I'll be back... in 8 hours

Some of my readers have wondered why I have been off the blogosphere in the last few months. The reason is that we brought a house and the move from the apartment to our new place has been a bit of a time sink.

First we went through the four stages of home buying:

Stage 1: What the &%#@ do you mean they accepted our bid? I thought you said we were lowballing?

Stage 2: When you use words like "downpayment", does this involve us giving you a check?

Stage 3: I assume roof's are like batteries? Meaning, when you say it is at the end of its life, there is still 80% left, right?

Stage 4: Keep repeating after me: "Owning is better than renting" and, please, stop asking "why?"!

Then we realized that owning a house also meant owning things like leaf blowers and lawn mowers! So, when I saw that there was a battery-powered lawn mower, I jumped at the chance to push my favorite technology forward.

I was looking forward to using my expertise in batteries to maintain and extend the life of my lawn mower for many years to come.

I was not particularly looking forward to mowing the lawn, but owning a battery-powered mower seemed to make up for that.

Until I realized that the top-rated battery-powered mower uses lead-acid batteries.

Lead-acid!!! really! How old school can one really get.

My first thought: Start a battery company to make Li-ion batteries for lawn mowers.

Then I started thinking about this some more. There must be a catch here. So I started digging into what it was.

Let us do some math: The battery for this lawn mower cost ~$270/kWh. That is one expensive lead-acid battery. Presumably, it is a deep-discharge battery, and so it is better made than a car battery.

And there is probably a significant markup.

Did I mention that the mower was on sale for $300! More like a mark-way-way-up.

Considering a typical Northern California growing cycle, this mower will probably be used ~50 times a year (once a week). These deep-discharge batteries can probably go a few hundred cycles. So I'm thinking 4-5 years easy.

But the bigger problem is going to be the calendar life. Sulfation can kill these cells.

Having said that, the battery is probably going to be at the top-of-charge pretty much through its life (think about it: Mown for 1 hour. Keep it plugged in all week). And remember our rule for lead-acid batteries: Keep them charged. So I think we can expect to get ~3 years from these cells.

Lets do some math for a Li-ion battery. I bought one a few weeks ago. This battery will probably last me 3 years and get me ~300 cycles or so. So it has similar specs to the lead-acid battery.

The Li-ion battery, on the other hand, cost me a whooping $2300/kWh!! No... really. This is what it cost me.

Now... knowing Apple, a new word has to be coined for their level of markup. But still this is one expensive battery.

At this price, a Li-ion lawn-mower-battery would have cost me $800!

All you MBA-types are probably cringing because you all know that cost and price are very different from each other and that the price is dictated by what the market is willing to pay (my wife is an MBA and she gave me this spiel).

Granted. So let us do a cost-differential comparison. This comes out to be ~$50/kWh more expensive for a Li-ion cell. Using this, the cost of a Li-ion battery for this mower would be higher than the lead-acid battery by... the price of lunch!

Not at Chez Panisse. But at the LBNL cafeteria (same quality, but at a much lower price?)

So why not use a Li-ion instead. After all it is almost 5x the energy density.

The mower that I have been talking about in this blog is a push mower. So the battery does not need to drag itself. All it has to do is turn the cutting blade. And space is not a big constraint. The mower's size is dictated by the size of the blade anyway.

So why bother using a new type of battery when you don't really see much of a benefit?

Frankly, although I have not looked at the life-cycle, I'll take a bet that it is probably better for the environment to use a lead-acid battery considering how much of this lead is recycled. In comparison, all you recycle in a Li-ion is the high-value metals in the cathode. The rest, literally, goes down the drain.

Maybe the math will change for a self-propelled mower. Or if the weight of the mower is an issue. But I don't see any reason to jump to using a Li-ion battery for the mower I was looking at. This business plan does not appear to have much legs.

So what did I do? I got the corded version of this mower. It cost me $100 less.

I guess the price that I'm willing to pay to push my favorite technology forward is less than $100!

In my guilt I decided to do my part for the technology by buying a Roomba.

I'm not sure if you guys know about this amazing robotic vacuum. It is pretty interesting to watch. It has sensors that make it slow down when it approaches objects, detect dirt, and prevent it from falling off of stairs.

It is not particularly good at vacuuming. But it feels like it is cleaning the floors and isn't that what's important!

But here is the kicker. The one I have has a Ni-MH battery.

I suppose I should be thankful it is not a lead-acid, but Ni-MH? COME ON!

The last time I checked, the cost of Li-ion and Ni-MH batteries were pretty comparable. And the energy density of Li-ion is 2-3x greater. And remember that in this machine (unlike the mower) the vacuum has to drag its battery along. And space is a big deal. The smaller the footprint, the smaller the space it can vacuum.

So why use a 20th century battery for a 21st century machine? Strange.

A web search revealed that Li-ion Roomba's have apparently died prematurely. If this is the case, then iRobot (the company that makes the Roomba) needs to change battery suppliers.

The vacuum I got goes through 2 rooms and then runs out of juice. It could have finished my house in one charge if it had a Li-ion of the same size. Or you could have a better vacuum on it so that it actually picks up dirt instead of moving it around and still only vacuum 2 rooms. You get the point.

When I first saw this machine, I thought that all the things on Terminator (the movie) were coming true. Jokes apart, it really is a pretty decent robotic cleaner which does indeed find its way around. You can, pretty much, set it and forget it.

But then I realized that we had nothing to worry from the machines as long as battery technology evolves the way it has been in the past.

When Arnold Schwarzenegger's character said "I'll be back" on T2, he (it) actually meant "I'm running out of battery and I need to go find an outlet and charge for 8 h. I will then come back to look for you for the 1/2 hour my battery lasts. Wish me luck".

So much for the machines taking over.

If you constantly complain about how batteries are not evolving fast, did you ever consider for the second that maybe we are out to save the world in our own way?

Venkat

This week in sulfation

First off, I have to apologize to all 7 of my regular readers for not making a post in a while. I have been busy and hence the hiatus. Between watching reruns of Curb your Enthusiasm, election ads, and the Giants world series games, I need more time in a day. But all good things have to come to an end. I hope to resume regular postings starting now.

This week, I was rudely reminded that just because you know how batteries work does not mean that you can deal with a battery problem. But first some background.

Subaru's are excellent cars and I would highly recommend buying one. They are well made, they run great, and they hardly give you any problems (touch wood). But Subaru's also have small quirks that can be bewildering and a bit infuriating. Like not having a remote trunk opener or having only 3 wiper speeds. Or having this one light that does not turn off when the ignition is off and is so hidden that you can't tell that its on in the first place.

The light serves no real purpose, except to make sure to keep the battery industry thriving. Long story short, "someone" left that light on on Sunday (and I'm not saying who).

A bit more background: My car is 7 years old. My battery is (or was as of yesterday, but I'm getting ahead of myself) the original battery.

Monday morning at 7 AM I get to my car to start driving to San Francisco Intl. to catch a flight to Washington DC to be at the center of the action during election day (Tuesday). I was using a meeting on energy storage for grid applications (think storing solar and wind electricity) as an excuse to get someone else to pay for the trip. I was going to be gone all week.

Except my grand plan was crumbling because my car appeared to have a dead battery. I quickly realized what had happened.

When professionals are thrown into a crisis, they do not think, they react. Think Jason Bourne coming across an assassin or Neo coming across Agent Smith (in the Matrix).

Just so there is no confusion: I consider myself to be a professional.

In a matter of seconds the pieces started to fall into place in my brain without any real thinking. I knew my battery was probably pretty sulfated (and corroded with lots of shedding, for that matter). I opened the hood, took one look at the battery and I could literally see the sulfation on the plates (through the grime that covered the hard polypropylene battery shell). I knew that if I jumped the car right away I had a chance of saving the battery. I also knew that if I left it in the discharged state all week, I would probably sulfate it more and maybe end up killing the battery altogether.

To remind you, when you discharge a lead-acid battery you form lead sulfate. Lead sulfate is pretty insulating. But it has a bit of solubility in the electrolyte (sulfuric acid). This leads to some Pb2+ ions in solution. When you charge, these Pb2+ ions allow you to get back to the charged compound. As you react the Pb2+, more lead sulfate dissolves.

Sulfation is a process by which the lead sulfate starts to agglomerate together to form large crystals. This leads to a loss of surface area. What this means is that you don't have enough area for the electrolyte to dissolve the lead sulfate to form Pb2+. So charging the battery gets harder. Sulfation takes a while; maybe in the order or days. Not hours.

Sulfation is reversible. But you have to charge the battery very very (very) slowly so that you slowly dissolve the large crystals of lead sulfate to Pb2+ and convert them back to the charged state. But, who has the patience to wait all week to charge a battery!

What this means is that discharged lead-acid batteries are bad news. You need to avoid them like the plague.

So the path was clear: Get a jump start, drive the car for the next hour and half to SFO and hope that the battery gets its life back. Like I said, a professional does not think. He/she reacts.

But a professional also recognizes his/her limitations. In my case, my limitation was that I was really cheap and had not invested in jumper cables or AAA insurance. A professional also typically has no real friends. I knew that none of my "friends" would be willing to stop by to give me a jump at 7 AM.

So I reacted to this problem by taking BART (i.e., public transport) to the airport.

I dont know if you guys encounter this, but sometimes when you have something on your mind it seems like the whole world is starting to think about the same thing. So here I was thinking about sulfation of lead-acid batteries and at this conference on grid storage, there was all this talk about the UltraBattery (this is a trade name) as a means of preventing sulfation. There were actually two talks completely dedicated to it.

I spoke about this advance a little while ago. The concept is to add some carbon into the negative plate when you make the electrodes. Everything else stays exactly the same. But for some magical reason the sulfation decreases significantly.

The talks were showing data where the battery was being cycled a little bit (like a Prius battery does) where the battery without the carbon tanks after a few hundred cycles but the one with the carbon works much better (thousands of cycles). Apparently, the carbon solves the sulfation issue.

There is a variant of this idea which involves connecting the negative plate in parallel to an electrode made of activated carbon. Activated carbon by itself is used to store energy in the double layer (i.e., a double layer capacitor). So essentially you are connecting a battery electrode in parallel to a capacitor electrode in both these configurations.

What was amazing was that they really had no idea why this was happening (although its been a few years since this first came out). DOE is actually funding a project to try to get to the bottom of this advance. Its sad but true- there is a fundamental change in a 150 year old technology but we have no idea why this change occurred.

Some things in life are really hard to reconcile (e.g., is there a God? why does EVERYBODY make more money than Scientists?) but this problem is not in the same class. I believe that this lack of understanding comes down to limited funding for research combined with the "herding" of research topics.

Lead-acid batteries are cheap. The one that I ultimately replaced my car with cost me $165/Kwh. Assuming I got ripped off (is that even an assumption for things car related?) it stresses the point that these batteries are cheap. Adding a few percent of activated carbon to the paste is not going to break the bank. All through the conference I kept wondering why lead-acid's are not being made with a bit of activate carbon to ensure that sulfation is not a problem.

As an aside, the cheapest battery we know of is the lead-acid battery. For grid applications, the thought is that we need batteries that cost less than $100/kWh. Something pretty radical has to happen for us to make these cost numbers. Making batteries the way we have made them in the past will not allow us to get there. But this is for another post.

So would adding a bit of carbon result in the battery lasting the life of the car (say 15 years)? The company that is commercializing this says it could last 7 or maybe 15. Meaning: they have no idea! Problem with not knowing why something works- there is no way to predict how long it will work. Moreover, the data they show is based on Prius-like cycling. My car does not have the same cycling profile. Unless we understand what is going on, we have no way to knowing how changing conditions will change performance.

In time, this data will show up and we will know if we will can make a better lead-acid car battery. But wouldn't it be better if we understood what was going on?

Truth be told, I'm not at all convinced that my battery was sulfated and could have been saved with a timely jump start. It lasted 7 years! That's a long time. Plates corrode and they shed. Maybe I ought not to expect more from my energy storage devices.

But being able control failure modes in batteries is crucial in ensuring that we can move to electric drive. And they will be crucial if we ever want to store the electricity from the sun. Solar panels last 25 years; do you really want to change your batteries every 5 or 7 years?

But to control the failure, you first need to understand them. And in many cases, we don't quite have life of batteries pinned down. They change with chemistry, multiple failure modes occur at the same time, and in some newer chemistries we just haven't had the time to collect the data to see what the failure mode is.

So we have clues and hypothesis for failure, but we can't quite predict failure rates with cycling conditions. There is no cycle/calendar life simulator for batteries.

Actually that is not true. There are many many life simulation tools for batteries. But none of them actually work!

To me this incident highlights the importance of fundamental research to understand real-world problems. There is quite a bit of fundamental research; and there is no dearth of real world-problems. What we don't do enough of is linking the two (The Program at Berkeley is unique in that its the only one that does this in the whole field of batteries).

It also highlights the fact that I really ought to stop being so cheap, spend some money, and get jumper cables.

Venkat