EVE LF280k Vs 304

[Quattrohead]

The bottom line is there are a lot of underperforming shit reject cells now being pawned off onto us in various disguises, buyer beware.

 

[Ahjwinth]

The cells where purchased from Docan power and advertised as typical 270-280ah discharge. No grade was mentioned, they came with the single bolt connection.
And agree with above statement. I just want to make people aware. I took a gamble with the purchase and somehow expected this. The intend for me is just to make people aware and take above information into account when ordering cells. I have enough capacity for my ESS so I’m not worried directly if it is 260ah or 280ah.

 

[Alkaline]

I have purchased 64x LF280K in October, they arrived just before Christmas and currently testing all the cells. So far we have tested 23 cells. Charge to 3.65. Discharge to 2.70 volt With 40a. The best cell is 280ah (1) and the worst cell 250ah, the others are between 260 and 270ah On average. When all cells are tested I will post the Full results but this will take some time.
All the cells look good, QR codes are valid and no dents or damage. The sides are nice and flat, no signs of any swollen cells.
The plan is to take the 8 worst cells and install these in my camper and replace them with 8 other cells to that my 4 banks of battery average around 260-270 ah.

Please make sure to check date codes with qr scanner app to see when they were made. What tester are you using?

 

[RCinFLA]

The 304 AH cells have some doping to the electrodes. There are a couple of possible candidates for doping.

One is lithium added to graphite negative electrode. This is to provide self charge forming of the Solid Electrolyte Interface (SEI) layer around graphite. Its primary benefit is to reduce manufacturing time and resource cost of the time-consuming charge forming process. It also reduces the amount of positive LFP cathode electrode lithium used up during normal 280 AH cell charge forming process. This yields about 5% more cell capacity. Downside is there may be a small amount of cell bloating as the cell is sealed during the self-charge forming process and the ohmic resistance will be slightly greater. The small amount of bloating gas will eventually diffuse out through the plastic grommet cell terminal seals.

Silicon can store a lot of lithium-ions, about 5x that of graphite. Problem is it gets so stuffed and expanded it fractures the silicon lattice. Some manufactures put a small amount of silicon dust into graphite neg electrode to increase capacity. I think it is a temporary boost in capacity that fades away in a couple hundred cycles of cell.

Other candidate is small amount of graphene added to LFP cathode. Graphene additive adds cost but significantly improves the access yield to lithium from the positive cathode by improving electrical conductivity of LFP cathode. Adding 1% of LFP weight of graphene to LFP cathode improve capacity of LFP by about 125%. A 280 AH cell relies solely on fine carbon black powder to improve LFP cathode electrical conductivity to a useable level. Have not heard any downside other than cost of graphene.

My guess is the 304 AH cells have cathode electrode graphene doping. Negative electrode lithium addition alone will not yield enough capacity increase. Curious thing is the 304 AH cells have slightly greater ohmic resistance compared to 280 AH cells, usually over 0.20 milliohms. Graphene alone should reduce ohmic resistance. Maybe both electrodes are doped.

Whenever additives are introduced to electrodes or electrolyte there is always the possibility of unforeseen long term detrimental side effects. Graphene doping has low risk of any detrimental side effects since it is very similar to existing carbon powder which has been used for years.

Only time will tell.

 

[Ahjwinth]

Please make sure to check date codes with qr scanner app to see when they were made. What tester are you using?

I tested all codes with a QR scanner and all where valid. All cells where build on the same day 3/June/2022 and the serial numbers are all in a range of 200 following numbers.
The tester I’m using is the EBC-A40L. Once all cell testing is completed I will post the full results on the forum

 

[SolaroSsaurius]

The 304 AH cells have some doping to the electrodes. There are a couple of possible candidates for doping.

One is lithium added to graphite negative electrode. This is to provide self charge forming of the Solid Electrolyte Interface (SEI) layer around graphite. Its primary benefit is to reduce manufacturing time and resource cost of the time-consuming charge forming process. It also reduces the amount of positive LFP cathode electrode lithium used up during normal 280 AH cell charge forming process. This yields about 5% more cell capacity. Downside is there may be a small amount of cell bloating as the cell is sealed during the self-charge forming process and the ohmic resistance will be slightly greater. The small amount of bloating gas will eventually diffuse out through the plastic grommet cell terminal seals.

Other candidate is small amount of graphene added to LFP cathode. Graphene additive adds cost but significantly improves the access yield to lithium from the positive cathode by improving electrical conductivity of LFP cathode. A 280 AH cell relies solely on fine carbon black powder to improve LFP cathode electrical conductivity to a useable level.

My guess is the 304 AH cells have cathode electrode graphene doping. Negative electrode lithium addition alone will not yield enough capacity increase. Curious thing is the 304 AH cells have slightly greater ohmic resistance compared to 280 AH cells, usually over 0.20 milliohms. Graphene alone should reduce ohmic resistance. Maybe both electrodes are doped.

Whenever additives are introduced to electrodes or electrolyte there is always the possibility of unforeseen long term detrimental side effects. Graphene doping has low risk of any detrimental side effects since it is very similar to existing carbon powder which has been used for years.

Only time will tell.

Thanks for your lengthy response!

If you were buying cells right now, which cells (vendor?) would you bet?
(in my case, it's for a mobile application - campervan)

 

[RCinFLA]

Thanks for your lengthy response!

If you were buying cells right now, which cells (vendor?) would you bet?
(in my case, it's for a mobile application - campervan)

I am probably conservative, so I would likely buy the 280K's. Let the 304 AH's cells get some track record history.

When I get new cells, I am more concerned about their overpotential voltage slump with load current than how many AH's they have. It only takes 3-4 minutes to test each cell's overpotential load slump voltage so can run through testing the cells fairly quickly.

If a cell has significantly more voltage slump under load it likely means it is a used cell and will have degraded AH capacity.

Match cells for series stacking that have similar voltage slump for same load current.

You need to be able to pull 0.2-0.4 C(A) load current test to get a good voltage slump indication. A 10-20A load tester isn't really enough load for a 280 AH cell.

 

[Alkaline]

280K and 304 came out around the same time. The track record for the 280K is not as good as the 304. Would have been better for EVE to call these 270K...

 

[SolaroSsaurius]

I am probably conservative, so I would likely buy the 280K's. Let the 304 AH's cells get some track record history.

When I get new cells, I am more concerned about their overpotential voltage slump with load current than how many AH's they have. It only takes 3-4 minutes to test each cell's overpotential load slump voltage so can run through testing the cells fairly quickly.

If a cell has significantly more voltage slump under load it likely means it is a used cell and will have degraded AH capacity.

Match cells for series stacking that have similar voltage slump for same load current.

You need to be able to pull 0.2-0.4 C(A) load current test to get a good voltage slump indication. A 10-20A load tester isn't really enough load for a 280 AH cell.

Thanks for your detailed response!

All I have access to is a 40A charger / tester, so barely reaching the minimums for such test...

 

[Sverige]

My guess is the 304 AH cells have cathode electrode graphene doping. Negative electrode lithium addition alone will not yield enough capacity increase. Curious thing is the 304 AH cells have slightly greater ohmic resistance compared to 280 AH cells, usually over 0.20 milliohms. Graphene alone should reduce ohmic resistance. Maybe both electrodes are doped.

Whenever additives are introduced to electrodes or electrolyte there is always the possibility of unforeseen long term detrimental side effects. Graphene doping has low risk of any detrimental side effects since it is very similar to existing carbon powder which has been used for years.

Only time will tell.

Does the difference between the LF280 and LF304 have to be doping? You’ve said before electrode thickness affects Ah capacity and peak cell current. Maybe the LF304 is simply a thicker electrode version of the LF280k? The stated max current on the LF304 spec sheet is 250A, which seems low for a cell which is rated for 304Ah, so that would be consistent with a thicker electrode material.

 

[RCinFLA]

Does the difference between the LF280 and LF304 have to be doping? You’ve said before electrode thickness affects Ah capacity and peak cell current. Maybe the LF304 is simply a thicker electrode version of the LF280k? The stated max current on the LF304 spec sheet is 250A, which seems low for a cell which is rated for 304Ah, so that would be consistent with a thicker electrode material.

If you just add more LFP and graphite material you have to increase weight and volume of cell.

Capacity is primarily the amount of total LFP positive electrode (about 160 mAH/gm and 2.7 gm/cm3) and graphite negative electrode (about 370 mAH/gm and 2.3 gm/cm3).

These numbers are for raw material. In practical use, the electrodes also have other additives like binders (glue) to keep the 'cake' together, fine carbon power to improve inter-granule conductivity, and manufacturer specific 'special' additives that may provide some benefit for other attributes of operation, or ease of manufacturing. The net mAH's/cm3 volume and mAH's/gm weight will be less than raw material due to the additives. There is also electrolyte, metal foil current collectors, separator, and cell physical casing, all adding some volume and weight but do not contribute to capacity of cell. About 80% of total cell weigh is due to LFP and graphite material.

Doping with graphene in the LFP cathode improves the access of electrons to LFP giving a greater effective mAH/gram of LFP material. Usually, cells are built with an excess of graphite storage since it is cheap material and, in LFP cells, is subject to the most degradation over life of cell, so giving some excess graphite improves cell longevity. You could skimp on providing excess graphite and replace it with a bit more LFP to get a little more capacity to the detriment of shorter cell longevity.

Doping the graphite negative electrode with pre-lithiated material helps the formation of SEI protective layer during initial charge forming. This reduces the approximately 5% LFP lithium from cathode LFP consumed during the initial charge forming process leaving more useable lithium for cell capacity.

Twenty years ago LFP was recognized as a strong lattice material to be used for cathode in LiIon batteries, but its poor electrical conductivity and lower electrode voltage caused it not to be selected as a viable candidate. With further development, the powered carbon addition made LFP practical. Graphene,(more expensive) takes that a step further.

 

[Ahjwinth]

I Promised to share my testing results from the Docan power supplied EVE LF280K cells which is finally completed. I ordered 64 cells in total Which were specified as 280 to 270AH cells. I tested all cells with a EBC-A40L tested with the following cycle. All testing was done around 15degC ambient temperature.

• Charge to 3.65 volt / 40A
• Wait 30 min
• Discharge to 2.70volt /40 A.
• Charge to 3.4 volt cutoff at 20A.

The discharge Amp/Hour current was noted.

• From the cells tested 39 cells where in the range of 282ah - 270 ah.
• 19 cells where in the range between 270-260ah
• 6 Cells where below 260 with the lowest cell 255ah.

This means 40% of the cells are not inside the range which was advertised By Docan power.

All cells had valid QR codes and were manufactured 3 June 2022. Link to one of the cells: Gobalpower
Cells internal resistance was between 0.21 - 0.19 ohm, all cells arrived around 3.292 and 3.295 volts. None of the cells where swollen, well packed and the packages included 4 busbars and nuts.

 

[Sverige]

Not great results @Ahjwinth. I’ve also been reading another Eve 280k capacity test today (by the guy behind the Electrodacus BMS) which concluded they were below the rated storage performance.

 

[Alkaline]

I Promised to share my testing results from the Docan power supplied EVE LF280K cells which is finally completed. I ordered 64 cells in total Which were specified as 280 to 270AH cells. I tested all cells with a EBC-A40L tested with the following cycle. All testing was done around 15degC ambient temperature.

• Charge to 3.65 volt / 40A
• Wait 30 min
• Discharge to 2.70volt /40 A.
• Charge to 3.4 volt cutoff at 20A.

The discharge Amp/Hour current was noted.

• From the cells tested 39 cells where in the range of 282ah - 270 ah.
• 19 cells where in the range between 270-260ah
• 6 Cells where below 260 with the lowest cell 255ah.

This means 40% of the cells are not inside the range which was advertised By Docan power.

All cells had valid QR codes and were manufactured 3 June 2022. Link to one of the cells: Gobalpower
Cells internal resistance was between 0.21 - 0.19 ohm, all cells arrived around 3.292 and 3.295 volts. None of the cells where swollen, well packed and the packages included 4 busbars and nuts.

How can this be? If you have a valid qr code made after March 2022 then it would be grade A. Post pictures of your QR codes, in fact peel back the black cover and post pictures of your qr code and your ZKE resuls

watch this video:

See if you have a fake RE-LASERED QR CODE (notice the scratches from sanding marks) If so then you got scammed. Did you get a test report?

 

[RCinFLA]

I Promised to share my testing results from the Docan power supplied EVE LF280K cells which is finally completed. I ordered 64 cells in total Which were specified as 280 to 270AH cells. I tested all cells with a EBC-A40L tested with the following cycle. All testing was done around 15degC ambient temperature.

• Charge to 3.65 volt / 40A
• Wait 30 min
• Discharge to 2.70volt /40 A.
• Charge to 3.4 volt cutoff at 20A.

The discharge Amp/Hour current was noted.

• From the cells tested 39 cells where in the range of 282ah - 270 ah.
• 19 cells where in the range between 270-260ah
• 6 Cells where below 260 with the lowest cell 255ah.

This means 40% of the cells are not inside the range which was advertised By Docan power.

All cells had valid QR codes and were manufactured 3 June 2022. Link to one of the cells: Gobalpower
Cells internal resistance was between 0.21 - 0.19 ohm, all cells arrived around 3.292 and 3.295 volts. None of the cells where swollen, well packed and the packages included 4 busbars and nuts.

One of the largest variations in the manufacturing process is the printing thickness of graphite and LFP electrodes ink on the copper and aluminum foil, respectively. They measure the final thickness of a foil/electrode after baking on electrodes for the manufacturing lot foil print roll and add or subtract a few turns of laminate wrap to roughly compensate for electrode thickness variation to yield similar AH capacity. It can be based on final lamination roll weight to ensure similar amounts of graphite/LFP material which is closest to telling final cell AH capacity.

For these cells, the electrode thickness is in the 130-170 um thickness range. That is only 5.1 to 6.7 thousandths of an inch range so it is not easy to have perfectly uniform electrode thickness.

Thinner electrode thickness will have lower cell impedance but require more layers to lamination wrap to get the same AH capacity from cell.

Matched cells are selected to be from same lot run to have similar layer thickness. Automotive customers sometimes also require grading for similar self-discharge current which requires another time-consuming test at elevated cell temperature to speed the testing of self-leakage rate.

 

[Alkaline]

Bro,

You sound more and more like the Phd_in_everything guy on reddit just sayin...?

 

[B.Babs]

I'm pretty happy so far with my Grade B LF304 purchase from 18650. All the cells have tested out at 315 Ah on my little fan tester with its 20A max load. Even the Wh capacity seems pretty good despite voltage drops in my connections and in the electronic load. Testing was done on individual cells. I'll be doing a high capacity test once the battery is built and hooked up to a 2000W inverter.

 

[Ahjwinth]

I peeled the top of one of my cells. See below pictures of the extra QR code I found. I can not judge if the cell has been recoded since the other QR code is invalid at the side of Gobalpower. Maybe it is an extra code used during manufacturing, as it is a shorter number.
I will post the results from all cells later on when I have them together.

 

[Alkaline]

Clean off the residue I can not not tell unless the entire top area is clean. Peel it off all the way and clean it see if you find sanding marks.

EVEN the worst grade A 280K are able to get 276AH so when you say cells with intact QR codes made in June of 2022 are getting 260Ah something is not making sense.

 

[Nelius]

I guess i can share my results as well.

Eve LF280K, energy storage grade. (Call it A or B, they are not certified for automotive use, but dont have "B" on the QR code)
Bought from Shenzhen Qishou @ 140USD each included VAT and shipping to DK.
Production date 11 august, 25 august and 30 august 2022

13 cells tested so far:
Testing from 3,55 down to 2,65.
5 mins resting between charge / discharge.
50A charge / discharge

Lowest cell were 280,05 AH, highest were 284,66 AH.
Average 282,68 AH

Lowest cell group are the 4 cells with production date 11 August.
Those from 25 August and 30 August are clearly higher.

Very happy with the cells so far.
Before the tester arrived i setup a temporary 16S battery for the inverter, charging and discharging at around 120A.
No swollen cells, no problems.