Battery not always fully charging

[trevor brooker]

I was going to post to ask with help in balancing the battery, but today it fully charged for the first time since February.
Battery was described as 52v 17.5ah Lithium battery pack using Panasonic or Sanyo NCR18650GA cells
sticker on battery says 2017, but I purchased in January 2018
I connect charger to battery, turn the battery on then switch the mains on.
I leave the charger on until the leds show its finished
Why does it mainly charge to around 90% then very occasional fully charge?

 

[Nealh]

If sometimes you only get 90% then a cell group is out of balance and charges to 4.2v faster then other groups, balancing may takes ages to sort it self out as typically 50-100mv is the balance rate. It is not uncommon for the first or last cell group to be out of balance with some BMS so the issue might just be a crappy BMS.
I would power up the charger before connecting to the battery (already switched on) to reduce the spark at the connector.

If 90% charge occurs again check the cell group V's via the bms balance wire connector to see which group is low, Balance can sometimes take two or three attempts for it to correct it self if balance voltage isn't close.

Hopefully now it is balanced and the BMS has finally done it's job it will remain balanced time will tell, if not think about replacing the BMS with another from a supplier with a better reputation for quality.

 

[trevor brooker]

"check the cell group V's via the bms balance wire connector to see which group is low"
I'm sorry I've never built a battery so i'm a bit lost with the jargon. I've got a multi-meter & happy to learn.
On the BMS I have at one side Yellow Red Black wires
on the other lots of white wires (I presume 14 as 14 series x 4.2v = 58.8v) so are the white wires the balance wire connectors
So to check I measure the metal pins beside the white plastic plug & where do I put the other probe?
I did try to remove the white plastic plug but did not want to force it.
If the bms contines to misbehave any suggestions for what company to use, & what type of BMS (or are they all fairly dumb)
ps thanks for correcting my charging procedure, I had heard about a possible spark & thought I had it the right way around

 

[Nealh]

Correct Trevor the little white wires are the sense/ balance wires. One to each cell group, it should simply pull out though may be tight.
The Black is 0v and is soldered to cell group #1 negative all the other sense wires 13 white and 1 Red are soldered to the positive of each group.
The first White is 4.2v and so on until you get the last group #14 which is the Red.

You can measure cell group voltages singularly;
#1 Black probe to thin Black sense wire & Red probe to 1st thin sense White wire.
#2 Black probe on 1st White sense wire & Red probe on 2nd White sense wire.
#3 Black probe on 2nd White sense wire & Red probe on 3rd White sense wire.
And so on till you have 14 individual voltage readings between 3.2 - 4.2v.
The 14th reading will Black probe to 13th White sense wire and Red probe to thin Red sense wire.

Or you can measure and have accumulative voltage reading.
Black probe remains on Black sense wire 0v and the Red probe moves along to each of the 13 White sense wires with the Red last sense wire being #14.

Write down 1 -14 on paper vertically and then write the voltage next to each one as you measure it, then you can refer to see where/if any unbalance may be.

 

[trevor brooker]

after the battery was fully charged to 58.8v (& therefore balanced?) I used the bike, with the battery reading 23.4v when I got home.
The next day I left it on charge, but this time it only reached 57.6v.
So one of the five rows of cells has reached 58.8v, with some of the other four not, so that the cumulative total of the five is 57.6v?
Is there anything I can do

 

[Nealh]

Not quite each row of 5 cells is a parallel group of which you have 14, each group (of 5) will only charge to 4.2v. 14 x 4.2v =58.8v. The 5 cells in parallel voltage wise act as one so only 4.2v, voltage only increases when connected in series.
Capacity wise the cells are 3500mah, in parallel the capacity = x 5 to give 17.5ah.
It looks like you probably may have a weak cell group, the only way to be sure is by checking the 14 cell group voltages via the BMS connector.
43.4v is what I think you meant = SOC of 3.1v per cell group.
When charging make sure the battery switch is on as this also turns on the BMS.

 

[vfr400]

after the battery was fully charged to 58.8v (& therefore balanced?) I used the bike, with the battery reading 23.4v when I got home.

Do I read this right? You ran your 52v battery down to 23.4V???

 

[trevor brooker]

"Do I read this right? You ran your 52v battery down to 23.4V???"

sorry typo - I should have checked my fat finger typing - it went from 58.8v to 53.4v

 

[trevor brooker]

"It looks like you probably may have a weak cell group, the only way to be sure is by checking the 14 cell group voltages via the BMS connector."

I managed to slide out the white plug, exposing the BMS connectors, but failed to obtain consistent reliable readings, as I was unable to maintain good connections with both probes simultaneously.

So assuming I have one of more weak cells I have two options.

1- Do nothing - is this safe?
2- Arrange for re-cell battery (Can you do partial replacements?)

 

[vfr400]

Don't assume. Always measure.

There are slots in the side of the connector that should allow your probes to make good contact, otherwise use a bit of pinstrip that you can get from EBay for 99 pence.

 

[trevor brooker]

my lack of knowledge is a hindrance - can I just check the basics

I pulled the white plug with its white wires off the BMS

This exposed the BMS pins, which is what I tried to measure

But should I instead have used a pinstrip into the white plug, so that I measured the white wires?

 

[vfr400]

The white connector has holes in it for the pins on the BMS. It also has slots on the top or bottom face, where you can see little silvery contacts through the slots, where you can attach your meter probes.

The connector in the BMS is basically a strip of pins. You can buy that on Ebay, break off two pins (it's designed to break at whatever length you want) and insert them into two holes at a time to measure each cell without risk of shorting anything.

Pardon our interruption...

www.ebay.co.uk

Alternatively, you can use jumper wires. select male to male.

40Pin 30cm jumper wire cable line for arduino F/F F/M M/M RDR | eBay

The dupont jumper wires which can be used over and over again. Every unit contains 40 jumper wires. These jumper wires can be used on electronic projects, on a solderless breadboard and direct to circuit board headers.

www.ebay.co.uk

 

[trevor brooker]

many thanks for the link to the pin strip - I've ordered it & will try to measure the voltage correctly!