Lithium based chem, "7,000 full charges"

[bob_ninja]

I have been following the new Lithium based batteries being developed for tools and cars. In particular, A123 is one of the vendors competing for the GM Volt contract.

In this article there is a reference to

7,000 full charges

Since these tests are for cars which usually control SOC between 40% and 90% I am guessing he refers to cycles within these parameteres. So they don't do deep discharges. Still 7K cycles sounds pretty amazing

I am surprise these new batteris and their new chemistry has not been discussed here more.

I recently bought a set of new Black&Decker VPX tools that is based on these A123 batteries. A single VPX 7V battery is based on a pair of A123 cells. So far I've been using the mini-saw tool a fair bit. At times the load was high and I actually felt the battery warm up. So far they are holding up to all the abuse I throw at them. Of course, the main question of lifetime will take some time to answer.

Oh, and it didn't explode .... yet

Anyway comments?
Even at half as much, say 3K cycles they blow away NiMH

Here are more comparisons:

Eneloop AA: 1.2V 1950 mAh, 2.34 Wh, $2.50
VPX: 7V, 1100 mAh, 7.7 Wh, $20

At this point VPX looks bad. These are prices I paid this year.

As I don't drain my cells 100%, I will use an average 30% - 95% SOC range, which means I use 65% of capacity.

Expected lifespan

Eneloop

2.34 Wh x 65% = 1.521 Wh
500 cycles = 760.5 Wh, 0.33 cents / Wh
1000 cycles = 1521 Wh, 0.16 cents / Wh

VPX

7.7 Wh x 65% = 5 Wh
2,000 cycles = 10K Wh = 0.2 cents / Wh
3,000 cycles = 15K Wh = 0.13 cents / Wh
4,000 cycles = 20K Wh = 0.1 cents / Wh
5,000 cycles = 25K Wh = 0.08 cents / Wh

It should be noted that Eneloops are older and NiMH infrastructure is long established while A123 is brand new stuff. So I expect their prices will come down in the future.

Still, according to this their cost relative to total energy supplied over lifetime looks better than one of the best NiMH. In other words, if they indeed last as long as claimed then they actually end up being cheaper.

Now just to clarify, I am more interested in higher drain apps for these cells. So you are not going to power your smoke alarm with these cells that can supply many kWh over their lifespan. Instead, things like power tools or powerful lights, etc. In particular, they are supposed to be more tolerant to high current drains. I got VPX toolkit for experiment, see how much abuse theiy can take and how is lifespan affected.

As an example, I used the small saw tools on some very hard wood, likely beyond the intended design. I want to see what happens to batteries, if they last. The old NiCd packs tended to die after a few years of abuse and/or neglect (not charging entire winter).

I imagine that flash light people would be interested as they are supposed to be safer (no explosions). Anyway there are plenty of high drains apps.

For general electronics and such NiMH is perfect.

Anyway thoughts?

 

[bob_ninja]

More links:

http://www.slkelectronics.com/DeWalt/packs.htm
http://www.slkelectronics.com/DeWalt/index.htm

 

[LuxLuthor]

Bob, these and Emoli have been discussed a lot here, in many threads. Here is a thread looking at the various 18650 size cells which are less common than the 26700 sizes.

 

[bob_ninja]

Interesting read, thanks Lux

However, I am also looking for longer term use results, like after 1000 cycles it is such and such .... I suppose it is still too early

I found some RC discussions, but they don't care much about lifespan so abuse them a lot.

Also I am curious if these new Li chemistries will replace NiMH and NiCd in some apps. For power tools for instance NiCd is dead. These new cells are way superior.

Also, NiMH used to be safer than Li chemistry so NiMH still has a wide use, even for some high current apps. Now that the safety issue is resolved, I wonder if the new Li cells will start replacing NiMH as well in some applications.

Anyway I am just looking for comments accross chemistries.

 

[LuxLuthor]

Definitely they are replacing older battery chemistries in tools....but it is really important that these cells are not over-discharged, and need to have pack balancing or they won't hold up. Part of their initial detraction in comparison to Lithium Cobalt is their reduced mAh capacity in the same size form factor. That becomes a marketing challenge to switch the public from ever increasing battery mAh capacity over to an emphasis on safe chemistry as a trade-off. Most people who use Lithium Cobalt cells still have little awareness of fire/explosion risks....so you have to come up with some good reasonsto convince a person not aware of Li-Cobalt dangers to buy a lower capacity, more expensive Lithium battery.

The unexpected fire/exploding issues with Lithium Cobalt packs likely slowed down manufacturing enthusiasm for new Lithium chemistries until they have been used and abused for a good while. I am not aware of any size variations in A123/Saphion & Emoli/Sony-Konion smaller than 18650 which would be another required step for their widespread adoption.

It is very likely that the RC crowd, and uses like our direct drive flashlights are the few "unapproved" applications that are pushing/testing these cells beyond their normal pack PCB regulated production. This is just starting to be a possibly acceptable new battery chemistry.

Power tools and electric transportation where performance, durability, safety are tantamount are the cutting edge....which is why we have to cannibalize tool packs to get them.

In any case, you are right to be enthusiastic about them.

 

[bob_ninja]

I am glad that RC people are pushing them into other apps, besides power tools. I am looking forward to a more wide spread use. Even besides fire, the older Li packs just cannot handle abuse. Hence most laptop batteries loose their capacity after a couple of years. It is funny when I have a meeting and everyone with laptops is rushing to plug in. So much for portability.

In power tools area, I was frustrated by the size of older NiCd packs. My current packs are 19.2V which would be 16 sub-C NiCd cells. That is a lot of cells and odds are high that 1 or 2 will go bad sooner rather than later. It is a pain to replace one or two bad cells. People have a similar issue with Roomba robots that use a decent size NiMH packs. They don't last unless you pay attention to avoid deep discharges. In fact robots were changed to remove the "Max" mode that would deep discharge.

This new VPX system has simple tiny 2 cells in series. So the maximum damage is only 2 cells. In the worst case if one fails you are only throwing away a single good cell. When more power is needed you just use 2 packs (like their VPX drill). Not to mention it can sit on a shelf for a long time and not die due to SD.

Anyway I am pretty excited by this new Li chemistry and looking forward to more formats or perhaps more devices using the new format.

Good stuff

 

[bob_ninja]

From RC ppl:

Discussion - A123 long term: 1st 1,000 CYCLES done, 2nd +800cycles highC done, 3rd to 0V going on

Summary

Here is the original thread. He has over 1000 cycles with only like 10-20% loss in capacity.

Pretty amazing considering how harsh RC ppl are on cells. Long threads, so still reading ...

 

[LuxLuthor]

I think the other tests are those where individual cells without protection circuits are over-discharged (which can ruin the cells quickly), and those that are left sitting in the charger continuously (which people tend to do), and see how long they last....or those that are fully charged and left sitting, or those that are thrown in the trunk or glovebox during a freezing winter, or those that are dropped and dented.

 

[James35]

I'm one of those "RC people"


I've been using the A123 cells in my cars for practice. Serious RC racers (like myself) know batteries. We know how to get the most out of packs, what hurts them, and we have routines for lengthening their life, equalizing cells, etc... Long story short, A123's are the best battery technology out there in my opinion.

• 3.3V per cell, 2300 mAH
• Charge them 1000 times and only lose 5% of your initial capacity!
• Safest battery: Chemically designed so they cannot explode. (A123 Systems has a patent on this.)
• Designed to take a 10 amp charge!
• They only drop 3% of their charge per month! They can sit for a full year and still have 69.5% of their charge left!
• Can output a continuous 70 amp discharge. (120 amp peak)
• Wide operating temperature: -30C to +60C
• Only $14 a cell (for the larger 26650 version)
• 10 year shelf life

Here is a my first flashlight mod thread.

 

[LuxLuthor]

James, it's really great to get experienced RC users sharing their expertise. I get lost reading and asking questions on RC Groups forums which is one of the sites I keep on my bookmark toolbar.

I agree with you totally about the A123, on all aspects. The one thing that is unique however with flashlights--especially direct drive incandescents--is many times needing higher voltage in the limited diameter of hand-held lights.

It is extremely rare that we need more than 10-12 sustained Amps for lights, but pretty common to require above 7-9V....so I believe in this flashlight hobby, the Emoli/Sony-Konion are more practical. It is also a side benefit that they hold more mAh, and can use more readily available 4.2V/cell Li-Ion/LiPo chargers.

I did this thread showing both cells in the 18650 form factor here, especially since there are the larger diameter "Elephant" bodies shown, allowing a much higher voltage, and some new sales threads coming out:

• FiveMega's Elephant-II

• H22A's Mammoth

• Modamag's Colossus

Again, thanks very much for your post and expertise!

 

[bob_ninja]

Thanks James

"RC people" was meant with respect, of course.
I would still give up some energy capacity for safety of A123 even without high current.

 

[James35]

It is extremely rare that we need more than 10-12 sustained Amps for lights, but pretty common to require above 7-9V....so I believe in this flashlight hobby, the Emoli/Sony-Konion are more practical. It is also a side benefit that they hold more mAh, and can use more readily available 4.2V/cell Li-Ion/LiPo chargers.

Do you mean 3.6V? To my knowledge Li-Ion's are 3.6V with the exception of A123's at 3.3V. I was looking over your post about why you like the Emoli/Sony-Konion. Is the main reason because you like the voltage and capacity better? Would 3 A123's be too much voltage (about 9.9V)?

Since I already have the ability to charge A123's, I don't have to worry about that, but many other people might. So that is certainly a valid point. But I think the A123's have the Emoli/Sony-Konion beat in 2 key areas: cyles and self discharging. The airplane guys who do these charge/discharge cycle testing on the A123's did them under harsh circumstances. If you charge the A123's 1000 times at room temperature (77 degrees F) they only drop 5% of their initial capacity. A123 spec sheet. That's incredible to me, that after 1000 charges, your batteries are 95% as good as new! But more importantly, my wife and I agree that we really wanted a flashlight that will always retain a charge in it whenever we need it. The A123's only drop 3% of their charge per month which is extremely attractive to us. That is huge.

"RC people" was meant with respect, of course.

Yep I know RC racing is the greatest hobby ever. It incorporates so many aspects into a single hobby (attention to detail, mechanical skills, physics, electronics, sportsmanship, fellowship, developing consistency, documenting results, forum research, hand/eye coordination, reflexes, and self improvment). However, when it comes to the latest technologies, it's the RC airplane guys who adopt fastest. They were the first for brushless motors, LiPo's, and A123's. This is because (generally speaking) they don't have rules like RC racing, and we tend to adopt technologies a little slower. But I fell in love with the A123's as soon as I started using them. They have saved me a lot of money. For those of you who have a LiPo/Li-Ion charger, you can purchase a $40 Dapter. It has the programming to reduce the charge voltage to 3.3V per cell for the A123's.

The most popular charger that RC Racers use is the Duratrax ICE or Team Checkpoint 1030 charger.

They charge nearly everything. 1-10 cells, NiMH, NiCad, Lipo, Li-Ion, A123's (attended). Or add a Dapter for A123's unattended charging. Linear, pulse, reflex, and step charging methods. I use this to charge anything, including our squirrel-proof bird feeder, rejuvenate old NiCads in things like cordless drills, etc.. Nothing like having a nice linear graph to see what's going on too.

 

[LuxLuthor]

Do you mean 3.6V? To my knowledge Li-Ion's are 3.6V with the exception of A123's at 3.3V. I was looking over your post about why you like the Emoli/Sony-Konion. Is the main reason because you like the voltage and capacity better? Would 3 A123's be too much voltage (about 9.9V)?

Since I already have the ability to charge A123's, I don't have to worry about that, but many other people might. So that is certainly a valid point. But I think the A123's have the Emoli/Sony-Konion beat in 2 key areas: cyles and self discharging.

James, it depends on your source for the is the Emoli/Konion voltage specification, (here is the Canadian Molienergy reference that also lists them as 3.8V) but most of us have not been splitting hairs between them and Lithium Cobalt Ion which are listed as 3.7V, and because both can be charged with more available same Li-Ion/LiPo charger (vs. A123 charger spec--which most flashlight users don't have or know about yet), and charge up to virtually the same 4.2V full capacity.

Again, I'm not disputing the better construction, durability, and performance that you underscore of A123, but charged voltage of 4.1+ vs. 3.3+ V in the two chemistries gives Emoli a higher voltage needed for many hotwire applications (see link in my sig). Both chemistries have adequate amp performance for lights, so that's not as big of a factor, as it is in RC.

The # of charge cycles is valid, but I have not seen adequate comparisons with Emoli with average run cycle testing (that avoids the extreme abuses). Assuming Emoli is lower cycle life, when we are going crazy overdriving our bulbs, I think cycle life is less of a factor...within reason. I can't imagine needing to charge a set of flashlight batteries 500 times, let alone 1,000 before I feel I got my money's worth.

Another issue in the 18650 size, the Emoli/Konion seem to be more available & affordable than A123 found mainly in the 2 cell VPX packs.

So in summary, for flashlights, I think the advantage is towards Emoli/Konion because of higher voltage range, more readily available Li-Ion chargers that many have been using with their Li-cobalt cells, cheaper/easier to find in 18650. Once you go to the 26700, and are limited to a 3-4 C/D length Maglite, you have to contend with lower performing bulbs in those pack voltage ranges.

I have not found the Emoli are dropping their charged voltage signficantly, but then I don't let my lights sit for a year. In fact, out of curiosity, I just checked 6 of these Konion that I got and charged in January shown here, but have not used, and they are all showing voltage of 4.139 to 4.143 V !!! That surprised me, and resolves that question.

One caution with that Duratrax or ICE charger which I looked at is their limitation of NiMH pack cell numbers. We routinely use 13-15s NiMH (2/3A cells which fit 3 wide in a tri-bored D Maglite) packs....and I even have a 24s (Elite 4500mAh SubC) in a hand held spotlight, powering a 747 Aircraft Landing Light (GE Q4559X 28V 600W) bulb...so for me, the Hyperion 1210i was the better choice (adding on the LBA-10 balancer). It goes up to 30s NiMH packs. :naughty:

 

[bob_ninja]

....
The # of charge cycles is valid, but I have not seen adequate comparisons with Emoli with average run cycle testing (that avoids the extreme abuses). Assuming Emoli is lower cycle life, when we are going crazy overdriving our bulbs, I think cycle life is less of a factor...within reason. I can't imagine needing to charge a set of flashlight batteries 500 times, let alone 1,000 before I feel I got my money's worth.
.....

Strictly speaking of lights alone everything you said is basically correct. However, I hate having to deal with different apps using different formats. My ideal scenario would be a single general purpose format/cell type that can be used in everything, from simple electronics to light, power tools, toys, etc.

Well Ok, maybe 2 formats: AAs/AAAs for simple/small electronics and bigger A123 for bigger apps (lights, power tools, toys, RC, etc.)

In that case you could be charging it every day, so 500 cycles is less than 2 years. I still agree that even 500 cycles is fantastic. Just pointing out that depending on usage it may equate to less time than you think.

For instance, with VPX I use them outside in power tools (saw and drill), then inside in a portable vac (messy kids) and also in a light. That way I am seeing much more use of them than other formats that are more specialized.

Also this way having several VPX packs means that there is always at least one charged for light in case of pwer failure. Compared to AAs sitting in lights for many days possibly not being used (I use lights much more during winter).

 

[LuxLuthor]

But I don't actually know what the practical number of Emoli cycles are if not abused like RC guys often do. I am also mostly talking about the 18650 form factor, which requires they be taken out of the VPX packs. So you need to include their higher price, unique charger setup (although the $40 Dapter that James mentioned is an easy solution--but not one that most flashaholics are going to bother with), price per cell, and lower voltage/mAh that I keep bringing up with this application.

I don't think it is fair to extrapolate characteristics of 26700 form factor over to 18650 in either chemistry unless there is independent testing to back that up. For example the A123 18650 cells use a steel can, instead of the lighter aluminum in the 26700. Personally, I found that aluminum can was a hassle for welding/soldering.....and you end up having to use their battery pack heavy tabs which I didn't like....but this different can feature alone may affect performance. I just don't know if A123 has the same cycle life expectations in the 18650, etc.