Eneloop Self Discharge study

[SilverFox]

Hello DualMonitors,

The discharge method depends on what you are testing for...

If you want the largest capacity number, and have an application that depends on the batteries being used "hot off the charger," you don't want any rest time.

On the other hand, if you are trying to check your cells capacity according to the IEC standards, a rest time after charging is required.

RC people generally use their batteries "hot off the charger," and if that is your intended use, it would be better to test in a way that simulates your usage. Most flashlight use is done some time after the batteries have been charged, so a rest period would be better when you are trying to check performance.

Hot chemistry is also able to hold higher voltages under load, so if your application is voltage sensitive, you would want to test immediately while the batteries are hot.

As you can see, there isn't a single answer to your question...

Tom

 

[Turak]

Hello Turak,

You got most of the way the C9000 functions correct, but you forgot to mention the top off charge...

After it terminates the main charge, it charges at 100 mA for 2 hours before droping to its 10 mA trickle charge. Depending on the capacity of the cell, this usually ends up with the cell voltage in the 1.5 - 1.55 volt range and it being fully charged.

A major difference between the BC900 and the C9000 is that the BC900 begins the discharge immediately after charging, and the C9000 incorporates a rest period before discharging.

Tom

Hi Tom, DualMonitors, ....

DualMonitors - It does not matter which charger you get, the BC900 or the MH-C9000, they are both excellent chargers and both charge the batteries to within about 97-99% of their capacity...which is about the best your going to get with todays technology. I am just nitpicking some of the fine details about both chargers.


Tom - Hehehe.... No I didn't forget .... but i needed to leave more stuff for us to talk about....

Now that you have mentioned it.......

I agree that after the MH-C9000 terminates, typically at 1.47V (maxV) that it starts to apply a 100mA 'top-off' charge for 2 hours. I too have seen it terminate at 1.47V and then drift up to anywhere from 1.50 to 1.55V with the 'top-off' charge. But that's about it and with anything other than an extrmemly low mAh battery or the newer LSD batteries, the so called 'trickle charge' of 10mAh is just ridiculous. While the 10mAh rate might be adequate for a LSD cell, it should be closer to 25-100 mAh to keep up with a regular NiMh batteries self discharge.

Now.....as to the BC900. Well I have watched them terminate at 1.52V or even 1.53V occasionally, and then it proceeds to start a 'trickle charge' which is based on the charging rate you selected. Which it then proceeded to drift up to anywhere from 1.55V to 1.60V.

The manual for the BC900 says the trickle charge rate is around 5% of the charge rate which translates into anywhere from 10mAh for a 200mAh rate to 50mAh for a 1000 mAh charge rate all the way to 90mAh for the 1800mAh rate. Now this IS enough to keep up with the self discharge rates of regular NiMh batteries.

Personally, I wish BOTH of them had an adjustable trickle charge rate, along with an adjustable top-off charge rate and timer. Along with an adjustable maxV setting and an adjustable -deltaV rate.

I AM NOT trying to say that the BC900 is 'better' than the MH-C9000 or vice-versa. I am just pointing out some techinical details that I have tested and verified at my little table at home with 2- BC900's and 2 - newer MH-C9000's.....which is;

Given the following scenerio and the fact that you are using some newer batteries that have been exercised at least 5-10 times AND ARE WITHIN 50mA of each other when discharge tested (both are to improve testing accuracy).

If you put a BC900 (firmware 33) and a newer (rev G or H) MH-C9000 charger side by side for your test. It DOES NOT matter whether you grab the batteries off both chargers immediately after they both terminate (yes I know that your supposed to give the MH-C9000 an extra two hours) or even if you do wait the extra 2 hours (but this means leaving them on the BC-900 the extra two hours also to be fair) then test them......The BC-900 will almost always give you a fuller charge on a regular NiMh cell, but not necessarily on all the LSD cells (they seem to vary, even Eneloops).

Not to mention the fact that with the MH-C9000's two hour 'top-off' charge to get a 'full' charge.....seems to give the BC-900 an advantage in that overall it makes the BC-900 a 'faster' charger given that you use the same sharge rate in each charger. Mostly because.....the BC900 is terminating on -deltaV most of the time and applying a higher trickle charge afterwards.

Another thing to think about.....

If you follow the IEC standard of letting the cell rest an hour before testing it for capacity (which the MH-C9000 does).....then the self-discharge rate of the individual cell becomes a factor that skews your capacity test results that you are not accounting for.

Probably the most interesting thing I find is that when I am comparing the manufacturers ratings against what a BC-900 rates a batteries capacity at and what an MH-C9000 rates a batteries capacity at.....the BC900 is closer almost ALL the time. Really makes me wonder if ALL the manufacturers aren't running thier tests on the batteries 'hot off the chargers'.....hehehe.

 

[Codeman]

Hi Tom, DualMonitors, ....
...
If you follow the IEC standard of letting the cell rest an hour before testing it for capacity (which the MH-C9000 does).....then the self-discharge rate of the individual cell becomes a factor that skews your capacity test results that you are not accounting for.
...

This is true from an intellectual basis only in regards to battery usage in flashlights, where the cells are probably going to be resting for periods of longer than an hour anyway. In that respect, niether charger will give an accurate estimation of the real capacity during use, but the C9000's rest period will result in a closer approximation.

Besides, any cell that looses enough charge during a single 1 hour rest period to be of concern probably can't hold a charge long enough to be useful anyway.

The 2-hour top-off period is of no concern to me. I prefer to pull my cells as soon as the C9000 indicates "done" because I'd rather have a slightly increased cycle life than that extra bit of charge. After all, the cell(s) is going to lose capacity regardless of the existence of a top-off period.

Which charger is "better" is far more dependent on the intended cell usage than any specific difference between the chargers without regard to cell usage.

 

[SilverFox]

Hello Turak,

Keep in mind that cells left in a charger with a higher trickle charge will end up beeing cooked to death. NiMh chemistry is more sensitive to this than NiCd chemistry, so this may indicate that Maha has finally introduced a charger for NiMh chemistry in particular, and LSD chemistry in particular...

Another thing to keep in mind is that when cells are tested by the manufacturers, they are brand new cells. With the exception of the LSD cells, most of the cells we get are aged. Aged cells tend have less capacity than new cells, unless they are properly taken care of.

With this in mind, I may entertain an argument that if a charger is giving you rated capacity or over with aged cells, it is overcharging them... :devil:

This may also be the reason that the standard testing is done after the standard 16 hour charge at 0.1C. This removes the "personality" of the charger from the equation.

Tom

 

[Turak]

Hello Turak,

Keep in mind that cells left in a charger with a higher trickle charge will end up beeing cooked to death. NiMh chemistry is more sensitive to this than NiCd chemistry, so this may indicate that Maha has finally introduced a charger for NiMh chemistry in particular, and LSD chemistry in particular...

I totally agree with that. Leaving batteries 'trickle' charging on chargers for days at a time is surely slowly degrading that batteries performance and the overall number of cycles you will ultimately get out of it.

When I mentioned the trickle charge rates, it was not with the intention that you would leave a battery 'trickle' charging for any type of extended period of time. I personally, like to pull mine as soon as possible after they complete their charge cycle which is typically anywhere from right away up to about 3 or 4 hours. It was more from the perspective of truly overcoming the self discharge rates of a typical high capacity NiMh cell, which on a poorly perfoming cell can easily be 100mA per hour. Basically, the 10mAh rate on the MH-C9000 is useless other than with the LSD cells or low capacity batteries.

Another thing to keep in mind is that when cells are tested by the manufacturers, they are brand new cells. With the exception of the LSD cells, most of the cells we get are aged. Aged cells tend have less capacity than new cells, unless they are properly taken care of.

With this in mind, I may entertain an argument that if a charger is giving you rated capacity or over with aged cells, it is overcharging them... :devil:

Ummm....the term 'overcharging' seems used very loosely when it comes to battery charging.

Referring to NiMh cells.... technically, you can not force extra capacity into the battery by 'overcharging' it. Once a NiMh cell has reached its maximum capacity, continuing to charge it or 'overcharging' as some like to call it, does not give you any more mAh.

What it does do however, is to cause a small voltage drop in the potential voltage. Also the internal chemical reaction going on starts forming more gas than can typically be reabsorbed back into the battery, not to mention the heat that is being created because it can not absorb any more energy. If this continues for any length of time, the battery can be slowly 'cooked' to death or even 'vent' the pressure in extreme cases.

As far as 'aged' cells go.....that's kinda reaching for it.

For the first 3-10 complete charge/discharge cycles of a typical NiMh battery, it will actually increase its overall capacity as the chemicals and reactive materials become 'fully' activated/energized.

Depending on a few factors (i.e. were they ever charged, temperature stored at, if they were charged how low did you let them go before charging, etc.), a quality NiMh cell can easily go 3-5 years and still come within 5% of the manufacturers capacity ratings.

This may also be the reason that the standard testing is done after the standard 16 hour charge at 0.1C. This removes the "personality" of the charger from the equation.

You could say the 'personality' of the charger is what I was referring to, in a manner speaking....

One point I was trying to make, was it seems that even though the MH-C9000 chargers 'break-in' function is supposedly closer to the IEC standard of testing the batteries capacity.......it is interesting that the BC-900 comes closer to the manufacturers rating about 80-90% of the time...at least in my limited testing.

Personally, I think that the fact that it uses a 2A load that is simply modulated using a duty cycle to give you a 'simulated' or 'averaged' overall charge rate has something to do with why the MH-C9000 is 'always' low on the capacity measurements when compared to manufacturers ratings.

Another point I was trying to make was ......

Because of the changes to the MH-C9000 awhile back to alleviate the heat build up in the cells (especially cells with slightly higher internal resistances) when charged at their default 1A rate, the MH-C9000 no longer teminates on -deltaV for 'most' batteries. It terminates on maxV most, if not all, of the time now.

A good side affect of this is that I think it makes it a bit gentler on the batteries overall, probably helping to give more overall cycles in the life of the battery. A bad side effect is that it tends to slightly undercharge the batteries when compared to a charger that is terminating on -deltaV (i.e. BC900) most of the time.

 

[Bones]

...
When I mentioned the trickle charge rates, it was not with the intention that you would leave a battery 'trickle' charging for any type of extended period of time. I personally, like to pull mine as soon as possible after they complete their charge cycle which is typically anywhere from right away up to about 3 or 4 hours. It was more from the perspective of truly overcoming the self discharge rates of a typical high capacity NiMh cell, which on a poorly perfoming cell can easily be 100mA per hour. Basically, the 10mAh rate on the MH-C9000 is useless other than with the LSD cells or low capacity batteries.
...

Conversely, a 100mA trickle charge would grossly over-compensate for the self-discharge of most decently performing regular cells along with the low self-discharge cells.

In this regard, I think MahaEnergy got it right with the MH-C9000, especially considering the self-discharge rates will almost certainly continue to decrease as the low self-discharge technology evolves.

 

[SilverFox]

Hello Turak,

This brings up another consideration...

In evaluating various chargers, I have run comparisons of several chargers. I did these comparisons with a "test set" of cells that were giving me very similar performance from cycle to cycle. The chargers were used to charge the cells, then I let them rest for 30 minutes, then I ran a constant current discharge on my CBA II.

The results I came up with indicated that the C9000 charged the cells a little more than the BC900 did. The C9000 charged cells came in at 2.462 watt hours and the BC900 cells came in at 2.355 watt hours.

I have not checked the accuracy of the discharge values from either the BC900 or the C9000, so I can't comment on the effects of their various discharge methods.

Tom

 

[curtis22]

Sanyo advises to charge in the upright position. It's
suggested to do so because of the following reasons stated
in the previous email.

Best Regards,

Customer Service
Sanyo Energy (U.S.A.) Corp.
http://www.eneloop.com
http;//www.sanyobatteries.com

One wonders which side is supposed to be "up". It isn't labeled.

 

[Turak]

Conversely, a 100mA trickle charge would grossly over-compensate for the self-discharge of most decently performing regular cells along with the low self-discharge cells.

Quite true. There are a few 'better' methods they could have used though. One would have been to simply scale the trickle charge rate up based on either the charge rate or better yet, the battery capacity. Another nice method would have been to allow us to set the trickle charge and top off charge rates. The 'best' method would be to have the charger actually calculate the self-discharge rate and then automatically adjust the trickle charge rate based on that.

In this regard, I think MahaEnergy got it right with the MH-C9000, especially considering the self-discharge rates will almost certainly continue to decrease as the low self-discharge technology evolves.

Like everything...Maha got a few things right but they also got a few things wrong too, in my opinion. Probably the number one thing I dislike the most would be the display. Lacrosse got it right with the BC900's display method of showing all 4 batteries at 1 time and no irritating auto rotation of the display. Maybe put in a selector to toggle between the two display methods....if anyone actually likes the current one....hehe. I also wonder what they were thinking with the 'high beam' display backlighting (it puts out more light than a couple of my flashlights). Another thing they got a bit wrong was the impedence test before charging....wayyyyy too conservative. Maybe make that a programmable option too or allow us to set the voltage threshhold. Their 'fix' to alleviate the problem where batteries with slightly higher internal resistances were coming off the charger a tad bit warm when charged at their default 1A rate that had a dramatic impact on the way the charger terminates now....wasn't the best approach either.

Don't get me wrong....I like the charger (actually have 2 of them). Those are just a few things that keeps a 'good' charger from being 'the perfect' charger.

In evaluating various chargers, I have run comparisons of several chargers. I did these comparisons with a "test set" of cells that were giving me very similar performance from cycle to cycle. The chargers were used to charge the cells, then I let them rest for 30 minutes, then I ran a constant current discharge on my CBA II.

I'd like to hear bit more about what Silverfox thinks about the CBA II.

 

[SilverFox]

Hello Turak,

The CBA II is a great hobby device. It doesn't totally live up to its specifications, but it does a very good job at lower discharge rates.

Once you get it properly calibrated, and understand the influence of the lead wire resistance, it is also very accurate.

It is unfortunate that after 4 years they haven't addressed all of the initial issues, but when you are basically the only game in town you don't have to be motivated to improve your product.

I still like it and recommend it, but it would be nice to see some competition for them that would drive improvements.

Tom

 

[rdh226]

Not sure if this thread is still considered alive or not, the last post being 6 months ago...

Sanyo started the ball rolling with low self discharge NiMh cells with the introduction of their Eneloop cells. I, along with several others, have been testing these cells for roughly a year now and they seem to be holding up to the manufacturers claims.
...
I am only going to project out to the year mark because I am going to be using my cells. Someone else can do the 1 year testing…
Tom

OK; I decided to buy some and try them out myself. Bought 16 Eneloops from Thomas Bros.
Being Thomas Bros, I couldn't help but notice the Imedions too...so I bought 16 of them too.

Initially, I wasn't all that impressed. Off the shelf, all the batteries took a substantial charge;
For the Imedions, varied from 431mAh (not bad) to 1566mAh (not good at all); For the Eneloops,
varied from 621mAh (not particularly good) to 1263mAh (pretty poor, IMO). Although the
Imedions had the worst datapoint, overall they fared better than the Eneloops.

I then measured the first full-charge-cycle discharge capacity of each cell, then proceeded to
cycle them several times to break them in, as NiMH typically show a healty growth in capacity
with several break-in cycles. While neither set of batteries showed a dramatic improvement after
several cycles, the Imedions again won (slightly) over the Eneloops, showing only about 5%
growth in capacity, versus about 10% for the Eneloops (or, rephrased, the Imedions were closer
to providing full capacity more quickly).

I then put the fully-charged batteries aside, intending to test charge retention after a few months.

The days passed; the moons waxed and waned; the rivers rose and fell; the snows came and
went; Spring arrived. Merde! My batteries! After a full year, I "remembered" my batteries. Figuring
I had now toasted 32 expensive (well, relatively speaking) LSD batteries, I retrieved them and ran
them through their paces.

I stand amazed. After 1 year sitting neglected and ignored, my 16 Eneloops all boasted 1725mAh
to 1780mAh residual charge! One Year! Talk about consistent performance, Wow! The Imedions
fared almost as well, with remaining charge varying from 1665mAh to 1895mAh, with one outlier
cell "failing" with less than 300mAh remaining charge.

All testing was performed on my LaCrosse BC-900s (0.5A max discharge, etc., and so forth).

In the table below, the columns mean:

* Cell -- Cell (or "battery") number, 1 to 16
* Initial -- Initial 200mA charge
* cy1 -- First capacity cycle: 200mA discharge from full charge, recharge at 500mA
* cy8 -- Eighth capacity cycle: 500mA discharge, recharge at 1000mA
* 1yr -- 500mA full discharge capacity after 1 year after cy8 charge
* 1ycy1 -- 500mA discharge capacity after 1yr recharge

Here are my Eneloop/Imedion charge capacity and charge retention results.

Maha Imedion AA 2100 April, 2008 buy

Cell.Initial..cy1....cy8....1 yr..1ycy1..
----+------+------+------++------+------+
#01 | 0813 | 2250 | 2320 || 1770 | 2300 |
#02 | 0468 | 2240 | 2340 || 1730 | 2290 |
#03 | 0446 | 2210 | 2300 || 1730 | 2240 |
#04 | 1566 | 2280 | 2280 || <300 | 2260 |
#05 | 0433 | 2250 | 2360 || 1845 | 2300 |
#06 | 0441 | 2270 | 2380 || 1895 | 2310 |
#07 | 0930 | 2250 | 2410 || 1550 | 2330 |
#08 | 0520 | 2200 | 2300 || 1720 | 2250 |
#09 | 0541 | 2250 | 2340 || 1650 | 2270 |
#10 | 0523 | 2250 | 2370 || 1855 | 2310 |
#11 | 0559 | 2290 | 2390 || 1825 | 2340 |
#12 | 0541 | 2240 | 2340 || 1720 | 2300 |
#13 | 0800 | 2270 | 2340 || 1810 | 2310 |
#14 | 0429 | 2180 | 2310 || 1830 | 2260 |
#15 | 0431 | 2220 | 2320 || 1665 | 2260 |
#16 | 0814 | 2200 | 2260 || 1855 | 2230 |



Sanyo Eneloop 2000 April, 2008 buy

Cell.Initial..cy1....cy6....1 yr..1ycy1..
----+------+------+------++------+------+
#01 | 1188 | 1943 | 2170 || 1780 | 2140 |
#02 | 0643 | 1870 | 2110 || 1760 | 2060 |
#03 | 0908 | 1920 | 2130 || 1765 | 2090 |
#04 | 0733 | 1980 | 2120 || 1770 | 2100 |
#05 | 0768 | 1930 | 2150 || 1750 | 2110 |
#06 | 0621 | 1842 | 2100 || 1720 | 2060 |
#07 | 0748 | 1888 | 2120 || 1725 | 2080 |
#08 | 1210 | 1971 | 2150 || 1720 | 2110 |
#09 | 0720 | 1890 | 2140 || 1730 | 2120 |
#10 | 0724 | 1892 | 2140 || 1745 | 2090 |
#11 | 0650 | 1920 | 2170 || 1755 | 2120 |
#12 | 0730 | 1960 | 2150 || 1735 | 2110 |
#13 | 1263 | 1920 | 2150 || 1755 | 2120 |
#14 | 0687 | 1871 | 2120 || 1730 | 2080 |
#15 | 0692 | 1947 | 2170 || 1775 | 2130 |
#16 | 0713 | 2000 | 2190 || 1765 | 2150 |


I was very favorably impressed with how both the Eneloops and the Imedions performed, excepting
one "bad" Imedion (well, 31 out of 32 batteries ain't too bad, in my experience).

I don't care if "traditional" NiMH can provide higher per-cell capacity, this breed of LSD cell/chemisty
is clearly a winner, and far more practical to use, effectively matching LiIon cells for overall charge
retention (set aside and forget capability).

Overall, I give "first place" ever so slightly to the Eneloops for their more-consistent cell-to-cell per-
formance. But the Imedions (well, 15 out of the 16 of them) did manage about 10% more charge
capacity, which is significant and would I suspect place them "first place" in most folks' opinions.

Yay!

-RDH

 

[BentHeadTX]

Thank you for the one year results, you are either very patient or forgetful!

I have converted to all Eneloops myself and will remain true to their cause. My family understands that WHITE cells are rechargable so all is well.

Good to hear the numbers on my favorite battery, they work well and maintain their charge for a loooong time.

 

[rdh226]

Thank you for the one year results, you are either very patient or forgetful!

Forgetful gets most of the credit here...sigh. Still, 'twas a useful forgetful.

-RDH

 

[SilverFox]

Update:

In March of 2006 I ordered and set aside some Eneloop AAA cells. I just got to the point where I needed them, so I decided to see how much capacity they had left.

The cells were stored at room temperature for these 3 years.

They ended up with about 550 mAh of capacity. This is about 74% of their original capacity.

Not too bad...

Tom

 

[Black Rose]

You're the master of understatements

Were those cells charged before being stored or was that capacity out of the package? Either way, that's impressive.

 

[Burgess]

Wow ! ! !



That was before i even joined CPF !



Good information, SilverFox.


:goodjob::thanks:

_

 

[digitor]

Whoops, I got it wrong....

 

[ICUDoc]

Great data thanks rdh226 and silverFox.

 

[Bullzeyebill]

Tom, would you say that they were about 85-90% capacity when you got them, reading in the neighborhood of 1.33 volts or so each? If so then that is very good to have lost so little in three years.

Bill

 

[SilverFox]

Hello Black Rose and Bill,

With the original release of Eneloop cells in Japan, the cells were fully charged before being distributed for sale. I just took them out of the package and discharged them to see how much was left.

Tom