Saturday, November 30, 2019

An electric car economy

Recently, Brad Templeton posted a column at Forbes examining whether an electric car economy could handle peak-travel holiday loads. The major problem, of course, is that electrics have limited range at a time when, in Perry Como's words, "From Atlantic to Pacific, the traffic is terrific." So everybody is on the road and needing a recharge at the same time. In holiday traffic, you often wait in line at the pump to fill up, but with gas, that's not too much time; with a half-hour recharge, it's completely impractical.
Brad thinks it would be possible with a major build-out of recharging facilities (and it would also require a major upgrade of power generation and transmission lines). But what if that is looking at the problem backwards? What if we could reuse the current infrastructure, and do electric cars the way we do gas ones? The problem vanishes, and we could gain quite a few extra advantages.
So in the current system, you go to the service station and put 50 pounds of gasoline into the car, taking a few minutes since it has to flow in through a smallish hose. In an average car, that will get you 250 miles or so before you need to fill up again, and you'll need a rest stop long before then anyway. Why not simply have removable batteries, and pop new ones in instead of letting the car sit while you recharge the old ones? That's the way most high-powered hand tools work today, for example.
EGO system power tool batteries
There's one major problem with this for a car. The battery in a Tesla weighs 1200 pounds and when fully charged, can power it about 300 miles. You won't be "popping in" one of those.
How big a battery could you pop in? The average man can handle 50 pounds (what typical lead-acid car batteries weigh) for the few seconds it would take to take it off a service cart and onto a connector mount like the one on a power tool easily enough, and for other people a service station attendant could do it. But 50 pounds of Tesla battery would only take you 12 miles.
But those are not close to the best batteries we have. Instead of 4 pounds per mile, a lithium metal battery holds enough energy to go one mile per pound. Popping in 5 of them would get you 250 miles of range and take less time than filling with gas.
The problem, of course, is that the lithium metal battery -- it's the chemistry used in watch batteries, for example -- is not rechargeable. But so what? You're not trying to recharge it. The batteries you swap out go back to the factory to be recycled. This is just like the current car-fueling infrastructure. The service station has a supply of gas brought periodically by trucks. New batteries the same; the only difference is that the trucks go back full, of old batteries, rather than empty.
Each battery could be about the size of a briefcase; there would easily be room for ten of them under a typical car hood. You wouldn't normally drive around with it full, because less weight is more efficient; but you could load it up for the big trip. And for extra range or emergencies (e.g. the station you were counting on has run out), you could throw a few extra in the trunk.
The battery factories/recycling plants can be anywhere. Do you know where the refinery is that your gasoline comes from? They would be sited for cheap power, and maybe even intermittent power. Most people today really have no idea how cheap high-volume manufacturing processes are; recycled batteries might well be cheaper than gasoline per power provided.
One more advantage to the multiple modular battery scheme suggests itself. All the batteries don't have to be the same kind. You could have some rechargeables, some high-density, some of whatever new chemistry or new fuel cell happens to be invented next.

Sunday, October 6, 2019

Exploding galaxies and climate change

There was recently posted on the physics preprint server Arxiv.com, a paper soon to be published in The Astrophysical Journal about a huge radiation explosion in the center of our galaxy, probably associated with Sagittarius A*, the enormous black hole there.

The paper itself, by Bland-Hawthorn, Maloney, Sutherland, and Madsen, is here.
A more digestible overview for the general reader is here.

The explosion event appears to have happened about 3 and a half million years ago, and lasted maybe 300,000 years, for reasons outlined in the paper.

We now turn our attention to theories of climate change. The two leading paradigms for causes are CO2 and the Sun. CO2 is more popular today, but as recently as the 1980s leading climatologists favored the Sun, for example because the depression in solar activity called the Maunder Minimum coincided with the period of unusual cold known as the Little Ice Age.

It is worth point out that the two theories are by no means exclusive or contradictory. Excess CO2 might warm the Earth, by a well established pathway of radiative physics and the greenhouse effect; but at the same time the Sun might also have an effect.

Over the past two decades it has become clearer how this effect might work. The Sun's magnetic field becomes stronger and weaker with solar activity levels as revealed by sunspots. A strong magnetic field protects the Earth from galactic cosmic rays, a weaker one lets more of them through. Galactic cosmic rays seed cloud formation by means of a complex but laboratory-tested process.

Anything that affects cloud formation might thus have an outsized effect on climate. Consider for example the classic diagram of energy flows in the Earth atmosphere from Kiehl and Trenberth:


The atmospheric window (on the right) is a relatively minor energy channel, and affected to the tune of a few percent by CO2 variation. Clouds (left) manage some 4 times as much energy; a smaller variation would have a noticeably greater effect. So the interesting question is one of the relative sizes of the effects.

It thus occurred to me to wonder, when I heard about the radiation explosion, whether we might see some indication of the huge wave of radiation that must have swept over the Earth (for 300,000 years or so) in the neighborhood of 3 and a half million years back. So I went looking for a temperature record with the appropriate range. The first thing I found was a 5 million-year reconstruction by Lisiecki and Raymo (2005) using oxygen isotope ratios in deep ocean cores. This appears to be quite well-regarded; it is even referenced in the Wikipedia article about the technique.

Here's the record:


Lo and behold, right at 3.3 million years ago there is a significant dip. Here it is enlarged:

Indeed, it's so distinctive that L&R have a closeup of it in their original paper:
(note that my graphs read advancing dates left to right, theirs right to left)


(Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene-
#Pleistocene stack of 57 globally distributed benthic d18O records,
#Paleoceanography,20, PA1003, doi:10.1029/2004PA001071, Fig. 9)

Now I'm sure there are other explanations for this out there; careers can be made in science for even an incorrect, if well-argued exegesis of so salient an anomaly. But I hadn't ever heard of this one. Given what I did know, though, I went looking for it, and there it was.