Cycle Testing Observations…

[js]

Tom,

Fantastic work as usual! Thank you so much for all of the info that you selflessly provide to us. It has been invaluable to me on more than one occasion (as you know )

I assume that when you say "It seems that once you get over 2000-2200 mAh capacity, things change" that you are talking about AA cells only, right?

Because a 2000 mAh sub-C NiMH would definitely be a LOW-capacity cell, and a 2000 mAh AAA cell would be incredibly HIGH-capacity. My point is too obvious to belabor further. Unless I'm wrong, of course, and then please feel free to belabor it.

Anyway, Tom, really, THANK YOU for this. Incredible information and very enlightening and revealing. I never would have thunk it.

 

[david]

look at this
New Products

Vapex Offers New VTE2000AAP NiMH Battery

January 15, 2004

Vapex Technology Ltd. (Hong Kong) unveiled its new VTE2000AAP rechargeable nickel-metal hydride (NiMH) battery, which has a high energy density that is 15% over the normal 1,800mAh battery, a high drain performance of up to 3C, and a high shelf life that is 80% of the initial capacity storage for two years.


WHAT do you say ? Anyone has it ?

 

[bob_ninja]

Re: Cycle Testing Observations…

First thanks for the effort, we appreciate it.

Two chargers were compared. The Energizer 15 minute charger was the fast charger, and the Sanyo NC-MQH01U was used as the slow charger by utilizing one of the slow charge slots. The Sanyo charger charges at roughly a 1.0 amp rate in that slot.....

Tom

I am bit puzzled by your terminology. I would call the 15 min charger "super-fast" and the Sanyo @ 1 / 2.5 = 0.4C a "medium speed". I thought that 0.1C (250mA) used for the forming stage is a "slow speed". Not to split hair, the reason I bring this up is that some people prefer longer cell life to 15 min convenience. For instance, I simply have a set of backup Alkiline cells for the times when I am in a rush and just don't bother with "super-fast" or even "fast" speeds.

When I got my BC900 I used the default 200mA rate - "slow". Due to the issues brought up here I switched to 500mA rate which is still "slowish" according to my terminology. My objective is to get a longer cell life.

In fact, many uses are a low to medium power draw, such as wireless mice, wall clock, MP3 players, etc. They are not as demanding as

then discharging it at 2.5 amps down to 0.8 volts.

Now I realize this method was chosen for speed, so don't suggest you test other "slower" scenarios, fair enough. My question is if there is a way to extrapolate from your tests or some other trick to derive from your tests results for this other scenario?

"Slow" (0.1C to 0.2C) charge rate and "low" power draw rate (max 0.3C)

For me at least this scenario would represent the majority of uses, hence I am more interested in it. It is especially relevant to the multitude of electronic gadgetry that doesn't have electric motors or other moving parts.

Again, not to diminish your work, it just doesn't represent most of my use scenarios. The tests seem to show that after 150-200 cycles the cells were degraded, unusable for most applications, compared to claimed 500-1000 cycles. I thought NiCd handles about 1000 cycles and NiMH handles about 300-500 cycles (might have read it at the battery university site). I wonder if the "slow/low" scenario would get much closer to the 500 figure.

As an example from real-life slow/low scenario, I have solar lights in my driveway. Each has a pair of NiCd AAs and a LED light. Clearly it is a slow charge rate as solar panel is small. I assume it is a low power draw as LED is tiny compared to reular flashlight lamps. From the original set most of them have suffered from "accidents" (lawn mower and such). The 3 remaing lights still work very well. I think they are about 3 years old, which is 3 x 365 = 1000 cycles. In fact, they survived temperature range from -30C to +30C.

Hence cells can last a very long time and even exceed specs.

Again great work on your part and I am not attacking/criticizing it. Just wondering if there is a way to estimate slow/low scenario cycle life without spending yet more time/effort.

thanks

 

[SilverFox]

Re: Cycle Testing Observations…

Hello Bob,

I believe most of the test data that suggests 300-500 cycles has been done with lower capacity cells. I need to do another round with 2000 mAh cells and see if they hold up better.

I believe the only way to get the information you are looking for is to test using 0.1C charging and 0.3C discharging. Unfortunately, several hundred cycles would take a lot of time...

I got about 5 years out of my NiCd cells in my solar yard lights. I don't think they were full cycling every time, but in the winter they never completely charged up. However, they were only 900 mAh cells. I replaced them with 1000 mAh cells and they are working fine.

Tom

 

[VidPro]

Silver thanks for taking the effort to do this and represent the data.

 

[SilverFox]

Update:

I am doing a test on Titanium 2000 mAh cells. I am using the same chargers and will follow the same test procedure, however since I am seeing a similar drop in voltage I will be adding a "forming" charge every 25 cycles. This should give the "best" chance for the most cycles.

Testing in progress...

Tom

 

[jtr1962]

I just looked at this thread for the first time. Very interesting results. I would be interested in seeing if there is any improvement in cycle life using reflex (also known as burp) charging, and also if this charging method can revive the cells which have lost their capacity. Problem is I'm not aware of any commercial chargers which use reflex charging. Also, while the number of cycles you're getting seems low you're stressing the cells much harder than most normal usage. Still, the results are encouraging. Even with fast charging you get 100+ cycles before significant capacity is lost. In a situation where someone recharges their cells once a week that's two years of use. From a cost standpoint (compared to disposable cells) that comes to $0.03 per cell (assuming the Sanyos cost $3 each). That's still an order of magnitude cheaper than disposables.

 

[Anders]

Thankyou Silverfox for the hard work you are doing for CPF.

Your threads is the first i read every day and in most cases the most interesting ones.

Anders

 

[MrAl]

Hi there,

Just a note for some that were talking about partial charging their cells
more often...

A partial charge does not count as a full cycle. You can probably top off
your cells many times before it equates to having used one full charge cycle.

 

[SilverFox]

Hello JTR,

The ICE charger can do reflex charging, but it won't bring the cells back to life. I have tried several chargers (including the Schulze and ICE, and others) and several charge rates and have not been able to get the test cells to "recover."

Tom

 

[jtr1962]

The ICE charger can do reflex charging, but it won't bring the cells back to life. I have tried several chargers (including the Schulze and ICE, and others) and several charge rates and have not been able to get the test cells to "recover."

Well, I had to ask. I haven't had any luck trying to get bad cells to recover, either. I have a few cells from a batch of 50 1800 mAH Nexcells which have the same symptoms as your cycled test cells (lower discharge voltage, much less capacity). They didn't go through that many cycles, either.

To add to the mix of possible variables you might test for I wonder if fan cooling will extend the number of cycles significantly. Lately I always use a fan regardless of charge rate. It seems like it delays the charge termination a bit while preventing the cells from getting anything beyond slightly warm, even at 1000 mA. I'm wondering if heat is the primary failure mechanism for NiMH/Nicad. Might be something worth testing for.

I've thought of doing something like this myself on occasion when I have the time. I'd probably want to automate the process somehow. Too bad the BC-900 doesn't have the ability to go through some set number of cycles.

Anyway, great work. :thumbsup: This thread is yet another valuable resource which you have contributed to CPF. :twothumbs

 

[Billson]

Even with fast charging you get 100+ cycles before significant capacity is lost.

jtr,

What rate is considered fast charging? I have a couple of Sanyo 2500's that have virtually died after less than 50 cycles. Their capacity have decreased to around 2100mah and self-discharges to below 1.2V in about 3 days. I have never charged it at more than 1A on the La Crosse charger so I'm quite curious. I have dropped the cells a few times though.

 

[SilverFox]

Hello Bill,

Fast charging is usually charging in 1 hour, or less.

Tom

 

[Billson]

Hi Tom,

Thanks for the clarification. I must have gotten some bum cells then.

 

[SilverFox]

Update:

I am nearing 150 cycles on the Titanium 2000 mAh cells. They are handling this charging and discharging test a lot better than the Sanyo 2500 mAh cells did.

With 2000 mAh, or less, cells, I can believe you could get over 500 cycles with "less stringent" use. These test cells may still be "usable" after this test. I will have to charge them up and check the self discharge rate... mmm, just what I need, more testing...

I only have a few more cycles to go, then I will post the results.

Tom

 

[JimH]

Silverfox, Thanks for all the effort you are putting in on this project. The biggest problem with this kind of testing is the amount of time it takes. By the time you come up with useful results on a particular battery, that battery is obsolete (at least for those of us in constant quest for the latest and greatest).

I think the most useful results of your testing would be the ability to say that, on average, batteries produced by company "C" perform better than batteries produced by "company A" (subject to constant reevaluation).

 

[paulr]

Failure mode I've encountered with Sanyo 2500 mAH is they don't lose capacity but they develop a rather fast self-discharge (a few weeks instead of several months). I've killed several NiMH cell phone packs without heating them much, by simply overcharging at 0.2C for a day or two (stupid Motorola so-called "Intellicharger" has no smart charge termination at all). Overall I think all these claims of 500-1000 cycles are just marketing fantasy, like 100,000 hour LED life, maybe under lab conditions but not in the real world. In a decade or so of using cell phones, laptop computers, etc. I've never gotten more than 100 or so cycles out of any kind of packs.

 

[WildChild]

I think the Sanyo had problems with high self-discharge at the end of 2004, early 2005. Same thing with Energizer 2500 mAh made in Japan. All mine developped high self-discharge within 5-6 cycles. I have some Sanyo 2500 mAh manufactured at the end of 2005 and so far, no problem with high self-discharge after over 10-15 cycles.

 

[NiOOH]

Thanks Tom for all the time, effort and money you have put into these tests!Interesting results indeed. Your results and observations are similar to mine, i.e the newer high capacity cells are not very durable. Here is my example. I have 4 AA 1600 mAh Kodak cells (Made by Sanyo), which I use since early 2003. They have estimated a minimum of 300 cycles, at least 100 of them deep. They have been charged on various chargers (Lenmar Mach1, Conrad CM, Lenmar Pro66, LP 4KN, BC-900 etc)These cerlls are testing 1450-1500 mAh at 0.5 A discharge current, maintaining above 1.2 V for 2/3 of the discharge time. For comparison they tested little above 1600 mAh when they were new. In short, I consider these cells reasonably healthy and still use them. In contrast, my 2500 mAh Sanyo Industrial cells are showing signs of age after only a year of use and about 100 cycles. One of them self-discharges faster than the others, not cathastrophically, but still faster. 2 other cells show 10% lower discharge capacity compared to when new. I am planning to match 4 cells and retire the the other 4 alltogether. These cells were charged at max current of 2 A, i.e. not even 1C and often less, using mostly relatively gentle chargers.

And another question question to Tom. I see you are using NC-MQH01. Have you measured the trickle charge current of this charger? I bought one recently, to replace my broken Pro66. Do you have any other comments on that charger? To me it's quite good. Sanyo states that it uses PVD for termination and the cells (especially the outer slots) are not heating much. I only wish it had 4 separate charge indicators.

Thanks once again.

 

[SilverFox]

Hello NiOOH,

One of the "features" of the Sanyo is that it does not trickle charge. The charge stops when the light goes out.

I have noticed that severally mismatched cells will not fully charge in one cycle. This does not happen each time, but I have noticed it on a number of occasions.

I like the charger. I gave one to a friend of mine that killed his cells by leaving them to trickle charge for extended periods of time. It is nice to have the option of two charging currents and the battery checker seems to work reasonably well. It also seems that there is wider spacing between the charging slots to keep the cells cooler.

Tom