Design Check: 12Vdc portable system

[sallaben]

Hi! I'm designing a portable 12Vdc system and still in the planning stages. I'm starting with a smaller battery bank and hoping to build a truly portable system that can be used in many settings: to reduce my electrical bill while on-grid, to provide instant backup power with a transfer switch during grid outages, as well as enabling minor conveniences on car-camping trips.

The first part of my system will be the battery bank, 12VDC distribution and 120VAC distribution. The inverter I plan to use is an inverter/charger with a transfer switch. I'll work on solar and charging from DC later.

Is anyone willing to take a look at my design and point out any potential improvements and/or glaring flaws? I am new to working with power but always willing to learn. If there is any problem with the image below, please let me know.

One outstanding question I have is the size of busbar I should use. Will 250A cut it or should busbars always match/exceed the preceding fuse's amperage limit?

Thank you and have a great rest of your day!

https://imgur.com/ESzTxYA

 

[Rednecktek]

A couple initial thoughts:

A 30a SCC/charger is only going to get you a max of 360w of charging to your battery bank, at 270Ah * 12v = 3240Wh / 360w = 9 hours to charge the battery, or 2 average days of sunlight per full cycle of the batteries. That initial 80a charging from shore power will be fine though.

270Ah sounds a lot like a lead acid flavor of battery, you are aware that you can only discharge a lead battery to about 50% right, cutting you to 1620Wh but only needing a full day of no load draw to recharge. Is LFP an option?

The wire size between the batteries and inverters is going to be HUGE with that inverter wanting to pull 192a, that's welding wire size. Just to throw it out there, with your 12v loads is a 24v system with a hefty 24v->12v converter an option?

I'm assuming your 12v Load Center is a fancy way of saying fuse block?

Yeah, as a basic design it looks pretty good. I would step up the size of the MPPT controller at least so you can get a realistically usable charging rate without having to buy a 30a and then another 60a later on, but that's just me. As long as you're good with working with the larger wires I really don't see any problems.

 

[Bud Martin]

The small gauge wire used for powering the Shunt monitor should have small in-line fuse so if the monitor has shorted circuit the small wire will not catch on fire.

 

[sallaben]

A couple initial thoughts:

A 30a SCC/charger is only going to get you a max of 360w of charging to your battery bank, at 270Ah * 12v = 3240Wh / 360w = 9 hours to charge the battery, or 2 average days of sunlight per full cycle of the batteries. That initial 80a charging from shore power will be fine though.

270Ah sounds a lot like a lead acid flavor of battery, you are aware that you can only discharge a lead battery to about 50% right, cutting you to 1620Wh but only needing a full day of no load draw to recharge. Is LFP an option?

The wire size between the batteries and inverters is going to be HUGE with that inverter wanting to pull 192a, that's welding wire size. Just to throw it out there, with your 12v loads is a 24v system with a hefty 24v->12v converter an option?

I'm assuming your 12v Load Center is a fancy way of saying fuse block?

Yeah, as a basic design it looks pretty good. I would step up the size of the MPPT controller at least so you can get a realistically usable charging rate without having to buy a 30a and then another 60a later on, but that's just me. As long as you're good with working with the larger wires I really don't see any problems.

Thanks Tek I appreciate the detailed response.

> 270Ah sounds a lot like a lead acid flavor of battery, you are aware that you can only discharge a lead battery to about 50% right, cutting you to 1620Wh but only needing a full day of no load draw to recharge. Is LFP an option?

It's actually a 270Ah LFP unit, which I had never seen before either. All the cells in one case too!

> A 30a SCC/charger is only going to get you a max of 360w of charging to your battery bank, at 270Ah * 12v = 3240Wh / 360w = 9 hours to charge the battery, or 2 average days of sunlight per full cycle of the batteries. That initial 80a charging from shore power will be fine though.

Solar is one area I have not yet planned out, so I appreciate the pointers. I agree with your assessment. You suggested 60A further down in the post. While I don't plan to full discharge every day, would a 50A/600W charge controller be enough by your standards in this scenario? That works out to a theoretical 6 hour charge time. 60A would be 4.5 hours (which, maybe that's the difference between comfortable full charge and living Wh to Wh.)

> I'm assuming your 12v Load Center is a fancy way of saying fuse block?

Yep, just something to distribute the 12vdc to multiple points with fuse in between. I've heard it called a "DC subpanel" before, is "fuse block" a better term?

> The wire size between the batteries and inverters is going to be HUGE with that inverter wanting to pull 192a, that's welding wire size. Just to throw it out there, with your 12v loads is a 24v system with a hefty 24v->12v converter an option?

I considered the step down converter, but felt that it would be overkill for such a small capacity system. Maybe if I increase my bank size in the future.

As for wire size, I was just planning on using 2/0 gauge wire over short distance, while expensive it's only a foot or two for each segment. I'm starting to push the limits of 12V I suppose, but the availability of 12V components has me pretty certain that I want to stick with it. I am curious, how did you calculate 192A?

 

[sallaben]

The small gauge wire used for powering the Shunt monitor should have small in-line fuse so if the monitor has shorted circuit the small wire will not catch on fire.

Appreciate it Bud! I'll double check that a fuse is between the monitor and power source. Is that just the case for the shunt monitor, or does it apply to the inverter monitor too? If not, I am curious why?

 

[Bud Martin]

Any device that has power cable to run it, then the fuse should be chosen and used to protect the wires from catching on fire.
Is there another Volt meter or shunt monitor in that drawing? I see 'Monitor' in the circle with the line to the inverter/charger, does that monitor have small power wire for it to run the monitor? If it does then you need fuse for that too. What is the "Monitor'?
Basically if you have shorted circuit, the wire should not catch on fire.

 

[Rednecktek]

Thanks Tek I appreciate the detailed response.

> 270Ah sounds a lot like a lead acid flavor of battery, you are aware that you can only discharge a lead battery to about 50% right, cutting you to 1620Wh but only needing a full day of no load draw to recharge. Is LFP an option?

It's actually a 270Ah LFP unit, which I had never seen before either. All the cells in one case too!

Aaahhh, never seen those before. Good call!

> A 30a SCC/charger is only going to get you a max of 360w of charging to your battery bank, at 270Ah * 12v = 3240Wh / 360w = 9 hours to charge the battery, or 2 average days of sunlight per full cycle of the batteries. That initial 80a charging from shore power will be fine though.

Solar is one area I have not yet planned out, so I appreciate the pointers. I agree with your assessment. You suggested 60A further down in the post. While I don't plan to full discharge every day, would a 50A/600W charge controller be enough by your standards in this scenario? That works out to a theoretical 6 hour charge time. 60A would be 4.5 hours (which, maybe that's the difference between comfortable full charge and living Wh to Wh.)

50 would be much better but I think the sweet spot is the EPEver 60a units for about the same price. Since you always want a little overhead for cold voltage spikes the 200v or 150v units buy you a little headroom. Remember that if you have a cloudy/krappy day or two before your panels get to actually do anything you're going to have 3 days of discharge to make up and if you can do that in 1 day it's better than doing it in 3 days.

> I'm assuming your 12v Load Center is a fancy way of saying fuse block?

Yep, just something to distribute the 12vdc to multiple points with fuse in between. I've heard it called a "DC subpanel" before, is "fuse block" a better term?

Yup, fuse block is the more common term.

> The wire size between the batteries and inverters is going to be HUGE with that inverter wanting to pull 192a, that's welding wire size. Just to throw it out there, with your 12v loads is a 24v system with a hefty 24v->12v converter an option?

I considered the step down converter, but felt that it would be overkill for such a small capacity system. Maybe if I increase my bank size in the future.

One thing to consider is that most pre-built batteries only have a 100a discharge rate from their BMS's so at 12v that's only 1200w of continuous power from a single battery. To push that 2Kw inverter you're going to need at least 2 batteries in parallel to feed it. Going to a higher voltage buys you more wattage for the amps. I.E a 100a BMS feeding a 12v system theoretically maxes out at 1200w, whereas at 100a on a 24v system it's maxing out at 2400w. If you were to jump to 48v right out the gate that would net you 4800w of max inverter before things start giving up the greasy orange smoke.

As for wire size, I was just planning on using 2/0 gauge wire over short distance, while expensive it's only a foot or two for each segment. I'm starting to push the limits of 12V I suppose, but the availability of 12V components has me pretty certain that I want to stick with it. I am curious, how did you calculate 192A?

According to everyone's favorite Blue Sea Ampacity Chart, 2/0 should be fine, it's just thick and heavy. If you want to increase the inverter capacity though you're either going to have to re-wire all your main wire or go to a higher voltage where a step-down would be useful. At 2000w you're really pushing what 12v can do.

As to the math, my basic napkin math formula was 2000w * .85% efficiency = 2300w / 12v = 191.66 amps. Your fuse should therefore be 240a or so for optimal performance. At 300a you risk the wires lighting off before the fuse blows.

The general consensus in the solar world is that unless you really need a lot of 12v load OR you have physical space limitations, going to a 48v system from the start buys you a lot of headroom for both performance and expandability as most anything over 3Kw is going to be calling for 48v and almost every rackmount battery out there nowadays is built as a 48v unit. The 12v and 24v rackmounts have come and gone for the most part and the rackmount form factor is really becoming the most popular option, so going with anything else may end up limiting you in the future for expandability just on the grounds of availability.

Did that make sense?