Aims inverter split phase shock hazard

BarryFields
BarryFields Registered Users Posts: 17 ✭✭
in Solar Forum News and Announcements #1
My 6000w Aims inverter has a "neutral bonding" disconnect that does nothing. As there is no neutral pass thru, the inverter becomes your "new" utility panel and should be bonded.  I moved the wire on P7 of the control board to P6 on the filter board. I moved the disconnect to the inside of the unit and left it connected. I have found no one who can rationalize un-bonding this unit.

Tagged:
· Share on Twitter

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,617 admin
    #2
    Can you supply a link to the AIMS unit you have? There appear to be several 6 kW models out there (just to make sure we are talking about the same thing):

    https://duckduckgo.com/?q=6000w+Aims+inverter&ia=web

    The transformer to the "left"? Is that part of the Aims, or something else? I am a bit confused...

    If the transformer is part of the AIMs, then the "neutral" is simply the transformer center tap and provides the 120/Neutral/White wire (and only 240 VAC input). And the left hand set of windings (240VAC only?) go where? No N+G reference?

    In general, the lifting/connecting Neutral to Ground bond "switch" is for mobile applications where you may have AC inverter 120/240 VAC power--And other times an "alternative" power source such as Grid Connected shore power, AC genset, etc....

    In AC Inverter mode--The local system N+G bonding is required for NEC/general system safety (various reasons).

    In Shore Power mode, the N+G bonding is provided by the shore power system (normal NEC wiring).

    With generators, N+G may be present in the genset (typically larger >~3.5 kWatt--I think) or floating AC output (smaller AC gensets).

    AC inverters may come with with N+G bonded or not (only true for PSW/TSW AC inverters--Never N+G bonding for almost all(?) MSW inverters which will short out if DC battery bus and AC output are both "ground bonded"). Again, smaller inverters seem to "float" Neutral. Larger inverters seem to do N+G bonding or have switched N+G bonding if there is an internal transfer switch.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • BarryFields
    BarryFields Registered Users Posts: 17 ✭✭
    edited February 2022 #3
    Aims PICOGLF60W48V240VS  This conversation only applies to split phase units.
    The left side go to the main power board.
    Whether  this unit is inverting or passing power thru, The CT on the transformer generates the neutral.
    In shore power the shore power bonding has NO way of reflecting thru the inverter because the neutral is not even connected to the unit.


    · Share on Twitter
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,485 ✭✭✭✭
    edited February 2022 #4
    This is a common issue and debate on LF inverters with single output transformer.  The issue comes down to the output of the inverter transformer becomes an auto-transformer across the mains input when AC input pass through is engaged.  

    If Inverter transformer secondary has its neutral commonly connected to grid neutral any imbalance in grid L1-N-L2 will try to be corrected by the inverter output transformer.  This results in some of inverter transformer power handling capability being used in attempt to balance grid.

    Some inverters recommend not using grid neutral.  This means pass-through is just grid 240vac and imbalanced L1-N-L2 120 vac output loads are carried by transformer just like normal inverter off grid operation.  This can limit pass through for 120 vac loads although it is rarely an issue.

    It also means the grid neutral potential can be slightly different than inverter provided neutral.  With main grid panel grounded this can create a safety issue.  Not having the inverter output auxiliary panel box commonly ground bonded to main panel would be a code violation.  This would mean aux box neutral would have to remain floating and separate from main box neutral, which creates another problem as all outlet's ground prong can have difference in potential with all aux box branch outlets neutral.

    Usually, any grid imbalance is minor and no more than one or two amps max will flow between output transformer center tap neutral and grid neutral.

    Personally, I would always common connect grid neutral to inverter output neutral for safety reasons.

    By the way, this also occurs when two 120vac inverters are used with series connection.  The L1-N-L2 grid balancing current is just passed between the two inverters via the common battery node.  It usually only lasts for a short period until inverter reacts to regulate grid to battery voltage ratio.  Any calibration variations in the two inverter's charging float or absorb voltage setting will also push current between L1 and L2 sides through the batteries.  This may cause an issue with zero export control by inverters.

    · Share on Twitter
  • BarryFields
    BarryFields Registered Users Posts: 17 ✭✭
    edited February 2022 #5
    Main point is that that little green disconnect does NOTHING. It will not bond the neutral. Refer to schematic.


    · Share on Twitter
  • jdobush
    jdobush Registered Users Posts: 18 ✭✭
    edited February 2022 #6
    I use a similar model aims split phase inverter, so I'd like to understand this better. 

    So the way I understand from your note and schematic is that one side of the green disconnect wire is simply not attached to anything? What is the source of this schematic? As far as having the option to bond the ground to the output neutral, it makes sense to me to have the option of using the inverter as your bonding point or if you are feeding a panel that is has it's own neutral to ground bond. 

    There is a warning in the manual to NEVER connect input neutral to output neutral on split phase models. This green wire should just be bonding the output neutral to the ground at the inverter.

    FWIW my experience with AIMS has shown me that while they are built very stoutly, the design of many products leave a lot to be desired so I wouldn't be surprised had they overlooked something like this. My manual shows an automatic neutral to ground connection and schematic, but of course it only applies to the single phase models so as far as the split phase ??? I would just like to verify this, so I can correct my setup if needed.
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,617 admin
    #7
    As far as I can tell (guessing) about the Sigineer and Neutral+Grounding....

    There is no direct manual documentation about the transfer switch--So this is an educated guess (i.e., confirm with vendor, schematic, or inspection of the inverter in question).

    First... For the Sigineer, it appears that all versions shown in the manual have an isolated AC output transformer (H+N for 230/120 single phase, H1+N+H2 for 120/240 Split Phase).

    For AC input power, the inverter only uses the two load inputs plus green wire ground (i.e., H+N or H1+H2). It does not use H1+N+H2 (does not 'pass thru' split phase 120/240 from AC input to AC output (inverter in standby, transfer switch taking AC input to AC output power.

    All AC outputs are the output of an isolated transformer... H1+N, H1+H2, or H1+N+H2 (the manual applies to all versions, 120 or 230 single phase input and output (N America or "overseas"). Or to 240 VAC in and 120/240 VAC output (N. America).

    Since the AC output is always the output of a 2 wire or 3 wire isolation transformer... The only way a "Neutral" can be created or supplied is if the Neutral+Ground bond is done inside the AC inverter (Neutral to Ground Bond). If there was no N+G bond, then you would have two "floating" AC outputs (for 120 or 230 VAC single phase) or threes floating outputs (H1+center tap+H2) for N. America 120/240 VAC split phase.

    Since the AC output transformer is always in the circuit (AC mains power transferred to AC output) there is no N+G bonding (inverter to AC output) vs AC mains input to AC output--The output is always isolated, so the AC "Neutral" is always bonded to Inverter AC safety Ground.

    And there is never a "conflict" between "shore power" N+G bonding and the question of N+G bonding or not with the AC output.

    If the transfer switch output was "past" the AC isolated output transformer (direct AC mains input to AC output--No transformer) like many other AC inverters--Then there would be a need for the N+G transfer switch (i.e., shore power vs AC power, or AC Generator to AC inverter input and question of AC genset N+G bonding or not).

    Example 240 VAC in (no input Neutral used) to 120/240 Split Phase out. AC inverter's output/isolation transformer provides the 3 wire/split phase 120/240 VAC output.

    L1======) (====== H1
    N              ) (======N (N+G bonding here) (AC loads)
    L2======) (======H1

    Example of Transfer switch (like Sigineer) where Inverter's output transformer is always in the circuit (with mains/AC inverter transfer :

                                       AC inverter Xfrmr alway in circuit
    AC Mains In======[ \  /]L1======) (====== H1
                                    [  \/ ]N               ) (======N (N+G here) (AC loads)
    AC Inverter out ===[ /  \ ]L2======) (======H1

    Example of Transfer switch where inverter is completely bypassed by Xfer switch (inverter transformer is only there on "AC inverter output"--L1/L2/N supplied by home panel/utility power. This may be what the AIMS is doing:

    AC mains supply
    L1============== H1
    N=(N+G at panel)==N
    L2==============H1
                                                Transfer switch here H1+N+H2 output (inverter bypass). To  AC Loads
    AC inverter output
    L1======) (====== H1
    N              ) (======N+G bonded inside inverter
    L2======) (======H1

    Here we have the N+G bonding ALWAYS at the main panel in the home. And we have an AC inverter that need the N+G "somewhere" (inside the inverter, or at the transfer switch where N+G (supplying local grounding to AC inverter) or floating N only (N+G provided by shore power). Inverter AC transformer is only "used" if AC inverter output is powered and connected to local AC loads.

    Note that N+G switching with transfer switch can be done several ways... An alternate is to switch N+G "connected" or floating depending on the AC input source to the transfer switch.

    Sorry the drawings are not great... But this is how I understand the question to be.

    The 
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,485 ✭✭✭✭
    edited February 2022 #8
    120vac loads by your house, and/or your neighbors 120vac loads that are also fed from same utility pole transformer/cable run feed, can cause a loading voltage slump on one phase leg of split phase 240vac from utility transformer causing AC voltage of L1 to neutral to be slightly different from L2 to neutral.

    Most of the imbalance is caused by wiring voltage drop from utility transformer to your house and finally to your inverter. Utility pole transformers are fairly high power, so the actual utility pole transformer doesn't contribute much to L1-N-L2 imbalance.  Utility usually uses smaller gauge cable for neutral support cable because they don't expect much current to be carried by neutral return to their transformer.  Typical maximum current on neutral utility return is less than 30 amps although it could theoretically be service breaker amperage maximum.

    When inverter is sync'd and connected to grid, the inverter output transformer is connected in parallel with grid.  Having the inverter output transformer neutral also connected to grid neutral will result in inverter transformer acting like an auto-transformer put across utility L1-N-L2 and it will try to correct any grid L1-N and L2-N voltage imbalance.  This results in inverter transformer carrying some current not normally part of inverter operation, leaving less power handling capability for actual inverter function.  In the worst case it could overload inverter transformer.  Remember any split phase grid imbalance may be contributed to by your neighbors 120vac loads.

    Some LF inverters, that allow connecting a 120vac only generator AC input, have a relay that allows inverter transformer neutral to be connected to AC_in neutral to allow only 120vac AC input. Only one side, L1 or L2 AC input, is connected for this situation.  The inverter output transformer does function as an auto-transformer in this case to provide 240vac inverter output with only 120vac generator input.

    Grounding translates to a neutral connection issue since utility neutral is connected to ground in the main breaker panel. 

    All metal cases must be connected to ground for safety.  This includes main breaker panel, any aux breaker panel, and inverter case.  If inverter output neutral is connected to ground and main panel utility neutral is connected to ground, then inverter output neutral is connected to utility neutral.  To keep inverter output neutral separated from utility neutral all inverter output neutral connections must be floating.  This includes aux breaker panel neutral bus and any aux box branch circuit outlet's neutral.

    NEC code requires an aux breaker box case be grounded to main box, the aux box neutral bus must be floating from aux box case ground, and floating neutral bus in aux box be connected back to main box neutral bus.  The last item is in conflict with this inverter installation and creates the safety issue.



    · Share on Twitter
  • BarryFields
    BarryFields Registered Users Posts: 17 ✭✭
    #9
    The schematic is self generated after much time and tediousness.
    In my  system, the inverter feeds a totally separate sub panel that feeds totally separated outlets.
    Therefore, my inverter should be neutral bonded.
    · Share on Twitter
  • BarryFields
    BarryFields Registered Users Posts: 17 ✭✭
    #10
    BB  In my analysis, when the inverter is passing thru, the inverters center tapped transformer creates the new neutral.

    RCinFLA  these inverters should not be connected to the grid.
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,617 admin
    #11
    Barry,
    I am not sure I understand... The AC Input (really 240 VAC single phase) is NOT isolated to the 120/240 VAC split phase output? (Where is AC input "transfer" to AC Inverter output "managed"?).
    If non-isolated (Auto Transformer output):
    I am not sure--But I believe that having an AC input (genset, mains, etc.) can be an issue with basically an Auto Transformer output (non-isolated output center tapped "Auto transformer").
    If AC L1 and L2 are "floating" (no "neutral to ground" bonding). Then Center tapped auto transformer should not have any issue. This is not an normal North American setup (with AC Mains, the L1 and L2 are always "ground referenced" through a N+G bond). Larger gensets and inverters may/typically (>=~3.5 kWatt) that have a "Neutral" will have the N+G inside the genset (or inside the AC inverter output section). The genset and inverter may have an "easy" method to make/break N+G bonding.
    If AC  L2 and L2 have a Neutral and it is N+G grounded--I think there is the possibility of "circulating" currents between the AC L1/L2/N transformer (or AC alternator from genset) and the N+G bonded output auto transformer.
    With an Autotransformer H1/H2/N output and a genset/mains transformer with L1/L2/N input to "Inverter" and L-N and H-N connected to "ground"--There is a big current loop between the 3 wire input and 3 wire output where a "circulating" AC current can "develop" (through resonance, or unbalanced H1/H2/N currents--as a couple examples).
    Anyway--My guess--No expert here.
    If isolated output (or input) AC transformer with L1+L2 in and H1/H2/Neutral out:
    Or is the AC Input to AC Output isolated (and connecting Center Tap to (White wire, Neutral) to local Ground is appropriate? Basically there is an internal transfer switch that Switches the "input to the output stage" between AC in and Inverter in(???).
    I feel for you--I know how difficult it can be to trace complicated circuits to even come up with a "basic" functional diagram/schematic. Stuff inside is not always "obvious" and you have to really pay attention to get it "right" (or at least "right enough" for your/our needs).
    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • BarryFields
    BarryFields Registered Users Posts: 17 ✭✭
    #12
    The schematic in the my post is indeed accurate. At 73 with 20+years in diagnosis and repair and engineering support of electronic hardware, I was not easily confused by the transfer circuit in that post.

     In utility mode, only H1 and H2 are passed through the unit. When H1 and H2 are connected to the output transformer, the center tap generates a "new neutral". That new neutral should be bonded to ground.
    QED
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,617 admin
    #13
    Yep.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter

Categories