Molicel P50B variable charge-rate cycle life test

Test eqiupment:
- NEWARE BTS-4000 test system with T-type thermocoupler module
- The cells are connected to the tester by Micro-TIG welded Cu strips with a cross section of 6 mm2 with separate sense wires for 4W connection
- Tested in room with relatively stable temperature RT: 20 ±3 °C

Test settings:
- The CC-CV method is used for charging, but the charging process is terminated by a time here, for example 4C charging is terminated after 15 min, 3C after 20 min, .., 1C after 60 min. The consequence of using this method is that the cell is never charged to 100% SoC due to the internal resistance of the cell. The higher the charging C-rate the lower the SoC is achieved.
- For discharging, the standard CC method (3C discharge-rate for all tests) is used, with common termination voltage of 2.5 V

1000 cycle chart




Chart below shows dT = Tend - Tstart (The difference between the cell body temperature at the beginning of the charging which is RT: 20 °C with up to 3 °C tolerance and the body temperature reached at the end of the charging). As a Temperature sensor is used NEWARE T-type miniature (0.2 mm) thermocouple, attached with kapton tape.


2000 cycle chart




Chart below shows dT = Tend - Tstart (The difference between the cell body temperature at the beginning of the charging which is RT: 20 °C with up to 3 °C tolerance and the body temperature reached at the end of the charging). As a Temperature sensor is used NEWARE T-type miniature (0.2 mm) thermocouple, attached with kapton tape.


Chart below shows nominal capacity@energy test (according to IEC, the fully charged cell is discharged with 0.2C discharge-rate, for P50B it is 1 A). This test is inserted each 50 test cycles (100 test cycles above 1000 test cycles). The nominal capacity test is used, among other things, to determine the SoH, where the chart shows a level of 70% SoH (for the P50B, 100% SoH corresponds to 4.85 Ah or 17.5 Wh)





CC-CV charging: Voltage & Current & Capacity & Temperature charts

Test eqiupment:
- Professional Battery Test System with T-type thermocoupler module
- The cells are connected to the tester by Micro-TIG welded Cu strips with a cross section of 6 mm2 with separate sense wires for 4W connection
- Tested in professional Temperature Chamber (TC) with 100 l of internal volume with no air flow. Cell sample was pre-conditioned to test temperature which was very stable during the test TC: 25 ±0.5 °C (different temperature than the cycle life test)

Test settings:
- The CC-CV method is used for charging to 100% SoC, the charging process is terminated by cut-off current 50 mA here,
- before each charging test, the cell was discharged by 5 A current to 2.5 V cut-off voltage.

New:

Amazing data Pajda. Thank you for taking all the time and effort.

Small update:

- added 1C charge-rate charts

Small update:

- added more cycles

If these get to the $6 range for a hundred cells by next year I'll probably sell the 100 lightly used Samsung 50s I recently bought and swap them into my latest battery pack. Turns out I prefer Molicels- better ebike top end.

Small update:

- added more cycles

Small update:

- added more cycles

Small update:

- added more cycles
- 500 cycles at 4C @ 15min continuous charge-rate with no problem

Small update:

- added more cycles

Wow! Thank you so much for this amazing testing! I have been curious about this for a long time, so really happy to see testing done on this! Cycle life still seems amazing at 20A charging.

I'm curious as to why the charging isn't stopped at say 1A of current instead of with a set timer, is there are technical reason behind this?

Charging with a fixed current is never really optimal, so if the cell can handle being charged at 20A up to ~80% charge with good cycle life it can probably handle much higher currents in the 0-50% charge range. Will you be continuing this with 5C and 6C charging tests as well in the future?

Simon676 said:

I'm curious as to why the charging isn't stopped at say 1A of current instead of with a set timer, is there are technical reason behind this?

Click to expand...

This is more of a commercial reason , where the average user is not interested in the principle of electrochemical cells, but only in the charging time. I called it as “True C-rate” charging test.

Simon676 said:

Charging with a fixed current is never really optimal, so if the cell can handle being charged at 20A up to ~80% charge with good cycle life it can probably handle much higher currents in the 0-50% charge range. Will you be continuing this with 5C and 6C charging tests as well in the future?

Click to expand...

I ended up at 4C with P50B because of the limitation of the current tester to 20A. For higher C-rate I would have to switch back to the ZKE tester, but then we lose the temperature data (I am thinking about 5C which is the maximum continuous charging current by Molicel datasheet) Another thing is that I was more interested in the comparison between modern HP cylindrical cells, where it turns out in my other tests that even 3C charge cycling at >80% DoD is still just too much for most of the competition. So as you suggest, other competitors will need use of the step-charge with high C-rate only up to 50% SoC, ideally combined with preheating the cell to ca 50°C before charging.

Small update:

- added more cycles

Simon676 said:

Wow! Thank you so much for this amazing testing! I have been curious about this for a long time, so really happy to see testing done on this! Cycle life still seems amazing at 20A charging.

I'm curious as to why the charging isn't stopped at say 1A of current instead of with a set timer, is there are technical reason behind this?

Charging with a fixed current is never really optimal, so if the cell can handle being charged at 20A up to ~80% charge with good cycle life it can probably handle much higher currents in the 0-50% charge range. Will you be continuing this with 5C and 6C charging tests as well in the future?

Click to expand...

Outside of what Padja already wrote, stopping charging at max voltage serves as a sort of CC only charging, which does help with battery lifetime and with overpotential, serves as a sort of SOC limiter.

Small update:

- added more cycles

Bigger update:

- added more cycles
- 4C charge-rate test will continue up to 2000 cycles or sudden-death
- added new charts: V,I,C,T,dT, measured in Temperature chamber (TC) with very stable temperature without any airflow.

Small update:

- added more cycles
- started 2C charge-rate test

Thoughts and comments:
I'm still thinking about how to run a 5C charge-rate test, the biggest limitation now seems to be that ZKEtech testers does not support Time cut-off condition for charging. So I would have to set 8A charging current as a cut-off condition. This value corresponds to termination after 12 minutes of charging at 5C when the cell is new. It will slightly skew the results for direct comparison but it is worth a try.

I would also like to expand the data to compare HP cells fast charging. A first attempt at 3C @ 15min charging showed that this value is still too high for most of the market, however as an eternal optimist I would like to try 2C @ 30min next.

Bigger update:

- added more cycles
- added 2000cycle charts
- added nominal capacity test charts

Interesting, so if I'm looking a this right there is no much of a penalty for quicker charging?

riba2233 said:

Interesting, so if I'm looking a this right there is no much of a penalty for quicker charging?

Click to expand...

Exactly, not only the P50B technology allows you to charge practical application range 5-90 % SoC in just 15 minutes, but also very important feature is that you can do this under RT conditions (20 °C in my test). So you do not need to preheat the cells to cca 50 °C before fast charging for maintaning decent cycle life. Where this preheating process cost you not only the wasted energy, but more important it still takes ca 30 minutes even with modern battery pack temperature management systems.

The only problem is that due to the conventional tab design, the cell generates a lot of heat, so powerful cooling is needed to keep the battery pack charging temperature below 70 °C.

Pajda said:

The only problem is that due to the conventional tab design, the cell generates a lot of heat, so powerful cooling is needed to keep the battery pack charging temperature below 70 °C.

Click to expand...

Have you ever tried fast charging it without any cooling at room temperature ? Most of us charge our battery packs in our garages without any AC on.

Pajda said:

The only problem is that due to the conventional tab design, the cell generates a lot of heat, so powerful cooling is needed to keep the battery pack charging temperature below 70 °C.

Click to expand...

Now if only Molicel would hurry up with their implementation of tabless cells we could have the best of both worlds!

Elios.pk said:

Have you ever tried fast charging it without any cooling at room temperature ? Most of us charge our battery packs in our garages without any AC on.

Click to expand...

I thought about the question for a while and then realized that my description of environmental conditions is misleading. So all the results in this particular thread are done without active cooling. The conditions are further divided into:

Ambient Cooling RT 20°C this means that the all cycling of cells is done in the room temperature with relatively stable ambient air of 20 °C average (no direct forced air cooling applied) but there can be some air flow in the room from the testers fans and also the doors and windows are not airtight.

Ambient Cooling TC 25°C this setup is used only for Voltage & Current & Capacity & Temperature charts. The diference is that higher temperature of ambient air 25°C is applied and the tests was performed inside the temperature chamber with 100l of internal volume preconditioned to 25°C and just bedore test the fan is turned off. (no forced air cooling applied either). TC conditions are so slightly worse in terms of cooling than RT.

For active air cooling I am using FC (Forced Cooling) shortcut.