Must-see lecture on lithium-ion construction and behaviour

Malcolm

Malcolm

10 kW
Joined
Jan 26, 2007
Messages
755
Location
Newcastle upon Tyne, UK
75 minutes long, but well worth watching if you're at all curious about how lithium-ion batteries work:

"Jay Whitacre, a Carnegie Mellon professor with a joint appointment in Materials Science and Engineering and Public Policy, gives a quick overview of Lithium-ion batteries. Starting at the chemical level, he explains the properties and mechanics of the battery which give rise to macroscopic behavior"
http://www.ri.cmu.edu/video_view.html?video_id=60&menu_id=387
 
Hehe...
"...stay under 140 degrees C..."

Awesome Malcolm. Tks for this.

FYI others here, terrific details on the theory, science, economics of LiFePO4 cells, design and construction... Touches on super-caps also. EVen PbA. Really interesting about pulse charging and discharging to realize high C rates (w/out electrolyte breakdown/shortened life)...
1ock
 
Another thank you, very informative.
 
Outstanding lecture! Thanks for the link. A must watch. 49 minutes in some neat statements on the formation cycle. He says it is more important than the chemistry for the characteristics we desire! Says A123 may have optimized it. Interesting, as I always thought they mastered the chemistry, and it may "just" be the conditioning...

Makes me want to go back to school, as long as there are no exams!
 
Thanks, well worth the time.

Interesting that pulsed performance could exceed steady performance.
 
So the only thing that degrades cell life is heat? Does that mean I could water cool my ping pack and run it at what ever C rate I want? :p

I thought his stance on bms was odd. He is against them for economic reasons, which I can understand. Individually monitoring and balancing all 7000 some odd cells on the tesla roadster is out of the question, but what about each parallel group? What about a system of 100ah thunder sky batteries? If you only have one series string, a bms is cheap!

It's definitely a trip to see someone who put a lithium battery on mars refuse to pay BMS more than a few sentences. Perhaps he is unaware of how affordable Chinese bms tech is?
 
auraslip said:
Perhaps he is unaware of how affordable Chinese bms tech is?
Click to expand...

Affordable, yes, but often unreliable.

If I were to put together any kind of "got to work every time" pack, or one made of expensive cells and/or enough cells of other types to be a really expensive pack, I would not want to use any of the BMS boards I've seen so far from China, as their failure modes often seem to be construction-related (there are some with design issues but those don't count as they would already be known or findable in general before committing to using it).

I'd rather use something I'd have to put together myself from scratch or a kit (like the F/G BMS boards), than use stuff of dubious construction quality with probably counterfeit parts of unknown origin/quality. :)


It's possible that the tiny coverage of BMS in there is for that kind of reason. Also possible that it simply isn't the subject of the lecture, and therefore only mentioned in passing, as it were.
 
Affordable, yes, but often unreliable.
Click to expand...

I completely agree.

On the other hand I just got a brand new laptop for $300. The sophistication and reliability of that is a few magnitudes larger than even the most complicated bms on the market. It's possible to cheaply produce decent electronics. You just need good QC, and a large enough market.

I often wonder of the differences between the quality of products on the Chinese domestic market and the quality of stuff the export. Part of me thinks maybe they ship the crappy stuff over here because we can't do anything about it.
 
auraslip said:
I thought his stance on bms was odd. He is against them for economic reasons, which I can understand. Individually monitoring and balancing all 7000 some odd cells on the tesla roadster is out of the question...
Click to expand...
But they do individually monitor all the cells in the Tesla. Their system can even eliminate bad cells from the charge/discharge loop.
 
A late post, just got to say I learned a lot here ! Non of this really gelled, until I started to precharge my headway batteries, before assembly! 1st attempt charge in groups of 4 ie 12 volts N F G volts all over !! Next i ganged the whole lot in paralell and charged til 3.7 v, it took days, remove charge let them settle til no change 3.3., repeated this until they stayed at 3,5 volt. the last charge to get them to 3.7 less than an hour, mind you this is with a meager 2 amps over 24 cells in paralell!!
This tells my the capacity increase between 3.3 to 3.7 isn't nothing
I have run the bike hard, all cell read the same, the numbers say above 50 percent capacity, and matched
Have yet to explore the bottem side of discharge, i am guessing it the same
Do other lithium behave the same ?? Intriged by thoso lopo's

Hope to hear all opinions

Pete
 
movie was excellent information ( even though about 500mb ) for me most interesting points were:
1. the lower the dsicharge rate the more energy is available from a cell: this could translate into the lower discharge the more kilometeres you can expect from a pack
2. lower discharge rates will mean higher number of cycles per pack, in fact it seems very marked increase in pack longevity can be got by going lower discharge rates.
Conclusion for me: use higher ahr pack than is needed and run at lowest amps possible
Although these things are already known it was very interesting to see physical explanations of why these things occur.
Brief end section was also interesting on using internal resistance and capacity graph as a relatively simple means to find similar cells.
 
I don't know about anybody else but to me this is the kind of presentation that should be added to the "stickified" thread.

eh?
 
On the other hand I just got a brand new laptop for $300. The sophistication and reliability of that is a few magnitudes larger than even the most complicated bms on the market. It's possible to cheaply produce decent electronics. You just need good QC, and a large enough market.
Click to expand...

I have been thinking that very thing for a while. I have a LiMn cordless drill with embedded electronics. If I harvest the cells, the disconnect is recorded and the warrantee is voided. If I leave the cells intact in their original form, I am limited to the amps the embedded BMS will allow. The smart-charger the kit was supplied with, keeps a constant measure of the packs State-Of-Charge during the programed charging cycle.

For the purposes of pro-rating the battery for a warrantee return, the number of charge cycles is recorded. If I return a pack because it will not hold a charge, it is tested. If it fails for the dealer, I get a new pack. But, if it has 70% of its rated cycles logged, I will have to pay 30% of the price of a new pack, although I do get a new 100% life cycle warrantee. All the afore-described electronics are designed and performance-tested by some of the best people in the business.

As a result, a new Makita 28V LiMn pack with a tiny 3-aH size is $130, but it has a good track record in the real-world of construction workers, and it comes with a rock-solid 3 year warrantee.
http://www.toolbarn.com/milwaukee-48-11-2830.html
 
That was really interesting informations.. but as many other say... he contredict some truth we all know for a while...

Many pack on teh industry have a BMS..ALL LiCo pack like laptop DOES HAVE A BMS and have cell monitoring!

The only pack i know that dont have BMS that do an individual cel monitoring are the old gen Makita pack and some really cheap crap ebike battery from chineese i ever seen.

When he say it is too expensive and the industry do not choose this BMS option.. I completly disagree!! ..

YOU CAN NOT HAVE LIPO OR LICO PACK SOLD IN THE INDUSTRY WITHOUT CELL MONITORING THAT MAKE NON SENCE! :roll:

Doc
 
i think his point was for a car size ev, being sooooo many cells you cant monitor each single cell cost effectively from a manufactures point of view,
personally i gave up on bms years ago, I just check cell voltages every so often and balance individually, with problem cell isolated, on the down side........!!!! i've had a couple of fires!!! because i'm slack checking voltages.......using cobalts.
 
Very interesting video! Thanks.

Some things he said that are different from what I've heard around here:

You won't damage the anode(or was it cathode?) of LiFePo4 by over charging, only electrolyte damage at around 4.3V, so just stay below 4.0v on charging.

You won't damage LiFePo4 on over discharging, don't reverse polarity and electrolyte damage at under 1v.
 
You won't damage the anode(or was it cathode?) of LiFePo4 by over charging, only electrolyte damage at around 4.3V, so just stay below 4.0v on charging.
Click to expand...

I personally have over charged 4 out of 16 cells in a ping pack to 4.1v - they are weaker and now have only 80% the capacity of other cells.
 
You must log in or register to reply here.

Similar threads

Lock
Georgia Tech High-performance lithium-ion anodes
Replies
0
Views
378
Lock
Lock
MitchJi
Metal-AirBatteryBreakthrough? energy density >= 900wh per kg
Replies
1
Views
577
def215
def215
Back
Top