Supplier defends lower capacity as being within 5% of stated capacity

RGIvy

Solar Enthusiast
Joined
Dec 9, 2020
Messages
77
Hi folks

I'm curious if this is quite common.
We decided, partly out of curiosity, to check out a few Alibaba sellers of cells.
We've received a set of 4x300ah cells a few weeks ago which have tested out at 285ah (four tests including high current at around 0.5C and low current at around 0.04C, three done at a controlled temperature of 25°C and one at a lower temp of 21°C).

Anyway, we got hold of the seller and challenged him. His response: as long as it's within 5% of stated capacity that would be acceptable.

Comments?
 

DerpsyDoodler

Photon Sorcerer
Joined
Jan 10, 2021
Messages
2,270
Hi folks

I'm curious if this is quite common.
We decided, partly out of curiosity, to check out a few Alibaba sellers of cells.
We've received a set of 4x300ah cells a few weeks ago which have tested out at 285ah (four tests including high current at around 0.5C and low current at around 0.04C, three done at a controlled temperature of 25°C and one at a lower temp of 21°C).

Anyway, we got hold of the seller and challenged him. His response: as long as it's within 5% of stated capacity that would be acceptable.

Comments?
He is wrong. The stated capacity (or more) that the cells were sold and billed at is the only acceptable capacity.
 

DerpsyDoodler

Photon Sorcerer
Joined
Jan 10, 2021
Messages
2,270
Maybe just one cell is well below the rated capacity.
Could also be human error. Maybe they weren’t charged or top-balanced properly. I am capacity testing my cells individually to eliminate potential confusion if one doesnt test well. I have no idea how OP charged or performed his test.
 

Ampster

Renewable Energy Hobbyist
Joined
May 3, 2020
Messages
5,298
Location
Kenwood, California
Anyway, we got hold of the seller and challenged him. His response: as long as it's within 5% of stated capacity that would be acceptable.
Right or wrong is a matter of great debate. I think it boils down to expectations. I never expected pefection at the price I have paid per kWh. Consequently I only expected to use 80% of capacity because I knew the cells were not matched. There is one vendor offering matched cells for a premium and that may be an acceptable alternative to some. It all depends on where you are standing.
 

RGIvy

Solar Enthusiast
Joined
Dec 9, 2020
Messages
77
They were top balanced on the usual way to 3.65v
We then attached the BMS with the cells in series. The BMS has been set to cut off at 2.5v, one of the cells reaches that while the others are around 2.7 or 2.8
While discharging from around full soc to around 10% the cells are very close in voltage. Typically way less than 0.1v delta. As they go below 10% one of the cells runs down a bit faster.

edit: the inverter cuts off before the BMS when doing the high current test
 
Last edited:

smoothJoey

Ding Ding!
Joined
Nov 30, 2019
Messages
11,852
They were top balanced on the usual way to 3.65v
We then attached the BMS with the cells in series. The BMS has been set to cut off at 2.5v, one of the cells reaches that while the others are around 2.7 or 2.8
While discharging from around full soc to around 10% the cells are very close in voltage. Typically way less than 0.1v delta. As they go below 10% one of the cells runs down a bit faster.

edit: the inverter cuts off before the BMS when doing the high current test
If you want to beef with the supplier, charge and capacity test the cells individually.
The usual test is
at 25 celcius
charge to 3.65 volts until saturation
wait one hour
discharge to 2.5 volts
If you don't get the advertised amps the cell is not to spec.

A vendor that would try to argue the 5% is close enough is not one I would do business with.
Now whether they should pro-rate the cell or pro-rate all the cells, that should have been discussed up front.
Everyone treating these cell purchases like some back alley drug deal is not doing the community any good.

UPDATE: check the cell specification for specifics of the test.
 
Last edited:

smoothJoey

Ding Ding!
Joined
Nov 30, 2019
Messages
11,852
5% really annoys me.
The difference between a 272ah cell and a 280ah cell is under 3%.
Please test cells indivually starting with the weakest and the strongest and report back.
Publish results and vendor name.
If 5% is really ok then so be it.
 

45North

Let it shine!
Joined
Jan 2, 2020
Messages
534
Location
Canada
Sounds reasonable subject to the Consumer Protection laws that apply in the relevant jurisdiction.
I assume, like any manufactured product, cell voltage follows a Normal Distribution statistically. This means you can expect about 66% of the readings to be within +/- 1 standard deviations of the mean, and that 95% of the readings would be withing +/- 2 SD.
Hard to call foul on such a small sample like this: mebbe you just happened to get a few on the low side. But for the average (mean) to be 300 then someone else must have been lucky to get some on the high side!
Unclear what the SD would be for these - depends on how consistent their manufacturing process is and how good their QC is.

I noticed recently that the stated weight on a bag of potato chips I purchased was 198g. Seems odd. You'd think it should have been a nice round number like 200g. But then I thought about it. Mebbe they were aiming for 200g but some of them were coming up short (as per the statistical distribution described above). Easy work-around solution - just label with a lower weight and more of them will make the grade. Satisfying consumers and any regulatory agencies that enforce these things!

Edit: We need a sample size of about 30 to make a meaningful conclusion. Who's gonna take one for the team and buy 30 of these to test this out?
 

Hedges

I See Electromagnetic Fields!
Joined
Mar 28, 2020
Messages
10,681
Is it expected that any of these sellers would test capacity of cells they got from manufacturer or other source?
Maybe spot check. But I would think they are just a warehouse/internet retailer operation. They wouldn't have capability to 100% test capacity.
Random sampling and math would be the best approach.
They could advertise, "This batch of cells was tested to 282 Ah capacity, +/- 2 Ah 3 sigma, with xxx confidence level."

Sounds reasonable subject to the Consumer Protection laws that apply in the relevant jurisdiction.

:ROFLMAO: :ROFLMAO: :ROFLMAO:
 

smoothJoey

Ding Ding!
Joined
Nov 30, 2019
Messages
11,852
Hi folks

I'm curious if this is quite common.
We decided, partly out of curiosity, to check out a few Alibaba sellers of cells.
We've received a set of 4x300ah cells a few weeks ago which have tested out at 285ah (four tests including high current at around 0.5C and low current at around 0.04C, three done at a controlled temperature of 25°C and one at a lower temp of 21°C).

Anyway, we got hold of the seller and challenged him. His response: as long as it's within 5% of stated capacity that would be acceptable.

Comments?
Please post a picture of these cells and the name of the vendor.
 

Johnson

Solar Addict
Joined
Nov 21, 2020
Messages
676
I got 272Ah cells that tested at 274Ah. They were sold to me as 280Ah cells and one cell was slightly damaged during shipping (which was not the fault of the vendor) I decided to keep them, against the advise of others here and the seller offered me a discount on a subsequent order. I shouldn't have because if they get away with it, they will just keep doing it. However, they are still fine for my application and I didn't want to ship the cells back and wait for new ones. (pressures at home)

The supplier telling you that 5% lower is "normal" is outright lying, which is a really dumb move. The best thing you can do for every one is name the supplier here and leave a review on alibaba. Simply state that you did not receive what you paid for and that the vendor believes in their communication with you, that this is normal for them.
 

blutow

Solar Addict
Joined
Dec 20, 2020
Messages
337
They were top balanced on the usual way to 3.65v
We then attached the BMS with the cells in series. The BMS has been set to cut off at 2.5v, one of the cells reaches that while the others are around 2.7 or 2.8
While discharging from around full soc to around 10% the cells are very close in voltage. Typically way less than 0.1v delta. As they go below 10% one of the cells runs down a bit faster.

edit: the inverter cuts off before the BMS when doing the high current test
Are you saying the inverter cut off the test and you are using that capacity number? What was the pack and individual cell voltages (not inverter voltage) at the end of the test? Ideally, you should have a DC load that can go down near ~10v or it's unlikely you will get an individual cell down to 2.5v. It would be interesting to understand exactly how you tested these cells and what you measured with.
 

Johnson

Solar Addict
Joined
Nov 21, 2020
Messages
676
Is everyone using calibrated test equipment?
In my case, even though I checked the voltages and currents with (two) multi meters and got no deviation, I wouldn't consider them calibrated. Only one of the multi meters can read amperage and even then it can only read up to 10 Amp, which agreed with the BMS but under my test load I was drawing 41Amp. However, it should be "close enough", or am I wrong?
 

snoobler

Solar Honey Badger
Joined
Jul 10, 2020
Messages
8,395
Location
HBR, AZ
In my case, even though I checked the voltages and currents with (two) multi meters and got no deviation, I wouldn't consider them calibrated. Only one of the multi meters can read amperage and even then it can only read up to 10 Amp, which agreed with the BMS but under my test load I was drawing 41Amp. However, it should be "close enough", or am I wrong?

What is the stated accuracy of the devices you used? Correlation is encouraging, but it's not an indication of accuracy.

I'm of the mind that paying 1/3 to 1/2 the price for cells vs. those obtained through official distribution channels carries with it the expectation that the cells do not meet spec. Period. Any expectation otherwise is unrealistic. My 9 EVE cells measured 265-273Ah on correlated equipment (3 sources). I'm tickled.

I would be far more concerned about the as-received resting voltage and internal resistance measurements.
 

smoothJoey

Ding Ding!
Joined
Nov 30, 2019
Messages
11,852
Test the pack from dead full until the weak cell hits 2.5 volts.
Then test the weak cell individually.
See how different the results are.
I would guess that the difference will be within 1%.
 
Top