Eneloop & Enduro LSD Duration Testing

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
Eneloop & Enduro LSD Duration Testing (60 days update)

cumulativedatazi8.jpg


capacitylossjj7.jpg


energizer2500nl8.jpg


titanium2600fg8.jpg


titanium2700cb8.jpg


titaniumenduro2100eu3.jpg


sanyoeneloop2100xn6.jpg


=== Testing Details ===
1. Cells repeatedly formed @ 200mA for 16 hr to determine "max" capacity.
2. Charging method:
- LaCrosse 200mA pre-charge (allow minimal resting time difference)
- Triton final-charge with 5mV dT (assure the same end charge condition)
3. Standard storage at room temperature (Bay Area/CA) 65-80F (18.3-26.7C).
4. Frozen storage at refrigerated temperature 10F (-12.2C).
5. Frozen cells are thawed for 8-10 hrs prior to testing.
6. Discharge rate of 1.0 A
7. CBAII Details:
- Software version 1.0.8.1
- 0.019V offset
- (-)8mA offset
- Temperature
- Powerpole to Battery Test Clamp


=== Interpretation ===
Vavg - voltage average - higher value result in brighter output in the same "unregulated" light.
Delta Vavg - how much Vavg change over storage time.
Vsd - voltage standard deviation - larger value result in more constant output.
Delta Vsd - how much Vsd change over storage time.
Current Capacity - how much capacity you'll really get @ 1A discharge.
Delta CC - how much CC change over storage time.
Power Capacity - how much power capacity you'll really get @ 1A discharge.
Delta PC - how much PC change over storage time.
Runtime - how long you'll get the constant 1A discharge.
Delta RT - how much RT change over storage time.


=== FAQ ===
 
Last edited:

CroMAGnet

Flashlight Enthusiast
Joined
Sep 4, 2004
Messages
2,540
Location
Los Gatos, CA
Re: Low Self Discharge Duration Testing

I like my Titanium 2700 NiMHs but let's see you leave them in a light in the closet for six months! But isn't that what L91's are for? :shrug:
 
Joined
Feb 14, 2006
Messages
2,724
Re: Low Self Discharge Duration Testing

modamag said:
Well, all this hupla about low self discharge cell have got me going.
I was lucky enough to get a small sample from Amondotech (Thanx Wayne) for testing.

I want to conduct compare them versus your typical 2500 mAh cell of the day to see where the strong and weak points are.

I will be adding new data set as time goes on. Let's see if these are really low discharge.

OK, let the graph speaks ...

day01pq.jpg

Pretty weak start for the Enduro 2100 mAh but let's see if how they fair in two months.
Look at those Titanium 2700, they're a whopping 5% better than the previous generation.

* All batteries acquired are brand new in retail package.

What is the charge method? Quick charging can vary in termination point and introduce error.

I always use low current, timed charge method. For the purpose of testing, I think it is a great idea to use the testing current used by the IEC, which is 0.1C (270mA for 2.7Ah cell, 250mA for 2.5Ah and so forth) for 16 hours.
 

wptski

Flashlight Enthusiast
Joined
Jan 18, 2004
Messages
2,987
Location
Warren, MI
Re: Low Self Discharge Duration Testing

Handlobraesing said:
What is the charge method? Quick charging can vary in termination point and introduce error.

I always use low current, timed charge method. For the purpose of testing, I think it is a great idea to use the testing current used by the IEC, which is 0.1C (270mA for 2.7Ah cell, 250mA for 2.5Ah and so forth) for 16 hours.

Are you going to use low current all the time? How C/D low current cycles do you perform? Better off charging the cells the way you are normally going to charge them! You only need to charge at 1/10C the first cycle anyway.
 
Last edited:
Joined
Feb 14, 2006
Messages
2,724
Re: Low Self Discharge Duration Testing

wptski said:
Are you going to use low current all the time? How C/D low current cycles do you perform? Better off charging the cells the way you are normally going to charge them! You only need to charge at 1/10C the first cycle anyway.

The number of cycles is debatable as the datasheets do not specify number of cycles before full capacity is achieved. For testing purpose, its best to use laboratory conditions that is easily duplicated elsewhere.

Person A tests brand A 2500mAh using his XYZ smart charger.
Person B tests brand B 2500mAh using his ABC smart charger.

Person A produce a report of 2300mAh with 500mA discharage to 0.9V
Person B produce a report of 2500mAh avg with 500mA discharge to 0.9V

The problem here is that because charging conditions are different, we don't know which batteries are actually superior. The difference could be a result of the difference in charge algorithm.

If on the other hand, you use C/x rate for x hours, the charge method is reproduceable anywhere, allowing more accurate assesment of true battery performance under identical test conditions.
 

Scimmia

Newly Enlightened
Joined
May 22, 2005
Messages
7
Re: Low Self Discharge Duration Testing

I'm not sure I see the point of this graph. You're comparing a 2100mAh rated battery against 2500-2700mAh rated batteries? Is anyone suprised they didn't last as long? The Enduro registered about 7% less than it's rated capacity on your test, while the Energizer was about 6.7% less, the Ti-2600 was about 13.8% less, and the Ti-2700 was about 8.4% less. I'd say things don't look very good for the Ti-2600, everything else is as expected.

What I really want to see is a test of the low self discharge feature.
 

wptski

Flashlight Enthusiast
Joined
Jan 18, 2004
Messages
2,987
Location
Warren, MI
Re: Low Self Discharge Duration Testing

Handlobraesing said:
The number of cycles is debatable as the datasheets do not specify number of cycles before full capacity is achieved. For testing purpose, its best to use laboratory conditions that is easily duplicated elsewhere.

Person A tests brand A 2500mAh using his XYZ smart charger.
Person B tests brand B 2500mAh using his ABC smart charger.

Person A produce a report of 2300mAh with 500mA discharage to 0.9V
Person B produce a report of 2500mAh avg with 500mA discharge to 0.9V

The problem here is that because charging conditions are different, we don't know which batteries are actually superior. The difference could be a result of the difference in charge algorithm.

If on the other hand, you use C/x rate for x hours, the charge method is reproduceable anywhere, allowing more accurate assesment of true battery performance under identical test conditions.
The problem with using C/x is that you must discharge the cell(s) fully before using this method of charging but not everyone has the means to do so!
 

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
Re: Low Self Discharge Duration Testing

I'm currently using a 200mA charge rate, 5mV dV then 15mA trickle for a total of 20hrs.

Typically you want to do low current C/D cycles until the capacity no longer increase to record the maximum cell capacity. This can range anywhere from 3-8 cycles in my experience.

Guys also keep in mind that there is also cell-cell variation which I can not account for in this test due to limited sample size.

I would also like to observe the temperature effect on self discharge so 1/2 of the test will be conducted at room temperature ~70 F while the other half is held in refrigerated environment < 30F.

The main purpose of this study is self discharge among different cells with variation of storage temperature.


Scimmia: Keep in mind that I'm doing 1A discharge to reflect more closely of how we use the cells not 0.1C (or was it 0.2C) discharge, what the manufacturer usually use for capacity gauge.

This is only the begining of the test to see the "low self discharge" and where/when it's more appropriate to use LSD vs high capacity NiMh regardless of $$$.
 

Scimmia

Newly Enlightened
Joined
May 22, 2005
Messages
7
Re: Low Self Discharge Duration Testing

I realize that you're draining them much faster than they were meant to be, and I wasn't complaining about the capacity being less than rated, I was just pointing out that everything was just as should be expected, and I don't think your comments about a "Pretty weak start for the Enduro 2100 mAh" were warrented.

I'm very interested in seeing the final results of this study. Do you have any guess what the running time will be on it? I guess I'm not expecting any real results for atleast a couple of months.

If I may ask, how are you planning on testing? Checking voltage every week (or few days, or whatever)?
 

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
Re: Low Self Discharge Duration Testing

My bad not clearly understanding the writing.

I guess I had a "high expectation" where I was hoping that the labeling is a mistake.
2100mAh (standard) was technology of 2 yrs pass. I was expecting a little more :D

Discharge curve will be posted weekly for the first two months and then every other week.

I plan to carry this on for 4-6 months duration. Typically the majority of my cells are cycled periodically w/in 6 month period.

Just as a clearification, the main comparison is single cell over time not model vs models :D
 

JimH

Flashlight Enthusiast
Joined
Apr 8, 2004
Messages
2,714
Location
San Jose, CA
Re: Low Self Discharge Duration Testing

Jonathan,

What are you using to measure wh?
 

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
Re: Low Self Discharge Duration Testing

Simply math my friend. <giggle>

Wh = Ah * Vavg
Wh = I(discharge) * Time (hours) * Vavg

I wish I have that fancy contraption that you recently acquire. Maybe I'll head over your house with couple water guns ... start shooting your dogs ... as they bark and you go outside to check what's all the fuss about ... I'll climb up the rear balcony, then down the stair to borrow that fancy unit </end grin>

BTW: glad to see you're back fromt he 18 hrs work day in 1 piece.
 

Scimmia

Newly Enlightened
Joined
May 22, 2005
Messages
7
Re: Low Self Discharge Duration Testing

Ok, thanks for the clarification, but I guess I'm still not understanding something here.

I assume low self discharge means exactly what it says, if you charge them and let them sit on the shelf, they won't discharge as quickly as a regular NiMH battery would, as in dead a few months later. If I'm wrong in this assumption, I sincerely apologise.

Assuming I'm right, do you have a large number of these cells that you're testing? A 4 month trial like you're describing would require 12 of each cell, and even then there's no way to account for differences in the individual cells.
 

JimH

Flashlight Enthusiast
Joined
Apr 8, 2004
Messages
2,714
Location
San Jose, CA
Re: Low Self Discharge Duration Testing

modamag said:
Simply math my friend. <giggle>
Wh = Ah * Vavg
Wh = I(discharge) * Time (hours) * Vavg
Aw shucks - I thought you had bought some new toy for me to drool over or at least were using calculus to integrate the area under the curve.


modamag said:
BTW: glad to see you're back fromt he 18 hrs work day in 1 piece.
Survived the last 2 (back to back), but today is another one.
 

big beam

Enlightened
Joined
Jan 14, 2006
Messages
521
Location
upstate ny
Re: Low Self Discharge Duration Testing

I was going to get some ti's but looks like eneloop's hold higher voltage under load.Maybe I should wait for more testing befor I buy some
 

billgr

Newly Enlightened
Joined
Feb 16, 2004
Messages
174
Location
Harrisburg PA
Re: Low Self Discharge Duration Testing

i've bookmarked this thread. i'm most interested in finding out whether these are the "real deal"!!
 

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
Re: Low Self Discharge Duration Testing

*** Update ***
I'm sorry for the lack of updates recently. An unexpected work travel have tempered my weekly update plan. So I'll try to update the data every other week instead.

Original post have been updated with pure data and testing methodoly.
I have also received a small sample (4x) of Sanyo Eneloop cells from my friend dat2zip for testing.


*** Discussion ***
So far after two weeks of sitting on the shelf, all the standard (chemistry) cells still outperform the LSD Enduro. Most likely by mid-week three the LSD will overtake the 2500s.

The interesting result so far have been the refrigeration process. Frozen standard chemistry cell retained their capacity even better than LSD @ minor cost to the grid bill.

The most surpising result is that LSD doesn't seems to like the frozen process, Vavg drop significantly when this happened. It might be too early to conclusively tell at this point, we can be more certain once we get the next couple sets of frozen data.
 

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
Re: Low Self Discharge Duration Testing

*** 30 days Update ***

*** Discussion ***
I'm quite disappointed with all this low discharge hopela. If you look at the summary the percentage of self discharge of the Enduro & Eneloop are pretty much the same as the other Titanium.

What's the deal? Have anyone else with LSD cells seen similar phenomenom?

The Eneloop cell are really nice. The voltage underload for these things is exactly as Silverfox reported earlier. They maintain their load pretty well at high current draw.

cumulativechart8zu.jpg
 

altis

Newly Enlightened
Joined
Oct 26, 2005
Messages
124
Location
Warrington, UK
Re: Low Self Discharge Duration Testing

modamag said:
The most surpising result is that LSD doesn't seems to like the frozen process, Vavg drop significantly when this happened. It might be too early to conclusively tell at this point, we can be more certain once we get the next couple sets of frozen data.

Might be summat to do with the deionised water freezing:
http://www.stevesforums.com/forums/view_topic.php?id=96500&forum_id=51
 
Top