I’m assuming that 10S4P above means you have 4 sets of 10 cells? How do you tie the packs together? Do you charge all packs and then just close your eyes and attach the wires? Do you discharge all packs? I would think that any voltage difference between the packs could make for some pretty dramatic correction when you connect them?

Or is it not a concern? I ask because I just took delivery of my first set of 12 18650’s for a 12S1P but I could see a 2P in my near future.

Yea u can have a spark when connecting the packs of they are slightly different voltages. U could integrate the parallel connections within the series chain and avoid sparks that way. I’ll post a pic what I did later tonight.

Endless sphere for sale section has the best deal on awesome 3500mah cells but they are 10amp cells so you’ll need to make a bbbbiiiggg pack otherwise suffer possible lower cell cycle life and voltage sag.
Otherwise I think the of hg2 is a great choice. For smaller packs you’ll need higher discharge cells of
Course and they’re only possible in 2500mah

@DougM [quote=“DougM, post:1, topic:3032”]
I’m assuming that 10S4P above means you have 4 sets of 10 cells?
[/quote]

Not quite; it means you have 10 sets of 4 cells wired in paralell. Then you connect those 10 sets (with 4 each) in a series to increase voltage.

Ideally the cells begin with the same voltage and the same capacity, and then a BMS balances them during charging. You can test them with a multimeter before you connect them to makes sure there are no duds and to assure that they’re already close to balanced.

What if you did have 4 sets of 10 cells (10 cells in series then parallel them with 3 other sets)? Would you just have a pack that can’t be balanced by a BMS?

I picked up the LG HG2’s since I read somewhere they were pretty solid cells. The packaging says they are certified OEM but I’m a little suspicious because they are rated 3000mAh but when I drain tested them at 1A I got slightly less than that. I would have expected slightly more out of a brand new cell.

Thanks for the input - it does make more sense to do it the way @treenutter describes - 4 parallel then tie those together in series. That way the potential V delta is much much less than 12 series tied to another 12 series. I’ll do that

so far I’ve been balancing my LiFePo’s during charge with a balance charger, which is easy since I have a few very large cells, but I guess it’s time to start looking at BMS for this transition.

@Jinra OK I’ll take a shot at answering this, but I’m not the resident battery expert! Essentially, you’d have a 4s10p pack that has a discharge rate that is too high for our purposes. Let’s say you used Samsung 25R 18650 cells. The specs are 20A(continuous),100A(burst) 3.7V, and 2500mAh.

So if you built a pack as you described above, the pack would have a continuous discharge rate of 20Ax10= 200A, 10,000A burst rate, a capacity of 2500mAhx4=10000mAh, and a voltage of 3.7x4 = 14.8V.

So essentially you’d have a pack designed to fry your ESC

@treenutter He is talking about series and then parallel. Don’t put cells in series first. There is a whole argument on ES on why you shouldn’t, something to do with science. Parallel and then series.

It makes sense - the voltage differential of any 2 fully charged cells is maybe max 100mv but the voltage differential of 10 or 12 in series would be 10 or 12 times that much, so when you attach them in parallel they will try to equalize instantly, causing a lot of smoke and grief.

This is what I was trying to understand in my initial question

The only other option would be to attach the series packs via some very large resistors so they equalize more slowly.

A super simple way to visualize your battery pack is to look at the parts in terms of cells.
For a 10s4p pack, you will need 10 packs, wired in series. 4 batteries wired in parallel each pack.
So visualize your 4 battery parallel connection as one “cell”. This will determine your packs overall capacity in Mah as well as the total amperage that you can safely draw from the pack. More capacity, more amps!!
Once you have your “cell” planned out, you need to bring the overall pack voltage up to the desired level to efficiently operate your motors and ESC. Running 10s or 10x4parallel cells will give you about 36v. As long as you follow this formula, you can build any size pack at any voltage you desire.
Have fun, but be safe too!!

Hey guys, I was just on here looking for some battery build tips, and I’d like to point out that this comment from treenutter :“Not quite; it means you have 10 sets of 4 cells wired in paralell. Then you connect those 10 sets (with 4 each) in a series to increase voltage”… is wrong according to Jason Potter’s (onloop) site.

Regarding the Spacecell 10s4p he states:
The battery is 10S4P meaning it has a total of 40 Cells, 4 Parallel groups of 10 cells in series. Samsung 25R Cells.

So now as Ali G says “I’s well confused!”

10 Cells of Samsung 25r’s in series (nominal voltage of 3.6 per cell at 2500 mAh) would produce a single pack of 36volts and 2500 mAh (or 2.5wh) correct?

Wiring 4 of those 10 series packs in parallel gets me 36volts and 10,000 mAh (10wh) of power, correct?

Conversely, 4 of the same cells wired in Parallel gets me a 3.6volt pack with 10,000 mAh of power. 10 in series gets me to 36volts at 10,000 mAh of power.

So my question is, is the advantage to have more packs for a BMS to manage? The more micro-managing that can be done the longer the pack lasts?

Or am I missing an advantage in Discharge Rate? From what I’ve read, Samsung 25r cells have a 5c discharge rate. Thats 5 times the capacity of 2500, or 12.5 Amps. Multiply that times the number of parallel groups correct? This would be a max “c” discharge rating of 50Amps for the battery built with 4 parallel groups of 10 in series, and the same for the pack made with 10 4 cell parallel packs in series? Just remembering that cells in series increase voltage and cells in parallel increase capacity.

It seems to me there is no advantage either way? Am I missing something? I’d like to get started on a build but the conflicting information out there has me confused a bit…

@Racetored apologies if my comment caused any confusion; I think it’s a problem with “order of operations” when you read my statement. The way I read it, My comment and Jason’s are the same. Let me try to break it down:

“4 cells wired in paralell:” - each one of these is 4 cells wired in paralell:

in a 10s4p pack you have ten of these paralell packs, lined up like this:

…and from here you make the serial connections to bridge together all the packs:

Imagine my quote with these parenthesis :
“…you have 10 sets of (4 cells wired in parallel). Then you connect those 10 sets (with 4 each) in a series to increase voltage”

So parallel increases capacity, series increases voltage.

In the case above (10s4p) you get 4x the capacity of the cell (2500mah) to equal 10,000mah. Then you multiply the voltage of of each pack (4.2V max) by the number of packs in the series (10) so you get 42V max.

You could also express the voltage as nominal voltage, then it would be 36V.

Hope this helps, and I’m very open to being corrected so could someone pls double-check me?

I totally get that. Jason clearly says 10 cells in series in 4 parallel groups. After doing the math it reall seems like “half a dozen of this, six of the other.” The math all seems to add up the same, I was just looking for some clarity as to if there was a benefit to one over the other. There was an earlier comment here that a pack built on 4 parallel groups of 10 cells in series would produce a battery with too high of a discharge rate.

Btw, thanks for those great photos! Is there somewhere here with a detailed bms wiring description?

So, just to clarify a wiring diagram…this pack would be a 10s3p, or this would be a 3s10p? From what I gather, this is the wiring diagram used in the 10s3p Spacecell. Is this right anyone?

I’d love some clarity on how to wire a BMS as well! I have a charger for my RC helicopters that has balancing functions, but I don’t think it can go up to 10s.

I just realized that the parallel connection between pack 1 and 2 is wrong …I would just flip the arangement of the first 10 cell pack…I am aware but this is the idea…