advise on design

S

Scottietheyoung

Member
Joined
Oct 10, 2018
Messages
79
Hi guys,

I am after somwhere between 40-60kwh for my pack (aiming for the moon) and after some recomendation on the size of the banks.

have a solax 5kw inverter and charger. seems 14s is the way for a 48v system but what is the advantages/disadvantages of 100P vs 80P vs 150P??

keen on your advise.

Also thinking about storing them with cells in vertical position (to aid in cooling) with a side by side pack configuration (will upload a pic of my thoughts later)

Bus bar terminalswill need the same wire length where they connect? if so some funky chicken wiring will need doing to go side by side. I have the space and wonder if the passive cooling advantage would make it worthwhile?
 
The total "P" term is dictated by your bank capacity as you've fixed your system voltage at 14S (48v).

So 60kWh / 48V = 1,250Ah.
1,250Ah / 2.6Ah = ~480 cells. Assuming 2.6Ah (2600mAh) per cell.

14 x 480 = 6,720 cells. About 300KG worth, plus connections and mounting, so perhaps 350KG all in.

You can arrange this in one large bank so you only need one BMS, effectively giving you 14S480P, or you could split this up into strings, each with their own BMS system (at greater cost). Ie 4 strings of 120 cells as 14S120P with 4 BMS systems. The 4 strings only connect at the very top and bottom of the pack.

Multiple strings would give you redundancy allowing you to take a string out of service but leaving the others in the system. Single string is cheaper and simpler.

Either way, making 14 physical blocks of 480 cells wouldn't be much fun, so you will make smaller packs and then electrically connect them with wire to make a 480P pack.
 
Scottietheyoung said:
seems 14s is the way for a 48v system but what is the advantages/disadvantages of 100P vs 80P vs 150P??

Also thinking about storing them with cells in vertical position (to aid in cooling) with a side by side pack configuration (will upload a pic of my thoughts later)

Bus bar terminalswill need the same wire length where they connect? if so some funky chicken wiring will need doing to go side by side. I have the space and wonder if the passive cooling advantage would make it worthwhile?
Click to expand...

The larger the packs, the harder to move and the more difficult the work to assemble. You also need the space for it, obviously.
The smaller packs are easier to manage and quicker to build.
Mike actually has a really large P arrangement. I just don't recall how many. I know it's well over 100.

The wire lengths generally need to be about the same, yes. But they don't have to be "exactly" the same length. Difference in a few inches is fine. The thing to make sure of is that the connections on the packs are balanced. The central bus bars are equal distance from cells. This will come into place when large current draws are applied.
Mike and Pete both have adopted a variation, where they parallel 5 cells together, then fuse that small pack to the bus bar. This is fine as long as the connection from cell-cell in that small pack is large enough to keep them equally balanced as well. So you wouldn't use a 26awg wire to connect them as it would be too small.

Passive cooling is fine. Generally speaking, if you have enough cells together, you really shouldn't have a heating problem. If you have 100A charge, and 200 cells in parallel, then each cell is only getting 500mA. If you multiple strings together, say 5 strings total, then at a 100A charge, each cell is only getting 200mA per cell. As long as you keep each cell below 1A, you really shouldn't have a cooling issue.
Depending on where you live, you might have a 'heating' issue, though. Cells don't like being charged/discharged below 0C/32F and can shorten their life span. So if you are in a climate where this can occur, some have actually designed their sheds/boxes/etc so that the inverters/chargers while running actually keep the area warm enough for the cells to operate safely.
 
When you are trying to get up in the terms of capacity I would personally look at larger batteries than an 18650. Something like a Leaf pack or like electric car that has been wrecked can get you up there in kWh and not be really tough to manage.

When I go properly off grid when the kids leave the nest I will be looking at minimum 100Ah Single cells There are Calb cells or the LTO cells too.

I currently have about 5kWh of build 18650's and that's 560 cells (7s80p) then I have another 1120 cells tested and ready to build for an additional 10kWh and that is about as big as I want to go on such small cells. too many places for things to go wrong.
 
Korishan said:
Scottietheyoung said:
seems 14s is the way for a 48v system but what is the advantages/disadvantages of 100P vs 80P vs 150P??

Also thinking about storing them with cells in vertical position (to aid in cooling) with a side by side pack configuration (will upload a pic of my thoughts later)

Bus bar terminalswill need the same wire length where they connect? if so some funky chicken wiring will need doing to go side by side. I have the space and wonder if the passive cooling advantage would make it worthwhile?
Click to expand...

The larger the packs, the harder to move and the more difficult the work to assemble. You also need the space for it, obviously.
The smaller packs are easier to manage and quicker to build.
Mike actually has a really large P arrangement. I just don't recall how many. I know it's well over 100.

The wire lengths generally need to be about the same, yes. But they don't have to be "exactly" the same length. Difference in a few inches is fine. The thing to make sure of is that the connections on the packs are balanced. The central bus bars are equal distance from cells. This will come into place when large current draws are applied.
Mike and Pete both have adopted a variation, where they parallel 5 cells together, then fuse that small pack to the bus bar. This is fine as long as the connection from cell-cell in that small pack is large enough to keep them equally balanced as well. So you wouldn't use a 26awg wire to connect them as it would be too small.

Passive cooling is fine. Generally speaking, if you have enough cells together, you really shouldn't have a heating problem. If you have 100A charge, and 200 cells in parallel, then each cell is only getting 500mA. If you multiple strings together, say 5 strings total, then at a 100A charge, each cell is only getting 200mA per cell. As long as you keep each cell below 1A, you really shouldn't have a cooling issue.
Depending on where you live, you might have a 'heating' issue, though. Cells don't like being charged/discharged below 0C/32F and can shorten their life span. So if you are in a climate where this can occur, some have actually designed their sheds/boxes/etc so that the inverters/chargers while running actually keep the area warm enough for the cells to operate safely.
Click to expand...

jdeadman said:
When you are trying to get up in the terms of capacity I would personally look at larger batteries than an 18650. Something like a Leaf pack or like electric car that has been wrecked can get you up there in kWh and not be really tough to manage.

When I go properly off grid when the kids leave the nest I will be looking at minimum 100Ah Single cells There are Calb cells or the LTO cells too.

I currently have about 5kWh of build 18650's and that's 560 cells (7s80p) then I have another 1120 cells tested and ready to build for an additional 10kWh and that is about as big as I want to go on such small cells. too many places for things to go wrong.
Click to expand...

Thanks heaps for the input guys.

I have an internal garage for the battery system and feel confident in the fire safety of these systems after seeing how they are running in other applications. The garage never gets below 5C in winter as its fully insulted (better than our house actually.

I have a really good supply at no cost of 18650's so hence the desire to build that big with them. the leaf packs are a poor lifecyle chemistry option compared to other larger cell options. even still the used EV pack route is very expensive ultimately.
 
Modular is the way to go with a big system.
That way you can take a pack off line without taking the whole system off line with it.
Figure out your plans & space to build it & this will help you work out pack size.
 
So I am thinking of running 14S 120 P and have 4 strings to make up the size required.

My question I have is around the connection of the strings? Has anyone else got a thread with this many strings attached and pics of their design/ build?
 
Just go to y-boob lookup, average joe, HBpowerwall, jehu, many other 18650 related video's.
You'll find them,
 
Yeah.

Please, DO NOT CHARGE BELOW 0C.

It is extremely dangerous for your cells to do so, as it will kill them extremely quickly.
 
I would do an n+1 system. Ie 3 strings minimum where it can easily run on 1 string at max load and the last you can remove if needed.

With that in mind beware of that bigger strings = a bit harder to maintain but less strings = cheaper in terms of hardware around them like cables and BMS.

Less strings = equal larger packs to move and work with.
 
I watched many youtubes and noticed 80p packs etc. I eventually went with 60p for various non-technical reasons - which happened to be 130ah (average 2166mah per cell). Next I did 7s for 24v nominal for a while... and made sense for smaller (12/24v 60amp range) charge controllers. After a few months I began to understand that this is pretty small in terms of powering a house.

I kept the 60p pack but went 14s in conjunction with upgrade to a Midnite Classic 150s and set of 15 panels and 48v@3000 watt inverter. 48v reduces the amps (e.g. half as many as 24v) and matches the more serious (15-30 panels) and 80a + charge controllers etc. However, 48v requires city permit/conduit etc. When you get above 12/24v (depending on jurisdiction) you get into enough volts/amps to be 'more dangerous' and for my location that was 48v level.

Then I went Batrium which requires longmons. Longmons can handle 260ah (or even larger) packs so I still kept the 60p pack design... but doubled them up so effectively I now have 120p 14s and 3 of them which is about 40kwh. So making the packs modular is good! However, I now need 1,680 cells per '48v battery' - to add a 4th battery I have to buy/process 1,680 cells.

I happened to make my packs using 4 of those 4x5 cell holders - with blank spots on each end to protect the +/- pack connection posts. This allowed me to vary the number of cells but keep the same physical pack size and connection posts so they are all interchangeable. Some cell were 2900mah and others as low as 1900mah... so some packs have 45 cells (for 130ah) and some have 68 cells (for 130ah). Also, by leaving blank slots in your pack design, you can add a couple of cells to week packs later on once Batrium exposes them.

I like the comment about going with Nissan Leaf G1 batteries - price competitive to 18650 and so much less work that soldering up 5000 cells and I would have gone that way knowing what I know now... and prices now.

However - decisions can kind of lock you in. For example I could add Nissan Leaf cells to my 18650 bank OK but I would need a 2nd Batrium as these are 2s2p 'units' so they have different longmon/cellmon settings. So my 60p 18650 original decision is core to my battery system. Similarly, going with Tesla car batteries puts you in the 6s / 12s relm.. which limits inverter choices a little bit as the 7s/14s is more aligned with mainstream 24v/48v systems.
 
You must log in or register to reply here.
Back
Top