Hurricane hits off grid solar installation on remote beach which previously had performance issues

2»

Comments

  • richardimorse
    richardimorse Registered Users Posts: 98 ✭✭
    They can be purchased for much less than the sustainable price, this time around I am getting 24 * 2V and not 24 * 12V
  • richardimorse
    richardimorse Registered Users Posts: 98 ✭✭
    BB. said:
    Battery #24 looks weak at 12.58 volts... There is one way to help that battery... Get some 12 volt brake lamp bulbs and wire on some alligator clips. In this case connect the bulbs to the other 3x batteries that are "higher" voltage... And this will give you ~1-2 amps of bypass current to charge the weak battery (if it is "salvageable"). You will have to monitor the voltages across the other cells to make sure you don't take battery 24 over the state of charge of the other batteries. The series clamp meter current readings are either a bit strange, or simply the level of accuracy with the equipment you have (are you zeroing the DC Clamp meter every so often? DC clamp meters have zero drift over time and temperature). There do not appear to be any "cross connected" (or ladder connects) between strings? They are all wired in series and paralleled at the 24 volt +/- bus bars? There should only be "one" current measurement per string (each battery in a single strings "sees" the exact same current). One other thing to check... You want the battery cables in each string to match overall length. If one string has an "extra long" cable run, it will have more resistance for those batteries and cause them to supply/accept less current. You want to measure the "quality" of the cables and their connections under heavy current (either supplying loads, or charging a discharged battery bank). That is when any cable/resistance issues will generally show up. Many folks will terminate absorb charging when the charging current falls below ~2% to 1% rate of charge (i.e., a 200 AH battery string, terminate a normal charging cycle at ~4 to 2 amperes). AGMs should have low "leakage current" and will draw well under 1% rate of charge when full. If your batteries ever show a charging cycle tail of >1-2% rate of charge (the batteries never fall below 1% rate of charge with normal absorb set point voltage), the batteries are getting old and need replacing soon. Any batteries that "tail" at >2% rate of charge, should be replaced right away (or taken out of service). They will run hot and can overheat if >2% is pumped through them for hours on end). There is an "absorb" that some folks (and battery mfgs) recommend for their AGM batteries. Whatever the absorb charging set point is (say 14.4 volts per 12 volt battery or 57.6 VDC on the 48 volt bus bars), hold that voltage for ~8 hours every 6 months. That is the typical "Equalization Charge" for an AGM battery bank. Your post #7 vs post #14 voltages kind of have me wondering what is going on... post #14 looks pretty good with just battery #24 looking like it can use some help... Post #7 the batteries look all over the map. What is the absorb voltage set point (or was this floating)--What was the 48 bus charging voltage from the genset? -Bill

    The inverters were disassembled and a control board swap performed, the fault followed the control board into the 2nd inverter, so it is either a faulty control board or someone messed with the configuration badly, (unlikely to cause a red fault light).  The board is reporting battery voltage too high and ac power short circuit.  SO I am planning to order a new VFX3048E control board today together with a new VFXR3048E inverter, anyone know what the difference between a VFXR3048E and a VFX3048E is, can they be stacked together.  I am not planning to do that, but worth knowing that I can if I choose to.

    Not for the inverters, running two battery banks anyone know how stacking works when the battery banks are separate, can I make two inverters slave to one master for a synchronised sine wave with 1 master distribution DB board and run 2 battery banks from them with 2 charge controllers.



  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    I think the main difference between fx and fxr is the grid interactivity.  Also, fxr needs mate3 (my fx use the simpler mate).  Don't know if the grid stuff would interfere with stacking in an off-grid application.  It might, so I'd check with OB.  

    Likewise, check with them on the split bank thing.  In theory, I think it may work, but don't know what happens if (eg) a slave bank hits LBCO.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • richardimorse
    richardimorse Registered Users Posts: 98 ✭✭
    Estragon said:
    I think the main difference between fx and fxr is the grid interactivity.  Also, fxr needs mate3 (my fx use the simpler mate).  Don't know if the grid stuff would interfere with stacking in an off-grid application.  It might, so I'd check with OB.  

    Likewise, check with them on the split bank thing.  In theory, I think it may work, but don't know what happens if (eg) a slave bank hits LBCO.
    Last day here, and I have noticed something weird on the Inverter affected by the lightening strike, it has a flashing AC input light, so looks like the AC board and the control board could have been affected, looking into the warning, it mentions frequency out of range, here is an interesting thing:

    The inverter changes its humm (audible pitch) every 5 to 10 seconds from a low frequency to a higher frequency sound, so tested frequency on the AC input which is taken from the AC output when the generator is off (for equalisation), and it stays at 50H for 10 seconds then switches to 60 Hz for 5 seconds and repeats indefinitely.

    So I am going to build a working inverter today with no errors and take the other one back to South Africa for repair and replacement, leaving the system with one inverter and generator.

    I need to purchase 2 new inverters and 24 batteries and 9 to 12 solar panels and swap comm hub 4 for communications 10.3 hub plus 2 Outback DC isolator switches (one has failed), 3 Outback AC isolator switches for new inverter, I don't see a need to swap out the POWER TWO panel for a Power Three panel, so will just mount the 3rd inverter on the wall above the other two and use 50mm conduit to plumb it into the existing OUTBACK Power TWO.
  • richardimorse
    richardimorse Registered Users Posts: 98 ✭✭
    Phase I installation completed, this involved 3 * FXR3048E Inverters in parallel stacking, replacing the two damaged inverters, switching out the mate for a mate3s and the hub 4 for a hub 10.3 still using the old battery bank and generator which is showing its age after a few weeks of intensive work, seems to leak oil quite badly, Phase II will involve new panels and the implementation of surge suppression to prevent the lightning strike repeating the problem.

    In addition the load DB board was in the AC side of the Flex Power Two, not a great idea as it introduces lots of risk to the inverters.
    This will be given its own surge protected db board and the panels given a GDFI breaker for the first time. So lightning strike has been fixed.