diy solar

diy solar

24 Volt 560 ah Lithium Battery Pack Build

chrisski

Solar Boondocker
Joined
Aug 14, 2020
Messages
5,087
I am building two 24 volt batteries, each with 8 S 280 ah Eve Cells. I've decided on two Overkill 8S BMS 24 volt 100 amps. I have a SAMLEX 3000 watt inverter, of which, I do not plan on running moe than 2000 watt loads, like a microwave for several minutes a day. I expect that and other loads to max out at 75 amps from the batteries. Looking for feedback.

I have quite a bit of time until these cells show up, so I've started to plan the build which will be placed in a 2' by 2' battery slide out tray. Below is a mostly to scale done by power point drawing of the wiring.

1625798289915.png
My 24 volt inverter will see a variety of voltages, from 28 volts charging, to 24 volts from the battery, to as low as 20 volts if the inverter is pulling a lot of power from the batteries. So when the panels are pushing amps into the battery at 28 volts and the inverter is running its full 3000 watts, I could see 123 amps from the battery with inverter losses. But to push the inverter to the same 3000 watts if the voltage is only 20 volts, I could see 172 amps from the battery with inverter losses. So to fuse for that 172 amps with some headroom, I come out with a 215 amps, of which available is a 200 amp class T or a 225 amp Class T. I'm going with the 225 amp class T.

When choosing the wires sizes, some were pretty easy like getting the two pair 8 AWG upgrade for the Overkill BMS. Not much of a choice about what to install. At an inverter DC cutoff voltage of 20, this is some of the data for voltage loss I could expect at max continuous production of 100 amps for these Overkill wires:

1625798846188.png
For the rest of the build, I compared both Ampacity of a wire to Voltage loss:
1625798939250.png
Now from each positive battery to the positive busbar opted for 2 AWG wires because ampacity is less than I expect to see.
1625798781842.png
After that, I looked at ampacity of the wires and wire loss, and ampacity was more of a factor than voltage loss. Joining the batteries became a bit of a search, especially because I'm such a fan of Blue Sea, but at $100 for a busbar, I just happened to have a 48 volt 250 amp pair of 4 post busbars left over from my last build. Not Blue Sea but the price is right and they are in specs.

Once, joined at the Battery bus bar, there could be 200 amps continuous coming from the batteries, even though now I never plan to see more than 75 amps total. To reach from the battery busbars to the main bus bars, I have up to 10' of cable to run. Probably less than 4 feet on the positive side although this will terminate at a fuse prior to the busbar and 6 feet on the negative side, although it actually will terminate at the a Shunt located before the busbar. Based off the ampacity table, I'm planning on using 2/0 wire. This is some of the voltage loss data:

1625799267482.png
Out of everything I listed, I did not quite make the ampacity of 200 amp with the 2/0 wire, so I will may drop the fuse to 200 amps, or more likely use the remaining 4/0 wire I have and I'd be comfortably in the ampacity.

Part of a possible expansion would be adding a third battery bank, but honestly I do not see a need for so much power in my RV. Unless I get a functional solar AC, which is my 2022 project.....
 
Last edited:
I think 200A fuse is fine, 2/0 not so much, I'd goto 3/0 - think about tomorrow, it always comes and is different. I was all 2/0 and had to change to 4/0 and OUCH copper is not cheap and luckily I had use for the 2/0.

remember to keep your + & - wires the same length and TOGETHER to prevent RFI/EMI noise & DC Ripple.
 
I think 200A fuse is fine, 2/0 not so much, I'd goto 3/0 - think about tomorrow, it always comes and is different. I was all 2/0 and had to change to 4/0 and OUCH copper is not cheap and luckily I had use for the 2/0.

remember to keep your + & - wires the same length and TOGETHER to prevent RFI/EMI noise & DC Ripple
I’m really starting to like the idea of 4/0 better. Even though now won’t have more than 75 amps max across this wire, if I do get a working AC off this next year, the start up surge could be quite. A bit more as the rotor spools up.

For the length, seems my battery positIves to busbar are of unequal length and I’ll change those. I’m going to look at turning the batteries 90degree counter clockwise and that will make the battery bus bar to main bus bar connection even, if it fits. The compartment is not a perfect square, and that little but may make a difference. RIght now there’s less than a half inch on either side of the batteries and compartment wall. Once tilted would be slightly less.
 
I went from 24V/3000W Inverter to the Samlex EVO 4024 and then I changed up from the 2/0 to 4/0 because I realized I can hit 250A Draw given the right combo of things... I can run my 3HP Compressor & MIG Welder so...
Panasonic 1200W Inverter Nuke eats 75A, the coffee maker 50A and with a couple of big things on it really does not take much. But like right now this moment I'm only using 6.2A and most often am not pulling much more during the day (without fridge running which adds 4A) summer or winter. Don't have AC, gave it away wasn't needed after I moved into my home thanks to the design & efficiency.

I did have to be smarter & tried with wires that were not equal (2/0) and realized quickly that it was a foolish notion but one which had a solution that worked out fortunately. My cell phone verified the Noise issue (never had signal in the powerhouse) till I bound my wires together. Now at least there is a "bit" of signal and it's clear when I get it.

IF you go shopping for an AC or other, they are available with SoftStart now and anyone on solar power with batteries should only use Soft-Start systems. There is little to no difference in cost. Avoid the surges whenever possible.
 
I'll be following this as I'm doing something very similar. My batteries are 230ah cells and the two BMS I ended up with are JBD 60 amp units. My inverters are a 1500 watt Cotek for the house basics and a 700 watt Cotek for the pellet stove. My build will be easier as it's stationary and I'm using an existing 21" rack with shelves currently occupied by my AGM batteries. My background is in RF and I'm intrigued by the idea of keeping the battery leads bound together to avoid RFI. I've always used a snap on ferrite choke on the positive lead close to the input of the inverter and that seemed sufficient.
 
Direct from Samlex EVO Inverter/Charger Manual.
1.3.4 Electro-Magnetic Interference (EMI) and FCC Compliance
These inverters contain internal switching devices that generate conducted and radiated electromagnetic interference (EMI). The EMI is unintentional and cannot be entirely eliminated. The magnitude of EMI is, however, limited by circuit design to acceptable levels as per limits laid down in North American FCC Standard FCC Part 15(B), Class A. These limits are designed to provide reasonable protection against harmful interference when the equipment is operated in a residential environment. These inverters can conduct and radiate radio frequency energy and, if not installed and SECTION 1.3 | General Information – Inverter Related SAMLEX AMERICA INC. | 15 used in accordance with the instruction manual, may cause harmful interference to radio communications. The effects of EMI will also depend upon a number of factors external to the inverter like proximity of the inverter to the EMI receptors, types and quality of connecting wires and cables etc. EMI due to factors external to the inverter may be reduced as follows:
• Ensure that the inverter is firmly grounded to the Ground System of the building or the vehicle.
• Locate the inverter as far away from the EMI receptors like radio, audio and video devices as possible.
• Keep the DC side wires between the battery and the inverter as short as possible.
• Do NOT keep the battery wires far apart. Keep them taped together to reduce their inductance and induced voltages. This reduces ripple in the battery wires and improves performance and efficiency.
• Shield the DC side wires with metal sheathing / copper foil / braiding.
• Use coaxial shielded cable for all antenna inputs (instead of 300 ohm twin leads).
• Use high quality shielded cables to attach audio and video devices to one another.
• Limit operation of other high power loads when operating audio / video equipment.
 
Chris, my 4s batteries are wired very similarly to your diagram. I didn't use a bus bar to gather the batteries together, but the net effect is the same.

My two batteries join at the shunt and at the Class T fuse. A single fuse is frowned upon by some. A fuse for each battery is better. But I went with the single fuse like you did.

With regard to your wire size, how often do you expect to be at 20v? My system is likely to never see 12.0v. Heck, I doubt it will ever see below 13.0v. If you plug in 24v (maybe even 26v) instead of 20v, how does that change your voltage drop? Enough to stick with 2/0?
 
I considered two fuses, but had not come across an example like that. My assumption is the BMS would shut off if amps exceeded 110 for 6 seconds or whatever the cutoff is. If I can see something supporting two fuses, I will definitely do this, but right now if you don't put fuses in a 2S2P battery config, then what's different for lithiums. Does a reputable lithium manufacturer like Battleborn say in their instructions to fuse each of their batteries when placing in parallel?

I will likely never see 20 volts, but I plan for amps as if I were. I guess I left this off the original post. These are amps I'm likely to see based off volts and I compare those to my ampacity charts
1625878553949.png1625878594743.png

So what the excel document on the left tells me is that at 20 volts, I see 172 Battery Amps to power 3000 watts, and should be fused to 215. I need wire at least thick enough to cover the fuse, otherwise the wire itself could be the fuse. This is why I'm coming up to 4/0 for a 225 amp fuse.

At Night, perhaps my batteries will be 24 volts. Alright, they're actually higher like 25 or 26, but I will calculate at 24 to err on the side of caution. I fully expect 24 will be the lowest I will see on this system if I pull 3000 watts at anytime of the day with reasonably charged batteries, but I need to do the calculations for the conservative side So at a nighttime resting voltage of 24 volts, I see 143 amps drawn to power a 3000 watt inverter which I need to fuse to 180 amps, which is really a 200 amp fuse, and then I'm at 3/0, but perhaps I can "get away" with 3/0

Changing that to a charging voltage of 28 volts, I see a 123 battery amps pulled to power 3000 watts, and need to fuse to 154 amps, which depending on the fuse size I have is 1/0 wire or 2/0 wire.

I found that for fusing the short wires in my system, voltage loss had little to do with the size wire I chose. It is all ampacity. If this were longer, 20' or 40' away, voltage loss plays a bigger issue. A well installed inverter / battery such as next to each other in the same compartment probably does not need to worry about voltage loss, but ampacity is an issue. Now change that around where someone routes there Battery to the front of the RV, but for whatever reason puts the inverter in the back, and now voltage loss will change.

Here's the voltage loss I looked at to decide for my build, at 24 volts and 20 volts, I do not consider loss a huge deal. The four wires I looked at were 4 AWG, 2 AWG, 2/0, and 4/0. Of course 4 AWG would be the most loss and at 200 amps and 20 volts it was a 3.11% loss, which I may consider acceptable under certain circumstances, but 4 AWG wire only topped out at 95 amps. At that point my wire is now the fuse. At 200 amps and 24 volts, voltage dropped to 2.59%, but the 4 AWG wire is still only rated for 95 amps, so that becomes the fuse.

20 Volt Data
========

1625879643462.png1625879657848.png


1625879671457.png
24 Volt Data
========
1625879499324.png1625879467015.png
1625879442397.png
 

Attachments

  • 1625878517866.png
    1625878517866.png
    109 KB · Views: 3
No. Not in conduit. Is there a different version of this voltage loss calculator? It’s not turning up in my google searches. Every time I search I get the one with the conduit.
 
I'm not familiar with that calculator. I was thinking that running the calculations, with the assumption that the cable will be in conduit ,changes the outcome. I use this calculator.

 
Amazing. Slightly different calculations. I was trying to build my own calculator on excel and stopped. I do think the differences between the one you I use and the one i use have to do with significant digits and rounding errors.
 
Is the inverter low or high frequency?
The Samlex high frequency inverters are ~90% efficient with no significant surge capacity.
The Samlex low frequency inverters are >90% efficient with substantial surge capacity.
 
Is the inverter low or high frequency?
I can’t figure out if it’s high our low. My guess is high frequency.

It’s the SAMLEX PST 3000-24. Some of the specs area:

UL458 compliant
3000 Watt continuous power
6000 Watt Surge @ 6ms.
>88% efficiency

CEF84FBE-D179-4856-BA0C-CDB7E50B4F72.png
 
Its high frequency.
Its very rare to see a dedicated low frequency inverter.
Surge output power is <8ms which is ~half a cycle of AC current.
Which means its only relevant to the marketing department.
Still its a good inverter.
I have the 1500 watt version.
 
It is HF inverter, it can do the surge for only 6mS, that is only about 1/2 of the cylce (1 cycle of 60Hz is 16.67mS). LF Inverter can handle surge in seconds.
 
After a little more thought these are the updated plans with 4/0 wire and tilting the battery 90° to make wire spacing better.

1626042348734.png
I also looked at an active balancer, but decided against it based off this video and the specs sent from the supplier saying they are pretty well balanced.



1626042257733.png
 
3000 ac watts / .9 conversion factor / 24 volts low cutoff = 138.888888889 dc amps
138.888888889 / .8 fuse headroom = 173.611111111 fuse amps.
4/0 awg is overkill.
You could get by with as little as 2 awg depending on the length of the wire run.
 
4/0 awg is overkill.
I agree that 4/0 is more than needed. What I'm trying to avoid also is the wires heating up under a constant load.

I'm getting this to run one high wattage appliance as much as I want, like to run the microwave for 10 minutes straight pulling 2100 watts from the battery, or to run Air Conditioning at 1100 watts from the battery all day long. So I don't expect more than 75 to 80 amps tops continuous draw. I also expect it being next year that I'll have the Air Conditioner set up to run on solar.

The problem I have is the other three that are with me are not as careful as me when it comes to making sure only one high wattage device on at a time, especially when I'm not around. So like Steve mentioned in post #4, the right combination of things can get my draws up quite high. I would expect them to forget and turn two high wattage items on at once.

The inverter I have has a non-adjustable 20 volt DC low voltage cutoff. I don't think I saw any stand alone DC to AC inverters that an adjustable cutoff. I doubt I will ever see it dip below 24 VDC, but the inverter is spec'd for 20 VDC, so I use 3000 ac watts / 20 volts * 1.15 AC to DC conversion factor = 172 dc amps, multiply by 1.25 to get a 215 amp fuse. At some point I will delve into the NEC code to see the proper formula.

Having wires not heat up to 105°C is a big thing to me as it looks like 2 AWG could. I wish there was a formula to calculate how hot these wires get, but the more I google it, the more I get things like are the wires in conduits, are the wires in still air or is there a fan and no formula assigned to help me put this together. These batteries are in my pass through compartment now, and I don't want wires in there reaching a temp how enough to burn when touched. If I get to the point I'm running the AC all the time, I'll need a way to either cool that compartment down or move the batteries and inverter to under my sink.
 
Well 4/0 may be over-wire today but will it be tomorrow ? Future proofing while also having little to no resistance thanks to the larger cables not a bad thing and you won't have to upgrade the expensive copper later (which is only getting more expensive).

Just a wee side-note. When I built my packs, I had the dilemma of what wire to use from the batts to the batt lugs, considering I have them setup to handle 250A output (175A for the smaller packs). I ended up using the 10" of 2/0 Wire with 5/16 Tinned lugs on the cells and 3/8's tinned lugs on the batt lug side.

Here is one of my 280 Packs setup with lugs to cell terminals.
280ah-with-qnbbm-jpg.30604


Here are my 175AH packs. Because of their cell sizing, I used Aluminium Bars to extend out from the cells which I was worried about at first but there was no cause for concern. Again 5/16 Lugs at the cell side to bar and 3/8's lugs for the batt terminals.
2x-175ah-8s-with-qnbbm-8s-jpg.30602


The Batt Pack tops are Optix 10mm Acrylic Sheet used with piano hinges.
 
Back
Top