X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Installing E, S, W panels vs all South

    I have seen it suggested to use east and west facing panels vs all south facing so as to increase production time and decrease inverter size.

    I did some simple calculations on pvwatts where I assumed 2.4kw of e, w, and s facing panels (8 300w panels each) and it produces more kwh than all 7.2kw of panels facing south. I guess this makes sense, it is kind of like a poor mans version of a tracking system. My question is, why wouldn't this be a better option if your mounting situation allowed it? I have a shop with straight east, west, and south facing sloped roofs that would all 3 be very easy to install panels on. I realize some of the wiring would be a little trickier to connect all together and it might require a little more mounting hardware.

    I guess the next question is, how do I calculate inverter size with this kind of setup? I would need to figure out how much production from all 24 panels would overlap to figure out max inverter size. Inverter size is a little important because my coop has a simpler application procedure for systems of less than 5kw. And they don't specify a limit for production amount, just max system size.

    The 7.2kw of panels would produce 9606 kwh in my area using the e,w, & s orientation, and 9146 kwh for all South facing. and I'm assuming the production of the first scenario would be spread out more evenly through the day.

    Any ideas are appreciated. Still very early in the planning phase.


  • #2
    Every situation is different. However, with a properly sized inverter, unless you have managed to account for some south shading, or you have some unusual weather characteristics that produce less insolation around mid day, or your elevation angles on the E and W orientations are much more favorable than the south elevation, or you made an error on input, or all or some combination of those things and perhaps others, it's unlikely an E,W,S combination of orientations will outperform an all south orientation.

    All other things being equal, it's pretty unusual for an equator facing orientation with a reasonable tilt to produce less than any combination of other azimuthal orientations.

    Example only: My location, zip, 92026, all at 20 deg. tilt, east azimuth, 1,488 kWh/yr./STC kW. West azimuth, 1,561 kWh/yr per STC kW. South azimuth, 1,756 kWh/yr. per STC kW.

    Multiple array orientations may allow for a slightly smaller inverter, and that may be of some benefit if the goal is to minimize an inverter size, but if the goal is to maximize output per installed STC kW while getting the most cost effective system, a mostly equator facing orientation is almost always the best choice.

    I'd suggest checking your PVWatts input.

    Comment


    • #3
      Originally posted by J.P.M. View Post
      Every situation is different. However, with a properly sized inverter, unless you have managed to account for some south shading, or you have some unusual weather characteristics that produce less insolation around mid day, or your elevation angles on the E and W orientations are much more favorable than the south elevation, or you made an error on input, or all or some combination of those things and perhaps others, it's unlikely an E,W,S combination of orientations will outperform an all south orientation.

      All other things being equal, it's pretty unusual for an equator facing orientation with a reasonable tilt to produce less than any combination of other azimuthal orientations.

      Example only: My location, zip, 92026, all at 20 deg. tilt, east azimuth, 1,488 kWh/yr./STC kW. West azimuth, 1,561 kWh/yr per STC kW. South azimuth, 1,756 kWh/yr. per STC kW.

      Multiple array orientations may allow for a slightly smaller inverter, and that may be of some benefit if the goal is to minimize an inverter size, but if the goal is to maximize output per installed STC kW while getting the most cost effective system, a mostly equator facing orientation is almost always the best choice.

      I'd suggest checking your PVWatts input.
      You are right. Although my south facing angle is about 4 degrees shallower than the e-w, that wasn't enough to make up the whole difference.

      I just redid my calcs and got a different answer. Their site is a little buggy and if you don't click into another field after you change the angle it doesn't upate the angle for the calcs. So I suspect my south facing only w/ 7200watts was the west facing only number since it didn't update the azimuth.

      The correct results are 9606 kwh e,w, s and 10633 kwh south only.

      Also looking at the max production times for e, w, and s in the hourly results, it would only allow me to under size the inverter by about 500w so its probably not worth the effort.

      thanks for the input!


      oh btw- Looks like your area is about 12% more productive than mine. 74083.

      Last edited by df0rster; 06-14-2018, 01:22 PM.

      Comment


      • #4
        Originally posted by df0rster View Post
        Also looking at the max production times for e, w, and s in the hourly results, it would only allow me to under size the inverter by about 500w so its probably not worth the effort.
        I don't think I'd undersize the inverter and point in multiple directions to try to save money. That is really unlikely to make sense financially with net metering.

        If you're limited to a specific size inverter (ex. 10kW inverter is all that's allowed by utility without spending a bunch more money), that's where it can be beneficial to use E-W orientations instead of all-south.

        If you're dealing with roof mount system where you have the roof you have and need some in each direction to get the kwh you want - then it can definitely make sense. (I have that situation)

        I think you should start with "how much kwh do I need each year?"
        And "How does net metering / solar feed in work in my location/my POCO?"

        Good luck

        Comment


        • #5
          Originally posted by foo1bar View Post

          I don't think I'd undersize the inverter and point in multiple directions to try to save money. That is really unlikely to make sense financially with net metering.

          If you're limited to a specific size inverter (ex. 10kW inverter is all that's allowed by utility without spending a bunch more money), that's where it can be beneficial to use E-W orientations instead of all-south.

          Good luck
          Yes I was looking at a specific size limit, it is a little easier approval and paperwork wise to stay below a 5kw inverter. Although I don't know the extent of the additional work yet.

          Looking at the manual from the coop for larger than 5kw systems, I have to use the "long form contract" and provide things like one line diagram, testing records, inspection results. The contract is about 4 pages longer with various additional requirements like this.

          My usage is about 800kwh/month so a 5kw system would provide most of my need. But below 5kw I am mainly seeing 3.8kw inverters and I'm not sure yet if they will allow 5kw equipment or has to be <5kw for the simpler process.

          Comment


          • #6
            Originally posted by df0rster View Post

            You are right. Although my south facing angle is about 4 degrees shallower than the e-w, that wasn't enough to make up the whole difference.

            I just redid my calcs and got a different answer. Their site is a little buggy and if you don't click into another field after you change the angle it doesn't upate the angle for the calcs. So I suspect my south facing only w/ 7200watts was the west facing only number since it didn't update the azimuth.

            The correct results are 9606 kwh e,w, s and 10633 kwh south only.

            Also looking at the max production times for e, w, and s in the hourly results, it would only allow me to under size the inverter by about 500w so its probably not worth the effort.

            thanks for the input!


            oh btw- Looks like your area is about 12% more productive than mine. 74083.
            You're most welcome.

            Yea, I'm here for the sun. Greater production here vs. OK is mostly f(more sunshine).

            PVWatts is a decent 1st approx. for prelim. design. Not much of this business is rocket science. Given the variation in output from the weather alone, being to the gnat's tooth doesn't make a lot of sense for most of the prelim. work.

            BTW, a lot of users have found that using 10 % for the PVWatts system loss parameter vs. the 14 % default value often gives a better match of the model's output to actual production, at least for the first, maybe 5 - 10 years of production.

            Comment


            • #7
              So, in the end what did PVWatts tell you in terms of production for each face? For me if I use East (76deg) as the baseline, then South (166deg) produced 21.9% more than East and West produced 9.3% more than East. If I split it three ways evenly, it would have resulted in a 9.4% reduction in production vs a pure South.

              Now with Time of Use taken into consideration, the West got within 5.9% of value, while it was still 11.4% lower in electricity production, so TOU rates may need to be considered for when it's produced.

              As for doing this to go with a smaller Inverter, you have to ask yourself if the ~$300 savings is worth the reduction in solar production, but spread out a bit more in time. If you're in a Net Metering situation, that savings of $300 with a smaller inverter is likely to be gone within the first year or two, but you'll have to do the math to figure it out.

              Unless there's a real need to limit the size of the inverter (i.e. going to the next inverter size means you're going to need to upgrade your panel at the cost of $2000... is a good example) then I'd suspect in most circumstance, intentionally spreading the solar over E, S, W faces when net metering, would end up yielding a lower financial return very quickly.

              Comment


              • #8
                Originally posted by TAZ427 View Post
                So, in the end what did PVWatts tell you in terms of production for each face? For me if I use East (76deg) as the baseline, then South (166deg) produced 21.9% more than East and West produced 9.3% more than East. If I split it three ways evenly, it would have resulted in a 9.4% reduction in production vs a pure South.

                Now with Time of Use taken into consideration, the West got within 5.9% of value, while it was still 11.4% lower in electricity production, so TOU rates may need to be considered for when it's produced.

                As for doing this to go with a smaller Inverter, you have to ask yourself if the ~$300 savings is worth the reduction in solar production, but spread out a bit more in time. If you're in a Net Metering situation, that savings of $300 with a smaller inverter is likely to be gone within the first year or two, but you'll have to do the math to figure it out.

                Unless there's a real need to limit the size of the inverter (i.e. going to the next inverter size means you're going to need to upgrade your panel at the cost of $2000... is a good example) then I'd suspect in most circumstance, intentionally spreading the solar over E, S, W faces when net metering, would end up yielding a lower financial return very quickly.
                I don't remember the exact numbers. something close to what you have though, I know West was close to south, and east was the lowest.

                But I am not trying to undersize the inverter for cost, it is for ease of approval per the coop's rules. I think less than 5kw and I basically don't have to do inspections, plans, etc. I need to talk to someone about that for sure though. They just call it "short" vs "long" form application and the long form has a lot more requirements as far as design and legalese, etc.

                Comment


                • #9
                  I wouldn't just look at the PoCo's rules, but what are the laws in your area, find out who the AHJ is for your area regarding Solar/Grid Tie and see what their rules and restrictions are. You may find them more restrictive than the PoCo's.

                  Comment


                  • #10
                    Originally posted by df0rster View Post
                    Looking at the manual from the coop for larger than 5kw systems, I have to use the "long form contract" and provide things like one line diagram, testing records, inspection results. The contract is about 4 pages longer with various additional requirements like this.

                    My usage is about 800kwh/month so a 5kw system would provide most of my need. But below 5kw I am mainly seeing 3.8kw inverters and I'm not sure yet if they will allow 5kw equipment or has to be <5kw for the simpler process.
                    I know Solaredge has 5kW inverters.

                    I don't know how much DC wattage they allow you to connect - Probably 1.4x.

                    Hard to say how an oversized E/W setup will compare to a all-south in that type of situation.
                    With all-south it'll clip the peak production. But with E/W it will have less production because of less optimal orientation.

                    I'd probably look to just do the "long form" in your situation... A "one-line" you'll need/want anyway.
                    I'd definitely at least get some more info on what's required if you do a 7.6kW or whatever would normally make sense for you.
                    (7.6kW is match to a 40A breaker - which in my case is what works well with the service entrance / main breaker)

                    And I agree with Taz that I'd check on what AHJ as well as POCO requires, since you need both to sign off on your install.

                    Comment


                    • #11
                      Originally posted by df0rster View Post

                      I don't remember the exact numbers. something close to what you have though, I know West was close to south, and east was the lowest.

                      But I am not trying to undersize the inverter for cost, it is for ease of approval per the coop's rules. I think less than 5kw and I basically don't have to do inspections, plans, etc. I need to talk to someone about that for sure though. They just call it "short" vs "long" form application and the long form has a lot more requirements as far as design and legalese, etc.
                      If you will be getting billed to a T.O.U. schedule and have a true NEM that does not have tiers in it, and system production is less than consumption for a year, or perhaps any billing period, you can treat the array as a revenue producer separate from what you would be paying for without an array that can be used to offset a bill. By multiplying each hour's modeled system production from PVWatts or some other (hopefully) reliable model by the POCO's per kWh rate for that hour (the time integrated hourly production-hourly rate product), summed over 8,760 hours, you'll get a modeled estimate of what any array at any orientation can do to offset a bill. Redoing that method for different orientations will show which orientation will produce the largest bill offset separate from usage, again, provided the billing tariff is true T.O.U. and does not have tiers overlaid in it. If there are tiers within that tariff, the method will not work without further manipulation. I'm working on that one as time permits.

                      As a way of example only, around here, and under the common T.O.U. tariff that many/most new (non grandfathered) residential PV owners fall under (SDG & E's schedule DR-SES), the optimum orientation is ~ 200 - 225 deg. azimuth at about a 30 deg. tilt for an unshaded array using PVWatts as a model for output. Under that tariff (SDG & E schedule DR-SES) and using current rates and times of that schedule, that optimum orientation will offset about $455 - $460 /yr in electric bill per installed STC kW (less NBC = non by bypassable charges).

                      So, a 5 STC kW array will produce ~ $450per yr./kW * 5kW = ~ $2,250/yr. in offset for an electric bill (less NBC), regardless of the size of the bill, provided the array production is less than the usage, and again, the T.O.U. tariff is true T.O.U. and does not have tiers overlaid in it. (For CA users, the small amount of non bypassable charges of ~ $0.02/kWh for each kWh sent to the user must also be accounted for and are not included in this analysis. Those charges will probably amount to somewhere in the area of $20/yr. per installed STC kW. but will vary with usage pattern. The max. NBC per installed STC kW will be ~ $30-$33.)

                      FWIW, a west facing (270 deg. az.), 30 deg. tilt will offset about $410/yr. in electric bill per installed STC kW (less NBC). I still have to get around to the east of south orientations, but I'd expect them to be a few bucks per installed STC kW less than the 270 deg. azimuth.

                      More FWIW: In my ZIP, the 180 deg. az., 20 deg. tilt (with 20 deg. tilt being a fairly common, if not ubiquitous, roof tilt), the billing offset is ~ ~450/yr. per installed STC kW (less NBC), or about the same as the optimum orientation.

                      Until the orientations get past ~ 240 deg. az. or ~ past 40 deg. tilt, the billing offsets don't change much. That is, the penalty for going off optimum is not that sever for a wider range of orientations than one might at first think, at least in/around San Diego, and I'd also suspect for other sunny climates.

                      Bottom line: Other billing from other POCOs may be similar but will probably not be identical. Unfortunately, there doesn't appear to be an easy, one size/method fits every situation. The devil is in the details and from my experience, digging into POCO rate and policy is about as pleasant as getting a root canal by going in through the rectum, but I've not found an easier way than just digging into the details. Do the POCO's rules. You will not win otherwise, but you may find ways to game the system a bit.

                      Good luck.

                      Comment


                      • #12
                        Originally posted by foo1bar View Post

                        I know Solaredge has 5kW inverters.

                        I don't know how much DC wattage they allow you to connect - Probably 1.4x.

                        Hard to say how an oversized E/W setup will compare to a all-south in that type of situation.
                        With all-south it'll clip the peak production. But with E/W it will have less production because of less optimal orientation.

                        I'd probably look to just do the "long form" in your situation... A "one-line" you'll need/want anyway.
                        I'd definitely at least get some more info on what's required if you do a 7.6kW or whatever would normally make sense for you.
                        (7.6kW is match to a 40A breaker - which in my case is what works well with the service entrance / main breaker)

                        And I agree with Taz that I'd check on what AHJ as well as POCO requires, since you need both to sign off on your install.
                        Thats probably a good plan. I probably won

                        Comment


                        • #13
                          I'll have some input, soon as I take care of some other stuff today. Bruce (E-S-W) Roe

                          Comment


                          • #14
                            Where this really comes into play is with larger systems that are up against the 120% backfeed rule and limited to a 7.7kWac inverter so as to not have to modify the service entrance. By spreading the panels around at different orientations you can achieve some really large arrays with just a 7.7kW inverter.
                            BSEE, R11, NABCEP, Chevy BoltEV, >3000kW installed

                            Comment


                            • #15
                              Originally posted by solarix View Post
                              Where this really comes into play is with larger systems that are up against the 120% backfeed rule and limited to a 7.7kWac inverter so as to not have to modify the service entrance. By spreading the panels around at different orientations you can achieve some really large arrays with just a 7.7kW inverter.
                              I believe I understand what you're saying, but there are limits. One such (perhaps illogical) limit: I could have a bigger array with a smaller inverter if I faced all the panels away from the equator. I doubt however, that such an arrangement would be the most cost effective utilization of PV.

                              I always sort of figured that after safety, one big and common goal was most bang for the buck, or as a minimum, more long term cost effectiveness than the next most cost effective alternative (usually POCO power). That usually starts with best (most productive) orientation possible for the application within known limits.

                              Poor, or less than most optimal orientations for the sake of getting some required production without considering the (negative) impacts on cost effectiveness that those less than optimal orientations can have, particularly when better or at least more cost effective orientations are available, seems contrary to the goal of most bang for the buck.

                              As for end running rules, politics and bureaucracy (and safety) can and will get in the way, but down at the root of all those rules we all like to hate, at least to my experience, there are usually sound reasons why people can't always do everything they want. Considerations for things like service entrance mods, building codes, etc., are part of preliminary design and thus play a part in the preliminary cost effectiveness considerations. Sort of like: Spend $X on service upgrades or $X+ $1 on more panels/materials/labor on a less than optimal system.

                              Comment

                              Working...
                              X