First Battery Build 21700 Molicel P42A

kudos

10 kW
Joined
Aug 2, 2011
Messages
629
Location
Guernsey, Channel Islands, UK
After much deliberation I'm going to build my first cylindrical cell battery.

I've been using LiPo very happily for 12 years so I'm very familiar with putting them together using 6S packs.

My current 32AH battery made from four 16AH Multistar LiPo is still going after four years, cells never our of balance.
They stay this way because I charge at 6A and LVC set at 3.6V.
However, after 6k miles they are starting to get a bit puffy so I'm looking for my next pack.

If LiPo was readily available I probably would have stuck with what I know, but large LiPo packs are perma-unavailable so I'm going with 21700's.

I've ordered 72 Molicel P42A 21700 which I'm planning to put together in 12S 6P

I already have a charger for this voltage 50.4V

I've picked up 21700 spacers and I also have Nickel strip coming in 0.15mm and 0.2mm thickness', as well as 21700 size positive insulators and kapton tape.

I wanted to use the ladder style nickel strip but I could only seem to find 25 & 27mm widths which appear to be 18650 size, although I don't know for sure.

I picked up a cheapo spot welder powered by a small internal Lipo battery, I've tested with 0.10mm nickel and seems to work fine. The high power mode should be fine for 0.15mm although we'll have to see when it arrives.

Having never used a BMS and having read countless threads of BMS's ruining batteries, I'm on the fence about using one and would appreciate input from anyone who runs 18650 / 21700 batteries without a BMS.

I appreciate it's the norm to have a BMS and having never run a cylindrical cell battery before I don't know how they behave, so a BMS might be a must.

I would prefer to stick with what I know, make two 6S6P modules and combine them up. I can easily balance the cells (if required) in a number of ways including my ichargers.

This first battery will be my learning curve battery, I intend to follow it up with a very large capacity unit.
I'm in the 'you can't have a big enough capacity battery' camp.

I'll document my build here and hope to gain from the wisdom of those who have gone before.

Kudos
 
https://www.18650batterystore.com/collections/21700-batteries/products/molicel-p42a
Looked up your cell 5.99 and 45amp 6p 270 amps. What buss bars.
I build my batteries big enough and don't let anyone use my stuff so I don't need a BMS just balance wires and thru balance wire charger.
 
I have been using the 18650 VTC6 Batteries without BMS for a while now.
I had some made and some I made. 4p20s 36v. 2 of these in series gives 72V
and 5 in Parallel gives me 30 min of run time in an MX bike. Just make sure
your controller has a shut off before it get to the lower threshold and temperature
does not get to an over rated level. I made these so they come apart and can be
rotated or repaired if necessary.

Battery Buss Bars (3).jpg
 
My cells have just arrived.

IMG-2106.jpg


I have an Opus BT-C3100 v2.2 charger.

IMG-2108.jpg


Is the wisdom to capacity test every cell and use repackr.com to sort the order, or when buying brand new do folks just ensure they are the same voltage and start building?
 
I decided to do it properly and capacity test every cell.

I only have one charger so it takes 24 hours to do 8 cells. I have another charger on the way.

IMG-2118.jpg
 
It is good that you are testing each cell although they are new. I did the same when building my pack. The P42A may be what I use for my next pack. Have you considered the "per slot" deviation on those chargers? For example, I was using a couple of cell chargers similar to those but found that each slot reported slightly different capacity figures.

Good luck with your build.
 
It did occur to me that one capacity test was never going to be optimal, but how many should you do ? 2,3,4,5 ?

Due to my limited charger capacity, I'm going to go with these results and see how I fare.
 
just doing a internal resistance test with a accurate meter straight from the box is faster and yields the same/better result as manually measuring capacity. a deviation in IR tells you more then capacity when measuring fresh from the box. then you can be done in 24 minutes instead of 24 days.
 
Hi Flippy,

Sounds good.

What do you recommend as an accurate internal resistance meter and how would you use the resistance data to organise your packs?
 
kudos said:
Hi Flippy,

Sounds good.

What do you recommend as an accurate internal resistance meter and how would you use the resistance data to organise your packs?

just get a SM8124 and get a spare, switching the polarity instantly kills it. its accurate enough for this use case.

and its simple: start measuring and dont use the ones that read off. so buy extra. repeat the test when the cells are installed and measure the groups as a whole. they should be identical if your welds are good.
 
Good to know about the SM8124 or similar. I will probably pick up something similar for my next pack build.
 
just to be clear: dont do anything with them. pull them out of the box and measure.

if you already charged them then make very sure you charged them all to the same level. the voltage should not differ more then 0.05v between cells when measuring.

ps: the SM8124 does not have a low batt warning. so get a rechargable 9v and make sure its full when doing a run. the measurements get a bit wonky when the battery gets low.
 
I've been wondering the same thing about measuring P42A's before assembling. The meter is a good idea but unfortunately the SM8124 seems to be unavailable, would you have any other recommendations for a meter? Does anyone use Allosun?
 
My Opus charger can measure IR, however, I’m not sure what the value should be and what would be considered a big enough difference to exclude the cell?
 
Frank said:
I've been wondering the same thing about measuring P42A's before assembling. The meter is a good idea but unfortunately the SM8124 seems to be unavailable, would you have any other recommendations for a meter? Does anyone use Allosun?

Many available here (may take a while to ship from China):
https://www.aliexpress.com/wholesale?catId=0&initiative_id=SB_20210218131150&origin=y&SearchText=SM8124

kudos said:
My Opus charger can measure IR, however, I’m not sure what the value should be and what would be considered a big enough difference to exclude the cell?
The issue with the Opus and similar charger/testers is that the DCIR measurement is quite inaccurate and inconsistent. A 4 wire battery impedance tester like the SM8124 will (should) give useable results.

As for what the internal resistance should be; you would have to check the cell's datasheet. Be aware the IR charges with state of charge (resting voltage) and temperature.
 
The SM8124 seems to be readily available on Amazon when I just checked.
 
Now rocking double chargers...

IMG-2126.jpg
 
I have a question regarding how people attach the main positive and negative discharge leads.

I’ve watched countless videos on people building packs but 99% skip how they attach the main discharge leads for some reason.

I’ve seen PTFE high temp cable soldered to the nickel strip in between cells. This is the way I had intended on doing it.
I have plenty of 12AWG PTFE cable which I was going to use.

I’ve also seen discharge leads soldered to busbars. Seeing as I only run a 40A controller and so don’t ever see much over 45A spike I was not going to bother with busbars and just use plenty of nickel strip.

How do you attach main discharge leads ?
 
continue the series connections as if you were to connect another row of cells but double the length + 1cm. then fold them back over and weld the folded back strip on top of the cells. you now have a nice loop. put the wire you want to use in the loop and use some piers to push the strip closed so the wire is compacted as much. you can use a weld at very low power to keep the strip "closed" and then use solder with good flux (i recommend using a flux pen and real lead solder) and solder each nickel strip and wire together. if you have the right length strip your wire should fold flat against the side of the battery. put 2~3 layers of kapton tape and ducktape with cloth as a shield between the cell and wire so the wire can NEVER touch or rub against the cell walls.
 
For B+ and B- Ive welded to the nickel strip then soldered the wire on as well as welding a 3rd layer of nickel then soldering. Both worked fine. If it's a really high output pack I'll lay down copper strip on main B+ or B- with my Miller Dynasty 210 Tig set to single pulse with no background amps.

Never built a pack that didn't have a BMS so cant really comment on that. But I can't recommend the ANT BMS enough. Being able to see real time temps, IR, balance and being able manually switching the BMS off completely, gives real peace of mind.
 
Where do you get your ANT BMS’ from?
 
I have a question about pack build voltage.

After capacity testing the cells and checking a few other things my cells are now not the same voltage.

The easiest thing to do would be to charge them all up to max 4.2V but I don’t really want to build the pack fully charged.

My Opus charger does not have the option to charge to say 3.6V.

I have a couple of iChargers which I could use to charge to a specific voltage but then I can only do one at a time.

How do people get their cells to the same voltage for the first time?

Is it ok to build the pack at 4.2V per cell ?
 
kudos said:
How do people get their cells to the same voltage for the first time?

you dont. you pull them out of the box. measure them and put them in the pack.

and a shorting "oopsie" during the build is just as bad full as it is half full when you get them out the box. you will still frock up the area you shorted. its just the difference between "holy shit!" and "holy frocking frock!" when you short it out.
 
kudos said:
After capacity testing the cells and checking a few other things my cells are now not the same voltage.
The easiest thing to do would be to charge them all up to max 4.2V but I don’t really want to build the pack fully charged.

If you can figure out a way to build a jig to wire a bunch of the batteries up in parallel, they will self-balance. Some 1" x 36" or 48" aluminum U channel with bolts spaced @ 21mm, some wood strips, and some non-metallic bungie cords should do the trick.

M
 
Back
Top