Announcement

Collapse
No announcement yet.

1 year update with Sunpower system

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • 1 year update with Sunpower system

    My system has been up just over a year and I wanted to share some results. I have 16 SP 327 panels, oriented SW at 225, tilt is about 20-25 degrees, connected to a SMA6000 inverter. When plugging the numbers into pvwatts, I stated premium panels, fixed roof mount, 5% loss.

    I have sce and was on tou-D-T and the 12month bill was $240.
    Attached Files

  • #2
    So what did the system produce vs. the PVWatts estimate ?

    Comment


    • #3
      sorry I thought you could see the attached graphs, the production and estimates were around 8900 each. I need to figure out my yearly use in that period, but I think my last bill with the summary may already be tossed out.

      Comment


      • #4
        It looks like your are saying if you set pvwatts to 5% loss, the modeled production equals the actual production. That doesn't say anything in an absolute sense about the performance of the array... if the past year was sunnier or cloudier than the "typical" year used in the model, the difference could easily be 5 or 10%. In a relative sense, if you do the same kind of analysis for other nearby arrays over the same time period, it might yield more meaningful results.
        CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

        Comment


        • #5
          Unless the weather at your location for the year, or period in under analysis, including solar irradiance, temp. and wind, for the prior year matches the weather used by PVWatts exactly, - and that will not happen - actual system output cannot be compared to PVWatts long term estimates of system output as a way to estimate system performance relative to expected performance.

          Jiggling the system loss parameter so that PVWatts estimate of long term performance matches actual output for the period of analysis will perhaps provide some loose indication of how close the prior 12 months' weather may have been relative to the weather PVWatts used as input, but that's about it.

          You pretty much know the output if you monitor system performance, but probably don't know the actual input - something PVWattts or any model cannot provide.

          I also have 16-327's. To get to the 9,051 kWh my system generated for the prior 365 days, PVWatts I need to force a system loss parameter of 11.1%.

          SAM and modeling I've done get pretty close at ~ 9,100 kWh/yr. with about 4% average annual shading.

          Comment


          • #6
            SO i guess a good question is, what should the loss percentage be input as ?

            Comment


            • #7
              Originally posted by rhedayi View Post
              SO i guess a good question is, what should the loss percentage be input as ?
              And a reasonable answer might be: Whatever a decent estimate might be as you can gather from some study and some critical thinking about what each loss component term might be as listed on the PVWatts loss calc. pull down screen.

              Two ways (of many ways) to look at this: Either estimate/SWAG each term and use that result as a defined reasonable estimate of long term (several year) performance, OR, measure/record actual output and adjust the gross system loss parameter to match the PVWatts long term average estimate. Either way, if you are looking for a gauge of how well your system is doing relative to what was expected, PVWatts is not the tool to do that.

              Dividing one above method by the other may - or may not - give some anecdotal indication of how closely the weather for the period in question, including irradiance, matched the weather PVWatts used for its estimate. But given how PVWatts mostly uses synthetic and then concatenated weather data to begin with, that's not very useful.

              If you want to estimate how actual system performance compared to published data, you'll need accurate, nearby weather data, including irradiance data, and probably some way to convert horizontal irradiance to plane of array irradiance, and also accurate records of system parameters like output. That's the short description. Most folks have neither the equipment or expertize for that. Don't shoot me, I'm only the piano player.

              It looks like you are trying to estimate your system's performance for the past year. Because of the variability of weather, and the way PVWatts is designed, it's purpose and how it uses weather data, as well as the necessary simplifications it makes, it is an inappropriate tool for estimating short term performance with any sort of useful accuracy or precision more than about 10% or so - maybe more - depending on the length of time under consideration.

              Read the PVWatts help/info screens for more explanation on PVWatts intended purpose and limitations. PVWatts is a modeling tool to estimate long term (many year) output. It is not meant to be a predictor of short term performance (like a year).

              Comment


              • #8
                u guys must be in a sunny state- I put 17 SP 327's on my sons roof n our 1st year production feb 2014-jan 2015 was 7963- we are in new York- interestingly our best month was aug when we produced 989kwh- I thought that april or may would have produced the most- our best day was in april.. almost 41kwh

                Comment


                • #9
                  Originally posted by J.P.M. View Post

                  PVWatts is a modeling tool to estimate long term (many year) output. It is not meant to be a predictor of short term performance (like a year).
                  I think many installers use this to estimate your yearly output, its an estimate. Remember weather patterns can vary year to year, el nino, la nina, whatever but we also have drought years etc and the output will be more. Making this more confusing than it needs to be is not the intended point of the post. If it estimates long term (many year) output, it must mean that is an average of year after year and this could be an average year or not. Its a great estimator of ball park performance given a set of parameters, zip code and array specifics. It works fine for what it is.

                  Comment


                  • #10
                    Originally posted by paris401 View Post
                    u guys must be in a sunny state- I put 17 SP 327's on my sons roof n our 1st year production feb 2014-jan 2015 was 7963- we are in new York- interestingly our best month was aug when we produced 989kwh- I thought that april or may would have produced the most- our best day was in april.. almost 41kwh
                    I am in california, and I think JPM is also in california

                    Comment


                    • #11
                      Originally posted by rhedayi View Post

                      I think many installers use this to estimate your yearly output, its an estimate. Remember weather patterns can vary year to year, el nino, la nina, whatever but we also have drought years etc and the output will be more. Making this more confusing than it needs to be is not the intended point of the post. If it estimates long term (many year) output, it must mean that is an average of year after year and this could be an average year or not. Its a great estimator of ball park performance given a set of parameters, zip code and array specifics. It works fine for what it is.
                      Sort of off topic, but - Yes, peddlers do that, often. And in doing so they can and often do manipulate the input so the output is less than realistic by most any reasonable set of criteria. The solar ignorant, thinking the model represents reality, but being clueless about it or its purpose and limitations, can fall for the manipulation and wind up with more PV than is cost effective without knowing it. Happens all the time in my HOA. Caveat Emptor.

                      As you note, yearly or other period variability can be quite high and in any case unpredictable, making performance estimates of +/- a few % unrealistic.

                      Comment


                      • #12
                        2 year update......lots of rain this year and output was lower than expected ....$96 REFUND FROM SCE
                        Attached Files
                        Last edited by rhedayi; 05-27-2018, 11:33 PM.

                        Comment


                        • #13
                          Sort of fun to update the yearly production.....3rd year......8810 kw produced, $156 due to edison. The month of April tends to be the best month, you can see we had a $110 credit due to nice production and no a/c use.

                          Attached Files
                          Last edited by rhedayi; 05-27-2018, 11:38 PM.

                          Comment


                          • #14
                            A very important thing is demonstrated here. I think folks really need to consider AGAINST 100% production. My SCE usage for the year was 4539kw. I produced 8810 so my total usage was 13349 kw. My system produced 8810/13349.....66%. Bill over 3 years has been $300 total for 3 years due to TOU rates. Recently my neighbor had solar city come and install 22 panels on the roof. This guy has no EV, is super cheap and barely runs the a/c, has no pool (which is something I have) and has a presumably bigger system than what I have. People get ripped off on solar systems all the time. I wonder if SCE will kill the TOU plans in time....
                            Last edited by rhedayi; 05-27-2018, 11:52 PM.

                            Comment


                            • #15
                              Originally posted by rhedayi View Post
                              Sort of fun to update the yearly production.....3rd year......8810 kw produced, $156 due to edison. The month of April tends to be the best month, you can see we had a $110 credit due to nice production and no a/c use.
                              Your figures talk about KWhours, not KW. Bruce Roe

                              Comment

                              Working...
                              X