Inverter connections - Why 3 power connectors for 12vdc

[veggie]

A few months back, I bought a modified sine wave inverter to power some greenhouse and yard lighting.
Today I took it out of the box to give it a little test.
While connecting the wires, I noticed the unit has 3 positive terminals and 3 Negative terminals for the 12VDC power feed.
The unit also came supplied with 3 heavy red and black cables.
Unfortunately the unit came with a generic (and useless) instruction sheet which does not address this feature.
Can anyone explain why a 5000 watt inverter would have a total of 6 x 12vdc input cable connectors?

The pictures below show the unit powering a resistance load at 900 watts.
Also shown are the 6 power terminals

veggie

[BioHazard]

After all the trouble I went to testing my 3kw inverter, I wish it had all those power connections. I don't even want to try 10kw @ 12 volts....

Interestingly, I noticed on the inside of my 3kw Xantrex inverter, the single cable stud on the back is attached to the circuit board inside by six small wires bundled together. I'm guessing that's easier than soldering 2/0 cable to a circuit board! Maybe your inverter does it the same way, that just gives you a better access to them all.

[RogerAS]

A few months back, I bought a modified sine wave inverter to power some greenhouse and yard lighting.
Today I took it out of the box to give it a little test.
While connecting the wires, I noticed the unit has 3 positive terminals and 3 Negative terminals for the 12VDC power feed.
The unit also came supplied with heavy 3 red and black cables.
Unfortunately the unit came with a generic (and useless) instruction sheet which does not address this feature.
Can anyone explain why a 5000 watt inverter would have a total of 6 x 12vdc input cable connectors?

The pictures below show the unit powering a resistance load at 900 watts.
Also shown are the 6 power terminals

veggie

Hey Veggie,

My inverter has two such cable lug attachments. I suspect these can/could be used with big knife switches to allow for multiple battery banks (I don't use my this way). I also feel that if one is pushing the upper limit on output these could spread the load over these multiple lugs and lower potential heat buildup.

RS

[veggie]

Thanks for the info guys,
The inverter will be used on a 675 AH, 12 vdc battery bank.
(two series banks of three 6volt T105's in parallel)

I will use all 6 cables and terminate them at a common point on the + / - battery bank connections.

cheers,
veggie

[veggie]

As Jens suggested, it looks possible to use much smaller cable to parallel my 3 battery banks by using the Inverter terminals and the "parallel junction point".
Here is how I would wire it (see picture).
Anyone see anything wrong with this ?
One potential hurdle is battery charging. Could I connect the charger to the "junction point" and do all batteries together ?


veggie

[squarebob]

The lugs are like an internal buss bar. just hook the charger to any + and - lug on the inverter and you should be good to go. Make sure to use the shortest possible cables from the batteries to the inverter. They need to be the same length also.

Bob

[Ronmar]

The only thing I see wrong with the drawing is no fuses...  There needs to be a fuse on EACH series battery loop.  These of course need to be large enough to pass the peak demand current of each leg, and small enough to keep the wiring from cooking into a fire if one of the series loops should short.

As for three terminals, watts are watts regardless of the source.  Power in watts equals current times voltage.  So watts divided by volts equals current. 5KW divided by 12V equals 416 amps...  Be carefull, battery banks are some of the most dangerous things any of us will ever work with...

[veggie]

Got it.
I need a fuse and a disconnect on each of the (+) cables feeding the inverter.

There are 4 120vac outlets on the inverter.
One additional form of protection I can add is to run the 120 outputs through a pony panel of 4 X 10 amp breakers.
This way any overloads can trip the 120volt side before reaching the overloading levels of the inverter/cables.

veggie

[Ronmar]

I was also being kind with the calculation of 416A.  That is not taking into account the efficiency losses.  The inverter is probably only about 85-90% efficient at making that 5KW of AC, so it will take closer to 460A - 480A of 12VDC to achieve that 5KW AC output.  One of the many issues when using parallel banks is uneven load sharing.  If one bank should open or fail, the other two banks will try and take up the slack.  If two banks fail, that 480 amp could possibly be drawn from any one of the series battery pairs...  This makes fusing a bit tricky.  This uneven load sharing also makes charging a little more problematic.  A weak series string will suck up most all the current of a parallel charge regime, and overcharge, while the other two strings will undercharge...

[mike90045]

There really ought to be a law against bazillion watt inverters for 12V.   At 5KW, thats 416 amps @12V.   Starting surge for a motor could easily be twice that.  So of course, they have to provide several terminals, because you do need 6 batteries to fire the gadget anywhere near full power.  1% loss in the lines, and you get close to the cut-out voltage for the inverter.

But if you had a 48V inverter, you only need 104A, and much less copper investment.  4 beefy 12V batteries in series, no parallel paths, and you can use a cheap fuse.   Price that 12V 500A fuse.  The holder alone will cost you a pretty penny.

And if you had a pure sine inverter, you could run all sorts of things and not worry about mod-square wave frying them.

[squarebob]

That 416 Amps would be split among the 3 banks so you got 139 Amps in each leg. If you fused at 50% safety, then 3, 200 Amp fuses should do the trick.

Are these fuses suitable for our use:
http://www.amazon.com/Raptor-RANL2002-Fuses-Gold-Plated/dp/B0002EXJIY

Here is a holder for the fuse:
http://www.amazon.com/Scosche-EWFH-Single-Fuse-Holder/dp/B000KIR8M0/ref=pd_bxgy_e_img_b

Or do we need this type:
http://store.solar-electric.com/fb-200t.html


Sure is a big difference in cost

Bob

[veggie]

squarebob,

That's exactly what I need. Thanks for posting the fuse information.

I can add fuses and isolation switches on the 12VDC side.

I can limit the load on the inverter by running each of the 120v outputs through a pony panel containing 2 x 10 amp breakers and 2 x 5 amp breakers. ( or 4 outlets allowing only 3600 watts of total load). Quite adequate for my greenhouse, pump, and yard lighting needs.
This way any overloads can trip the 120 volt breakers before ever reaching the overloading levels of the inverter/cables.

veggie

[Ronmar]

That 416 Amps would be split among the 3 banks so you got 139 Amps in each leg. If you fused at 50% safety, then 3, 200 Amp fuses should do the trick.

In a perfect world, yes the load is split among the 3 banks.  But in my experience, battery banks rarely operate perfectly:)  In your example of 139A per series string, with a 50% overfuse using a 200A fuse, if a single string fails while under load, it's 139A of current draw will be divided among the other two strings and add to their total draw.  This will push the current draw of these two strings right up to their individual 200A fuse rating...  Due to cost, these are fuses you don't want popping on a regular basis:)  That is a real drawback of parallel battery strings in these high current applications.  To do it properly, each series string needs to be wired with a large enough wire gauge and fuse to be able to deal with the possibility of it being the ONLY string providing ALL the current to the inverter.

I think I am of the same opinion as Mike on this subject.  There ought to be a law:)  IMO, 12V inverters are for emergency use, or low power applications only.  Such as when you cannot get power there any other way, like running a drill or power saw out in the sticks, or running the portable TV or stereo or laptop out at the campsite.  To do it right for a high wattage application, the copper investment is going to be huge.  Done incorrectly, this is something that can quite easilly burn down your house!  Don't low-ball the battery wiring and fuses...  These high current loads are also murder on battery efficiency.  Battery AH rateings are typically based on a 20 hour discharge rate.  IE: a 100AH rated battery will give you 5A for 20 hours.  Because of internal resistance changes and loss of efficiency with load increase, it will not give you 20A for 5 hours, or 100A for 1 hour... 

I fully understand running what you have.  But If I was starting from the ground up, especially for a large wattage fixed installation, I would do this project with no less than a 24V inverter, and would lean greatly towards a 48V unit and a single series string of batteries.  As mentioned, 48V would cut the current draw by 75% for the same output.  Even at a nominal 3600W load that you described(4140W out of the batteries after inverter/conversion losses), at 48V that is 86A, compared to 345A out of a 12V bank.  86A is way easier to deal with and would require much less invested in the copper and protection devices, as well as moving you closer to some acceptable battery loading...

Godd luck    

[veggie]

But in my experience, battery banks rarely operate perfectly:)  In your example of 139A per series string, with a 50% overfuse using a 200A fuse, if a single string fails while under load, it's 139A of current draw will be divided among the other two strings and add to their total draw.  This will push the current draw of these two strings right up to their individual 200A fuse rating...  Due to cost, these are fuses you don't want popping on a regular basis:)  That is a real drawback of parallel battery strings in these high current applications. 

Hi Ronmar,

In your cascading fuse failure scenario did you consider that the load on the Inverter would be limited by the rating of the 120 vac breakers ? so an overload caused by the end-use devises is unlikely.
Therefore a failure or load big enough to blow a fuse would have to come from a short on the DC side or a failure in the Inverter.
Hopefully that would be a very uncommon occurrence so I would not mind paying $18 per fuse to prevent a blow up.
The idea being to keep the end loads much lower than the capacity of the system.

Recall that this is for a greenhouse and yard lighting. Not for house power.
The inverter cost me very little so I don't mind de-tuning the system to bring the power levels into a safe range for 12vdc operation.

Perhaps I could get some further comments on this modification ...

Revised installation based on forum comments:
1] Limit the end use 120 vac power to 20 amps (2400 watts) with two 10 amp breakers. This should still provide plenty of power for my needs.

2] Run single wires from the bank to the inverter with an in-line 300 amp in line fuse.
Max. demand allowable through the breakers would be 2400 watts X 1.1 losses / 12vots = 220 amps.
A 300 amp fuse and holder are $68 at a local solar store.

3] Run Bank-to-Inverter wires sized for 350 amps ? (50 amps above the fuse rating)?

thanks for everyone's comments so far,
veggie

[Ronmar]

That sounds fine Veggie, you might even be able to go lower than 300 on the wire and fuse setup.  I like the output breaker idea as it adds a relitively simple added safety and another layer of switched control. My comments were meant to bring out this line of thought.  Regardless of what your maximum intended output is, the same line of thought needs to be brought into play whenever contemplating paralleling 12V battery banks be it for 20A or 500A worth of DC load.  The system needs to be built accordingly.

With 2400W max output +15% for conversion losses means 2760W max out of the batteries.  Sadly, that inverter probably only reaches it's peak efficiency near it's maximum rated load, so at half load, it probably won't be quite as good as 85%.  At 80% efficiency, the current draw at peak will be around 240A, or 80A per string.