diy solar

diy solar

Quick confirmation on battery to inverter wire size and breaker

TNmountain

New Member
Joined
Jan 18, 2022
Messages
7
So I have 6 of the eg4-ll 48v 100ah batteries with signature solars rack that will be connected to a growatt 12k inverter. I have been looking at the forums and I lack the confidence to decide the wire size and fuse/breaker to go with due to my concern of safety.

The inverter manual says that the output ac side should have a 80amp breaker inline between the inverter and the panel box. I intend to put a 80amp double pole inside the panel as my breaker/disconnect.

On the battery to inverter size, I saw on this forum somewhere to use the 12000w/48/.85 = 294amps? I am not even sure if this is correct, but this would be I need a 300 (or 350 for overhead?) amp breaker/fuse. Is this correct? The manual says something about using a 3*2awg wire for the 12k inverter but does not seem to take in consideration anything about battery bank size. Just says at least 400ah batteries for the 8-12k inverters.

So, my questions are what size breaker and wire on the DC side. And what does the 3*2awg mean?
 
12000 ac watts / .85 conversion factor / 48 volts low cutoff = 294.12 service amps
294.12 service amps / .8 fuse headroom = 367.65 fault amps
Ignoring voltage drop 4/0 awg with a 400 amp fuse is indicated between the inverter and the busbars.
In a perfect world each battery should supply 1/6th of the current .
But the world we live in is not perfect so we need fudge factor.
I suggest 4 awg wire between each battery and the busbar with a 150 amp fuse.

Of course we should not ignore voltage drop so I point you to this calculator https://baymarinesupply.com/calculator
 
You are correct that a 12kW inverter at 48V would require 300A of current from the battery. This is basically the same as people using a 3kW inverter on a 12V system. For 300A you can use 4/0AWG wire. That's a lot easier than trying to run 3 2AWG wires for each connection. I would suggest a 400A fuse between the battery and inverter. You should actually have a fuse at each parallel battery. For a 48V LiFePO₄ you should use Class T fuses.

You battery bank size has no bearing on wire size and fuse size.
 
To be clear, you are suggesting separate 150amp fuses between the batteries and the bus bar on the racks that are supplied with 4awg wire. Followed by 4/0 wire from the bus bar to the inverter with a 400amp fuse, correct?

These batteries individually have 125amp breakers built into them. Are they not sufficient or is the extra 150 amp more of a precaution thing?

Also, thanks for such a quick reply!
 
Last edited:
Followed by 4awg wire from the bus bar to the inverter with a 400amp fuse, correct?
That would be 4/0AWG, not 4AWG from the bus bar to the inverter. And make sure your bus bars can handle 500A or more (the fuse must be the weakest link).
 
Code:
legend {
    {} { functional block }
    nnn|UUU| { un-fused busbar position where nn is wire guage in awg }
    nnn|NNN| { fused busbar position where nn is wire guage in awg and NNN is fuse ampacity in amps }
    <-> { bi-directional current flow }
    -> { uni-directional current flow }
    <- { uni-directional current flow }
    @ { back reference }
}
dc_domain {
    busbars {
        postive {
            4/0|400|<->inverter_charger.positive
            004|UUU|<->150A_class_t_fuse<->battery.1.positive
            004|UUU|<->150A_class_t_fuse<->battery.2.positive
            004|UUU|<->150A_class_t_fuse<->battery.3.positive
            004|UUU|<->150A_class_t_fuse<->battery.4.positive
            004|UUU|<->150A_class_t_fuse<->battery.5.positive
            004|UUU|<->150A_class_t_fuse<->battery.6.positive
            016|001|->shunt.positive
        }
        negative {
            4/0|UUU|<->shunt.negative<->inverter_charger.negative
            004|UUU|<->battery.1.negative
            004|UUU|<->battery.2.negative
            004|UUU|<->battery.3.negative
            004|UUU|<->battery.4.negative
            004|UUU|<->battery.5.negative
            004|UUU|<->battery.6.negative
        }
    }
}
shunt {
    postive@
    negative {
        line@
        load@
    }
}
 
Last edited:
I suggest 4 awg wire between each battery and the busbar with a 150 amp fuse.
4AWG assumes each battery will provide no more than 100A under normal circumstances, correct?

With 6 batteries and a 300W max load we can assume each of the 6 will provide 50A. Good so far. Let's do a walk through of a series of failures.

Let's say the inverter is being maxed out at 12kW - 300A. So each battery is providing 50A. Now let's say one battery fails for some reason. Now the remaining 5 batteries are each providing 60A. Still good. Oops - there goes another battery for some reason. The remaining 4 are now providing 75A each. Still good. Oops - a third strange failure (yes, the gremlins are busy). The remaining 3 batteries are now providing 100A each. Still good. It's a bad day and a 4th battery fails. The remaining two batteries are now being asked to deliver 150A each. One fuse now blows quickly leaving the last battery being asked to provide the full 300A and its fuse quickly blows and the whole system shuts down due to no more power.

In this unlikely chain of events, this seems like a safe response. There's a little gray area when down to two batteries. The 4AWG wire is being asked to provide 150A. The wires are likely going to get a bit warm if not hot but likely still safe until (and if) the fuses finally blow on those last two batteries.

Does this seem right? I'm just thinking this through to see if the safety is correct.
 
4AWG assumes each battery will provide no more than 100A under normal circumstances, correct?

With 6 batteries and a 300W max load we can assume each of the 6 will provide 50A. Good so far. Let's do a walk through of a series of failures.

Let's say the inverter is being maxed out at 12kW - 300A. So each battery is providing 50A. Now let's say one battery fails for some reason. Now the remaining 5 batteries are each providing 60A. Still good. Oops - there goes another battery for some reason. The remaining 4 are now providing 75A each. Still good. Oops - a third strange failure (yes, the gremlins are busy). The remaining 3 batteries are now providing 100A each. Still good. It's a bad day and a 4th battery fails. The remaining two batteries are now being asked to deliver 150A each. One fuse now blows quickly leaving the last battery being asked to provide the full 300A and its fuse quickly blows and the whole system shuts down due to no more power.

In this unlikely chain of events, this seems like a safe response. There's a little gray area when down to two batteries. The 4AWG wire is being asked to provide 150A. The wires are likely going to get a bit warm if not hot but likely still safe until (and if) the fuses finally blow on those last two batteries.

Does this seem right? I'm just thinking this through to see if the safety is correct.
That is what I going for.
Could also do 6 awg and 100 amp fuses and still have lots of redundancy.
Since we have not addressed voltage drop I over spec things a tad.
 
That is what I going for.
Could also do 6 awg and 100 amp fuses and still have lots of redundancy.
Since we have not addressed voltage drop I over spec things a tad.
The problem with 6AWG and 100A fuses is that the scenario I described results in total cascade failure sooner.

I guess it also depends on the max continuous discharge current of each battery's BMS. Do you want the fuse to blow before the BMS decides there is too much current and shuts down the battery? If the BMS supports 100A continuous it might support 200A for a few seconds. Should the wire and fuse be chosen to be 100A? 150A? 200A? I would say a 100A fuse is too low since that is still in the happy zone.
 
The problem with 6AWG and 100A fuses is that the scenario I described results in total cascade failure sooner.

I guess it also depends on the max continuous discharge current of each battery's BMS. Do you want the fuse to blow before the BMS decides there is too much current and shuts down the battery? If the BMS supports 100A continuous it might support 200A for a few seconds. Should the wire and fuse be chosen to be 100A? 150A? 200A? I would say a 100A fuse is too low since that is still in the happy zone.
Good point.
 
Oh yeah. And how does the EG4 battery's built-in 125A breaker fit into this and the choice of external fuse for each battery? I suppose using a 150A fuse would be a safe choice just incase the built-in fuse failed to trip at 125A. Or is fusing each battery unnecessary due to the built-in fuse in this case?
 
Oh yeah. And how does the EG4 battery's built-in 125A breaker fit into this and the choice of external fuse for each battery? I suppose using a 150A fuse would be a safe choice just incase the built-in fuse failed to trip at 125A. Or is fusing each battery unnecessary due to the built-in fuse in this case?
I'm waiting for definitive word from @RichardfromSignatureSolar on weather those breakers can be trusted to quench the arc from a dead short.
Until I get word I will assume the answer is no.
If/when the answer is yes, the class t fuses can be omitted.
 
Code:
legend {
    {} { functional block }
    nnn|UUU| { un-fused busbar position where nn is wire guage in awg }
    nnn|NNN| { fused busbar position where nn is wire guage in awg and NNN is fuse ampacity in amps }
    <-> { bi-directional current flow }
    -> { uni-directional current flow }
    <- { uni-directional current flow }
}
dc_domain {
    busbar {
        postive {
            4/0|400|<->inverter_charger.positive
            004|UUU|<->150A_class_t_fuse<->battery.1.positive
            004|UUU|<->150A_class_t_fuse<->battery.2.positive
            004|UUU|<->150A_class_t_fuse<->battery.3.positive
            004|UUU|<->150A_class_t_fuse<->battery.4.positive
            004|UUU|<->150A_class_t_fuse<->battery.5.positive
            004|UUU|<->150A_class_t_fuse<->battery.6.positive
            016|001|<->shunt.positive
        }
        negative {
            4/0|UUU|<->shunt<->inverter_charger.negative
            004|UUU|<->battery.1.negative
            004|UUU|<->battery.2.negative
            004|UUU|<->battery.3.negative
            004|UUU|<->battery.4.negative
            004|UUU|<->battery.5.negative
            004|UUU|<->battery.6.negative
        }
    }
}
Am I the only one who sees "Gobbledegoop"? Do I need to use a different browser?
Please do NOT HIJACK the thread to answer.
 
Code:
legend {
    {} { functional block }
    nnn|UUU| { un-fused busbar position where nn is wire guage in awg }
    nnn|NNN| { fused busbar position where nn is wire guage in awg and NNN is fuse ampacity in amps }
    <-> { bi-directional current flow }
    -> { uni-directional current flow }
    <- { uni-directional current flow }
}
dc_domain {
    busbar {
        postive {
            4/0|400|<->inverter_charger.positive
            004|UUU|<->150A_class_t_fuse<->battery.1.positive
            004|UUU|<->150A_class_t_fuse<->battery.2.positive
            004|UUU|<->150A_class_t_fuse<->battery.3.positive
            004|UUU|<->150A_class_t_fuse<->battery.4.positive
            004|UUU|<->150A_class_t_fuse<->battery.5.positive
            004|UUU|<->150A_class_t_fuse<->battery.6.positive
            016|001|<->shunt.positive
        }
        negative {
            4/0|UUU|<->shunt<->inverter_charger.negative
            004|UUU|<->battery.1.negative
            004|UUU|<->battery.2.negative
            004|UUU|<->battery.3.negative
            004|UUU|<->battery.4.negative
            004|UUU|<->battery.5.negative
            004|UUU|<->battery.6.negative
        }
    }
}
Am I the only one who sees "Gobbledegoop"? Do I need to use a different browser?
Please do NOT HIJACK the thread to answer.
 
I thought the
Am I the only one who sees "Gobbledegoop"? Do I need to use a different browser?
Please do NOT HIJACK the thread to answer.
I originally thought he accidentally put his reply in a code block. But after looking at it a second I recognized what he was doing, although I am familiar with some coding anyways. He has basically mapped where to put the fuses and what size. He included the legend to read it. The DC_domain is just the DC side, specifically the bus bar on my rack. Which is labeled positive and negative. And has where to put the fuses on the rack between bar and battery. Kudos to him for writing out like this. I assume it was a lot faster then drawing it
 
Code:
legend {
    {} { functional block }
    nnn|UUU| { un-fused busbar position where nn is wire guage in awg }
    nnn|NNN| { fused busbar position where nn is wire guage in awg and NNN is fuse ampacity in amps }
    <-> { bi-directional current flow }
    -> { uni-directional current flow }
    <- { uni-directional current flow }
}
dc_domain {
    busbar {
        postive {
            4/0|400|<->inverter_charger.positive
            004|UUU|<->150A_class_t_fuse<->battery.1.positive
            004|UUU|<->150A_class_t_fuse<->battery.2.positive
            004|UUU|<->150A_class_t_fuse<->battery.3.positive
            004|UUU|<->150A_class_t_fuse<->battery.4.positive
            004|UUU|<->150A_class_t_fuse<->battery.5.positive
            004|UUU|<->150A_class_t_fuse<->battery.6.positive
            016|001|<->shunt.positive
        }
        negative {
            4/0|UUU|<->shunt<->inverter_charger.negative
            004|UUU|<->battery.1.negative
            004|UUU|<->battery.2.negative
            004|UUU|<->battery.3.negative
            004|UUU|<->battery.4.negative
            004|UUU|<->battery.5.negative
            004|UUU|<->battery.6.negative
        }
    }
}
What is the 016 shunt? I have never used a shunt, although somewhat familiar. Is the 16 a gauge?
 
nnn|NNN| { fused busbar position where nn is wire guage in awg and NNN is fuse ampacity in amps }

016|001|<->shunt.positive

What is the 016 shunt? I have never used a shunt, although somewhat familiar. Is the 16 a gauge?
A shunt will have a small wire that needs to be connected to the battery positive. So that is showing it is a 16AWG wire with a 1A fuse. The shunt would normally provide that wire with the inline fuse.
 
I have pair of AIMS 12,000w. I use 250a breakers on each AIMS and an overall 400a Breaker/shunt-trip on the whole battery bank. 250a * 49v (low cut-off) = 12,250w. In theory, each AIMS can do 3 x for 20sec for 36,000w - but I don't have significant 'surge' since converting to modern heat-pump technology. The main source of surge would be a tool like a chop-saw and that's pretty small at this level.

The battery bank is 6 x 260ah 14s100p 18650 batteries but the cells are fused - so I don't have a breaker or fuse on each individual battery. There is an overall master SACE (shunt-trip) of 400a. Its unlikely with the consumption wiring that both AIMS will hit 250a at the same time needing 500a and so I did 400a to save $.

Here's a pic. The grey box has the bus-bars and shunts and breakers. Battery is wired with 4/0 AWG welding wire. Typically runs at 200a overall with certain days pushing 300a. All works smoothly going on 3 years now.
1645855782604.png


Side Note:
I had a 22yr old AC failing compressor other that caused the 250a breaker to trip. Has since been replaced with a new whole house heat-pump :)
1645855473583.png
Notice the 95.4a @ 196v surge in the power monitor log - this is 18,620w - and tripped an individual inverter 250a breaker!
image_wokzlb.jpg
 
Last edited:
I'm waiting for definitive word from @RichardfromSignatureSolar on weather those breakers can be trusted to quench the arc from a dead short.
Until I get word I will assume the answer is no.
If/when the answer is yes, the class t fuses can be omitted.
The breakers are OEM standard breakers from Chint. I can grab the model number on Monday. They are connected to the positive output of the battery. They will trip for anything exceeding this threshold upstream from them.
 
Back
Top