Inverter Grounding Question

Discussion in 'OnBoard Electronics & Controls' started by fireman_bob, Mar 3, 2015.

  1. fireman_bob
    Joined: Mar 2015
    Posts: 4
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: hartford

    fireman_bob New Member

    Hi, I'm installing an inverter/charger and have a question about the ground connection. According to ABYC the ground conductor must be of equal size to the DC+ conductor or at most one size smaller if the fuse is rated no more than 135% of the ampacity of the grounding conductor.

    My question is, why can't I tie the chassis ground connection of the inverter back to the DC- connection at the inverter? The DC- conductor is the same size as the DC+ conductor and goes right to the battery, where the ground would end up anyway but via the grounding terminal block, which adds resistance. In my case, the ground would require a 4AWG wire back to the ground point.
     
  2. sdowney717
    Joined: Nov 2010
    Posts: 868
    Likes: 33, Points: 28, Legacy Rep: 274
    Location: Newport News VA

    sdowney717 Senior Member

    http://www.tekrispower.com/pdfs/xantrex/Grounding Inverter-Chargers on Boats.pdf

    I run the inverter without the metal chassis grounded.
    Likely they want a separate ground wire just because they want it.
    Joining chassis to the negative is a ground.
    I suppose if your negative connection is bad, the inverter wont work very well.
    Supposedly what if the inverter develops an internal short, and seeing it is fused high, the 12vdc power must flow through the chassis, so IF you have a ground wire, and it is too small, then supposedly, the ground wire might burn and start a fire, cause the fuse wont blow. Or your 12vdc positive wire touches the chassis, all current will flow through the ground wire, so it has to be as big as the source for the fuse to blow.

    All of that would be unusual for it to happen, likely better not to ground the chassis at all, which is what I did. Or just do what your thinking, ground the chassis to the big negative 12vdc wire. Which they wont like, cause they dont like it.
     
  3. fireman_bob
    Joined: Mar 2015
    Posts: 4
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: hartford

    fireman_bob New Member

    Thanks for your reply. I guess the point is that if the unit is in inverter mode and a DC+ short to chassis develops, the return path would be through the AC ground (green) back to the DC- where all the bonding grounds connect (since in inverter mode the AC ground is connected to chassis leaving this ground off will certainly melt the AC ground wire). Since the DC+ bus can provide up to 200A, this would fry the much smaller AC ground wire.
    I just don't see why I can't use the DC- return back to the battery since it's already there with the correct sized cable.
     
  4. sdowney717
    Joined: Nov 2010
    Posts: 868
    Likes: 33, Points: 28, Legacy Rep: 274
    Location: Newport News VA

    sdowney717 Senior Member

    Good observation about the green AC wire having to return shorted current flow.

    I would just join them then.
    Likely that does not fit their definition of a 'ground wire' since a ground wire does not normally carry current except in a short. See what I mean?
    The inverter is internally bonded chassis to the dc negative connection, with a smallish wire.

    I have opened up two 3000 watt 6000 watt surge inverters, and they are attached to the positive stud with many smaller red wires of maybe 10 gauge that feed 2 circuit boards. And each wire runs to the mosfets and they are all fused to many 30 amp fuses. So I think an internal short would just blow out the 30 amp fuse and not burn up the AC ground wire.
     
  5. fireman_bob
    Joined: Mar 2015
    Posts: 4
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: hartford

    fireman_bob New Member

    One thing I found interesting in that this boat has a factory bow thruster with a 500A fused DC+/- fed with 4/0 wire. The ground wire is 8 AWG! I don't see how that satisfies the requirement. A DC+ short to motor case would melt that 8 AWG in a microsecond. And that's direct from Sea Ray.
     
  6. sdowney717
    Joined: Nov 2010
    Posts: 868
    Likes: 33, Points: 28, Legacy Rep: 274
    Location: Newport News VA

    sdowney717 Senior Member

    Yes same issue there it seems.

    I was wondering if the large inverter somehow did an unusual internal short, the little green internal ground connection wire is maybe 16 gauge and my AC ground wires are all 12 gauge, so I think the little green wire would melt with a few hundred amp flow before the outside AC wire grounds would say if the 30 amp fuses were not involved. Other thing is you could fuse the safety ground at the inverter and use a GFCI on the inverter output. I have a GFCI on my inverter output and it works fine, the ground wire is then not really needed.
     
  7. sdowney717
    Joined: Nov 2010
    Posts: 868
    Likes: 33, Points: 28, Legacy Rep: 274
    Location: Newport News VA

    sdowney717 Senior Member

    If it is like an engine starter, maybe the DC motor is internally grounded from its outer metal case to the DC negative wire by a direct mechanical internal connection?

    But if it is not, then yes, a short could potentially cause the little ground wire to burn.
     
  8. fireman_bob
    Joined: Mar 2015
    Posts: 4
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: hartford

    fireman_bob New Member

    You NEVER want to fuse a safety ground!
     
  9. sdowney717
    Joined: Nov 2010
    Posts: 868
    Likes: 33, Points: 28, Legacy Rep: 274
    Location: Newport News VA

    sdowney717 Senior Member

    About that never fusing ground, I only suggested it for the ground from the inverter along with GFCI to protect the inverter circuit, although the fuse may not react quickly enough to prevent damage to an already damaged inverter, it will prevent damage to AC ground wires from burning up.

    You could also simply not connect the AC green ground wire at the inverter and just use GFCI protection.

    http://ask-the-electrician.com/installing-a-gfci-outlet-without-a-ground-wire/

    Yes, a GFCI Outlet can be installed even if a ground wire is not available in the existing electrical circuit.
    The GFCI outlet must be marked with a provided label that the outlet is not grounded.
    The GFCI will still protect the user against ground fault just the same.
    This method is legal and compliant with the electrical code.
    This is ideal for these locations:
    Kitchen outlets
    Bathroom outlets
    Garage Outlets
    Outdoor outlets


    An AC ground is not needed if a GFCI breaker is in the circuit. If I had my choice between an intact ground wire or a gfci, I would choose gfci every time over the ground wire.
    Consider that the inverter is grounded by the chassis to the negative DC, so your good there for AC ground faults, and the GFCI is grounded to the green AC wire so your good there for AC ground faults.

    Breaking the ground wire from inverter to GFCI, your protected on both sides of both circuits, so there is no problem, got both circuit sides grounded-protected for AC ground faults.

    You might think the wire run between inverter and GFCI will then have no AC ground, BUT it does have one because it is connected at the GFCI ground lug.

    Since you still have an intact ground in all the AC wiring, no label is needed on any outlets, you're not breaking AC ground in the main boat wiring or the AC -DC ground so ground shorts will still go to ground running through the DC negative back to the inverter.

    I can get shocked handling worn grounded AC extension cords, even with their safety ground intact and I have been shocked with grounded cords. So you can protect the AC wires by putting them in conduit. The inverter also has overload protection built in, so a line short shuts it down. I have been putting wires in plastic conduit both solid grey and flexible ent in my boat to help protect them.
     
  10. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 146, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    An increasing number of inverters have DC- internally tied to the chassis, AC protective ground and one of the AC output wires, making that one the Neutral.

    There should be a big red warning label on such devices! Leakage current between AC live and protective ground should trip a GFCI, but that will not happen if the AC neutral is tied to the ground wire.
    This is also very annoying if there already is a ground connection in the external wiring or -even worse- the shore power ground is also connected to it.
     
  11. philSweet
    Joined: May 2008
    Posts: 2,287
    Likes: 163, Points: 63, Legacy Rep: 1082
    Location: Beaufort, SC and H'ville, NC

    philSweet Senior Member

    1. Because the battery - neutral - ground tie is often done at the battery terminal, and the battery is considered a serviceable item. Thus the ground path could be lost under "normal" circumstances.

    2. The inverter/charger itself can be considered a serviceable item as well, and disconnecting things in the wrong order may inadvertently remove the chassis ground. Less likely to happen if you are staring at a big fat green wire.

    3. The inverter is often used as a sort of distribution hub for AC and DC and you may have other chassis grounds arriving here looking for a proper ground. Commonly, a ground bus bar would be within a couple feet of an inverter charger.
     
  12. groper
    Joined: Jun 2011
    Posts: 2,467
    Likes: 123, Points: 73, Legacy Rep: 693
    Location: australia

    groper Senior Member

    Can you please explain that reasoning further?

    Same goes for sdowney...

    I see it like this, a gfci will not operate effectively without a protective ground. There must be a path of very low resistance in the form of a ground wire to route current to a point of lower potential. This is ideally back to the source, and is why the ground is tied to the neutral in most modern electrical installations. - it's easier for current to flow through a good wire conductor as opposed to the relatively high resistance of current flowing through the earth and soil etc. a low resistance wire ensures a rapid and effective trip time of the protective devices. If it were not there, only small amounts of current would flow to earth should a fault develop and the device will likely not trip.

    I have a stainless cooktop on my boat, and every now and again I used to get a little nibble from touching the stainless hob whilst on inverter power which had no protective earth going anywhere. I figured it was safe as the inverter is an isolation transformer essentially. There is a gfci protecting that same circuit, and it never tripped. So to resolve the problem, I tied the neutral and earth together at the inverter. Now when the salt builds up inside the cooktop terminals and the earth leakage fault presents itself, the gFci will now trip before I even get a chance to touch it and give myself a little zip... Make sense?
     
  13. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 146, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    The GFCI principle is that what comes in must be equal to what comes out.
    Current in phase and neutral are compared, if there is a difference it means somewhere current is leaking, so the GFCI trips.

    When you connect protective ground to neutral, leakage cannot be detected; you have in fact connected the metal parts of the cooktop to the AC neutral wire.
     
  14. groper
    Joined: Jun 2011
    Posts: 2,467
    Likes: 123, Points: 73, Legacy Rep: 693
    Location: australia

    groper Senior Member

    Exactly. ... However the second part of what you said is not accurate. The neutral and earth conductors should bonded on the supply side of the gfci so the faukt current bypasses the device and causes it to trip.

    So when there is a fault, such an active conductor leaking current to an extraneous metal part, provided the part is connected to a protective earth wire, then current will flow back to the source via the earth / ground conductor. Therefore the gfci sees this out of balance current because it only compares active and neutral. The current returning on the ground wire is what causes it to trip, and is why it must be a very effective conductor of very low impedance.

    If the earth and neutral are bonded upstream of the gfci, as it is in every transformer down the street in any first world city, the impedance of the earth return path through the mass of earth is not important because most of the faukt current can return to the supply transformer via the earth and neutral conductors due to the earth neutral bonding. The impedance of earth itself is quite high, so it cannot be relied upon to carry sufficient current which allows protective devices to operate. So, This earth to neutral link is extremely important to ensure a low impedance path so that circuit breakers and gfci devices trip quickly when any faukt condition occurs, protecting both wiring in the installation and people coming into contact with raised potential in most normal types of modern electrical installations.
     

  15. sdowney717
    Joined: Nov 2010
    Posts: 868
    Likes: 33, Points: 28, Legacy Rep: 274
    Location: Newport News VA

    sdowney717 Senior Member

    With a GFCI, what current goes through the hot into a device must equal what comes out of the device back through the neutral within a 5 milliamp tolerance, there is no ground needed. ground circuit is not considered in the current balance circuitry at all. Ground is just an extra safety say if the GFCI fails to function.

    Groper, I dont know why you got a tingle from your stove top unless you were somehow become part of the 'device'.

    GFCI will not trip if you are the device! Unless you leak current out to an alternate pathway, the current flow into you and out of you wont be lost (bypassing return through GFCI ) so the GFCI will not trip.

    If you are insulated from ground, touch the hot in one hand neutral in other and turn on the GFCI circuit, you will be shocked, since the flow is not lost, (bypassing the GFCI device) it will not trip off.
    That situation would be very unusual to occur, most electrocutions are you touching a hot electric current source surface and shunting the current through your body.

    Bypassing the return through the GFCI will cause it to trip, so then your not electrocuted. Somewhere, your completing the circuit to ground outside of the neutral current return wire.

    If you touch hot and your not grounded and not become part of the device, and no GFCI is in the circuit, you wont feel a thing, no shock. The current has no where to flow, cant go back to it's source.

    Anyhow, I dont suggest people not have ungrounded devices or get rid of the green safety ground in the distribution wiring.

    I am sure you can feel less than a 5 milliamp current. So you might get a tingle, but not trip a GFCI, but it is not enough to kill you. (most will survive that tiny current, maybe some that could still kill, maybe a baby wont?)
     
Loading...
Similar Threads
  1. alexandershudai
    Replies:
    4
    Views:
    1,553
  2. Squidly-Diddly
    Replies:
    2
    Views:
    686
  3. elyon67
    Replies:
    9
    Views:
    1,418
  4. 7228sedan
    Replies:
    14
    Views:
    1,727
  5. BertKu
    Replies:
    22
    Views:
    2,638
  6. BertKu
    Replies:
    19
    Views:
    2,817
  7. sdowney717
    Replies:
    12
    Views:
    3,339
  8. sdowney717
    Replies:
    4
    Views:
    8,261
  9. Jakegator
    Replies:
    12
    Views:
    11,894
  10. RoyHB
    Replies:
    38
    Views:
    12,700
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.