Announcement

Collapse
No announcement yet.

PV feeder voltage drop - California

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

    PV feeder voltage drop - California

    California electrical code requires that voltage drop does not exceed 5% total. This allows for 2% on feeders and 3% on branch circuits. I have seen countless PV systems designed with up to 5% drop. I'm guessing this is allowed because there are no branch circuit loads? Or have these systems all been in violation of the California regulations?
    TIA!

    #2
    The AC drop on PV systems is usually limited to around 2% so the inverter does not trip off on overvoltage. Sometimes PV designers list a total voltage drop number that sums the AC and DC voltage drops, why I don't know, and that can get over 5%. If you are seeing AC voltage drops of 5% then that's just poor design. They might have some really irritating intermittent inverter shutdowns.

    Comment


      #3
      That's the California Energy Code, not electrical code. It does not apply to residential installations. It is arguable that with no 'loads' involved, it does not apply to PV circuits. DC side voltage drop doesn't count (not a feeder or branch circuit). I too am a bit skeptical that you've seen 'countless' PV systems with 5% AC side voltage drop.

      Comment


        #4
        I’d have to do the math but it’s a pretty short run from the inverter to the panel (most systems) not real difficult to size the conductors for 1% on a typical residential <10kva system.

        I could be wrong though.

        Comment


          #5
          Well, here's the scenario: We're backfeeding 300A from 147kW commercial ground mount. The feeder length is a little over 1000' and it was designed with 350mcm. This is 480V 3-phase.

          When I say I've seen countless systems designed with up to 5% voltage drop, I mean that I worked with a team of designers/engineers that used 5% as a maximum on their designs but that doesn't mean that figure was actually routinely reached. Also, that could have been factoring in DC drop as well, I'm not sure.

          I'm an operations manager, not a designer or engineer. I'm following the design provided by an engineer but it was brought to my attention that the voltage drop might be too high on this 1000+ foot run because CA regs dictate no more than 2%. The inverters themselves won't start tripping until about 5% drop is reached and I believe this puts us somewhere around 3%. I'm more just concerned about compliance with regulations. Thoughts?

          Comment


            #6
            147 kW at 480V 3Ph = 177A line current. Are you sure your 300A figure is correct? My VD calculator is telling me that the VD for a 1000' 480V 3Ph 3W circuit with 300A through a #350 AL conductor is 6.56%. This would push the voltage up to 511V at the inverter if grid voltage is nominal. This shouldn't cause the inverter to shut down, but it seams a bit high. There may be additional VD between the feeder and transformer as well. If I change the current value to 177A, the voltage drop is 3.87%, which results in an inverter voltage of 499V.

            Comment


              #7
              Originally posted by PWDickerson View Post
              147 kW at 480V 3Ph = 177A line current. Are you sure your 300A figure is correct? My VD calculator is telling me that the VD for a 1000' 480V 3Ph 3W circuit with 300A through a #350 AL conductor is 6.56%. This would push the voltage up to 511V at the inverter if grid voltage is nominal. This shouldn't cause the inverter to shut down, but it seams a bit high. There may be additional VD between the feeder and transformer as well. If I change the current value to 177A, the voltage drop is 3.87%, which results in an inverter voltage of 499V.
              Your 177A figure is probably accurate in terms of actual production but we're using two SE100k SolarEdge inverters that have a max output rating of 120A each. Multiplied by 1.25 that puts it at 150A each.

              https://www.solaredge.com/sites/defa...tasheet_na.pdf

              The spec sheet shows "AC Output Voltage Minimum-Nominal-Maximum(1) (L-L) 422.5 - 480 - 529 Vac". So it looks like even the worst case scenario as far as VD will still keep the voltage within the manufacturers parameters. Not ideal but doable, yeah?

              Comment


                #8
                Section 130.5 of the 2019 CA Energy Standard covers electrical distribution system voltage drop. Source to load the combined voltage drop through feeders and branch circuits has to be 5% or less, not 2% on feeders and 3% on branch. This is not a requirement for energy production systems. There is a whole section on photovoltaic system requirements that does not mention voltage drop. Plus the industry standard design guidelines would be much less than 5% anyway. So this is not something you need to worry about.

                Worry much more about having an AC voltage drop above 2.5% or so. That can cost you big bucks when you have to replace all that conductor and conduit because the PV inverters keep shutting down under maximum production because the utility is supplying at the high end of the ANSI voltage range. In my experience 600 ft @ 480V 3ph is about as far as you can get with a reasonable conductor size even when paralleling circuits and keeping the voltage drop low.

                Comment


                  #9
                  Originally posted by MJennings View Post
                  YThe spec sheet shows "AC Output Voltage Minimum-Nominal-Maximum(1) (L-L) 422.5 - 480 - 529 Vac". So it looks like even the worst case scenario as far as VD will still keep the voltage within the manufacturers parameters. Not ideal but doable, yeah?
                  The ANSI B range upper end is 508V. That means you get at most 11V of voltage drop to work with before you hit 529V . Based on a nominal voltage of 480V that gives you 2.29% voltage drop. Even the less conservative A range upper limit is 504V allowing up to a 3.12% voltage drop. Get even close to these and you are rolling the dice on having to maybe replace all the AC conductor and conduit. Have a fat reserve fund if you want to play that game on a 1000 ft run.

                  Comment


                    #10
                    Originally posted by MJennings View Post
                    Your 177A figure is probably accurate in terms of actual production but we're using two SE100k SolarEdge inverters that have a max output rating of 120A each. Multiplied by 1.25 that puts it at 150A each.

                    https://www.solaredge.com/sites/defa...tasheet_na.pdf

                    The spec sheet shows "AC Output Voltage Minimum-Nominal-Maximum(1) (L-L) 422.5 - 480 - 529 Vac". So it looks like even the worst case scenario as far as VD will still keep the voltage within the manufacturers parameters. Not ideal but doable, yeah?
                    I can't imagine why you're oversizing your inverters so much, but that's irrelevant. There's no code requirement that would have you use 125% of the inverter output for a voltage drop calculation.

                    Comment


                      #11
                      Originally posted by pv_n00b View Post
                      The ANSI B range upper end is 508V. That means you get at most 11V of voltage drop to work with before you hit 529V . Based on a nominal voltage of 480V that gives you 2.29% voltage drop. Even the less conservative A range upper limit is 504V allowing up to a 3.12% voltage drop. Get even close to these and you are rolling the dice on having to maybe replace all the AC conductor and conduit. Have a fat reserve fund if you want to play that game on a 1000 ft run.
                      Having trouble following your math. Think you meant "at most 21 volts to work with."

                      Another way to put it is the inverter will go 10% over. How much will the utility be over and how lucky are you feeling, punk? Of course if the POCO's transformer has taps, and they are willing to change them, that would be an easy fix. My solar system would have issues if I didn't have taps (see the "my solar system thread").

                      I recall some papers on VD in PV systems (solar pro?) And the thesis was that keeping VD low is often not the most economical when you weigh conductor cost with losses. Of course keeping your inverters from dropping out from overvoltage is a separate issue.
                      Ethan Brush - East West Electric. NY, WA. MA

                      "You can't generalize"

                      Comment


                        #12
                        We want to look at the voltage window we have to work with, so we compare the maximum operating AC voltage of the inverter with the max AC voltage the utility is supposed to limit their system to in the ANSI standard. In this case if the max AC voltage for the inverter is 529V and the upper ANSI voltage is 508V then 529-508 = 21V. Looks like my other posting I got 11V by mistake. We can drop up to 21V in the conductor and the inverter will not cut out under worst case conditions. That's if everything is perfect, I would cut that down to be conservative to maybe 80% of 21V or 16.8V. And if we want a percent voltage drop then that's 16.8/480*100 or 3.5%. It's true that some people play too close to the edge and have to ask the utility if they can lower the line voltage and sometimes the utility will.

                        That article in SolarPro was interesting, but they were only looking at the financial and code aspects of conductor sizing and not any operational effects. Their idea being maybe it did not provide a return on investment to up the conductor size to get a voltage drop from 16% to 3% because the energy gain would not offset the cost of the conductor. They did not look at the system operational requirements like low and high voltage cutout.

                        Comment


                          #13
                          Originally posted by jaggedben View Post
                          I can't imagine why you're oversizing your inverters so much, but that's irrelevant. There's no code requirement that would have you use 125% of the inverter output for a voltage drop calculation.
                          I've found a lot of variety in doing the voltage drop calc. Since it's not a code issue we are left to choose values based on experience and how close we want to get to losing everything on the project. With the high DC/AC ratios of most designs these days hitting the inverter max output current can be assumed. If the inverter is oversized then we can find the maximum operational current based on the estimated array output.

                          If I were running 1000ft of underground AC conductor I would be so conservative. No way do I want to replace that run. If it's 10ft of above ground conduit I might be less conservative since it won't break the bank to pull larger conductor and I can oversize the conduit just in case.

                          Comment


                            #14
                            Originally posted by pv_n00b View Post
                            I've found a lot of variety in doing the voltage drop calc.
                            Why is that? If you know the current it 's just V=IR.

                            Comment


                              #15
                              Originally posted by ggunn View Post
                              Why is that? If you know the current it 's just V=IR.
                              I think he's saying he's seen a lot of variety in the choice of what current figure to use. However, if that includes using 125% of inverter output, well that just seems like someone not knowing what they're doing.

                              Comment

                              Working...
                              X