Suspect and odd voltages Dolphin case battery please join in and help

Batteries, Chargers, and Battery Management Systems.
Post Reply
leffex   1 kW

1 kW
Posts: 325
Joined: Apr 08 2012 4:39pm
Location: Sweden

Suspect and odd voltages Dolphin case battery please join in and help

Post by leffex » Mar 10 2021 11:45am

Battery

The problem. Unknown. What I suspect is that it has been manufactured in the wrong way at the minus poles as you shouldn't spot weld in the center of them. Can anyone give me any clue to what might be the issue. Any other hypothesises would be appreciated as main reason for this is sharing and be able to repair these kind of things.

Case: Dolphin type
Voltage full voltage 42v. Maximum current full voltage 38v
Capacity: 13AH. (10s5p) Each cell had 2500mah

Cell voltages:
4,10
3,68
4,07
3,64
3,66
3,46
3,81
3,68
4,07
4,18

Image

For three picture gallery. Main picture, zoom-in and voltages numbers for each cell group for two batteries:
https://imgur.com/gallery/56025M1
Salestech in an ebike store

goatman   1 GW

1 GW
Posts: 3195
Joined: Jun 23 2019 6:50pm
Location: Surrey, B.C.

Re: Suspect and odd voltages Dolphin case battery please join in and help

Post by goatman » Mar 11 2021 3:14am

thats pretty far out of balance
what cells are they?

i tie little magnets to 12v bulb and drain the high voltage p groups to the lower voltage p group to try and then charge to balance the pack

dont leave the bulb on the battery like this
IMG_20210304_184513.jpg
IMG_20210304_184513.jpg (122.39 KiB) Viewed 360 times
you have to watch the group voltages when charging, if theyre all charging up at the same p group voltages thats good but if one or more groups start charging at a noticeable difference theres probably bad cells, you dont want to see all the groups at 4.1v and another at 4.3v or 4.4v thats bad

leffex   1 kW

1 kW
Posts: 325
Joined: Apr 08 2012 4:39pm
Location: Sweden

Re: Suspect and odd voltages Dolphin case battery please join in and help

Post by leffex » Mar 12 2021 5:01am

Thanks goatman.

Reasons are to find ways to fault search batteriers as fast as possible as well as finding the correct fault as if this takes to much time they are sent to the grave and recycling yard.

I've been learning about 18650 and ebikes since 2012 but this still puzzles me why so basic faults are found in new/ unused batteries in 2021.

The most common faults may be water ingress destroying electronic parts or battery cell connections.
Water ingress
One cell or full cell group failure
Bad spot welds refering to spot weld positions
Bad spot welds ref. to bad connections

Regarding these batteries:
Yes it would be smart to drain and equalize all cell groups and then charge. While charging I'd maintain an check on each cell group and its voltages (could be done the opposite way as well fully charge all, then drain and cycle at least twice )

Are there any other issue with the cells more than that they are bad. Like how to se if one is shorted vs has a higher resistance or is this the same thing.

The cells are FST brand 2500MAH 10A chinese power cells. I don't remember its last name code.
Salestech in an ebike store

goatman   1 GW

1 GW
Posts: 3195
Joined: Jun 23 2019 6:50pm
Location: Surrey, B.C.

Re: Suspect and odd voltages Dolphin case battery please join in and help

Post by goatman » Mar 12 2021 12:01pm

i think trying to fully charge a battery to 4.2v with a bms in an older pack actually makes the pack more unbalanced,

i bottom balance and lower the charging voltage to balance a pack like in this post

viewtopic.php?f=14&t=109601&start=25#p1622842

im doing more cells right now, why force 2500mah/capacity into cells that can no longer hold it, if theyve degraded to 2375mah,2300mah, 2325mah they cant be top balanced at 4.2v but bottom balance at 3.0v and charge to 4.15v and they balance
IMG_20210312_084935.jpg
IMG_20210312_084935.jpg (160.54 KiB) Viewed 330 times

leffex   1 kW

1 kW
Posts: 325
Joined: Apr 08 2012 4:39pm
Location: Sweden

Re: Suspect and odd voltages Dolphin case battery please join in and help

Post by leffex » Mar 18 2021 12:44pm

So. Drained and then charged all cells after putting them at the same SOC. Results are however bad as cell groups differs from 4.26v to 4.02v in the fully charged pack

Some cell groups thus does carry more capacity than other groups and by a big margin.

The 4.02v group has the most capacity and or high internal resistance.

Another cell group of the first battery in name. the 3.81v dropped to 3.75 in a few days time before this test meaning high drain on cells that shouldn't fluctuate in voltage if for any reason nothing isn"t wrong with it. I suspect internal shorts somehow.

Questions have on this new unused battery. Why does this happend. Bad bms? Bad battery assembly (inlcluding bad battery cell capacity matching)
Salestech in an ebike store

leffex   1 kW

1 kW
Posts: 325
Joined: Apr 08 2012 4:39pm
Location: Sweden

Re: Suspect and odd voltages Dolphin case battery please join in and help

Post by leffex » Mar 25 2021 4:58am

Here's another Dolphin battery pack with a fault that is easily spotted. Pretty much not the same problem as the first packs in the first post.


48v 14ah measured capacity at 7,5% with one cell group showing imbalance or fault
Image
https://imgur.com/gallery/gHWeGwH
Salestech in an ebike store

leffex   1 kW

1 kW
Posts: 325
Joined: Apr 08 2012 4:39pm
Location: Sweden

Re: Suspect and odd voltages Dolphin case battery please join in and help

Post by leffex » Mar 25 2021 10:40am

So I think I have managed to find out what causes all these odd problems.

I have another 10 packs like this to handle and figure out. My assumptions after today are as following:

The cells are the culprit. I assume the cells are rewrapped and recycled LG M26 cells which is the worst cell ever created as their behaviour makes them not worthy of being mention and or used as batteries. They will self-discharge, I suspect, they age different and odd so one cell can be 500mah and or 1200mah whilst others can be around 2000-2500mah. A BMS can´t handle this as this creates too much voltage differences between the cell groups.

The issues that have arisen are similar to the LG M26 cell but might not be exactly that cell as I thought just a moment ago. I will open it up to see if there is any codes or letters on it.

The letters on the metallic are nonexistent. No letters or any writing on the cell surface. However, on the purple plastic, there is this text on a cell I choose: 02ICEAGZ3124A29C20002189 and also the positive side protection plastic has a light blue color.
Weight is 44g (1,552 ounces)
Estimated capacity of 2500-2600MAH from factory.

Looking at the positive pole it looks like LG M26 cell but no evidence more than that and that the positive ring color is blue like an intel inside core logo doesn't give enough to really really be assured that it is.

Before I said they were:
leffex wrote:
Mar 12 2021 5:01am
The cells are FST brand 2500MAH 10A chinese power cells. I don't remember its last name code.
Though that has changed as can be seen by the last post. Earlier batches had been the FST 2500. These are unknown at the moment.

Image
Salestech in an ebike store

Post Reply