64 eve lLF280k version3 cells in 8P8S configuation 24-volt build December 28th, 2023

[michael d]

there are 32 packages in the order. Please pay attention to the quantity when the packages are delivered
evidently >>>>2 cells per box.... being shipped out of the California warehouse. via Fedex
I am thinking ??? on no >>>> FedEx they never get the shipment on time or in one day and keep you wondering where the rest of the order is..... but have to wait and see if they surprise me with all of this shipment on one truck....or not???
UPS is more dependable on doing what they promise is my experience in the rural USA....???

 

[michael d]

http://www.watteredge.com/pdf/Ampacity_Tables.pdf

the bus bar ampacity table above and other resources could be helpful for high amp busbar design.

the one thing I question is >>>> just because a LiFePO4 ESS battery has a high amp potential when assembled.... if you are not drawing that amount of amps or ever intend to then >>> would you need the thick busbar >>> ????

7040-watt draw from the 25.6 volt LiPo4 battery is 275 Amps

How many amps can 4/0 Welding Cable handle?​

4/0 Welding Cable carries 310 Amps for 600 volt in-line applications.

4/0 Welding Cable is an extremely flexible cable with pure fine stranded copper. It contains an abrasion-resistant rubber EPDM jacket.

I realize the more amps drawn the hotter things get and the larger diameters allow an easier flow of the amps....

thanks for all....

I usually overdesign most things.

 

[michael d]

DIY busbars... 4/0 awg​

the above thread seemed relevant in this forum about busbar sizing

breakers are next on the research list for 64 cell 8x8 25.6 volt build

 

[michael d]

I found some tinned copper busbars for $42 each including shipping ( 4 inches by 12 inches x .25 inch thick with hardware with the holes already to assemble... hard to beat that price as they were brand new... so I bought two of them....
if that will help to keep that part of the assembly cool anyway...

 

[rhino]

I know you want to build it this way but maybe for others to think before following the same path.. having 8 cells in parallel with a single BMS lead to all 8 cells is effectively the same as someone having no BMS on a 24V battery and only using a volt meter to monitor it.

 

[740GLE]

I am trying to write this so you understand >>>> it is 8 cells connected in parallel first!!!! which acts as a single large cell and will be balanced by one of the small sense wires from the Electrodacus SBMS0...

then each of those large paralleled cells will be interconnected to the next set of 8 parallel cells in series, then the next in series, then the next in series etc... with the final series connections yielding 25.6 volts nominal...

54 series cable interconnections

then 16 more connections via copper cable to the busbar on each end (8 on each end)

yes, it is confusing but it has worked well in the 2P8S and 4P8S builds.....

I will increase the cabling sizes and the busbar sizes....
the factory bus bars work well for the parallel connections...
the series interconnections I use tinned copper lugs with short copper cables that I make up with a hydraulic crimper and use adhesive heat shrink to cover these lugs....

I may use 4/0 copper for the interconnections as I already have that cable and all the 4/0 lugs to fit....
it will take about 6 hours to make up those cables (approximately) (((((54 plus 16 is 70 cables to hydraulically crimp etc...)))))

64 cells at 5.4 kg weighs a lot >>>> 64x5.4 = 345.6 kg or (( for me 345 x 2.2 = 760.32 pounds)) Approximately not including the bus bars or cabling !!!!!!

the cells are connected in parallel 1st to make the 8P combined cell using 280Ah Eve Cells in this build..
64 cells will ultimately be an 8x8 block to make the 64 cell battery....

multiple inverters can be hooked to the same bus bars also,,,, and not the high overpriced Victron super dependable everyone raves about either....

I am still learning through many years of reading...//// studying.....DIY for 50 plus years farm boy for 60 plus years college 10 plus years.... still learning every day so suggestions are always read and studied before implementing....

but it still seems when one gets to adding up all the required charge controllers, pv combiner boxes, etc to control 18,000 watts of PV solar panels this still is doable a few parts at a time....

no combiner box is needed when assembling the 2 panels at a time in parallel at the solar PV array....

the LiFePO4 battery back-up is the most expensive part of my systems.... off-grid
I am not connected to the grid in any way on this....

That is for explaining it so my pea sized brain can understand.

Good luck with your build!

 

[toms]

I know you want to build it this way but maybe for others to think before following the same path.. having 8 cells in parallel with a single BMS lead to all 8 cells is effectively the same as someone having no BMS on a 24V battery and only using a volt meter to monitor it.

Not at all, it’s the same as someone buying an 800ah cell instead of eight 100ah cells.

I’ve seen a few setups using the very large cells, haven’t seen any problems with it.

You don’t monitor every single parallel segment of an individual 300ah cell, yet that also isn’t an issue. What do you see as the difference between segments in a 300ah cell, and segments in three 100ah cells in parallel?

 

[michael d]

my understanding is the 8 cells in parallel act as a single cell.
they are to be connected 1st. in an 8P configuration....
8P x 280Ah cell = 2,240 Ah ( a larger gas tank )
there is nothing saying you have to draw all 2,240 Ah at one time.... just that it is there...

every time you add the next series set of connections you essentially add another gas tank and the nominal voltage will increase >>>>

you can easily verify this as you assemble the multicell ESS with a multimeter (the increasing nominal voltage as each set of parallel cells are connected to the next set in series)

3.2 + 3.2 +3.2 + 3.2 + 3.2 + 3.2 +3.2 +3.2 = 25.6 volts nominal.

the cells in parallel kind of self-balance themselves (somewhat)
the BMS will help to balance the cells while the large battery is being charged from the solar PV panels...
The SBMS0 is a passive balancer.

some people use active balancers.... I do not at this time and have not for the past 3 years and have had no problems.... (knock on wood)

tesla battery packs are connected in the same manner >>> cells in parallel 1st >>> then they are series connected to get higher voltages....
so this is why I label them 2P8S (16 cells), 4P8S (32 cells); and the newest design 8P8S (64 cells) the cells are connected in parallel 1st... then series.

more than one inverter can be connected to each 25.6-volt battery build and 24-volt lights can also be directly connected to the batteries....

the HF inverters have a smaller standby draw (idle current) and are useful for light loads such as lights... IMO
the heavy LF inverters are better for the motors and the power tools requiring larger inrush currents to start and operate....IMO

thanks for all the replies....

 

[toms]

You are correct, the cells in parallel act as a single cell.

There is virtually no operational difference between a 300ah cell, and three 100ah cells.

 

[michael d]

happy New Year
7 degrees Fahrenheit this morning (January 1st 2024)..... winter is here.... the 8-10 inches of Christmas day snow is keeping it cold
the insulated off-grid solar power shed has 100-watt and a 50-watt light bulb keeping it warm from freezing...
working outside yesterday preparing for the 64 cells arrival..... it was 25 degrees and I was not liking that for more than a couple hours at a time...
the 32 boxes (64 cells)of the new version (version 3) LF280K LiFePO4 cells will have to sit on a pallet until I can start to thaw them for commissioning and charging (top balancing) etc inside where I prefer to be in the winter....
the PV Solar panels liked the cold and partially sunny weather yesterday though....!!!!

 

[Steve_S]

YES, 8 Cells in Parallel "appear as 1 cell" but are not in reality. Every single cell has its own personality if you will, and should one within the set be out of spec, the "set" is affected BUT due to paralleling this will not be too obvious "at a glance" and over time this can cause issues.

There is one thing MISSED - an LFP cell has the ability to DUMP @ 5C in a Short Condition. That is a 1400A Burst Dump ! from 1 cell. Do not underestimate the stored energy inside those polite blue cans. See EVE Specs, its not a joke.

Southwire's "Royal Excelene" Fine Wire Welding Cable table. (some of the best wire avail).

 

[toms]

YES, 8 Cells in Parallel "appear as 1 cell" but are not in reality. Every single cell has its own personality if you will, and should one within the set be out of spec, the "set" is affected BUT due to paralleling this will not be too obvious "at a glance" and over time this can cause issues.

There is one thing MISSED - an LFP cell has the ability to DUMP @ 5C in a Short Condition. That is a 1400A Burst Dump ! from 1 cell. Do not underestimate the stored energy inside those polite blue cans. See EVE Specs, its not a joke.

Southwire's "Royal Excelene" Fine Wire Welding Cable table. (some of the best wire avail).
View attachment 186191

Do you really believe that every layer inside a single cell (every section of the “jelly roll”), has the exact same characteristics?

They don’t, and the capacity of a single cell is limited by its weakest section.

Putting multiple cells in parallel extends the same limitation over all the cells in a group.

Your assumption that each section of an individual cell is identical is where you are mistaken.

 

[michael d]

right now I am acquiring NH3-type fuses and fuse blocks (made in Germany) for use between the battery and inverter....
the plan is to stay at under 275 amps draw or less depending on what inverter is pulling what load... on the 25.6-volt nominal build....

yes, the little blue cans do have my respect....

I use 4/0 cables to connect to the inverter posts on the LF inverter.... with a 400 amp DC fuse in line ( I call it the castrophy fuse ).

 

[michael d]

they make these in various amp ratings and are military grade,,,, perhaps an improvement over the class T fuses I have been utilizing...
at any rate, the idea is to do all the off-grid electrical installation safely....
I ordered 3 of the 160 amp ones (nh3-type fuses) to begin and trying to source some other sizes also....

 

[rhino]

Do you really believe that every layer inside a single cell (every section of the “jelly roll”), has the exact same characteristics?

They don’t, and the capacity of a single cell is limited by its weakest section.

Putting multiple cells in parallel extends the same limitation over all the cells in a group.

Your assumption that each section of an individual cell is identical is where you are mistaken.

I don't get this.. you are saying it is better to spread a problem across multiple cells until it becomes a serious enough issue to affect all of them instead of being able to detect the one cell having the issue as soon as possible?

 

[toms]

I don't get this.. you are saying it is better to spread a problem across multiple cells until it becomes a serious enough issue to affect all of them instead of being able to detect the one cell having the issue as soon as possible?

No, i’m not saying that. I’m sorry you don’t understand.

 

[740GLE]

I don't get this.. you are saying it is better to spread a problem across multiple cells until it becomes a serious enough issue to affect all of them instead of being able to detect the one cell having the issue as soon as possible?

I think he’s saying by doing 8p you put all your eggs in one basket, many people would rather have multiple banks of BMS controlling smaller number of cells, so if something goes south you just isolate a small % of the battery bank not the entire one.

But adding more BMS adds more costs so it’s a trade off each owner must weight their options.

 

[michael d]

As I read about all the problems and non-functional BMS's in the 48-volt 16-cell variety which would require 4 separate BMS's for the same amount of Wh vs one Electrodacus SBMS0 in the 8x8 64-cell design....

and all the non-compatible prebuilt battery units and inverter incompatibilities etc

the 8x8 64cell ESS build using one Electrodacus SBMS0 still looks like a good idea... and has to date been a good system for my off-grid farm.

as I have 4 of them already functioning in the 2P8S (16cells) and 4P8S(32 cells) configurations without all the problems that are routinely run into with the other setups all over the internet with unsolvable problems...

my redundancy is the other 4 multicell ESS batteries that are already in daily operation.

yes, with a small 16s battery with 25% of the Wh you may find the loose connection or faulty cell faster.... ( I can see that line of thought)
I have never had a faulty cell so far>>> knock on wood
the issue usually was a loose connection in all my electrical work
I bought torque wrenches specifically for that issue but over torquing them was more a concern than being loose...
just like in water compression fittings ( the water leak would be due to the under-tightened or over-tightened fitting) due to fear of ruining the fitting by it being too tight...

Mike G does the high voltage route with an (I forget but ) approximate 280-volt battery in series without the BMS but used active balancers...
that higher voltage I am not ready to deal with but I had an interest in it, unfortunately, the 30k low frequency"Sandi" inverter alone was a 7000 dollar investment for the high voltage route... as I was actively pricing those inverters...

and the real bottleneck (((super expensive part in a off-grid build))) is the energy storage for use at night and when the sun does not shine for extended periods of time...

 

[michael d]

if the cells get out of balance after a couple of years they can be balanced using active balancers or the Riden 6018/6012/6024 and LiFePO4 dedicated bench battery chargers ...
The LiFePO4 ESS will be cycled in the 20 to 85 percent ranges >>>> and the solar PV panels have been doing that without any active balancers without issues...

thanks, for all the replies and suggestions/ etc... all are welcome to better understand the issues and potential pitfalls...

 

[Partimewages]

As long as you know what your getting into!