Announcement

Collapse
No announcement yet.

Wire gauge advice request!

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

  • Wire gauge advice request!

    Hello!

    I'm about to install a system folks here helped me suss out last year. I'm pretty much all set except for the wiring and connectors. Here are the relevant variables at play:

    3 x 265W SunEdison MEMC-P265AMA pannels wired in series, each one contributing:
    • 44.2V (VOC)
    • 35V (VMP)
    • 8.15A Max

    35A Victron BlueSolar 150/35 MPPT charge controller
    2 x 105AH AGM batteries wired in series for 24v
    100' distance between panels and charge controller

    The big question, of course, is what gauge wire should I use to get from the panels to the controller. If I'm using the voltage drop calculator correctly, it seems that 10AGW or better works. Should I spend the extra money and go with 8AGW or even 6AGW, or are the benefits too marginal to worry about? I'm doing a direct burial and that seems to be tougher to source that in 100' continuous length for 6AGW.

    Thanks!

    Steve

  • #2
    10 AWG is more than adequate with 8 amps is 1.6%. Even 12 AWG is right at 2% and acceptable. 6 AWG is over kill and only yields 0.6%

    However I see a possible problem. You panel Voc = 44.2 volts x 3 in series = 132 volts which is over the limit for a 150 Voc controller. You have to multiply by at least 1.25 for temperature compensation for cold mornings, 1.25 x 132 = 165 volts So what Controller are you using?

    You may have to either loose a panel for 2 in series or gain a panel, for 2S2P and that would require larger wire (8 AWG) for 16 amps. There are controller that go higher in voltage than 150 volts. A midnight classic 150 will work, but a lot more controller than you need.
    Last edited by Sunking; 06-12-2018, 07:51 PM.
    MSEE, PE

    Comment


    • #3
      Thanks for the quick response. OK, 10AWG it is then. Or maybe not?

      Sadly I didn't know about the temperature adjustment before purchasing the Victron. It's purchased and opened so I'm stuck with it. However, I'm wondering if in my circumstances it matters. Here's the deal...

      This system will only get used during in the summer and a little bit in the Fall. Due to the way things are laid out and the arc of the sun this far north, the sun doesn't fully hit the panels until around 11:00 and goes out of direct contact around 15:30. Morning temps are in the mid 40s at worse during the beginning part of the season, not too much worse than that before we close up for the year. Temps by the time the sun gets fully onto the panels is rarely below 50 even in the Fall.

      With that in mind, how cold does it have to be to kick the downstream voltage to the controller over 150v? I'm wondering if by the time the sun gets to hitting the panels things will have warmed up to the point that temperature adjusted voltage will not exceed 150v.

      Your possible solutions brings up a question that I thought I could ask later, but it looks like I should ask it now as it could influence things...

      Our property is a small island that is basically a spit of land that runs east-west with a blob in the northeastern corner blocking the early morning sun. The panels are located on the south of a treeline (and accompanying vegetation) that has both southern and northern sun exposure. The sun spends roughly the first 1/3 of the day behind and oblique to the panels to the east, 1/3 of the day full on south, and 1/3rd oblique and behind to the west.

      Instead of dropping down to 2 panels or having two pairs (hybrid series and parallel) in the south, might it not be better to have one pair in series facing south and another par in series facing northwest? One set would be pretty much dark at any one time, so no worries about all four combining their full voltage potential. I could have each pair run to a junction box with 12AGW and then feed to the controller with the 10AGW I already have. A benefit of this setup is I can still collect good sun if the day starts off or ends up being overcast.

      Thanks!

      Steve
      Last edited by SolarSteve; 06-12-2018, 11:39 PM.

      Comment


      • #4
        What model Victron? Many are 100 VOC and you could not use even 2 panels in series. You would have to wire all 3 in parallel and with 24 amps at 100 feet are now looking at 6 AWG wire. Now you are starting to see and feel the effects of low voltage. You wallet is really going to feel it going from 12 AWG to 6 AWG.
        Last edited by Sunking; 06-12-2018, 11:30 PM.
        MSEE, PE

        Comment


        • #5
          I have the 150v 35A Victron. Spec sheet says this about VOC:

          150V absolute maximum coldest conditions
          145V startup and operating maximum

          For sure I can do two in series as even with the 1.25 multiplier I come out well under 150v. What's more up in the air is if realistically I'll ever exceed 150v VOC given the particular circumstances I've laid out. So how cold is "cold" for that full 25% multiplier to kick in?

          The 6AWG option is even worse given that I'm looking for a single piece 100' long that is also direct burial. That is not something I've been able to find on the regular market. Either way too short, way too long, or not direct burial.

          Steve
          Last edited by SolarSteve; 06-13-2018, 12:25 AM.

          Comment


          • #6
            OK, I did some math using a temp adjustment equation with 42F/5.5C as the low temp of the panels when first activated. Unfortunately, I don't have the specific temp adjustment coefficient for my panels as that isn't on the spec sheet I have (might be in the box, I'll check). For now I plugged in a 0.33% coefficient I saw used as an example and the result was 49.13v adjusted, which is 147.38v for the array. Adjusted for voltage drop on 100' of 10AGW I get 145.75v. That's below 150v max, but not by all that much.

            Not a really great margin for error, so I am going to have to do more homework to make sure my temperature and coefficient values are reasonable.

            Steve
            Last edited by SolarSteve; 06-13-2018, 01:52 AM.

            Comment


            • #7
              Originally posted by SolarSteve View Post
              OK, I did some math using a temp adjustment equation with 42F/5.5C as the low temp of the panels when first activated. Unfortunately, I don't have the specific temp adjustment coefficient for my panels as that isn't on the spec sheet I have (might be in the box, I'll check). For now I plugged in a 0.33% coefficient I saw used as an example and the result was 49.13v adjusted, which is 147.38v for the array. Adjusted for voltage drop on 100' of 10AGW I get 145.75v.
              No, No, No Steve. Voc = Voltage Open Circuit which means no current and thus no voltage drop. For voltage drop you use Vmp and Imp.

              The 125% number comes from NEC when you do not have the Temp Coefficient. If you do not have the controllerr yet a midnight solar classic 150 will work.
              Last edited by Sunking; 06-13-2018, 12:15 PM.
              MSEE, PE

              Comment


              • #8
                Ah, OK. Still learning the nitty gritty stuff as I've not had to dive into it before. OK, so if my input values are reasonable accurate I'm still under 150v, but not by a lot. I'm going to see about getting the coefficient from the box sometime today.

                The NEC number useless is an average rule of thumb designed for a wide array of possibilities over the course of an entire year. Correct? Therefore, it's not applicable in my situation since I'm not using it in the cold months. It's served it's purpose, though, in that it allowed you to quickly demonstrate that I have something I need to be concerned about. So big thanks to you for pointing it out to me.

                Yes, I already have the 150v controller. So that's a sunk cost and I'm therefore motivated to find a work around that doesn't involve chucking a $350 piece of equipment. I'm not trying to live off this system, just keep some beer cold on the weekends.

                My current plan is to go with what I have and see how it works out. I'll attach a direct temp sensor to the backside of the panels so I know what the temp of the panels are for real. If the nighttime temp gets down into the mid 40s I'll simply disconnect the panels, then in the morning connect again when the temp of the panels is cleared for operations.

                I still have the question about splitting panels for north and south facing, but I want to see how this works out first. Not every worry should result in a new credit card charge

                Steve

                Comment


                • #9
                  Originally posted by SolarSteve View Post
                  OK, I did some math using a temp adjustment equation with 42F/5.5C as the low temp of the panels when first activated. Unfortunately, I don't have the specific temp adjustment coefficient for my panels as that isn't on the spec sheet I have (might be in the box, I'll check). For now I plugged in a 0.33% coefficient I saw used as an example and the result was 49.13v adjusted, which is 147.38v for the array. Adjusted for voltage drop on 100' of 10AGW I get 145.75v. That's below 150v max, but not by all that much.

                  Not a really great margin for error, so I am going to have to do more homework to make sure my temperature and coefficient values are reasonable.

                  Steve
                  You can measure the open circuit voltage at a couple of different temperatures (first direct light on the panels in the morning on a cool day, right after noon on a hot day) and see if the voltage change matches your guessed-at temp coefficient. If you get that nailed down more precisely, and you are confident in the low temp being a true lower bound, then 147 V is all the margin for error you need. It's not like everything will instantly blow up if you happen to hit 150.1 V on a freakishly cold bright winter morning. You just shouldn't let it get that high as the risks of failure increase exponentially once you exceed the max rated voltage.

                  Voc is somewhat lower for lower amounts of solar insolation. For example, Canadian Solar's CS6K-280M datasheet indicates a 25 degree C Voc of about 38 V with 1000 W/m3 and about 34 V with 400 W/m3 (which is much more than you'd get in shade). If you're certain direct sun will never hit your panels on an early winter morning, do two calculations: the lower VOC at coldest temp (realistically, it's probably more like 50-100 W/m3 in the shade under blue skies), and Voc at the lowest temp you'd expect once direct sun hits the panels. Pick the highest Voc. This is not risk-free, but you can go crazy trying to eliminate all risk while designing a practical system.

                  And if the weather forecast does call for an unprecedented cold snap someday, just open your PV disconnect the night before to sleep better.

                  Comment


                  • #10
                    Originally posted by SolarSteve View Post
                    Yes, I already have the 150v controller. So that's a sunk cost and I'm therefore motivated to find a work around that doesn't involve chucking a $350 piece of equipment. I'm not trying to live off this system, just keep some beer cold on the weekends.
                    Perhaps you failed home economics and math in school? Oops, sorry public schools no longer teach math and budgets. My bad.

                    Where I live $350 buys 350 bags of ice. A bag of ice fills a cooler that last a whole weekend. 350 weekends is 7 years, and $350 is less than the cost of a 5 year battery.
                    MSEE, PE

                    Comment


                    • #11
                      Originally posted by BackwoodsEE View Post

                      You can measure the open circuit voltage at a couple of different temperatures... This is not risk-free, but you can go crazy trying to eliminate all risk while designing a practical system.
                      Excellent advice and exactly my thinking about the risk. I don't need to be 100% sure, I just need to be sure enough.

                      And if the weather forecast does call for an unprecedented cold snap someday, just open your PV disconnect the night before to sleep better.
                      Exactly. This might happen once or twice a year. With it flagged as something to be aware of, I'm fine with it.


                      Originally posted by Sunking View Post

                      Perhaps you failed home economics and math in school? Oops, sorry public schools no longer teach math and budgets. My bad.
                      Heh, no excuse for me. I went through school at a time when a simple calculator didn't fit in a pocket. We had to do math with little bits of wood containing graphite, which we then rubbed onto white or tan substances made from grinding up trees. Kids these days wouldn't believe me even if I could keep their attention for the 2 minutes it would take to explain it to them.

                      Where I live $350 buys 350 bags of ice. A bag of ice fills a cooler that last a whole weekend. 350 weekends is 7 years, and $350 is less than the cost of a 5 year battery.
                      I was kidding, of course We've done ice for decades and a propane fridge for extended stays. It gets old after a while. Plus, the battery bank has other purposes. It's just that the fridge is the only one that requires significant concern. If it was just lights and recharging allegedly necessary electronic devices, my previous system of 45w array with multiple 20Ah batteries was perfectly fine.

                      Sometimes it's not about the money spent, it's about what results from it. Er, to a point anyway.

                      Steve

                      Comment


                      • #12
                        "We've done ice for decades and a propane fridge for extended stays. It gets old after a while. Plus, the battery bank has other purposes. It's just that the fridge is the only one that requires significant concern."

                        Here in Hawaii a lot of off-gridders convert chest freezers into refrigerators because they use about 10% of the electricity of a traditional stand-up refrigerator (which uses half its power maintaining the frost-free components). You can get the temperature probe / power cycling component for the conversion on eBay or Amazon for $15-$30. It's an external device, less than 5 minutes to install. They are commonly used by homebrewers and cheese makers as well. I have one, waiting to score a good deal on a used chest freezer. Our current chest freezer is being used as a freezer. It only runs from 9-5:15 when we have sun. At about 5pm the chest freezer is about -8F, and in the morning when it turns on again it has only risen to about 12F. So in this application the freezer consumes no battery resources. The extra "energy" is stored thermally.

                        Comment


                        • #13
                          Thanks for the advice. Yup, I've looked into this quite a bit for my house (which is also off grid). I have good access to dead chest freezers and/or could build my own pretty easily. Definitely on my ToDo list for home to replace our two stand up freezers. But at our summer place... sometimes form outweighs function.

                          Our 1947 Servel refrig/freezer has been there since, er, 1947 I think. Losing it would be like losing a limb. Yes, there's a cost to saving it (I put in $1400 to covert it from gas to DC) and powering it, however in this case tradition seemed more important than saving money. Kinda balances out the rest of my life's philosophy

                          Steve

                          Comment


                          • #14
                            Update on where things are at.

                            WARNING! If you have a buddy who is an electrician and is super efficient, don't ask "hey, can you get me 100' of direct burial 6/2?" and that night decide to stick with 10/3. Because when you call up your buddy the next morning to say "I don't need it" you might get an answer of "too late, you'll have it a few hours".

                            Now that I have 100' of 6/2 direct burial wire (at a wholesale contractor price, thankfully) I'm planning on using it. This means Sunking's idea of 2 panels in series works from both a worst case VOC scenario as well as voltage drop (0.92% with 6AGW vs. 2.33% with 10AGW).

                            QUESTION -> if I go with the 2S2P setup, should I keep the batteries wired for series (24v) or go with parallel (12v)? The batteries are 105Ah AGM. All the important stuff I have can be either 12v or 24v, but 12v is certainly more flexible. Now that I'm dropping the voltage and increasing the amps, might 12v be a better choice?

                            Thanks!

                            Steve
                            Last edited by SolarSteve; 06-17-2018, 07:23 PM.

                            Comment


                            • #15
                              Maybe I'm missing something but wouldn't the 12/24 question be answered by the inverter you wish to use?

                              Comment

                              Working...
                              X