Eneloop discharge/over discharge (test to follow?)

This post will be lengthy and I apologize in advance.

I have read time and time again how it's better to only partially discharge a Li-Ion cell before recharging it. It's better for cell life. I cannot seem to find advice on this regarding Eneloops no matter how much I search. In short, what I want to know is, would it be harder on the Eneloop to discharge it waaay down then recharge it, or discharge it half way and charge it twice as many times?

To be more clear, let me give an imperfect example: 2 cells have a 2000 mAh capacity. Cell 1 is used and drained to 0, then recharged 100 times. Cell 2 is used and drained to 1000 mAh, then recharged 200 times. At the end, which cell would have a higher capacity remaining? I'm guessing the cell that was only half drained and charged twice as many times would be in better shape, but I can't find opinions in other threads to support this. These Eneloops are listed at 1500 cycles, but surely the cycle count has to be affected by the "depth of cycle"

I know Eneloop reversal is bad, so for the sake of the question, lets rule that out. Lets focus on single cell lights here (like the Zebralight EDC and H series lights)

I sometimes find it hard to gauge when to swap cells. This is especially true on my single cell AA lights. If I use it in 5 minute bursts, some on high, some on high 2 (which has double the run-time) and several times on lower settings, I lose track of how much juice is left. I might grab it out of my pocket, go for high, only to realize all I have left is medium and low. Is this hard on my cell or is this getting the most out of my cell? Is 1500 cycles 1500 cycles regardless of charge/discharge levels? Surely not, but maybe moreso than I think. Maybe you are better off letting it go at least 85% of the way down before a recharge. That's what I'm reaching to find out here.

Because this has been bothering me and I haven't found enough opinions in past threads I've searched, I'm considering doing a test. Taking maybe 4 new cells evenly matched in capacity, then doing some long term tests (at least 25, maybe 50 or 100 cycles or more) then taking results of capacity. The new capacity may be a gauge of cell life or how these circumstances wear on the cell. I can also look at the initial voltage given on the C9000. I don't really want to do this test. It will be time consuming and equipment consuming. I'd rather just get some of your opinions. Surely if I do this test I will share my results.

Has anyone done a similar test that I'm missing? Can you tell me your opinion on when to recharge cells for best lifespan? I appreciate your thoughts!

Google "NiMH depth of discharge vs cycle life" and you'll find a number of articles.

I haven't seen anything specific to Eneloops but the general case for NiMHs seems to be that partial discharges increase cycle life. The relationship appears to be non-linear, i.e., if you get N cycles going to 100% discharge, you get considerably more than 2N cycles by going to only 50% discharge.

If anyone is interested how do they find out those 1500 cycles:
Battery life based on IEC-61951-2 (7.4.1.1.).
Google this document and see yourself.

If you don't want to google it...the cell is 48 times cycled at charge:0.25C for 3hours 10minutes and discharged at 0.25C for 2 hours and 20min.
After those 48 cycles is discharged at 0.25C to 1V. Then is charged at 0.1C for 16hours, left standing in charged condition for 1 to 4hours and discharged at 0.2C down to 1V. Charged again at 0.1C for 16hours and again cycled 48times as stated above.

"Cycles 1 to 50 shall be repeated until the discharge duration on any 50th cycle becomes less
than 3 h. At this stage, a repeat capacity measurement as specified for cycle 50 shall be
carried out"

Btw I am little afraid that terminating charge at "fully charged" by dt/dV like all commercial chargers do will shorten the cycle life significantly as well as discharging it down to 0% every cycle. I am expecting <500 cycles when charged at standard charger even in best case scenario (Maha C9000 terminating properly and not overcharging etc..)
It would ve really interesting to have set of cells charged and discharged to various levels and see their performance over time. Unfourtunatelly I don't have time and resources to do this - I am already running long-term test concerning li-ions (different 18650 manufacturers cycle life comparsion and storage voltage and temperature aging factor). Expect first results here in ~1year (have only one charger :-( )

You would need a robot to actually test the batteries in any reasonable time.
I figure that after 10 years or so I can afford some more cooler and better cells anyway.
So If I do everything wrong several times a week , there should still be several years worth of
wrong doings.

Thanks guys for the valuable feedback and insight. I realize some of this seems obsessive compulsive in nature. Or maybe anal. maybe a bit of both. I know my cells will outlive need and I'll likely upgrade before these need to be replaced. I just find myself curious, right around the time "high" no longer works, how much I just damaged my cells. Did I just put a dent in my capacity? Is this going to be a bother in 3 months of "losing high" every second cycle? In a year, will my run-time be 7 minutes less? Those types of nagging questions harass my brain from time to time.

Your input gives a level of peace of mind, as well as some things to further research for myself. I still may do some "abuse vs. non-abuse" tests on Eneloops. I should have the means to easily do 3 cycles/week. I have spare lights, spare cells, and spare charging bays. A spare C9000 would make this much more scientific and simple. I wish the C9000 didn't terminate discharge at a predetermined voltage. If I could change that, running tests would be a no brainer. The real question is: Do I have the spare time and patience to accommodate my curiosity? I'm going to spend a week thinking it over and coming up with specific parameters. If it all pans out, maybe I'll have some info around the new year. Thanks again for the thoughts!

The problems I see with it is that by discharge less deeply you're likely to need to recharge them more often and that in turn will also add wear on them. Then of course there's the problem of having to monitor your cells even more closely.

I think you probably could stretch out their cycle life by not discharge nor charging them as fully at slower rates. The question then is: How practical and useful would they be having to do that?

The Eneloop FAQ recommends storing them charge so that they're ready to use and so that they don't need to be cycled as much to regain full capacity potential. Also, more fully cycling the cells seems to help keep up their performance in terms of how much current at voltage they can deliver.

So, there seems to be a balance between useability/performance and cycle life potential.

80% remaining capacity is generally regarded as the end of useful life of a cell, and cells will degrade exponentially in such a way that for most of their useful life they should have most of their capacity. Internal resistance/impedance is also an indication of cell health. Some of this performance can sometimes be regained by cycling a cell to rejuvenate it, more so if it's been sitting unused for a long time, but this of course can only do so much. Cells will invariably degrade over time and use.

Matching up Eneloop cells by capacity when I get them with my C9000s, using them in series cell sets where the cells only get used together, and having extra sets charged and ready to use works great for me so far.

Something additional to consider is the possibility that your AA to D adapters may be adding a bit of resistance, due to all the additional electrical contacts. That might affect your run time, even if only slightly.

WDG said:

Something additional to consider is the possibility that your AA to D adapters may be adding a bit of resistance, due to all the additional electrical contacts. That might affect your run time, even if only slightly.

Click to expand...

My Wifes radio is using six AA Eneloops in convertors and there is no problem as yet ... The set is drawing 190mA and the Eneloops easily last for ten hours ... For her (occasional) bedtime listening that is at least two weeks ... If I get 500 charges out of the Eneloops , that is about 1000 weeks (20 years) of use !
.