diy solar

diy solar

How to connect multiple high amp loads to battery bank

hipringles

Solar Enthusiast
Joined
Jan 31, 2021
Messages
214
I have (or will soon) have a 12V battery bank that has these 3 major loads:
1.) 75A (peak continous, 100A surge) 12V Air conditioner
2.) ~50-75A of misc 12V appliances (Water pump, heater, fans, lights, DC fridge, etc)
3.) 12V 3000W Victron inverter

I am trying to decide on the best way to connect these to the battery bank, and where the fuses should go... Here is what I think, and what am I questioning


1.) 4/0 Wire with 400A fuse between battery and inverter
2.) 0 or 2 gauge wire, 100A breaker between battery and Airconditioner
3.) 75A cb on 2 or 4 gauge wire to the fuse block from the battery

Thus 3 connections, all with their own breakers to the 3 major DC loads

Questions:

Q1.) How should I connect those 3 items to the battery bank? A bus bar would need to be a ~500A+ busbar? I am thinking possible 3 wires off battery positive? Or is a busbar required?
Q2.) Should there be no other loads between main inverter fuse and inverter (AKA, don't put item #3 on the output of the 400A inverter fuse)
Q3.) Negative return: I plan to have a 500A victron shunt, can I just attach the 3 main negatives back to this? Or is a 500A negative bar required, which then connects to the shunt? Could a smaller one be used?


Bonus:

Q4.) Cancel the 12V AC and spend the extra ~$700 to go 24V :)?
 
I would skip the 12V. Go to 24V. I'm at 48V myself and have no problem whatsoever running loads far higher than what you are mentioning.
I am leaning more and more towards it. To go 12V I think its impossible to have a big enough fuse to be a ""main"" fuse (100A for AC, 400 for inverter, and 50 for the DC bus).

I'd end up having to have 2 main positive leads off the bank, 1 for DC only, fused on 200A, then another on a 400A for the inverter.

If I can't get the AC order canelled, that ^^ approach would likely be my way forward

I'd go all the way to 48V if I could find a 48V/3000/120v multiplus thats not horribly overpriced.
 
I think DC "off-grid" appliances are way overpriced. I think you'll likely save money switching to a 120VAC lifestyle, even if you invest more money into a more robust solar system. Panels now are the single cheapest component. Go big!
 
I would repeat MichaelK's advice... Cancel then 12 and go 24, or better yet 48. BUT if you are still going to be set on 12v, then direct connections are your best bet. and 4/0 is only rated for 250A, that fuse is to big. Run a separate line to the battery for both positive and negatives.

what is "overpriced" in your department? 1700 for a 4400 watt Magnum inverter, A little on the expensive side, but with inverters you REALLY get what you pay for. My 4448 can handle my AC, fridge and either the microwave OR the PC... don't try to do both at once, tends to overload the inverter lolol.
 
Roof space is at an extreme premium :). Debating do a non-mounted setup, but yeah another discussion for another day.

The fuse size is based off this chart, from another thread (3000W 12 or 24V, 400A fuse, 4/0 wire). Will also links 4/0 wiring on the "2000 to 4000w" link at the top here.

For the double line advice: The idea is (if 12v) 2 lines, 1 for DC busbar (~150A) and 1 for inverter (400 per that chart))

Lastly, overpriced, i'd like to stay near the 24v multiplus price, which is about $1250. Space is also a premium so getting a yuge one isn't an option :/
 
The easiest way to make multiple connections is with a set of common bus bars. The bolt on the shunt has only so many threads, which means you can put only so many lugs on the bolt before you run out of threads. Run it too close and use too few threads and you risk stripping the threads.

I'll be the voice of dissent on 24v. Depending on what your 12v loads are in your RV (this is an RV, right?), it gets complicated and expensive to have to step up/down between 12v and 24v.

From what I hear, a DC fridge is an energy hog. But is it that much of a hog that it bloats your second item list past 50 amps? I really doubt it.

A set of 600 amp bus bars is probably overkill. Safe? Yes. But count up the amps of what you'll really be running concurrently. I bet it doesn't get anywhere close to 600 amps.
 
What's the battery? Can it handle these amperages? e.g. 2 280's = 560 amps for 1C max ... you're really beating on the battery with these amperages unless you have a lot of them in parallel. If you do, then you need large bus bars between cells and large wire etc etc ...
Personally I echo going to 24. The Victron 24/12 converter 70 amps is $140 to power all the stuff that can't be 24v and they can be placed in parallel if you need more than 70 amps.
 
The easiest way to make multiple connections is with a set of common bus bars. The bolt on the shunt has only so many threads, which means you can put only so many lugs on the bolt before you run out of threads. Run it too close and use too few threads and you risk stripping the threads.

I'll be the voice of dissent on 24v. Depending on what your 12v loads are in your RV (this is an RV, right?), it gets complicated and expensive to have to step up/down between 12v and 24v.

From what I hear, a DC fridge is an energy hog. But is it that much of a hog that it bloats your second item list past 50 amps? I really doubt it.

A set of 600 amp bus bars is probably overkill. Safe? Yes. But count up the amps of what you'll really be running concurrently. I bet it doesn't get anywhere close to 600 amps.
Thanks for the info.

here is the rough budget:

75A consistant on the Aircon
Inverter max is 3000w, its possible to get close with microwave + induction, plus inverter loss etc, I think its reasonable to assume this will pull 250-375A

The DC block will be:
5A (max) Maxxair fan
12A DC Fridge
5-10A Diesel heater (obv doesn't run when Ac does)
8-10A for Water pump
10-15A budget for things im not thinking of (lights, dc plugs, etc etc)

Thats how i roughly got 50.

That said, i think the realistic max pull would be: 275A from fully running inverter when cooking, 75A from AC (if its hot as shit) and maybe 25A DC (fridge/waterpump/misc). Bringing the total max pull to be around 375A

There would be 2 cables going into the shunt negative thread, so hopefully not an issue
 
What's the battery? Can it handle these amperages? e.g. 2 280's = 560 amps for 1C max ... you're really beating on the battery with these amperages unless you have a lot of them in parallel. If you do, then you need large bus bars between cells and large wire etc etc ...
Personally I echo going to 24. The Victron 24/12 converter 70 amps is $140 to power all the stuff that can't be 24v and they can be placed in parallel if you need more than 70 amps.
32 Eve 280s, 2p4s x 4 (each with its own 100A bms), they will be connected together with 4/0 (under this 12V assumption)

Still going to try for either 24 or 48... Calling AC order tomorrow to see if a replacement is possible
 
I have two 4s 280Ah batteries connected into the battery side of my Victron shunt and a single 2/0 cable on the load side of the shunt going to the negative common bus bar. Not a problem there. Can you fit four cable lugs? Maybe? If not, then you'll need to put a bus bar between the battery bank and the shunt. Four 2/0 battery cables into the bus bar, one 4/0 cable out to the shunt.

Are you in a diesel motorhome?
 
I have two 4s 280Ah batteries connected into the battery side of my Victron shunt and a single 2/0 cable on the load side of the shunt going to the negative common bus bar. Not a problem there. Can you fit four cable lugs? Maybe? If not, then you'll need to put a bus bar between the battery bank and the shunt. Four 2/0 battery cables into the bus bar, one 4/0 cable out to the shunt.

Are you in a diesel motorhome?
For the negative I was thinking something like that (4 BMS negative to busbar, busbar to shunt)

Nope, just a (big) van thats intended to go offgrid and be comfortable for some time. The Mrs has a hard requirement for Aircon (which is complicating things a bit, but thats life). The aircon is a DC. Adds price, but lets me avoid needing a 5000w inverter to be able to cook/cool at the same time. Also doesn't add inversion in-effeciency (plus heat), letting the batteries stretch a bit longer

The diesel heater is just the easiest way to heat on ski trips :)
 
The diesel heater is just the easiest way to heat on ski trips :)

OK. Figured I better ask. Sometimes diesel RV owners forget about some of the bigger amp loads in their rigs.

If I were building a custom RV from the ground up, I would be looking at 48v. But in an existing RV with lots of 12v loads, the higher voltage system sounds cool, but when you get into the weeds it starts to get complicated and expensive.
 
A victron lynx "power in" or distributer makes for a tidy bus bar/fuse combo. It's really easy to hang a battery switch and shunt directly off the side of the lynx.

I'm doing 2/0 from a couple battery banks into the shunt and disconnect switch and straight into the lynx from those. I'm doubling up lugs on both of those connections. They lynx has 4 spots for fused connections. For me, that's an inverter, solar charger, DC distribution block, and DC/DC charger. I don't know if you have 2 charge sources, but if you only had one, you could do a similar layout with the AC taking one of the connections.

 
A victron lynx "power in" or distributer makes for a tidy bus bar/fuse combo. It's really easy to hang a battery switch and shunt directly off the side of the lynx.

I'm doing 2/0 from a couple battery banks into the shunt and disconnect switch and straight into the lynx from those. I'm doubling up lugs on both of those connections. They lynx has 4 spots for fused connections. For me, that's an inverter, solar charger, DC distribution block, and DC/DC charger. I don't know if you have 2 charge sources, but if you only had one, you could do a similar layout with the AC taking one of the connections.



Are you using the full system? (Power in, shunt, distributer)

Are you putting a fuse between the distributor and the battery?
 
Last edited:
Maybe my answer for a decent inverter charger @48v!
I'm familiar/have contacts with the OEM for that unit. I can probably have one dropped to a US address for $600 or less. I wouldn't recommend it over a Victron though.

There is a new MultiPlus-II 48/3000/35-32 120V on the way but I'm waiting to hear when it will arrive stateside. I hope to have that information this week. If I had to guess on the price, I would put it in the sub $1200 range.

The Quattro 48/3000/35-50/50 120V would work. Can be had for sub $1700 but sounds like that is out of budget..

Lastly, overpriced, i'd like to stay near the 24v multiplus price, which is about $1250.
I can beat that for either a 24/3000/70-50 or a 12/3000/120-50. ;) Feel free to PM me if you are interested in a quote.

I also second the Lynx and 12V. It's my belief that losses on the DC-DC are likely going to be more than the losses sustained by the 12V -> 120V inverter. Besides the hassle installing them.

You can use up to an 800A ANN/CNN fuse with the Lynx Shunt. Up to 500A/32V or 300A/58V MEGA fuses with the Distributor as far as I know. No need for extra fusing with the Distributor (see diagram note below).

1614584015290.png

The Lynx isn't super expensive, about what a fully configured e-panel would cost. Around $137 for the Lynx Power In. Around $200 for the Distributor, more at some places.

The Lynx Shunt VE.Can is more pricey. Around $300, more at some places. A SmartShunt could be cheaper to install. Sounds like 500A should work. Around $120, more at some places.

32 Eve 280s, 2p4s x 4 (each with its own 100A bms), they will be connected together with 4/0 (under this 12V assumption)

Still going to try for either 24 or 48... Calling AC order tomorrow to see if a replacement is possible
IMHO a different cell configuration would be recommended. Eddy currents etc. Orion BMS has some great information on this. Make sure to check out page 7.

Strings, Parallel Cells, and Parallel Strings

IMHO a recommendation would be to parallel at the cell level. Similar to this (taken from the manual):

1614585307475.png
The likelihood of cell failure is quite low assuming properly configured/operated but there are some notes regarding that on page 7. If you want to have 2 packs, you could reduce to 2x 8-cell banks rather than 4x 4-cell banks. I do completely understand you may already have 100A BMS units. A contactor would work here to increase the amperage limit of the BMS.

Here is a quote about eddy currents:

"Eddy currents:

When two or more strings are paralleled together, currents will flow between the strings. These currents form due to differences in the total pack voltage between strings. The amount of current that flows is determined by the difference in total string voltages, resistance of each string, and the characteristics of the cells. With these currents, it is possible for one string to force charge a second string, which can lead to over-charging or over-discharging individual cells. A low capacity cell or a faulty cell can cause the force charging of an entire string which may result in over-charge and/or overdischarge.

While it seems counter-intuitive, it is possible (in fact, likely) when the charger shuts off due to a fully charged cell in one string, that string may continue to be charged by another parallel string (the same principal applies to discharge). For example, if string A contains a cell which is fully charged, it is possible that current could flow from a second paralleled string B into string A. This would happen every time the average cell voltage in string B was higher than the average cell voltage in string A, even though string A has a cell which has become fully charged (or fully discharged.) This can be compounded by the characteristics of the cells. Most cells have some “surface charge” which causes the cell voltages to artificially rise and drop depending on the recent history of the cell. For example, if it was charged, the voltage will be temporarily raised.

Common sense would suggest that these currents would form only when the packs are first paralleled together and then dissipate over time. However, because different currents flow through each string during charge or discharge and due to the surface charge phenomenon, these currents end up being present any time a pack has been charged or discharged. These currents present a significant challenge for managing paralleled strings. Because of the possibility of these currents, it is absolutely essential that each string MUST have a contactor, shunt trip breaker, or other automatic and redundant means of isolating the string from any other strings if a critical fault occurs. This is in addition to the standard over-current protection for the string. The designer must consider and ensure safe behavior in the event of a failure of any single component, including diodes, contactors, and BMS units."
 
Are you using the full system? (Power in, shunt, distributer)

Are you putting a fuse between the distributor and the battery?

No, I'm just using the lynx distributer. See pic below, this is my testing setup before installing in the van. I would have bought a "power in" instead of the distributer and saved some $ if I had known it could be used with fuses also. They are basically the same thing, but with electronics and hardware for the fuses in the distributer (the hardware can be added to the power in).

Yes, I am fusing the batteries at the positive terminals. https://www.bluesea.com/products/5191/MRBF_Terminal_Fuse_Block_-_30_to_300A

IMG_0475.jpg
 
I'm familiar/have contacts with the OEM for that unit. I can probably have one dropped to a US address for $600 or less. I wouldn't recommend it over a Victron though.

There is a new MultiPlus-II 48/3000/35-32 120V on the way but I'm waiting to hear when it will arrive stateside. I hope to have that information this week. If I had to guess on the price, I would put it in the sub $1200 range.

The Quattro 48/3000/35-50/50 120V would work. Can be had for sub $1700 but sounds like that is out of budget..


I can beat that for either a 24/3000/70-50 or a 12/3000/120-50. ;) Feel free to PM me if you are interested in a quote.

I also second the Lynx and 12V. It's my belief that losses on the DC-DC are likely going to be more than the losses sustained by the 12V -> 120V inverter. Besides the hassle installing them.

You can use up to an 800A ANN/CNN fuse with the Lynx Shunt. Up to 500A/32V or 300A/58V MEGA fuses with the Distributor as far as I know. No need for extra fusing with the Distributor (see diagram note below).

View attachment 39016

The Lynx isn't super expensive, about what a fully configured e-panel would cost. Around $137 for the Lynx Power In. Around $200 for the Distributor, more at some places.

The Lynx Shunt VE.Can is more pricey. Around $300, more at some places. A SmartShunt could be cheaper to install. Sounds like 500A should work. Around $120, more at some places.


IMHO a different cell configuration would be recommended. Eddy currents etc. Orion BMS has some great information on this. Make sure to check out page 7.

Strings, Parallel Cells, and Parallel Strings

IMHO a recommendation would be to parallel at the cell level. Similar to this (taken from the manual):

View attachment 39017
The likelihood of cell failure is quite low assuming properly configured/operated but there are some notes regarding that on page 7. If you want to have 2 packs, you could reduce to 2x 8-cell banks rather than 4x 4-cell banks. I do completely understand you may already have 100A BMS units. A contactor would work here to increase the amperage limit of the BMS.

Here is a quote about eddy currents:

"Eddy currents:

When two or more strings are paralleled together, currents will flow between the strings. These currents form due to differences in the total pack voltage between strings. The amount of current that flows is determined by the difference in total string voltages, resistance of each string, and the characteristics of the cells. With these currents, it is possible for one string to force charge a second string, which can lead to over-charging or over-discharging individual cells. A low capacity cell or a faulty cell can cause the force charging of an entire string which may result in over-charge and/or overdischarge.

While it seems counter-intuitive, it is possible (in fact, likely) when the charger shuts off due to a fully charged cell in one string, that string may continue to be charged by another parallel string (the same principal applies to discharge). For example, if string A contains a cell which is fully charged, it is possible that current could flow from a second paralleled string B into string A. This would happen every time the average cell voltage in string B was higher than the average cell voltage in string A, even though string A has a cell which has become fully charged (or fully discharged.) This can be compounded by the characteristics of the cells. Most cells have some “surface charge” which causes the cell voltages to artificially rise and drop depending on the recent history of the cell. For example, if it was charged, the voltage will be temporarily raised.

Common sense would suggest that these currents would form only when the packs are first paralleled together and then dissipate over time. However, because different currents flow through each string during charge or discharge and due to the surface charge phenomenon, these currents end up being present any time a pack has been charged or discharged. These currents present a significant challenge for managing paralleled strings. Because of the possibility of these currents, it is absolutely essential that each string MUST have a contactor, shunt trip breaker, or other automatic and redundant means of isolating the string from any other strings if a critical fault occurs. This is in addition to the standard over-current protection for the string. The designer must consider and ensure safe behavior in the event of a failure of any single component, including diodes, contactors, and BMS units."
Going the higher voltage route. Will be setup now in either 2p16s (48V) with 1 100A48V bms, or 2p8sx2 (With 2 100A 24V bms). Likely the 48V route.

PM'd about inverter stuff
 
32 Eve 280s, 2p4s x 4 (each with its own 100A bms), they will be connected together with 4/0 (under this 12V assumption)

Still going to try for either 24 or 48... Calling AC order tomorrow to see if a replacement is possible
That's a LOT of battery power!! What's going to be recharging it? There can't be enough room on the roof of a van to make a big dent in it. Dual 250 amp alternators on the engine?

Battery powered A/C in a van really does complicate everything. Propane for cooking and/or the refrigerator would lessen the load and make it possible to run a conventional A/C on Your 3000 watt inverter

By all means, forget about trying all this on 12 volts

Don
 
Back
Top