Battery Management Technology Could Boost EV Performance
Researchers from the University of California, San Diego have teamed with Bosch and Cobasys to create a battery management system that would enable EV batteries to be charged to their limits. (Source: UCSD)
Definitely a much-needed technology development, but a commercial version in three to four years? That seems a bit long given how fast this market needs to move. Is it the algorthim the cause for such a protracted commercialization schedule or are there other factors impeding its release?
Beth, that's an important point. For one thing, this is a research project. I have worked on DARPA projects in the past. If it is DARPA it is, by definition, more basic research and this probably further out. Frankly, this has not been done and the researchers are speculating at this point. That said, they might well make it work. On the other hand, in three to four years, the landscape may be completely different.
Timing is also important in light of events from yesterday. A123 Systems filed for bankruptcy protection. One of the hot battery makers in the US, funded with massive amounts of Federal money, A123 looked like a winner. One of the big problems they had was that the demand for electric vehicles has not materialized. Another area I noticed they were interested in was utility storage. They were a lithium ion maker, and I don't know how useful that technology would be at that scale. It was the automotive sector, though, that hurt them. Their assets are being picked up by Johnson Controls, who is a large supplier to the automotive sector. This is a technology that needs lots more work.
Thanks for the that context, Naperlou. I hadn't heard about A123 Systems and that's a huge ding for the battery industry and for the momentum around EVs. They were definitely touted as one of the pioneers with promising technology so it's disheartening to see them ending up in bankruptcy protection. Again, it speaks to the time-to-market pressures on startups in the industry, which are caught between needing to hammer out the very real technical and development challenges associated with this technology, but also feeling the heat from investors who want/need to see returns from a commercialized product.
As Naperlou pointed out, Beth, this is a research project. Given that, if they could productize this in three to four years (as they hope), it would be a major success. In truth, it's a tricky undertaking. In essence, they are squeezing more out of a battery by operating it to its very limits, which can be dangerous to the life of the battery. Today's electric cars typically don't come close to the operating limits of their batteries because automakers don't want to "brick" them (consider the guy who bricked his Tesla battery early this year and was told a replacement would cost $40,000). To make this work, they need to understand every specific battery chemistry very, very thoroughly.
Good point about A123, naperlou. The weak electric car market was their undoing. And more are sure to follow. The "last men standing" in the saturated EV battery market will be those who have enough capital to hang on until vehicle sales start rising. As you point out, A123 struggled, even with government funding.
Great idea, but where is the charging infrastructure coming from to enable these potential charge rates?
There's enough grumbling about the costs associated with putting in a 220V circuit in EV owners' garages, and the current rate of charge for existing technology is already capable of exceeding that source.
Even at an EV "filling station" the power feed required would be pretty remarkable. Which means even more expense and complications to create a point of load energy supply for them.
How about getting the cost of the batteries down first?
Nearly 20 years ago I dealt with a unique charging algorithm. We'd take a 1Amp/hr battery (suggested charge rate of 100mA peak) and inject 2.5Amp into the unit for roughly 0.5sec. This was followed by multiple 10Amp discharge pulses on the order of 1mSec in duration. This cycle was repeated continuously and formed the basis for the charging process. The result was a 20 minute time required to charge a fully discharged battery to 100% capacity. In addition the battery was cool to the touch upon completion of the charging cycle resulting in the electroytes were not being stressed due to excess heat. Batteries could be cycled through the discharge and charging process for more than 2000 charge cycles prior to showing a 10% reduction in capacity. This algorithm was implemented on nickel-cadnium or lead acid batteries and to a lesser extent nickel metal hydride batteries. The discharge pulses acted to redistribute the charged particles within the electrolyte reducing resistance. Efficiency in a normal charging process typically drops off in a matter seconds after the charge current is initiated. Here the discharge pulses act to "reset" the operating point on the efficiency curve to a much higher level thus allowing for the rapid charging process at a much higher efficiency. Less heat and longer life! Haven't seen or heard of the technology since then.
One of the possibilities that might occur from this effort is discovery of technology unconsidered at this time or previously. R&D sometimes gives us much needed but unforeseen new products. I certainly agree that four to five years is time enough for a changing landscape--politically and economically but I suspect the overall effort will yield value added.
I agree that that other discoveries could come from this, bonjengr. This, in itself, was a little bit of a surprise discovery. Virtually all reserachers have focused on changes in chemistry as a means of boosting energy and cutting cost. I'm encouraged by it, but I do wonder how the general conservativeness of auto companies will play into it. Auto companies are very frightened by the idea of warranty problems and they often prefer to build in big factors of safety to head off potential issues. The idea of operating a battery so close to its limits could be scary for them.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.