X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Howdy from Colorado!

    Budding solar fanatic who has read all the books, done all the drawings, ready to install! Just moved to a new home with 0 obstructions to the South and am looking forward to a DIY install. Super science geek and love everything from reading on what space based solar research and academic systems are up to all the way to powering my phone via solar. Love the math side of solar and getting it done right.

  • #2
    Do you know what P.O.A. irradiance is and how to estimate it ? Just wonderin'

    Comment


    • #3
      Sure! You take the direct irradiance multiplied by the cosign of the angle of incidence (PoA Beam), then add THAT by the PoA for Albedo (albedo and its angle of incidence) and add that by the diffuse radiation of solar scattering in the sky (PoA Sky) and bingo-bango, done!

      Caveat: I am a huge solar nerd but by no means a competent or solid engineer on this.

      PoA Beam -- For example, my roof gets about 5.46 kwh/m2 DNI per day (from PVwatts) on avg. It is currently 8 June 17 in Colorado Springs, so the solar AoI is currently 22.91 degrees at Noon today, the COS of which is .921 giving the PoA Beam of the Sun = 5.46Kwh/m2 * cos(.921) = 5.46 kwh/m2

      PoA Ground/Roof Reflected -- I have a rubberized aluminum roof, so let's be conservative and call the albedo roughly 75% of pure aluminum (0.85) and say it is .64. My roof has a 5 degree pitch, so lets plug that into the PoA formula with my GHI of about 4.83Kw/h = 4.83Kwh * .64 * (1 - cos(5))/2) = ~ .006Kw/h/m2

      PoA Sky -- I'm going to cheat and take 75% of perfect conditions, blue sky, and from the good old Intro to Solar Radiation (https://www.newport.com/t/introducti...olar-radiation), we see that perfect diffuse sky radiation is roughly 70W/m2, so we'll call it = 52.5W/m2 = .0525kW/m2

      So, for my roof in Colorado, a total swag for PoA Irradiance would be:

      5.46Kw/m2 + .006Kw/m2 + .0525Kw/m2 = 5.52 Kwh/m2


      All that came from: https://pvpmc.sandia.gov/modeling-st...oa-irradiance/

      Comment


      • #4
        Originally posted by coronad0 View Post
        Sure! You take the direct irradiance multiplied by the cosign of the angle of incidence (PoA Beam), then add THAT by the PoA for Albedo (albedo and its angle of incidence) and add that by the diffuse radiation of solar scattering in the sky (PoA Sky) and bingo-bango, done!

        Caveat: I am a huge solar nerd but by no means a competent or solid engineer on this.

        PoA Beam -- For example, my roof gets about 5.46 kwh/m2 DNI per day (from PVwatts) on avg. It is currently 8 June 17 in Colorado Springs, so the solar AoI is currently 22.91 degrees at Noon today, the COS of which is .921 giving the PoA Beam of the Sun = 5.46Kwh/m2 * cos(.921) = 5.46 kwh/m2

        PoA Ground/Roof Reflected -- I have a rubberized aluminum roof, so let's be conservative and call the albedo roughly 75% of pure aluminum (0.85) and say it is .64. My roof has a 5 degree pitch, so lets plug that into the PoA formula with my GHI of about 4.83Kw/h = 4.83Kwh * .64 * (1 - cos(5))/2) = ~ .006Kw/h/m2

        PoA Sky -- I'm going to cheat and take 75% of perfect conditions, blue sky, and from the good old Intro to Solar Radiation (https://www.newport.com/t/introducti...olar-radiation), we see that perfect diffuse sky radiation is roughly 70W/m2, so we'll call it = 52.5W/m2 = .0525kW/m2

        So, for my roof in Colorado, a total swag for PoA Irradiance would be:

        5.46Kw/m2 + .006Kw/m2 + .0525Kw/m2 = 5.52 Kwh/m2


        All that came from: https://pvpmc.sandia.gov/modeling-st...oa-irradiance/
        Not to rain on your parade, but my question was more rhetorical than academic.
        You got most of that response wrong. See Duffie & Beckman for why.
        A few things, briefly, and as examples only of many in your reply that are incorrect:
        1.) For average daily beam POA irradiance, the daily total is calculated from the time integrated day long total of the product (instantaneous beam irradiance * cos (angle of incidence) ), not from one angle measured at (I presume) solar noon.
        2.) The albedo comes from many sources, not just directly around a receiver. View factors are necessary between reflected sources and receiver. Depending on your array tilt, and view factor between roof and array, the roof albedo may be the smallest albedo component.
        3.) Commonly, diffuse radiation under clear skies is likely to be closer to about 15-20% of the GHI. Your diffuse component looks a bit light.

        I know I'm doing a poor job of not being a wise ass here, or seeming like what you get when you cross a hooker with a computer (An f-bombing know-it-all).

        However, if you're as enthusiastic as you say (and I believe you are), I'd respectfully suggest you either get a better understanding of the info in your cited sources, or put them aside until you get the basics down better - the Duffie & Beckman source I referenced is pretty much the bible of solar energy basics for those who are serious.

        I envy you. You are about to embark on what can be a very rewarding experience. I was in a position very similar to yours about 40+ yrs. ago. Thought I knew it all. Didn't know squat and still don't know much. But, FWIW, I had a good time. It also changed my life.

        Good luck.
        Last edited by J.P.M.; 06-09-2017, 09:53 AM.

        Comment


        • #5
          Awesome ... I was taking the widest swag as possible and was 50/50 on if I was close! I can not thank you enough for pointing me in the right direction, and its what makes this place great.

          As an aside, is POA Irradiance used for most installs or is it more applicable to solar farm/power plant level calculations? I'd imagine at exceptionally high costs, I'm going to want down to the minute what my potential harvest is going to be.

          Comment


          • #6
            What do you mean by "potential harvest"? NREL's System Advisor Model is a great resource.

            All modeling requires an estimate of POA irradiance at some level, although less rigorous approaches might make that implicitly using "sun hours" or something like it.
            Last edited by sensij; 06-09-2017, 11:07 AM.
            CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

            Comment


            • #7
              My guess is that with the change in the weather patterns a lot of the historical data from pvwatts and other solar calculators will no longer be as accurate as it once was.

              Whatever estimations anyone makes concerning "solar harvest" will require a pretty big fudge factor.

              Comment


              • #8
                Originally posted by sensij View Post
                What do you mean by "potential harvest"? NREL's System Advisor Model is a great resource.

                All modeling requires an estimate of POA irradiance at some level, although less rigorous approaches might make that implicitly using "sun hours" or something like it.
                With the availability and ease of obtaining much more refined estimates of irradiance in all forms - beam, diffuse, albedo, as well as pretty detailed models to get angles to just about any desired degree of accuracy of solar geometry, and ways to estimate clearness indices at just about any time increment, the use of "sun hours ", while probably useful for back of envelope type analysis, is just as probably somewhat anachronistic given the ease, availability and thus the approx. equal facility of those more detailed methods.

                Seems to me, most folks look for or incorrectly assume accuracy or precision in their output estimates that cannot be at all justified by using most any model, much less the "sun hours" concept.

                I'd suggest to at least get away from the term "sun hours" as, IMO only, I've seen it cause a lot of confusion. Calling it by what it really is: Daily average global horizontal insolation, often and more simply abbreviated as "daily average GHI" would perhaps help avoid the confusion and mistakes that happen when folks think because the sun is up 8 hrs. per day that their array sees 8kWh/m^2 per day. At least those who are unfamiliar with the more accurate term might then be encouraged to learn a bit about resource availability.

                "Sun hours" is an anachronistic term that causes confusion and errors, is unnecessary, and needs to go.

                For reasonable and easy resource estimates that are probably fit for most residential PV design, PVWatts is about as good a it gets at this time. decent or at least appropriate rigor, easy to learn and use, and probably fit for most residential work at least for preliminary design.

                Comment


                • #9
                  Originally posted by SunEagle View Post
                  My guess is that with the change in the weather patterns a lot of the historical data from pvwatts and other solar calculators will no longer be as accurate as it once was.

                  Whatever estimations anyone makes concerning "solar harvest" will require a pretty big fudge factor.
                  That may be true eventually, but from what I've seen the last couple years, the TMY files i have used are still representing the real world within expected variance.
                  CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                  Comment


                  • #10
                    Originally posted by sensij View Post

                    That may be true eventually, but from what I've seen the last couple years, the TMY files i have used are still representing the real world within expected variance.
                    Good to know that the data is still pretty accurate.

                    Comment


                    • #11
                      Originally posted by SunEagle View Post

                      Good to know that the data is still pretty accurate.
                      Accurate is probably a relative term. If that term implies reproducibility in an absolute sense, that's not what most weather data files will give you. The data, be it TMY, or other stuff, is usually and mostly historical in nature, and BTW (dirty little secret hiding in plain site), and contrary to common belief, the TMY irradiance data is mostly from models anyway, or good eyeball observation of things like % cloud cover estimates, or Campbell-Stokes sunshine recorder data that say nothing beyond "the sun was out at this time" information, and not actual data from irradiance measuring instruments. Besides, even what's considered good data from sophisticated instruments is probably +/- several to 5 % accurate or so, and only then if kept in good calibration. Hopefully, the instrument precision is a lot better, making repeated or continuous data acquisition helpful.

                      In the end, just like the weather, irradiance data as it's used for modeling is stochastic in nature. Is it representative ? Probably/mostly. Is it reliable ? Seems to be, at least within the variability of weather and within the limits of accuracy to measure such things.

                      The data is representative of what a PV or other system might expect to see over time and not a prediction of what will actually happen. That's most of the reason PVWatts stuff is good ~ +/- 10 % for an annual output estimate over many years.

                      The idea that global warming as it may effect a change in weather in a way that will have a noticeable, much less gross effect on the available solar resource and thus R.E. device performance may well be overblown or taken as more serious than will actually happen by folks who believe everything they read and can have their opinions easily manipulated by hearing/reading the same things in a more or less continuous nature. I don't think such effects on PV system performance have shown up yet. That is, I haven't seen a reproducible smoking gun (yet).

                      The TMY data, similar to U.S. NOAA and data from other sources is updated as time goes by. For example, TMY2 is different than TMY3, with data taken from different time periods. The files that are used by PVWatts and SAM also have "Solaranywhere" data available that uses different methods to get irradiance estimates than the TMY methods.

                      If the TMY data gave results that did not seem to fit reality, it would have been declared bogus a long time ago by folks knowledgeable in such things. Such is not the case, at least not so far as I'm aware.
                      Last edited by J.P.M.; 06-09-2017, 12:58 PM.

                      Comment


                      • #12
                        Originally posted by coronad0 View Post
                        Awesome ... I was taking the widest swag as possible and was 50/50 on if I was close! I can not thank you enough for pointing me in the right direction, and its what makes this place great.

                        As an aside, is POA Irradiance used for most installs or is it more applicable to solar farm/power plant level calculations? I'd imagine at exceptionally high costs, I'm going to want down to the minute what my potential harvest is going to be.
                        When you're swinging the estimate bat, be careful not to hit the baby in the head before you throw it out with the bath water.

                        POA irradiance as usually estimated from GHI data and taken together with ways to transform GHI to POA data are essential to getting g performance estimates for most so0lar processes.

                        See other posts on this thread as to "accuracy" of performance estimates.

                        Comment


                        • #13
                          Originally posted by J.P.M. View Post

                          POA irradiance as usually estimated from GHI data and taken together with ways to transform GHI to POA data are essential to getting g performance estimates for most so0lar processes.

                          See other posts on this thread as to "accuracy" of performance estimates.
                          I guess the word "accuracy" is not very distinct. Maybe I should have said the data was "relatively consistent" to previous measurements.

                          Comment

                          Working...
                          X