Announcement

Collapse
No announcement yet.

Electric panel with 2 mains, but not split single phase. Perplexed

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

  • Electric panel with 2 mains, but not split single phase. Perplexed

    I've got an electric panel with two mains, but the voltage across L1-L2 is 220 volts, and the voltage L1-N and L2-N 127 volts. 127 * sqrt(3) = 220, so this means the two lines are 120 degrees out of phase, not 180 degrees. Which means this is not split single phase service, but three phase service with only 2 of the 3 phases coming in to the main panel. WTF?

    So, I've got Enphase M250 microinverters that are tripping out-of-voltage errors, either due to 220 being too low (216 supposedly the threshold) or 127 being too high (130 supposedly the threshold).

    I can't use Enphase three-phase cable because I don't have 3 phases in the panel. WTF do I do?

  • #2
    The obvious fix, is get the PoCo to wire it with straight single phase power, and get the voltage down
    to the legal 120/240V limit. My string inverter could actually be set up to operate in the setup you
    described, but probably not your microinverter, unless purchased that way. Besides the voltage issues,
    a line to return monitor (which can be disabled on inverters here) would be confused.

    It seems extremely unlikely power would be set up with 2/3 of 3 phase, but not impossible. I'd
    check it on a scope or other phase measuring means. Bruce Roe

    Comment


    • #3
      I think it's near certainty it's three phase power but with only two of the phases brought into the house. How else could you explain L1-L2 = 220, and both L1-N, L2-N = 127? 127 * 1.732 = 220, so almost certainly 3 phase.

      Attached is the panel. Only two hot legs.

      What string inverter could handle this? Isn't the fact that L1-N + L2-N <> L1-L2 going to indicate to any inverter, micro or string, that this is three phase power and require 3 hots to output?
      Attached Files

      Comment


      • #4
        I use Fronius IG Plus which has 2 output connections and provision for a neutral lead
        voltage monitor. It can run on a 208V setup, 2 leads of a 3 phase, with voltage monitors
        that can be programmed to the situation. Won't help your equipment. Bruce Roe

        Comment


        • #5
          Originally posted by bcroe View Post
          I use Fronius IG Plus which has 2 output connections and provision for a neutral lead
          voltage monitor. It can run on a 208V setup, 2 leads of a 3 phase, with voltage monitors
          that can be programmed to the situation. Won't help your equipment. Bruce Roe
          I may need to go this direction. Even though Enphase can set the grid profile parameters such that 220 volts L1-L2 and 127 volts L1-N, L2-N are within the operational range for single phase, it doesn't work. The microinverters must detect three phase due to L1 and L2 being 120 degrees out of phase instead of 180? I'm not sure how, but pretty sure the microinverters have determined that it is three phase. The micros always output on two wires, but if three phase expect three coming in - A, B, C - and then the Envoy (I guess?) tell each microinverter whether to output A-B, B-C, or A-C. Because I have no C, even with the grid profile parameters set to accept 220/127 voltage, I get output on only 8 of 26 panels. Presumably these 8 are told to output on A-B, whereas the other 18 are told to output on B-C or A-C.

          Anyway, I've been reading the data specs on Fronius IG+ string inverters and not seeing anything that explicitly makes clear what you say. I hope you are correct, and assuming so, can you lead me in the direction of where I can learn more about how to set up the inverter to support running on just 2 leads of 3 phase? Also, do you know other string inverters that support this? In particular, does SolarEdge?
          Last edited by kny; 12-06-2017, 09:26 PM.

          Comment


          • #6
            Nothing to be perplexed about? You bought incompatible equipment for your electric service. Not sure what country you are in, but in the USA In cities where you have high-rise apartments it is not uncommon to have 3-phase service. In the USA 3-Phase service for residential is 208/120 wye, and commercial is 480/277 wye. So you have 208/120 and only have 2 of the 3 phases supplied. Pretty normal install. Ignore the voltage you see, it is 208/120 nominal. In the USA a utility is required to be +/- 10% and 220/127 is well within tolerance.

            You can make it work, the panels, you just need an Inverters made for Polyphase systems, You bought Inverters made for Single Phase 250/125. You need 208/120 wye Inverters.
            MSEE, PE

            Comment


            • #7
              Originally posted by kny View Post

              I may need to go this direction. Even though Enphase can set the grid profile parameters such that 220 volts L1-L2 and 127 volts L1-N, L2-N are within the operational range for single phase, it doesn't work. The microinverters must detect three phase due to L1 and L2 being 120 degrees out of phase instead of 180? I'm not sure how, but pretty sure the microinverters have determined that it is three phase. The micros always output on two wires, but if three phase expect three coming in - A, B, C - and then the Envoy (I guess?) tell each microinverter whether to output A-B, B-C, or A-C. Because I have no C, even with the grid profile parameters set to accept 220/127 voltage, I get output on only 8 of 26 panels. Presumably these 8 are told to output on A-B, whereas the other 18 are told to output on B-C or A-C.

              Anyway, I've been reading the data specs on Fronius IG+ string inverters and not seeing anything that explicitly makes clear what you say. I hope you are correct, and assuming so, can you lead me in the direction of where I can learn more about how to set up the inverter to support running on just 2 leads of 3 phase? Also, do you know other string inverters that support this? In particular, does SolarEdge?

              Feel free to contact off list - it doesn't look like this board does private messaging, so kyle@dcsolar.io
              There are likely many centralized inverters capable of 3 phase hookups, but I'm only familiar
              with the Fronius IG Plus. On pages 34 and 35 of the Operating Instructions are 6 diagrams
              of possible setups, 5 of them are using 2 load wires (per inverter) of a 3 phase system. I will
              try to scan these. Operating like this with micro inverters seems a lot less likely.

              Detection of 3 phase is possible when a 3rd wire connects to a voltage monitor, even if no
              power flows in that wire. Since your micros are not set up for this, the line to line and line
              to neutral voltages are preventing operation. Bruce Roe

              Comment


              • #8
                Originally posted by kny View Post

                I may need to go this direction. Even though Enphase can set the grid profile parameters such that 220 volts L1-L2 and 127 volts L1-N, L2-N are within the operational range for single phase, it doesn't work. The microinverters must detect three phase due to L1 and L2 being 120 degrees out of phase instead of 180? I'm not sure how, but pretty sure the microinverters have determined that it is three phase. The micros always output on two wires, but if three phase expect three coming in - A, B, C - and then the Envoy (I guess?) tell each microinverter whether to output A-B, B-C, or A-C. Because I have no C, even with the grid profile parameters set to accept 220/127 voltage, I get output on only 8 of 26 panels. Presumably these 8 are told to output on A-B, whereas the other 18 are told to output on B-C or A-C.

                Anyway, I've been reading the data specs on Fronius IG+ string inverters and not seeing anything that explicitly makes clear what you say. I hope you are correct, and assuming so, can you lead me in the direction of where I can learn more about how to set up the inverter to support running on just 2 leads of 3 phase? Also, do you know other string inverters that support this? In particular, does SolarEdge?
                I think you have some other problem going on here. The M250's should work fine across two legs of 208, when configured for it. Since you don't have access to the third leg, could the unbalanced generation be difficult for the feed to handle, and driving the voltage on those legs even higher than you've measured so far?

                SolarEdge and others can also run on two legs of 208 service, but if your effective grid impedance is too high, it won't matter.
                CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                Comment


                • #9
                  Originally posted by sensij View Post

                  I think you have some other problem going on here. The M250's should work fine across two legs of 208, when configured for it. Since you don't have access to the third leg, could the unbalanced generation be difficult for the feed to handle, and driving the voltage on those legs even higher than you've measured so far?

                  SolarEdge and others can also run on two legs of 208 service, but if your effective grid impedance is too high, it won't matter.
                  Please explain "The M250's should work fine across two legs of 208, when configured for it". Enphase support says you don't configure M250s for 240/208, that the cabling use determines that. Well, we're using 240 cabling (2 hots only - red,black), and we initially got nothing but AC Out of Voltage Range errors. Presumably this was because the 217-219 range we were getting was tripping the 216 lower limit. Enphase opened up the grid profile parameters to allow lower voltage (how low, I don't know) and then 8 of 26 M250s started working. I presumed this to be because the voltage was now in range, but that the M250s were determining it to be 3 phase (based on 120 degree rather than 180 degree shift?) and 2 out of 3 micros were attempting to output on the 3rd hot leg in the cabling, which does not exist.

                  Perhaps something is wrong in this assessment. Enphase support has not seemed terribly helpful, other than "if it's 3 phase service and you're using single phase cabling it won't work." "But, it's 3 phase service only in the 120 degree out of phase; if the L1-L2 and L1-N voltage ranges are in range, it should work as single phase, right?" Long pause and lots of uncertainty in response.

                  So, I would really love to hear how to configure M250 for two legs across 208. Does it require 3 phase engage cabling (red,black,blue) or single (red,black)?

                  Do you agree with my assessment that the M250s are somehow identifying 3 phase by a means other than the cabling used, which is why once Enphase opened up the grid profile parameters I started generating power on 8 of 26 panels?

                  I do appreciate all the assistance this forum provides.

                  Comment


                  • #10
                    No, I don't think the controllers are identifying single phase vs three phase. Each inverter is completely unaware of what the others are doing, except in their net effect on the line voltage. If 8 inverters works, they all should work. When those 8 inverters were working, did you measure line voltage?
                    CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                    Comment


                    • #11
                      Originally posted by sensij View Post
                      No, I don't think the controllers are identifying single phase vs three phase. Each inverter is completely unaware of what the others are doing, except in their net effect on the line voltage. If 8 inverters works, they all should work. When those 8 inverters were working, did you measure line voltage?
                      No, I was remote when Enphase altered the grid parameters and 8 of the 26 micros started producing. However, my read of this (https://enphase.com/sites/default/fi...ring208VAC.pdf) is that in 3-phase 1/3 of micros output on A-B, 1/3 on A-C, and 1/3 on B-C. So, while each micro may be unaware of the others, the Envoy must be telling each which of the two phases to output onto. If the micros are unaware of each other then a controller must be dictating the behavior of each.

                      So, my interpretation is that:
                      1 - something other than the enphase cabling (because I'm using single phase cabling) is telling the system that it 3-phase. Presumably this something has to be the 120 degree phase between L1 and L2, because L1-L2 and L1-N, L2-N are all within acceptable ranges. If the voltages are within range of 240 single phase and cabling does not include the 3rd blue wire for L3, I'm not sure how else the system would identify as 3-phase.

                      2 - the envoy, upon being informed by the microinverters that this is 3-phase setup, is doling out A-B, A-C, B-C responsibilities to each micro. If we could get the Envoy just to tell each micro to output on A-B and ignore C, then I'd be good. But apparently this is not possible with Enphase microinverters but is possible with some string inverters (Fronius at a minimum)


                      Or, maybe I'm flat out wrong and the microinverters do not auto-identify as 3-phase and the 8 of 26 working is a red herring. But, this would imply that even after the grid profile parameters were opened up, that L1-L2 voltages were still tripping out of range (low) or L1-N (high), but why then for just 18 of the 26? I really think somehow the M250s must identify this as a 208 Delta / 120 Wye service.
                      Last edited by kny; 12-06-2017, 10:19 PM.

                      Comment


                      • #12
                        Look at the installation manual. In a *balanced* three phase system, 1/3rd of the inverters output across each pair of hots because they are *wired* that way. There is no programming involved. Intentionally generating in an unbalanced way may not be very compatible with your connection to the grid, and causing voltage aberrations that prevent the others from activating.
                        CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                        Comment


                        • #13
                          Originally posted by sensij View Post
                          Look at the installation manual. In a *balanced* three phase system, 1/3rd of the inverters output across each pair of hots because they are *wired* that way. There is no programming involved. Intentionally generating in an unbalanced way may not be very compatible with your connection to the grid, and causing voltage aberrations that prevent the others from activating.
                          So you're saying if I looked at Enphase 3-phase cable that the plugs are different for the drops. That while the trunk cable has L1-L2-L3, that each drop has only 2 connections and these alternate L1-L2, L1-L3, L2-L3. Rather than that all drops have L1-L2-L3 and something is telling the micro which to output into. Interesting. And logical. It never occurred to me. I just assumed all drops were identical and had all of L1-L2-L3.

                          In that case, I don't see why this system would not operate with grid voltage of 220 +/- and 127 +/- both reasonably within 240v tolerance ranges. Enphase says range is 211 - 264 and 108 - 132, and they opened it up even more from there.

                          Comment


                          • #14
                            Originally posted by kny View Post
                            So, my interpretation is that:
                            1 - something other than the enphase cabling (because I'm using single phase cabling) is telling the system that it 3-phase. Presumably this something has to be the 120 degree phase between L1 and L2, because L1-L2 and L1-N, L2-N are all within acceptable ranges. If the voltages are within range of 240 single phase and cabling does not include the 3rd blue wire for L3, I'm not sure how else the system would identify as 3-phase.
                            Whatever Inverter you use does not care if it is 3-phase or not. It only has to match the voltage. Inverters can be made to work on both Single Phase or Poly Phase. However there must be a way to tell the Inverter the normal operating voltage. An Inverter does not care what the phase angles are only voltage

                            No Inverter is going to come with 3 current carrying wires. They may come with 3 wires, but only two carry current, the two that carry current connect to either L1/L2, L2/L3, or L3/L1. The third wire would be Ground which has no function other than safety. If you had all 3 phases would require 3 Inverters, one for each phase.

                            Takeaway is this. If you have Single Phase, you only need 1 inverter. If you have Poly Phase requires an Inverter for each phase, and in your case 2 Inverters. FWIW if you had a 3-phase Inverter which they do make requires requires 4 wires of L1. L2, L3,and G. Not 3 as you are thinking.
                            Last edited by Sunking; 12-06-2017, 10:50 PM.
                            MSEE, PE

                            Comment


                            • #15
                              My inverters have a voltage monitor attached to the 3rd wire, as well as across the 2 current
                              carrying connections. Satisfying all voltage monitors complicates the picture. I hope these
                              possible Fronius configurations come through. The voltage monitors aren't directly shown,
                              but show up in the setup procedure. Bruce Roe FronAC1.jpgFronAC2.jpg

                              Comment

                              Working...
                              X