18650 cell matching

[wannesd]

I've ordered 140 cells - sanyo 18650GA.
I plan to measure their capacity and internal resistance before doing anything with them.
Should I match them on ir or cap?
And should they be matched for series or parallel connection?

 

[Eteck]

What is your final cell arrangement going to look like ?

 

[wannesd]

What is your final cell arrangement going to look like ?

20s7p, but depending on final pack size I might order more cells to go 8 or 9p.

 

[999zip999]

More the better for battery life and distance. But can you fit it ? I'm like my muxus 3,000 and 24s of heavy A123. But Im it weighs 90 lbs. and can be dangerously heavy for bike frame.

 

[wannesd]

More the better for battery life and distance. But can you fit ?

Tested the 140 cells in my falconEV framebag, got room to spare.
I'll go 20s10p.

 

[999zip999]

I have 17.5ah of usable battery and can use it up 11/2 hours.

 

[wannesd]

Each 18650GA is 3450 mAh, so pack would be 34,5Ah.
But let's say 30Ah usable.
Should let me ride a long time at 45kph with pedal input...

 

[999zip999]

Now do you have a spot welder ? I like sense wires 4- 5s and bms for ez checking and. Lot's of bms choices. New genuine cells should be factory close and ez to work with these days. Use gasket on positive side and insulation between parallel strings.

 

[Samd]

I don't hear a lot of people cell matching on new high end cells. Think it's necessary on Sanyos? I'm assuming their QC would be perfect if bought from a decent vendor.

 

[wannesd]

I don't hear a lot of people cell matching on new high end cells. Think it's necessary on Sanyos? I'm assuming their QC would be perfect if bought from a decent vendor.

All bought from eu.nkon.nl, so should be top notch.

So you think I shouldn't bother? Would certainly be faster.

 

[redilast]

New grade A brand name cells are so good these days I don't think anyone matches them for capacity. If you do have an internal resistance meter you could just check all of them and make sure they are within 5-10% of each other. I've used an internal resistance meter on hundreds if not thousands of cells, and rarely see more than a 1-2 mili ohm difference between cells in the same batch. I believe GA's should be around 25 mili ohms @ AC 1Khz. But obviously if you are doing a DC resistance test, or use some cheapo resistance tester your results may vary.

 

[Samd]

Thanks man. That's what I would've expected. The factories would generally do I should think. Plus have tight tolerances for making.

I'd still test if I was recycling laptop batteries.


Sent from my iPhone using Tapatalk

 

[Ohbse]

Not worth doing if you're buying good cells new. They're likely to be closer in voltage than most chargers could reliably detect!

 

[flippy]

i bought 400 INR18650-29E and i got 2 duds. they were thankfully in a single block (i have a 13S30P pack) but they really did a number on the other 28 cells in the block. the internal resistance rose from 19 to 22.
in the end i replaced them by pullling apart a second bike pack i built at the same time and swapped the cells over.

i did not test all of them before building the pack and with hindsight i should have tested them all. pulling apart a 20kg pack is no fun.

 

[wannesd]

i bought 400 INR18650-29E and i got 2 duds. they were thankfully in a single block (i have a 13S30P pack) but they really did a number on the other cells in the block.
in the end i replaced them by pullling apart a second bike pack i built at the same time and swapped the cells over.

i did not test all of them and with hindsight i should have tested them all. pulling apart a 20kg pack is no fun.

That is exactly what I fear...
How did you test them?

 

[flippy]

from several discharge runs i know they are pretty well matched thankfully.

but i have bought that ebay tester linked here: http://www.ebay.com/itm/181743302391 and with all next packs i will do 100% checks.
i still plan on building a 10~20 cell parralel charger so i can do my tests with all full cells and i leave them standing for a few days to check self discharge. that way i can ensure i have a quality pack, taking apart a 100 cell block is no fun.

 

[Samd]

Wouldve been interested to know if the two duds had different voltages on arrival. And 29e are getting a bit older.

 

[flippy]

i am currently topping off all the cells from the block i removed and checking IR and self discharge. but i can only do 2 at a time so it is going to take a while.
i need to get some 18650 holders so i can parralel charge 10 of them in the future.

the IR for all the good cells is 19~20 mohms so no worries there. any duds i should stick out like a sore thumb...

29e might be "old" but they are very cheap for their capacity and in my case where i put 30 of them in parralel it is very hard to even get near the limit of them.
according to the spec sheet they can do 8 amps max (not recommended for lifetime) so that would mean i can get 250A at 50V (12.5kW!) continously if i wanted.
but my pack is built for range and lifetime, not smoking rubber from the line.

 

[Doctorbass]

Sinple rule:

The more you put i parallel the less you have to bother with cell matching :wink:

Statistics does nearly all the job!

Doc

 

[flippy]

yes i know. but the 2 duds were really eating up the power causing a massive inbalance those puny little balancing resistors can't handle. 87Ah is a bit too much to balance with a 80mA balance load.
i would manually charge up the block and the next morning the voltage would be at 4ish volts again while the rest of the 12 blocks are stiing comfortably at 4.16.
apart from the lost capacity/range it also causes a massive voltage drop under load as that block drains a lot faster.

all in all: take some time and test your bloody cells and you don't have to rip everyhting apart again....

 

[DVDRW]

i bought 400 INR18650-29E and i got 2 duds. they were thankfully in a single block (i have a 13S30P pack) but they really did a number on the other 28 cells in the block. the internal resistance rose from 19 to 22.
in the end i replaced them by pullling apart a second bike pack i built at the same time and swapped the cells over.

i did not test all of them before building the pack and with hindsight i should have tested them all. pulling apart a 20kg pack is no fun.

Bought cheaper cells Panasonic 18650PF 540pcs, all ok.
About 50 cycles, still no balance needed.
22s18p