE. On the other hand, they may not be getting that much of their irradiance from direct sunlight at that time of the day anyway, since they are oriented toward the SSE.

2. The end of the roof cutting off sun/shading the panels in the afternoon -- same question as above, I think, or perhaps not for the same reason as above.

Leaving for UK on Friday and probably won't have time to get out the 3x5 cards to make a model but believe I'm closing in on a quick and dirty solution.

The thing is, as Mr. sensi has generously pointed out to me by PM, if you look at the difference in solar energy generated (theoretically) between ENE and SSE, it is around 20% if I recall. So I'm thinking that tilting the array to SSE ought to close that gap significantly, even if we get some afternoon shading issues, and that might possibly be when they wouldn't be getting that much sunlight anyway.

JPM -- I'm thinking about how your panels are mounted a foot off your roof and how you might like them mounted a little higher so you can more easily do temp measurements. If you were ever to gain weight, that is. So I still am not clear on the extra expense to do tilting, as it seems just a variation on what you have. Or again is the expense in crunching the numbers to make sure it doesn't fly away with the roof ? Lol it just occurred to me the other day that my new GF is a structural engineer and she work it out for me. But again it is really just academic.

Part of it is I like to figure out quick and dirty ways of estimating things, when it is possible. Obviously that is not all the time. However, if one did this enough it would probably get easier and easier, I betcha.

When selecting an optimality criteria, TOU is often relevant for GT systems. It can shock even some of the experts.

http://www.solarpaneltalk.com/showthread.php?22302-45-Degree-Roof-Solar-Panels&p=166332&viewfull=1#post166332

[albert436]

Oh posplayer I didn't see your post before I posted mine, I'll have a look over it.

[albert436]

In fact, after letting this subject percolate and rest, let me take a stab at a quick and dirty solution, minus only 1.5 key factors.

Let's just say we work out the staggering to our satisfaction, which can't be that hard to do.

Then let's pretend that we are mounting them on a roof that is oriented in just that SSE direction.

Then we'll check out how much solar energy they "should" collect at that optimal angle at which we will tilt them. I don't know if the diffusion, irradiance, etc. is built into PVWatts or not, or if the think it just has to do with the sun shining directly at the panels, but let's forget that and let's see where we stand.

So . . .

The only things left to worry about are

1 The ridge of the roof cutting off sun/shading the panels in the afternoon -- to what degree does that affect panel output. On one hand, it will probably shade them more than a roof that is actually oriented toward the SSE. On the other hand, they may not be getting that much of their irradiance from direct sunlight at that time of the day anyway, since they are oriented toward the SSE.

2. The end of the roof cutting off sun/shading the panels in the afternoon -- same question as above, I think, or perhaps not for the same reason as above.

Leaving for UK on Friday and probably won't have time to get out the 3x5 cards to make a model but believe I'm closing in on a quick and dirty solution.

The thing is, as Mr. sensi has generously pointed out to me by PM, if you look at the difference in solar energy generated (theoretically) between ENE and SSE, it is around 20% if I recall. So I'm thinking that tilting the array to SSE ought to close that gap significantly, even if we get some afternoon shading issues, and that might possibly be when they wouldn't be getting that much sunlight anyway.

JPM -- I'm thinking about how your panels are mounted a foot off your roof and how you might like them mounted a little higher so you can more easily do temp measurements. If you were ever to gain weight, that is. So I still am not clear on the extra expense to do tilting, as it seems just a variation on what you have. Or again is the expense in crunching the numbers to make sure it doesn't fly away with the roof ? Lol it just occurred to me the other day that my new GF is a structural engineer and she work it out for me. But again it is really just academic.

Part of it is I like to figure out quick and dirty ways of estimating things, when it is possible. Obviously that is not all the time. However, if one did this enough it would probably get easier and easier, I betcha.

[posplayr]

Well let me put it this way.

There are many scenarios in which there are quick and dirty algorithms with which we can get a ballpark answer.

Or where by making a series of approximations and assumptions onewe can come to a reasonable range of values.

Maybe this situation isn't one that lends itself to that, or maybe it is beyond the ken of anyone on this site to do so.

As I've mentioned, I am involved in my own profession that doesn't deal with this particular type of science and so I'm probably a few steps behind some of you guys. I don't right now have the time to study up on it either.

In a nutshell what I am hearing is:

"it is too complicated to figure it out easily."

Which is also OK.

Simplifying assumptions and models are a good thing, but as you realize they are not a replacement for but are a compliment to the more full up models. In my experience from working on variety of engineering programs including very complex ones, your need both.

A common theme you will see echoed here is "how much is it really worth trying to optimize a configuration". This is an important question when you consider roof mounted systems as anything that is oblique in two dimensions (with represent to the roof) makes the mounting more complicated and as another poster has found out most installer just don't want to deal with it. There are many easier jobs to deal with that will go up much easier.

http://www.solarpaneltalk.com/showthread.php?22302-45-Degree-Roof-Solar-Panels

Having said that there are challenges of engineering to get an arbitrary mounting system to work and certified or stamped that is beyond the standardized (i.e. cookie cutter) mounts that are so widely used and have already gone through engineering reviews.

I think the original question of how to relate one optimal panel mount to another is simple and straight forward as I described. How easy or feasible it is to mount at those angles is a different matter. Bear in mind that your original impudence seemed to be aesthetics and so in none of the analysis you have see to date has used that value to justify the associated costs.

[albert436]

Well let me put it this way.

There are many scenarios in which there are quick and dirty algorithms with which we can get a ballpark answer.

Or where by making a series of approximations and assumptions onewe can come to a reasonable range of values.

Maybe this situation isn't one that lends itself to that, or maybe it is beyond the ken of anyone on this site to do so.

As I've mentioned, I am involved in my own profession that doesn't deal with this particular type of science and so I'm probably a few steps behind some of you guys. I don't right now have the time to study up on it either.

In a nutshell what I am hearing is:

"it is too complicated to figure it out easily."

Which is also OK.

Very interesting point about diffuse irradiance JPM, I haven't noticed anyone ever bringing that up before. I have often said that the sky is often so dang bright here when there's a little haze. I spent time as a lifeguard looking out over the ocean and am fairly protective of my eyes. So yes I can see that, but it had never occurred to me in terms of solar energy before.

About the up front costs -- I don't really get why a tilted array would be so much more costly. Or JPM are you suggesting the engineering/planning etc. to make sure it doesn't fly away on a windy day ? I mean the mounts themselves, would that really be such a significant part of an install ?

The thing is, I would think there are plenty of people whose homes might have a scenario where shading and orientation precludes an optimal southerly mounting. If there were an alternative way to do things that a company had, it might be good for all concerned -- or not.

OK thanks everyone for your comments !

[J.P.M.]

I don't understand what more you are looking for. There is no simple equation that generates the sideways tilt and rotation angles necessary to design the non-flush mount system. They can be determined by a model... be it 3D CAD, a miniaturized model, or detailed sketches on which the trig can be applied to determine the needed design parameters.

If you want a general analytical solution by evaluating the planes mathematically, a good start would be to read and understand the math necessary to perform operations in coordinate geometry. (Another example here)

Changed my mind - One/two or more considerations.

Albert:

Further, and adding to Sensij's post, any decent undergraduate level solar energy text will put most required solar geometry in a more useful and practical context in about 10 or so pages of text, usually starting with the eccentricity of the earth's orbit.

Also, and one of the important considerations no one seems to be writing about, much less thinking about, is that of all the irradiance being intercepted by a surface, using TMY3 Miramar data (probably the most appropriate for Albert's case), about 30+% of the total yearly P.O.A is diffuse. Under sunny skies, depending on the modeling used, and other things like the turbidity of the atmos., season, time of day, water vapor content and other stuff, a good portion of that diffuse MAY seem to have an origin "near" the sun, but a fair amount of it, again, depending on the modeling method, will come from all over the sky. under pt. cloudy skies the diffuse portion of the irradiance not originating near the sun will increase.

All that will affect the optimum tilt angle estimate. The diffuse fraction's variable angle of incidence will also impact the transmission of the diffuse portion through the glass, which also needs to be accounted for in a strict sense.

Strict geometrical relationships are useful for beam P.O.A. calcs., but they are not enough.

From what I can guess, you are after incidence angle calculations with the idea of estimating an optimum array orientation and thus annual output. If that's the case, I'll say what you are seeking is necessary but not sufficient information. You will need more and different data.

I'd forget the sawtooth, run PVWatts, and take what is probably a small penalty for parallel to the roof, flat (non sawtooth) orientation, and save time, money and trouble. Any increased savings from increased production (neither of which, BTW, are a lock - sawtooth may well induce a penalty), may well be easily exceeded by increased up front costs, as noted elsewhere.

PVWatts will take all of the above info, and other things you are still not considering, and spit out an estimate of annual output by the hour, if you want, faster and with a higher probability of being more likely to reflect reality than the path you are currently on.

PVwatts will not do shading from any source, including self shading of sawtooth arrays. Also as Sensij notes, you'll need SAM for that.

Take what you want of the above. Scrap the rest. Like most everything else around here, it's all opinion. Just that some of it's more educated by experience.

[sensij]

Purely a mental exercise just to stretch the brain.

I don't understand what more you are looking for. There is no simple equation that generates the sideways tilt and rotation angles necessary to design the non-flush mount system. They can be determined by a model... be it 3D CAD, a miniaturized model, or detailed sketches on which the trig can be applied to determine the needed design parameters.

If you want a general analytical solution by evaluating the planes mathematically, a good start would be to read and understand the math necessary to perform operations in coordinate geometry. (Another example here)