Am I Crazy?

[J.P.M.]

I will use PVWATTS and give it a try. I don't see what my SAVINGS will be. Need your help. So it spits out a column of numbers (see attached jpg) called ENERGY VALUE for a 12 month period listing Jan-Dec. What are these numbers? the $ I will spend on electricity or the $ I will save on electricity. Then what about 20 years? All I want or anyone else is HOW MUCH WILL I SAVE OVER 20 YEARS? JPM, Please help me understand the results of your critically acclaimed PVWATTS. thanks

I input $2.8/watt as my DIY system I am looking at is a 10 panel 250w/panel system (2.5kw) costing roughly $7000: $7000/2.5kw = $2.8/watt

[ATTACH=CONFIG]4573[/ATTACH]

Congrads. You just had liftoff.

The answer to your first question is in the PVWatts documentation. A lot of other answers to questions you will likely come up with are there for the reading and learning.

Go to PVWatts and hit the help screen.

Read the input. All of it.

On the economic questions:

How much you will save over a period of time is a problem in life cycle costing, a part of something called process economics. Two common goals of such an analysis are to estimate how much a proposed course of action (in this case, getting solar) will cost, usually in "today's" $$'s vs. how much it will save over X # of years, or to compare different ways to invest money by making estimates about the future with the usual goal of selecting (estimating) the most cost effective option, thus maximizing the return on investment.

With respect to how much you will save over 20 years: As much as we all like one word or one sentence simple plug/chug answers, how much something saves or costs in the long run, or how much one may save by any particular course of action is a function of many inputs and one's general situation - in the case of how much a solar electric system will save or, in an analysis of comparison of alternatives whether after, say, 20 yrs. you would have been better off investing the financial resources in real estate, the stock market, or whatever, is a function of many inputs.

A FEW of those inputs are: How your POCO billing structure works, how much energy you use, whether you buy/lease/ppa, cash/finance, interest rates, terms of any loan, tax considerations, resale value, estimates of future inflation, energy rate inflation, How long the analysis, etc,etc, with as many variables as the user thinks important or as few as necessary or as the patience of the user can tolerate. Same for the alternative investments with different parameters (such as estimates of stock market returns, for example).

The cost or economic analysis part of solar knowledgeability is some of what I was alluding to in my epistle of last nite. Some things just take some work. I can't do a whole lot for you there. You've got to do the heavy lifting.

That many folks choose not to spend a few hours to understand some relatively easy economic facts of life to help make a better choice in the solar go/no go decision, or to optimize a go solar choice and perhaps save thousands while at the same time bitching and crying the blues with every over-dramatized POCO rate increase rather than doing something (anything) that looks like proactive attempts at reducing their profligate energy use is something that, as hard as I try, escapes my attempts to understand it using common sense and logic. I guess that may be where my problem lies.

So, Datacruncher, as I suggested previously, get a copy of Duffie & Beckman. Add to that advice read/study the chapter on solar process economics. That's a good start to finding answers to your questions.

[J.P.M.]

I will use PVWATTS and give it a try. I don't see what my SAVINGS will be. Need your help. So it spits out a column of numbers (see attached jpg) called ENERGY VALUE for a 12 month period listing Jan-Dec. What are these numbers? the $ I will spend on electricity or the $ I will save on electricity. Then what about 20 years? All I want or anyone else is HOW MUCH WILL I SAVE OVER 20 YEARS? JPM, Please help me understand the results of your critically acclaimed PVWATTS. thanks

I input $2.8/watt as my DIY system I am looking at is a 10 panel 250w/panel system (2.5kw) costing roughly $7000: $7000/2.5kw = $2.8/watt

[ATTACH=CONFIG]4573[/ATTACH]

Post understood. Fixing my A/C just now. Sort of physician heal thyself time. Need 24 <= 24 hrs.

J.P.M.

[Datacruncher]

I appreciate what you are saying about back of envelope. I use that method myself a fair amount. That method however, still requires sound logic and some knowledge of the task at hand if the desired and expected outcome is to empower one to know what and in what ways things are figured out.


Why not take advantage of what a lot of knowledgeable people have already done ? Your way reinvents the wheel and it's coming out square. I'm only the guy telling you it's not as round as you think.

I'm of the opinion that most folks do not want to get terribly involved with as much number crunching as you and I have done. If so, PVWatts seems a logical choice for those folks, perhaps their only choice given their need to avoid confusion.
You, who have a shot at knowing what your doing, or at least a direction, got, as you state "analysis paralysis" and still got the method and logic wrong. Most folks don't know or care much how to get a # - they just want an answer and - as you seem to imply, a path to saving money. For the sake of reading some intro screens and about 20 min. of their time, PVWatts will puke out something they may be able to understand, probably for less time and effort than your faulty, or at least very questionable methods will involve, and perhaps be closer to reality. Or, they can use your scheme, spend time and if frustrated enough, give up, or perhaps worse, get it wrong the next time around. Which is worse: being ignorant and knowing it, or having incomplete and incorrect knowledge and imparting it to others ?

I appreciate your # crunching. Been there - done that. A lot. Honest. I'd urge you to continue. But I gotta tell ya': The # crunching I've seen so far leads me to think your spinning your wheels a bit.

As for being forced to go by past utility rate increases because the future isn't here yet, that's always been the case. But, it's not absolutely necessary, may not be the best thing to do, and may well lead to bad outcomes if it's the sole input. I'd suggest rather than being forced to do that by itself, one can also look around and see which way the wind is blowing with respect to legislation as it affects energy policy and rate structuring , technical advances and, on a personal level, what one wants and what future plans hold. Besides, while POCO rates tend to go up some in the long run, they do, believe it or not, go down once in awhile, as well as very large increases which can skew things for several years, further throwing future predictions based on the past out the window.

At the end of the day, the future is a big crap shoot. Relying on the past is no better, and maybe not as good as using other, or several, methods. It's certainly not the only method of crystal balling it.

I will use PVWATTS and give it a try. I don't see what my SAVINGS will be. Need your help. So it spits out a column of numbers (see attached jpg) called ENERGY VALUE for a 12 month period listing Jan-Dec. What are these numbers? the $ I will spend on electricity or the $ I will save on electricity. Then what about 20 years? All I want or anyone else is HOW MUCH WILL I SAVE OVER 20 YEARS? JPM, Please help me understand the results of your critically acclaimed PVWATTS. thanks

I input $2.8/watt as my DIY system I am looking at is a 10 panel 250w/panel system (2.5kw) costing roughly $7000: $7000/2.5kw = $2.8/watt

pvwatts_results_DIY.jpg

[J.P.M.]

JPM, back of envelope formulas are the best for ones that are able to use them, it empowers one to know what and how things are figured out. The PVWatts may be a better calculator but its all done behind back doors and no one knows how the answer was derived. Granted most will want to use that method seeing how its confusing to figure #of panels = Sys_size/(watts per panel). I certainly gave myself analysis paralysis by going through all this, and feel most get lost in the basic reason that we are all here, to save MONEY, and that formula at first seems basic [SAVINGS = (cost without Solar)-(cost with solar) ] but quickly becomes very involved. with the biggest unknown being what rates will do in the future. The only thing to go on is past rate increases, and that has been very difficult for me to find. The utilities say its on-line but its soo convoluted and diced up into multiple little pieces that its not attainable. I have managed to put together a good rate history for SCE from 2005-2014 but wont post it on this thread as its not the proper place for it.

I appreciate what you are saying about back of envelope. I use that method myself a fair amount. That method however, still requires sound logic and some knowledge of the task at hand if the desired and expected outcome is to empower one to know what and in what ways things are figured out. Respectfully, I think the way you get to a reasonably close # when compared to what a bunch of canned programs call an estimate is more luck than sound analysis. Rather than empowering one to know how things are figured out, your method uses a lot of questionable numbers out of nowhere in ways that do not make a lot of sense, and faulty assumptions that, at least in my experience, often and usually can lead to a false sense of knowledge about what is going on. Ignorance of a subject leads to faulty logic. Faulty logic leads to faulty empowerment.

All of the methods PVWatts and other tools use are not done behind back or closed doors. Believe me, it's all open literature stuff available for the looking. It ain't magic. I wrote stuff long before PVWatts and didn't use anything but openly available literature. FWIW, my stuff is pretty close to PVWatts, SAM, and TRNSYS which commonly give close but somewhat different results, depending on the application. It takes a little more number crunching than you have done, but not too much more, and in the end, it's not that much more involved than you have done. It's just done in different ways. Doing it using analytically derived, academically vetted, peer reviewed and corrected methods, verified by testing are what the canned and spreadsheet stuff are doing. It can be done by hand if you choose. About 8,760 times per iteration. I've found a few times is usually sufficient, but necessary.

Why not take advantage of what a lot of knowledgeable people have already done ? Your way reinvents the wheel and it's coming out square. I'm only the guy telling you it's not as round as you think.

I'm of the opinion that most folks do not want to get terribly involved with as much number crunching as you and I have done. If so, PVWatts seems a logical choice for those folks, perhaps their only choice given their need to avoid confusion.
You, who have a shot at knowing what your doing, or at least a direction, got, as you state "analysis paralysis" and still got the method and logic wrong. Most folks don't know or care much how to get a # - they just want an answer and - as you seem to imply, a path to saving money. For the sake of reading some intro screens and about 20 min. of their time, PVWatts will puke out something they may be able to understand, probably for less time and effort than your faulty, or at least very questionable methods will involve, and perhaps be closer to reality. Or, they can use your scheme, spend time and if frustrated enough, give up, or perhaps worse, get it wrong the next time around. Which is worse: being ignorant and knowing it, or having incomplete and incorrect knowledge and imparting it to others ?

I appreciate your # crunching. Been there - done that. A lot. Honest. I'd urge you to continue. But I gotta tell ya': The # crunching I've seen so far leads me to think your spinning your wheels a bit.

Without trying to sound condescending, if you want to become technically informed, I'd suggest if you look around some in the literature. With some, but not too much rigor, you'll understand where I'm coming from. Start with Duffie & Beckman. It's to Solar Energy what Thomas is to Calculus or Resnick & Halliday is to Physics.

As for being forced to go by past utility rate increases because the future isn't here yet, that's always been the case. But, it's not absolutely necessary, may not be the best thing to do, and may well lead to bad outcomes if it's the sole input. I'd suggest rather than being forced to do that by itself, one can also look around and see which way the wind is blowing with respect to legislation as it affects energy policy and rate structuring , technical advances and, on a personal level, what one wants and what future plans hold. Besides, while POCO rates tend to go up some in the long run, they do, believe it or not, go down once in awhile, as well as very large increases which can skew things for several years, further throwing future predictions based on the past out the window.

At the end of the day, the future is a big crap shoot. Relying on the past is no better, and maybe not as good as using other, or several, methods. It's certainly not the only method of crystal balling it.

The rub with predicting the future of anything is that to do so tends to feel good because it leads to a false sense of safety and comfort. There's a certain comfortable simplicity in extrapolating the future from the past. About the best that can be hoped for is a higher probability of a predicted outcome based on knowledge and understanding of the driving inputs - but no where near a guarantee of anything.

Sort of like the difference between weather and climate - Climate's what you expect, weather (like the future) is what you get.

Rate structures from the past are on the net. I have all the CPUC rate sheets for SDG & E going back to 1997. All of them. I've seen similar for SCE and PG & E but don't need them for my stuff. I do not know if they are all there, but I suspect they are.

Not everyone who tells you what you want to hear is your friend. Not everyone who tells you what may be unpleasant or seems insulting is your enemy.

[Datacruncher]

back of envelope formulas

1.) You are correct. I had brain flatulence about the load. 13,500 kWh/yr. it is.

2.) I'm still confused as to why all the back of the envelope methods being used here. Get the roof slope and azimuth, run PVWatts with a .84 or so derate factor and stick a fork in it. It usually gives better, more accurate and faster long term estimates than using methods that are somewhat irrelevant and #'s of questionable meaning. Read the FAQ's and help screens before you start for limitations and info.

3.) Not to sound condescending, but while there are lots of threads that discuss cost effectiveness, due to my lack of articulation, the subject is a lot broader than I discuss in a thread.

JPM, back of envelope formulas are the best for ones that are able to use them, it empowers one to know what and how things are figured out. The PVWatts may be a better calculator but its all done behind back doors and no one knows how the answer was derived. Granted most will want to use that method seeing how its confusing to figure #of panels = Sys_size/(watts per panel). I certainly gave myself analysis paralysis by going through all this, and feel most get lost in the basic reason that we are all here, to save MONEY, and that formula at first seems basic [SAVINGS = (cost without Solar)-(cost with solar) ] but quickly becomes very involved. with the biggest unknown being what rates will do in the future. The only thing to go on is past rate increases, and that has been very difficult for me to find. The utilities say its on-line but its soo convoluted and diced up into multiple little pieces that its not attainable. I have managed to put together a good rate history for SCE from 2005-2014 but wont post it on this thread as its not the proper place for it.

[J.P.M.]

JPM, thats what was given, 13500 kwh/year is what the original thread was asking to produce, so he already sized what he wanted to produce. So I take it you did not find any errors in the basic formula calculations then? correct?

"may not be Cost effective offsetting 100% of load " - I am in agreement there. Is there another thread that discusses that? I've got some calculations for discussion on that.

1.) You are correct. I had brain flatulence about the load. 13,500 kWh/yr. it is.

2.) I'm still confused as to why all the back of the envelope methods being used here. Get the roof slope and azimuth, run PVWatts with a .84 or so derate factor and stick a fork in it. It usually gives better, more accurate and faster long term estimates than using methods that are somewhat irrelevant and #'s of questionable meaning. Read the FAQ's and help screens before you start for limitations and info.

3.) Not to sound condescending, but while there are lots of threads that discuss cost effectiveness, due to my lack of articulation, the subject is a lot broader than I discuss in a thread.

[Datacruncher]

Before you start with sizing estimates, most folks start by knowing/determining how much electricity they want to produce. Also, many are of the opinion it is not a slam dunk that offsetting 100% of an electric load is cost effective.

Go to and use PVWatts with a .84 correction factor. Before you do however, read the info, FAQ's and other stuff associated with it. Respectfully, it looks like you're not on firm knowledge ground. It'll be faster, easier and more accurate with PVWatts. Don't shoot me, I'm only the messenger.

JPM, thats what was given, 13500 kwh/year is what the original thread was asking to produce, so he already sized what he wanted to produce. So I take it you did not find any errors in the basic formula calculations then? correct?

"may not be Cost effective offsetting 100% of load " - I am in agreement there. Is there another thread that discusses that? I've got some calculations for discussion on that.

[J.P.M.]

I have been looking at a DIY system vs leases for a month now n have learned much in the process. I am always open to corrections

System Size = (# of panels) * (kw/panel) = kw

Estimated Production = (sys size * 5.5 solar hours/day * 365 days) - Loss = kWh/year

5.5 is the solar hours/day for my area. Typical pitch is 22-26 degrees for my area.

I have found through quotes that LOSS is 25-30%. (over 5 diff quotes) This is from solar contractors. I thought it is high but after seeing multiple quotes n reverse calculations, this is the number I'm seeing.

So then; 13500 kwh * (1.3) = 17550 kwh annual production before 30% loss

Sys size = 17550/(5.5 * 365) = 8.742 kw system. = 8742 watts.

1). 8742 watts/ 250w per panel = 35 panels. 35x250w= 8.75 kw system

2). 8742 watts/ 305w per panel = 29 panels. 29x305w = 8.85kw system

8.75 * 5.5 * 365 = 17565 kwh/year , - 30% loss (5269) = 12295 kwh/year est production

8.85 * 5.5 * 365 = 17766 kwh/year, -30% loss (5330) = 12436 kwh/year est production.

U now have the equations n can play with the what ifs.

Before you start with sizing estimates, most folks start by knowing/determining how much electricity they want to produce. Also, many are of the opinion it is not a slam dunk that offsetting 100% of an electric load is cost effective.

Go to and use PVWatts with a .84 correction factor. Before you do however, read the info, FAQ's and other stuff associated with it. Respectfully, it looks like you're not on firm knowledge ground. It'll be faster, easier and more accurate with PVWatts. Don't shoot me, I'm only the messenger.

[Datacruncher]

Formula to calculate panels n production

I have been looking at a DIY system vs leases for a month now n have learned much in the process. I am always open to corrections

System Size = (# of panels) * (kw/panel) = kw

Estimated Production = (sys size * 5.5 solar hours/day * 365 days) - Loss = kWh/year

5.5 is the solar hours/day for my area. Typical pitch is 22-26 degrees for my area.

I have found through quotes that LOSS is 25-30%. (over 5 diff quotes) This is from solar contractors. I thought it is high but after seeing multiple quotes n reverse calculations, this is the number I'm seeing.

So then; 13500 kwh * (1.3) = 17550 kwh annual production before 30% loss

Sys size = 17550/(5.5 * 365) = 8.742 kw system. = 8742 watts.

1). 8742 watts/ 250w per panel = 35 panels. 35x250w= 8.75 kw system

2). 8742 watts/ 305w per panel = 29 panels. 29x305w = 8.85kw system

8.75 * 5.5 * 365 = 17565 kwh/year , - 30% loss (5269) = 12295 kwh/year est production

8.85 * 5.5 * 365 = 17766 kwh/year, -30% loss (5330) = 12436 kwh/year est production.

U now have the equations n can play with the what ifs.

[J.P.M.]

Ok here is what I think I understand now based on the great responses:

13,500 A/C annual kwh (Desired output)
1.6 DC annual Kwh (effective panel output per rated system watt)

8,437 DC watt annual system size requirement

250W DC panel yields 33 required panels

305W DC panel yields 27 required panels

Now the second part of my question is still stuck in my head... if the inverter max AC output is 250W how do I know the real impact of switching from a lower panel output to a higher? Thanks folks!

It sounds like you are (assuming) using micro inverters. That is certainly a choice, but may or may not be the wisest choice. Again respectfully, more reading and study will help answer your question and unstick your head.

[Ian S]

Ok here is what I think I understand now based on the great responses:

13,500 A/C annual kwh (Desired output)
1.6 DC annual Kwh (effective panel output per rated system watt)

8,437 DC watt annual system size requirement

250W DC panel yields 33 required panels

305W DC panel yields 27 required panels

Now the second part of my question is still stuck in my head... if the inverter max AC output is 250W how do I know the real impact of switching from a lower panel output to a higher? Thanks folks!

You rarely will see full rated output from a panel. The power rating is for ideal conditions that will rarely be seen in real life. That said, you may still see "clipping" where the panel will try to output a higher power than the inverter can handle. The inverter will in that case simply output the maximum it can. That could result in the loss of a tiny percentage of what your system is capable of producing. In the majority of cases, that won't likely amount to enough to offset the additional cost of the higher power inverter but it's very hard to predict that beforehand.

[sandiegolloyd]

starting to sink in !

Ok here is what I think I understand now based on the great responses:

13,500 A/C annual kwh (Desired output)
1.6 DC annual Kwh (effective panel output per rated system watt)

8,437 DC watt annual system size requirement

250W DC panel yields 33 required panels

305W DC panel yields 27 required panels

Now the second part of my question is still stuck in my head... if the inverter max AC output is 250W how do I know the real impact of switching from a lower panel output to a higher? Thanks folks!

[J.P.M.]

Yes I got the 5 hours from the site that you suggested. Did you happen to have a better formula for calculating the number of panels based on using a different panel output even though the inverter remains the same? Thanks!

Around here (N.C. San Diego) for reasonably well oriented, reputable equipment that is installed by a reputable vendor, you can expect about 1.5 to 1.7 kWkrs./yr. per nameplate (D.C) Watt of solar panel, pretty much regardless of panel manufacturer.

So, using 1.6 kWhr./yr. per Watt installed: 13,500/1.6 = 8.4 kW system size, +/- some, first cut. Your method got you close, but the logic was a bit off. I'd Respectfully suggest you spend some time w/ basics of solar and then use PVWatts - scrap the default rerate factor and use something like .82 to .86, depending on your shade situation.

[Ian S]

Yes I got the 5 hours from the site that you suggested. Did you happen to have a better formula for calculating the number of panels based on using a different panel output even though the inverter remains the same? Thanks!

"5 hours" is a rather meaningless number. Use PV Watts to do an actual calculation that will give you expected monthly production over the entire year. PV Watts uses averages that account for weather variables as well as physical layout. Just change the derate factor to something more typical such as 0.83-86. Just vary the DC power number to wind up with what you want in total annual production. Once you have the DC power value, then you can figure out the number of panels assuming you have the power specs for the individual panel.

[sandiegolloyd]

Sun Area

Yes I got the 5 hours from the site that you suggested. Did you happen to have a better formula for calculating the number of panels based on using a different panel output even though the inverter remains the same? Thanks!