Announcement

Collapse
No announcement yet.

SMA Sunny Boy grid voltage requirements and a lot more...

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

  • SMA Sunny Boy grid voltage requirements and a lot more...

    I have a really crazy setup.

    My "main" system is Outback power, 2 Radian 8048 inverters, 5 old MX60 charge controllers, 32 280W Chaori panels (4 sets of 8) and 10 285W GCL panels for a total of 11,810W . It's backed up with a propane generator and 4000 gallons of propane on site.

    I decided to expand, so bought 648 of the GCL 285W panels (long story). Due to costs of hybrid system components, I figured I'd do an A/C coupled grid tie inverter. Then I though more about it and wondered why can't I just do a second system. One hybrid system with generator backup, and one direct grid tie. If power goes out, chances are my 11.8kw is enough to survive on and the grid tie panels can go silent.

    So.....

    Grid voltage 240V. Hybrid system wants to sell, it has to push the power at more than that, so it decides to go to 248V.... Grid tie system sees 248V, needs to push higher, so sells at 250V. Outback sees 250V, goes to 252V, etc, etc, etc, etc. Voltage high for the day is recorded at 262V, and I miss an estimated 25kw of power from the SMA SB6000TL-22 because it has shut down.

    I'm getting the 262V max reading from the Outback system. Today, high volt 262V, low volt 241V.

    Now the confusing part, because this doesn't happen all the time. Most days, both systems play fine together and I get full expected numbers.
    The SMA manual says nominal AC input voltage from 211 to 264V.
    The SMA fault when it is down is 1501 "Reconnection Fault Grid" which is described as "The changed country data set or the value of a parameter you have set does not correspond to the local requirements. The inverter cannot connect to the utility grid. Corrective Measures: Ensure that the country data set has been configured correctly. Check the setting of the rotary switches A and B or select the parameter Set Country standard and check the value.

    So, theoretically, the error actually has nothing to do with the voltage climb. It's only experience that tells me that it is related.

    OK. Next diagnostic step is to spend $250 on the remote control breaker set up for the Radians and hook this SMA up as an A/C coupled inverter. That will negate the A/C input voltage question.

    But, that only allows me to fix this immediate issue (assuming my assumptions are correct, lol). If I want to increase the system size exponentially (target is 80kw), then I cannot do it through A/C coupling. I will need to run a direct grid tie system beside the hybrid system, or spend $50,000 on additional equipment and batteries.

    I guess this is the question. At what point does the voltage difference not matter. That is, if the grid is sitting at 244V, and my system comes on-line and starts pushing 248V to the grid, what if my neighbor has a system? Does his end up pushing 252V, which then forces mine to push 256V, etc, etc, until one of us shuts down? There are two reasons I ask this.
    1. My area has horrible voltage control. I have had to call out the electric company twice in the last few months for voltage issues. We have seen has high as 283V. Geeze the reacted quick on that one. They were here within hours and while they won't tell me what they did, they dropped the voltage to the 250V range.
    2. I have the ability to have the grid tie system feed in about 280 linear feet away from the hybrid system. At what point does the voltage normalize when feeding back. Load will effect this, but fact is, if there's 10 houses feeding into the grid on any given street, the voltages are not going to climb into the stratosphere, and I bet none of those systems are going to shut off due to voltage. There has to be some point where voltage normalizes over distance, with load.

    OK, enough rambling. Anyone care to comment?

  • #2
    Are both systems fully permitted and approved by your power company?

    Your premise is not right... If the grid was solely 10 houses feeding and none consuming, voltage would skyrocket. When you see the voltage rise to 264V, it is basically telling you the grid impedance is too high to accept all the power you want to push.

    bcroe has experience with this, and his solution has been to turn more panels east and west, which lowers peak power but still gets energy from the panels.
    Last edited by sensij; 10-18-2017, 12:30 AM.
    CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

    Comment


    • #3
      Is this fully permitted? How is it interconnected?
      OutBack FP1 w/ CS6P-250P http://bit.ly/1Sg5VNH

      Comment


      • #4
        Originally posted by sensij View Post

        Your premise is not right... If the grid was solely 10 houses feeding and none consuming, voltage would skyrocket. When you see the voltage rise to 264V, it is basically telling you the grid impedance is too high to accept all the power you want to push.
        That was not the premise I offered. I was not referring to a closed grid of ten houses on solar. I am talking about 10 houses in a street with grid tied solar. 49,000 other houses drawing off the same grid. If my issue is caused by the inverters from each system competing with each other on voltage output, then the question is, how does this not happen between neighbors. The assumption is that at some point on the line, the load will draw the voltage back down. The question is, how far down the line does this happen? Right now, the two systems are tied to the feed at the same point, about 24 inches from the hybrid inverters and 6 feet from the grid tie inverter. So, if I re-route the output from the grid tie inverter to have to travel approx 250' before it hits the feed, will that distance be enough for load to even out the voltage.

        I guess the simple test will be to check voltage at the meter when the hybrid inverter is showing 260+V 250 feet away.

        Comment


        • #5
          Originally posted by koalamotorsport View Post

          That was not the premise I offered. I was not referring to a closed grid of ten houses on solar. I am talking about 10 houses in a street with grid tied solar. 49,000 other houses drawing off the same grid. If my issue is caused by the inverters from each system competing with each other on voltage output, then the question is, how does this not happen between neighbors. The assumption is that at some point on the line, the load will draw the voltage back down. The question is, how far down the line does this happen? Right now, the two systems are tied to the feed at the same point, about 24 inches from the hybrid inverters and 6 feet from the grid tie inverter. So, if I re-route the output from the grid tie inverter to have to travel approx 250' before it hits the feed, will that distance be enough for load to even out the voltage.

          I guess the simple test will be to check voltage at the meter when the hybrid inverter is showing 260+V 250 feet away.
          Most likely the voltage is going up because of your interconnect wiring not being of suitable size, not the grid itself getting pushed up.
          OutBack FP1 w/ CS6P-250P http://bit.ly/1Sg5VNH

          Comment


          • #6
            Originally posted by koalamotorsport View Post

            That was not the premise I offered. I was not referring to a closed grid of ten houses on solar. I am talking about 10 houses in a street with grid tied solar. 49,000 other houses drawing off the same grid. If my issue is caused by the inverters from each system competing with each other on voltage output, then the question is, how does this not happen between neighbors. The assumption is that at some point on the line, the load will draw the voltage back down. The question is, how far down the line does this happen? Right now, the two systems are tied to the feed at the same point, about 24 inches from the hybrid inverters and 6 feet from the grid tie inverter. So, if I re-route the output from the grid tie inverter to have to travel approx 250' before it hits the feed, will that distance be enough for load to even out the voltage.

            I guess the simple test will be to check voltage at the meter when the hybrid inverter is showing 260+V 250 feet away.
            There is no universal answer for "how far down the line". The grid is not a homogeneous entity to which you can tap into equally in all locations. That is one reason larger systems are usually subject to more review at the time of interconnect. You've ignored the question about how your system was approved...

            Voltage doesn't "even out" over distance. Adding more copper will reduce the impedance... it doesn't matter much whether you route that copper separately or upsize the existing feed. What size wire are you using for the combined inverter output? How much current are you asking that 250 ft run to carry? You have a lot of control over the wiring on your own property, but once you tie to the grid, you could be running into wiring/equipment/design that isn't capable of taking all the power you are trying to push.

            As you said, measuring the voltage at the meter will tell you if the drop is in your own wiring, or external to your property.
            CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

            Comment


            • #7
              Originally posted by koalamotorsport View Post
              then the question is, how does this not happen between neighbors.
              It does. If your neighbor has a huge load (a big A/C compressor) your voltage will dip a bit when it starts up. As others have noted, it's all mediated by impedance.
              The assumption is that at some point on the line, the load will draw the voltage back down. The question is, how far down the line does this happen? Right now, the two systems are tied to the feed at the same point, about 24 inches from the hybrid inverters and 6 feet from the grid tie inverter. So, if I re-route the output from the grid tie inverter to have to travel approx 250' before it hits the feed, will that distance be enough for load to even out the voltage.
              Distance doesn't "even out" voltage. Impedance does, by the usual laws of electricity (Kirchoff's Law, Ohm's Law.)


              Comment

              Working...
              X