Testing of B grade lithium ion lipo soft pouch cells

wire.rat

100 mW
Joined
Aug 25, 2016
Messages
39
Hi all,

I will be conducting tests to find out about the self discharge characteristics of the high capacity lipo cells that you can get directly from china. I will be buying 3.7v 20Ah cells directly from this link
https://item.taobao.com/item.htm?spm=a1z09.2.0.0.2ed6e797pTlNV3&id=539888371314
for about $8 usd each per cell and get them shipped to my doorstep in Singapore for about $3.10 usd per kg.
Heres some photos of the cells from the item page.
ayFDbtb.png

HmzBfJo.png


Here's a bit of background of myself. I am a high school student in Singapore, where its close proximity to china allows me to get Cells shipped to me for cheap.
I first discovered about endless sphere when I was trying to find out how to build my own battery packs using 18650 cells for my electric bicycle with a 1 kw motor. Since then i have gone through 2 spot welders and made at about 10 batteries for other people. All of these batteries were made using Brand new sealed in box 18650 LG F1L 3400mah cells which had very good specific energy, but they came at a high cost at about $2.56 USD per cell, about $0.752/Ah. this is almost double of the $0.400/Ah of the 20Ah Lipo cell. Thus these lipo cells will be about half the price compared to 18650 battery and will be easier to put together as compared to spot welding 18650 cells.

The purpose of this thread is to try to find out what quality of the cheap B grade cells that you can buy directly from china from sites like aliexpress or ebay.

Firstly we will start with a explanation of the grades of the batteries.

In china, the quality of batteries that you can purchase as a random person on the internet are divide into a few grades.

A grade: Brand New untouched cells that came directly from the factory. Examples of these that you can buy online are the boxes of LG/Sony/Sanyo/Panasonic 18650 cells that are sealed in box. These are of the highest quality and meet the specification that they are given. Cheap lipo cells of this grade are hard to find and you can usually only find high discharge ones that are expensive, like the ones in the high powered lipo Batteey sale post .

B grade Brand new cells that do not meet the exact specifications. They may have higher Internal Resistance, higher self discharge after production, or have physical imperfection on them such as light dents or scratch on the surface of the lipo cells.

C grade: Used cells that are taken from used battery packs from cars/buses. These normally are the prismatic or pouch cells that you can find online with short tabs remaining on them.

Here's what i know about the cells from talking to the seller:
They are 3.7V 20Ah
Their dimensions are 225*68*13mm
They weigh about 500 grams
They have thin tabs of only about 12 mm wide, so the max discharge rate should be 2C instead of what is advertised.
they are B grade
they have more than 10 thousand in stock

Next, the testing methodology.

The testing method used will be identical to the one used by NASA when they tested several Li-ion Pouch Cell in 2012. It is called a Soft Short or a OCV(open circuit voltage) bounce back test. Basically, they discharge the cells to <10%soc (2.8V?) and let them sit for 10 days and measure the changes in their voltage. Cells with declining OCVs have high internal impedance and high self discharge.
Below are photos taken off this pdf. https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20120000040.pdf

XXHt9Nd.jpg


Vendor A is A123
QaoZaJJ.jpg


Vendor C is EIG
73GlcFi.jpg


The tests will be conducted in a air conditioned room at about 26 degrees Celsius.

Results:
I have just ordered 15 cells and im waiting for them to arrive
To be announced in 3 weeks time , around 29 September after i receive the cells and test them
 

Attachments

  • 1368225 规格.pdf
    319 KB · Views: 55
Interesting test method. I haven't seen that one before.

It might also be good to perform a standard capacity test. There are several possible failure modes for these cells.
 
It will be interesting to see the results, and obviously there are many other tests you can do such as fetcher mentioned.
Capacity, DCIR, charge retention, etc.
But, my concern with such cheap cells from uncertified sources is consistency of any result.and random defects.
15 cells is a good initial sample size, but i dont think you can draw good statisticly sound conclusions from that one sample to know if there is a risk of random irregularities.
100% inspection and testing of every cell you put in a pack is the only answer.
 
Yeah I will definitely do a capacity test at 1C, 20A discharge to find their capacity and temperature rise.

Yes I agree that a sample size of n=15 is really not a good representation of the overall quality of cells, but I would be quite happy if at least 66% if the cells do not exhibit any signs of high self discharge from the soft short test.
 
Different tests check different characteristics,
The DCIR test has been found very telling of a cells ability to perform,..
.. and obviously repeated cycling to gauge capacity loss with use . But that is very time consuming.
A "Stress Test" at discharge rates of 150% or 200% of your planned maximum discharge would also be wise.
PS : these cells are not too dissimilar to the Hobbyking "Multistar" cells some of us have been very happy with.
 
Yeah I'm planning to stress test one of them at 2 or 3 C (40A or 60A). I have ordered two 0.10 Ohm 300W resistors along with the cells to discharge them at high currents.
For the comparison to Hobbyking "Multistar" cells, erm i think these 20Ah cells can probably only sustain 2C discharge because of their small tab size.
I have yet to get 30Ah cells with much larger tabs like the one below to test yet.
FqJKIAi.png
 
Wish buying off Taobao was easier for non-Chinese speakers (readers/writers).

I'm quite happy with B grade for stationary solar applications. High internal resistance isn't an issue, because you tend to massively oversize, self discharge is meaningless when you cycle every day.

Unfortunately, I haven't found B grade cells anywhere near that cheap on Aliexpress/Ebay.

Good on you for this project though.

By the way, the tab size has nothing to do with the C rate. It's mostly ratio of anode to cathode. I can't remember which, but making one bigger gives you more energy density, and going the other one gives you more power density. So low C rate cells tend to be smaller and lighter than the same capacity high C rate cells.
 
Hopefully you can get some school credit for thsee experiments. It's am emerging technology that will likely power most of life on Mars!
 
Update 5 September 2017.

Received 2 cells.

Will be testing for Capacity first.

They are recieved at 3.92V and 1.35 mOhm of AC impedance.
 
This is great, do you have a livefeed :lol:

I also ordered 2 cells (without speaking Chinese, not sure where they will arrive !)
 
nunux59 said:
This is great, do you have a livefeed :lol:

I also ordered 2 cells (without speaking Chinese, not sure where they will arrive !)

Nah I'll keep updates posted here. Now I'm letting the cells rest for 12 hours to see if they self discharge differently before testing for capacity. I have 5* 1 Ohm large ceramic resistors , so i should be able to get at least 0.5 C discharge current to 2.50V :)
 
I'll be testing the cell temperature rise during 3C 60A discharge when I receive my 0.10Ohm resistors later this week

I will be using a 0.10 and a 0.20 Ohm resistor in parallel to give me 0.07ohm, which will give me 60A discharge current at 4.20V and 40A at 2.80V. So the cell will be going through at least 2C discharge for the close to entire test.

Here is the Datasheet that the seller sent me

Important Parts:

Capacity: 20Ah
Capacity can range from 19.8-20.3Ah
Nominal voltage : 3.7V
AC impedance: 2mΩ (1000Hz frequency)
Max charge voltage: 4.25±0.05V
Charge End current(Cut off): 0.01C
Discharge End voltage: 3.0V
Max Continuous Discharge: 60A
Peak Discharge: 100A, 5 seconds
Weight: 400g
Measured weight is 396G with a kitchen scale
 
Doing discharge test at 3A current.
Starting voltage: 4.171V x2 = 8.342V
End voltage: 2.500V x2 = 5.000V

B6Ho9LL.jpg

mIXoHsm.jpg
 
Hi, what are the results?
How are they holding so far?
Seller has 3.7v 40ah, what do you think about those?
 
Read this article and than decide for yourself, if you want take the risk :

https://www.mpoweruk.com/china_batteries.pdf
 
Back
Top