What do cell ratings actually tell us? Part 7B: Cycle Life

Status
Not open for further replies.

Mooch

Electron Wrangler
ECF Veteran
Verified Member
May 13, 2015
3,897
13,566
This is the second part of a two part post.

Example Three
Let’s look at two cells that perform very similarly, the Samsung 40T and the Molicel P42A. Knowing that they perform about the same (P42A a little better but let’s ignore that) we might look at how long each cell might last to help us decide which one to use, especially if one costs less.

The Samsung 40T is rated to deliver at least 250 cycles at room temperature before dropping down to 60% of its original capacity when discharged at 35A and charged at 6A. Another way to look at it is to say that after 250 cycles this cell will still have at least 60% of its original capacity. It might be 60.5%, it might be a lot more.

Cycle life ratings are set as minimums. It’s not a number where every one of those cells suddenly reaches the end of its life and starts performing like crap. It’s a number that all of those particular cells should reach though without dropping below the capacity they mention.

After 250 cycles some 40T’s will have dropped to 60% of their original capacity. Most will have dropped to a slightly higher level. Some though could still have a lot more capacity left and could be run for many, many more cycles before dropping to 60%.

On to the Molicel P42A…there are three cycle life ratings…
After 500 cycles at 4.2A discharge/4.2A charge the capacity will be at least 80%.
After 500 cycles at 10A discharge/4.2A charge the capacity will be at least 80%.
After 500 cycles at 20A discharge/4.2A charge the capacity will be at least 70%.
All cycled at room temperature.

Not sure why first two ratings have the same capacity percentage…datasheet typo?

At first glance the 40T might seem like it has a really short cycle life, only 250 cycles before dropping to 60% (worst case) of its original capacity. The P42A was only dropping to 70% after 500 cycles. But these cells were discharged under very different conditions!

Both the 40T and P42A were charged quickly, 6A and 4.2A respectively, so we can say that was roughly the same (remembering that that 40T was charged faster though).

But the 40T was discharged at 35A for its cycle life testing. This high discharge rate, along with its high charge rate, is typical for a cell being used for something like a handheld vacuum cleaner. Those devices need to stay light so they have small battery packs that are run and charged very hard.

The P42A was only discharged at 4.2A, 10A, and 20A for its three cycle life ratings. This results in much lower cell temperatures and a lot less internal damage to the cell (less aging). This can significantly increase a cell’s cycle life.

How can we compare these two cells then? Can we estimate which might last longer?

If we can find the manufacturer’s technical presentation document, often created when a new cell is released, they often have cycle life graphs that can help us estimate cycle life at different charge/discharge rates. Lots of other good info too.

If we only have the datasheets specs then we can try to spot a trend that would help estimate the P42A’s cycle life at where the 40T was cycled, at 35A.

The 42A stayed above 80% at 4.2A and 10A so we know that 10A has essentially no extra aging effect on the cell versus 4.2A. At 20A it drops to 70% (at least) capacity after 500 cycles.

So a 10A increase in the discharge rate led to a 10% decrease in the minimum capacity left after 500 cycles. Going further, we might assume then that going from 20A to 30A would drop it another 10%, to 60% after 500 cycles. Then going another 5A beyond that, to 35A, might drop us to 55% after 500 cycles.

But the temperature increase from running the P42A at 20A versus the 40T at 35A is huge. At 20A you are not at a high enough temp to start breaking down some of the chemicals inside the cell. At 35A you definitely are. This speeds up the aging of the cell.

How this actually affects the cycle life of the P42A at 35A we can’t say. But it will mean that the P42A will be at a lot less than that 55% capacity mark we just calculated in my opinion. I would put it at 40% or less.

So we have a P42A at 40% after 500 cycles at 35A.
The 40T was rated at 60% after 250 cycles at 35A.

If the loss of capacity per cycle is fairly linear (it is after the first few dozen cycles and before the cell reaches the end of its life) then that means we can assume the 40T would drop to about 30% after 500 cycles.

This leaves us with…
The P42A at 40% after 500 cycles at 35A.
The 40T is at 30% after 500 cycles at 35A.

They are not too far apart! But the datasheet specs made these cells seem like they had very different overall life expectancy (if you didn’t look at the details).

Since we made so many assumptions though I don’t think we can definitively say that the P42A will have a longer cycle life than 40T. Especially since the 40T was discharged so hard when setting its rating. It might last a lot longer than the P42A when used at lower discharge rates.

The only thing that will let us accurately compare the cycle life of these two cells is testing, which is exactly what the manufacturers would tell us to do.

The cycle life numbers in the datasheets are just a guide. Unless your application matches the charge, discharge, and temperature conditions used to set the cycle life rating you can’t say that the datasheet number is what you will get.

Unfortunately this means that, like so many things associated with using li-ion cells, the details matter and we have to spend a bit more time checking into things if we want to get the most out of them.
 
Status
Not open for further replies.

Users who are viewing this thread