Announcement

Collapse
No announcement yet.

Hypothetical questions : affordable grid-down preparedness options

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

  • Hypothetical questions : affordable grid-down preparedness options

    I have a 5kw Solaredge HD inverter and have been exploring affordable very basic grid-down options, just out of curiosity.

    Originally, Solaredge intended to offer "Optional: Self-Sustaining Power - produce up to 1.5kW directly from PV during grid failure" but now this option is in doubt. goo.gl/WBG6hJ
    This would have been the perfect backup solution for me. Serve basic needs - since it covers the worst case scenario with minimal up front costs.

    Looking at AC coupled Battery solution - what *if* my battery system size is so small, say 1kwh:

    1) What would happen to the inverter on a sunny day when my load is almost zero and the battery is full?
    2) Idea: set the Solaredge inverter to observe the zero export rule and connect Battery : Wattnode : Load. Would this work?

    It's clear that the Solaredge inverter could dynamically manage it's output. I can't help but think that using the zero export settings could achieve the desired results.

    Note, for the sake of simplifying the discussion and not be sidetracked by other challenges, such as safety etc. Let's assume that we are turning a grid-tie system into an off-grid system (*already physically disconnected from the grid*) in a major disaster event with an extended power outage scenario. Yes, I am aware of other options such as cheap battery inverter, ICE generators etc. The core of the question is how to leverage my PV system to get sustainable "basic" power to serve me and my neighbors' very most basic needs.
    Last edited by inspron; 09-13-2017, 01:32 PM.

  • #2
    Originally posted by inspron View Post
    I have a 5kw Solaredge HD inverter and have been exploring affordable very basic grid-down options, just out of curiosity.

    Looking at AC coupled Battery solution - what *if* my battery system size is so small, say 1kwh:

    1) What would happen to the inverter on a sunny day when my load is almost zero and the battery is full?
    2) Idea: set the Solaredge inverter to observe the zero export rule and connect Battery : Wattnode : Load. Would this work?
    OK well a few things here.

    1) Generators or cheap UPSes are the easy solution, but sounds like you know that.
    2) With SolarEdge you're sort of locked into their solution. There will be no power from the panel without the inverter. The StorEdge will allow connection to the battery, but the standard SolarEdge inverters will not. So once power goes down you don't have any real options to use your array.
    3) If you DO have a StorEdge the inverter will take care of limiting battery charging when the battery is close to full.
    4) If you have a standard SolarEdge inverter then you have to spoof it to get it to work - which means generating 240VAC with a 5+KW inverter. For example you could get a Radian GS8048 with a ~400AH 48V battery and provide the SolarEdge with 240VAC. The SE will then start up and start generating, and the Radian will accept it and charge the battery. There is no way to limit battery charging in this situation; you have to hard-disconnect the inverter (via relay or something) when the battery can no longer accept the power the SE is generating. To say this would not be cost-effective would be an understatement.

    In your case sounds like your only real option is to upgrade to a StorEdge.

    (BTW you can't do zero export rule without the SolarEdge meter; do you have that?)

    Comment


    • #3
      Originally posted by jflorey2 View Post

      (BTW you can't do zero export rule without the SolarEdge meter; do you have that?)
      I don't at the moment but I plan on installing a CCS Wattnode so ...yes? So in this case, hypothetically speaking, (off-grid) 240v from a very small battery AC coupled to the standard HD Wave inverter that is set with zero export rule. That should theoretically work with minimal costs correct? Essentially the inverter limits the output by monitoring the load to ensure zero export (since we are physically disconnected from the grid, all this mean is that the inverter won't back feed the battery - charging an already full small battery. ).

      Comment


      • #4
        Originally posted by inspron View Post
        I don't at the moment but I plan on installing a CCS Wattnode so ...yes? So in this case, hypothetically speaking, (off-grid) 240v from a very small battery AC coupled to the standard HD Wave inverter that is set with zero export rule. That should theoretically work with minimal costs correct? Essentially the inverter limits the output by monitoring the load to ensure zero export (since we are physically disconnected from the grid, all this mean is that the inverter won't back feed the battery - charging an already full small battery. ).
        I think you might have a misconception about how zero export works. It does NOT mean that if you disconnect the grid that the inverter will keep running, just outputting zero to the grid. If the grid goes out, the SE's anti-islanding detection system will detect the grid outage and shut down the system.

        For AC coupling to work, you need an entirely separate bidirectional inverter. That means an inverter that can both generate 240VAC to supply loads and accept power fed back to the inverter. You cannot "trick" a standard grid-tie inverter to keep running without another inverter to create the waveform it needs to sync to.

        Let me anticipate your next question, which is "what if I use a really small inverter on the 'grid' side of the power meter, so that the SE turns back on, and then the SE never exports to it?" A few problems there:

        1) If you do that you'll need a transfer switch on the grid side of the power meter.
        2) The speed of the zero-export control loop may be insufficient to guarantee that it NEVER exports - and if you use a standard (non-bidirectional) inverter, a few cycles may be all you need to blow it (or best case cause the SE to trip off-line.)
        3) The AC battery will also have requirements in order to stay on-line since it''s not allowed to export unless the grid is present either. Both the SE inverter and the AC battery will be expecting a very low impedance connection to the grid; the high impedance that your small inverter provides may not be stiff enough to allow normal operation. Again, best case is that both trip off-line when voltage or harmonic content go out of range.

        Also, are you certain that the SE inverter is able to use the CCS Wattnode instead of a Solaredge meter?

        Comment


        • #5
          Originally posted by jflorey2 View Post


          Also, are you certain that the SE inverter is able to use the CCS Wattnode instead of a Solaredge meter?
          The CCS Wattnode *is* the SolarEdge meter... just a branded label.
          CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

          Comment


          • #6
            Originally posted by sensij View Post
            The CCS Wattnode *is* the SolarEdge meter... just a branded label.
            Ah, that explains it, thanks.

            Comment


            • #7
              Originally posted by jflorey2 View Post
              I think you might have a misconception about how zero export works. It does NOT mean that if you disconnect the grid that the inverter will keep running, just outputting zero to the grid. If the grid goes out, the SE's anti-islanding detection system will detect the grid outage and shut down the system.
              I definitely understand this very clearly and that's why I repeatedly mentioned - in the scenario whereby I am *already physically disconnected from the grid* (so that the conversation isn't side tracked by linemen risk or transfer switch etc).

              During operation, I understand the SE inverter response time is not perfect (fast) and even if the zero export rule in the software is running, it would still export some amount of power when the load change significantly one way or the other - while being monitored by the Wattnode meter. In a grid down situation, again already physically disconnected from the grid, would a small battery and inverter work? In essence, the small battery and inverter are only responsible for creating the waveform for the SE to sync to and buffering the spikes that the SE inverter can't respond fast enough. The thinking is, if this is doable, then the battery and inverter don't have to be massive to be good enough for a true emergency.
              Last edited by inspron; 09-13-2017, 07:53 PM.

              Comment


              • #8
                Originally posted by inspron View Post
                I definitely understand this very clearly and that's why I mentioned - in the scenario whereby I am *physically disconnected from the grid*. I understand the SE inverter response time is not perfect and even if the zero export rule in the software is running, it would still export some amount of power when the load change significant one way or the other. In a grid down situation, again physically already disconnected from grid, would a small battery and inverter work? In essence the battery and inverter is only responsible for the buffering the spike.
                Well, you'd first need the inverter. Smallest bidirectional inverter I know of is the GTFX2524 (2500 watt) and its minimum battery size is 100ah. And it's a 120V inverter so you'd need an autotransformer to boost to 240 volts. So you'd be looking at about $2200 for this. And for $2200 you can just get some solar, a cheaper battery bank and a standalone inverter.


                Comment


                • #9
                  I am getting hung up in a logic loop on this.

                  To get the SolarEdge inverter to activate, the AC coupled battery (presumably through an inverter/charger) needs to create the appearance of the grid, possibly including some kind of impedance test.

                  Once the SolarEdge system activates, if PV generation potential is greater than the loads, the zero export should throttle the PV output. If PV generation is less than the loads (or load surges), the difference needs to be made up by the batteries.

                  If PV generation is greater than or equal to the loads, the inverter/charger is supplying no power, since all loads are supplied by the PV system.

                  If the inverter/charger is supplying no power, the "fake grid" is no longer active... won't the SolarEdge inverter discover that and shut down?

                  If the inverter/charger *is* supplying power, that means at some point, the battery needs to get recharged. To recharge the battery, you'd need to relax the zero export requirement, and tune it to whatever current the battery can accept. (During battery recharge, it clearly is not continuing to support the "fake grid")

                  Modifying the export power limit requires manually changing the setting in the inverter. There is a way it can be signaled (RRCR), but only in a few discrete intervals. This is a lousy way to charge a battery.

                  Edit: For what its worth, response time for the export control is about 2s, and subject to the communication time with the Wattnode meter, according to SolarEdge's documentation.

                  Last edited by sensij; 09-13-2017, 07:57 PM.
                  CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                  Comment


                  • #10
                    Originally posted by sensij View Post
                    I am getting hung up in a logic loop on this.

                    To get the SolarEdge inverter to activate, the AC coupled battery (presumably through an inverter/charger) needs to create the appearance of the grid, possibly including some kind of impedance test.

                    Once the SolarEdge system activates, if PV generation potential is greater than the loads, the zero export should throttle the PV output. If PV generation is less than the loads (or load surges), the difference needs to be made up by the batteries.

                    If PV generation is greater than or equal to the loads, the inverter/charger is supplying no power, since all loads are supplied by the PV system.
                    Yes, all this.

                    Originally posted by sensij View Post
                    If the inverter/charger is supplying no power, the "fake grid" is no longer active... won't the SolarEdge inverter discover that and shut down?
                    I am not sure on this. I am thinking that it wouldn't? But I haven't tried.

                    Originally posted by sensij View Post
                    If the inverter/charger *is* supplying power, that means at some point, the battery needs to get recharged. To recharge the battery, you'd need to relax the zero export requirement, and tune it to whatever current the battery can accept. (During battery recharge, it clearly is not continuing to support the "fake grid")
                    My thought is that during an extended grid-down situation, most of the power is only being used during the day directing from the PV. If the small battery, say 100Ah, is only handling momentary spikes, it should last a long while. Especially, if only small loads are consistently being used. The zero export rule is not truly zero, you can have a smallish value to back feed to the battery for charging purposes.

                    Originally posted by sensij View Post
                    Modifying the export power limit requires manually changing the setting in the inverter. There is a way it can be signaled (RRCR), but only in a few discrete intervals. This is a lousy way to charge a battery.
                    Agreed, I am aware of this option but I don't see how it could be used to dynamically adjust to match load, even small ones. The manual adjustment would be too slow and difficult.


                    Originally posted by sensij View Post
                    Edit: For what its worth, response time for the export control is about 2s, and subject to the communication time with the Wattnode meter, according to SolarEdge's documentation.
                    Thanks.

                    This is just a thought experiment. During grid-down situation, I can physically AC disconnect the PV system from the house and doing something like this. The key obviously is tricking the SE inverter to turn on, and secondly not drawing too much power from the very small battery pack/inverter.

                    Yes, the very expensive and obvious thing is to buy a Storeedge, obviously, that isn't a very economical solution and not very satisfying to the curious mind.
                    Last edited by inspron; 09-13-2017, 08:21 PM.

                    Comment


                    • #11
                      A small inverter and battery can provide a stable 60Hz for the inverter to lock onto, but it will likely fail the Pulse or Impedance test, where the GT inverter "tests" the grid to make sure its not a fake grid (just like you are trying to do)
                      Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
                      || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
                      || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

                      solar: http://tinyurl.com/LMR-Solar
                      gen: http://tinyurl.com/LMR-Lister

                      Comment


                      • #12
                        Originally posted by Mike90250 View Post
                        A small inverter and battery can provide a stable 60Hz for the inverter to lock onto, but it will likely fail the Pulse or Impedance test, where the GT inverter "tests" the grid to make sure its not a fake grid (just like you are trying to do)
                        Yep. So the obvious solution is a better (higher power, lower impedance) inverter. Which means you get into serious $$ for both inverters and batteries.

                        Comment

                        Working...
                        X