[PLUG-TALK] Electric cars and Wall Street

Keith Lofstrom keithl at kl-ic.com
Tue Sep 27 08:06:58 PDT 2011

On Mon, Sep 26, 2011 at 04:20:46PM -0700, John Jason Jordan wrote:
> On Mon, 26 Sep 2011 13:36:35 -0700
> Keith Lofstrom <keithl at kl-ic.com> dijo:
> >On Fri, Sep 23, 2011 at 08:36:14PM -0700, John Jason Jordan wrote:
> Some EVs are being designed as direct replacements for petroleum
> powered vehicles (e.g., Nissan Leaf), others are not. As for batteries
> and temperature, both the Chevy Volt and the Ford Transit Connect have
> cooling and heating systems to keep the batteries at the optimum
> temperature. I don't know about the Leaf, but I expect it does as well.
> As for batteries never replacing a gas tank, all I can say is that
> never is a very long time. There are two things to consider: 1) The cost
> of petroleum as political, economic, and supply factors make it rise
> and, 2) Improving technology. 

"Never" as in "cash dollars will never fall from the sky" as in
don't buy acreage in order to catch billions.

A car gas tank refills at speeds on the order of 20 gallons per
minute.  A gallon of gasoline has the energy equivalent of 38.3
kilowatt hours, so you "charge" your gasoline car at a rate of
767 kilowatt hours per minute - or 46 megawatts .  Of course,
only a fraction of that becomes mechanical energy.

Meanwhile, the average American consumes 10 kilowatts.  A similar
fraction of that power is delivered as electricity to your home,
and your residential neighborhood of homes.  Yes, your house may
have 200 amp 2 phase (50 kW) service, but your neighborhood of
1000 homes does not have 50 MW service, they are counting on 
averaging.  So, unless the residential electric grid is expanded
by about 4000x (a huge investment in aluminum wire), electric cars
will always charge much more slowly than a stop at the gas station. 
As is, charging every car every night, for 4 hours each, will
require  a huge expansion of the power grid, and more
generation, and ... 

And don't ever ever forget to charge your car at night, or you
will be walking the next day.   And don't run out of "gas"
while travelling; hauling a two inch extension cord from the
nearest charging station will be somewhat more difficult than
a two gallon gas can. (*)

If you are on a longer-than-average trip (say driving out to the
woods to shoot animals) you can plan on a four hour stop every
hundred miles or so, in order to recharge.  Unless there are 
huge power lines along those forest roads.  

To use those huge power lines, you will need fast charge batteries. 
Which, as a rule of thumb, will not survive as many recharge cycles
as normal, slower charge batteries.  Fast discharge also wears out
batteries sooner, so although I can easily build an electric 
quarter-mile dragster, I won't get many races out of it.

That is why I proposed a way to deliver the energy in a separate,
autonomous unit, available everywhere, automatically, based on
network awareness.  What actually happens will be better than
that, but it will have those characteristics.

Lithium batteries store 150 watt hours per kilogram.  Gasoline
contains 13,500 watt hours per kilogram, of which about a third
turns into mechanical energy - call it 4500 watt hours per kg.
So a 20 gallon, 120 pound gas tank stores as much useful energy
as a 3600 pound battery.  The battery actually operates somewhat
longer, because of regenerative braking, but you are also
starting and stopping a lot more vehicle weight, and fighting
a lot more rolling resistance, so for long distance ( > 50 mile )
vehicles, the regeneration is not significant.

Cost?  My laptop battery costs about $100, and stores about 100
watt-hours.  Approximately a buck per watt hour.  So if I wanted
to store the energy equivalent of 20 gallons, 760 kilowatt hours,
those batteries would cost $750K.  Obviously, the batteries will
store a lot less energy than that, and they will be made cheaper
than a laptop battery, but there is no magic way to improve that
by orders of magnitude.  Given the vast investment in laptop and
cell phone batteries, we have already found the easy improvements.

One of the things that makes batteries hugely expensive is that
they must store that energy in metal electrodes, separated by a
thin gap.  If the gap is too big, they perform poorly (and are
heavier).  If the gap is too narrow, they can short out, turning
the entire energy content of the battery into heat very quickly. 
A LOT of heat, as in white hot flames and holes burned through
concrete.  It is like storing gasoline whipped with oxygen, a few
degrees below the ignition temperature.  It surprises the hell 
out of me that they let me on an airplane with one, since it
can melt through the floor, and fill the cabin with poison gas.
One disaster such as that, and your lithium battery company
stock price will plummet.

Because of the extra weight, an electric car has longer stopping
distances.  When there are a lot of electric cars (right now,
there are hardly any), the collision rate becomes noticable.
Expect battery cookoffs and immolated passengers, as well as
asphyxiations.  Many new and legally fraught incentives to sue
manufacturers.  Which any smart manufacturer is going to factor
into the sales price.

KL> >Companies that try to sell ultra-expensive, low performance
KL> >electric vehicle using batteries will have very limited sales,
KL> >unless customers are forced to buy at gunpoint. 

JJ> It will be very interesting to see sales reports on the Volt and the
JJ> Leaf a few months from now. The Transit Connect doesn't count, however,
JJ> as the fools in Ford's marketing department decided to put a $60,000
JJ> price tag on it. The gas version of the Transit Connect sells for
JJ> $24,000, so it doesn't take a genius to calculate that the added value
JJ> from gas savings won't come close to bridging the price difference.

So some companies are subsidizing their electrics more than others,
or are not including the legal risks.  My friends are similarly
foolish to Ford, because they don't rent houses for $5/month,
even though they would attract a lot more renters that way.  Still,
there are probably more houses renting for $5/month than there are
plug electric vehicles.

                    Write this in big letters!
***  Electric cars, intended as functional equivalents of    ***
***  gasoline cars, do not make economic or practical sense! ***

We CAN design an electric powered individual transportation system
that is more convenient and cheaper to operate than individual
gasoline powered vehicles.  But such vehicles will match the
strengths and weaknesses of the technology, and will be quite
different than a gasoline car, in expectations and operation. 

The convenience and cheapness will be due to using technologies
that improve rapidly.  Computation and interconnection, 30X per
decade, perhaps even molecular biology, which is advancing 

Major improvements will NOT occur in things like batteries.
5X per century in the past, with very little future improvement
probable.  Chemistry is chemistry, and they won't invent a new
kind of atom or a higher energy atomic bond.  They might invent
new ways to make others pay for it, but that doesn't scale -
you end up with Solyndra, and you end up sharing a jail cell
with an Enron executive.

So, do you want to invest in Google, or Western Union, or Ponzi?

I gave some hypothetical examples of technologies that might
deliver more transportation per buck than automobiles.  What
actually succeeds will work better than those, because there
are millions of people smarter than I am out there.  But there
are billions who are stupider. 

You should invest in smart, imaginative, production and
market-minded entrepreneurs, rather than the ones who get
their technology ideas from watching algore on TV. 
algore is not an engineer, and watching TV does not lead to
entrepreneurial success.  The revolution will not be televised.

Science, engineering, and economics do not get suspended
because it would be convenient if they were, or because
powerful leaders promise or demand it, or because thinking
is hard and we prefer sound bites.  

We have big problems to solve.  We will not solve them by
thinking inside the box, by assuming that we already know
what the solutions are.  And we won't solve them by expecting
free ice cream and ponies, either.  But the problems will get
solved.  Looking backwards from 2050, the solutions will
appear obvious to average minds, no genius required.  Your
job as an investor is to anticipate 2050, not extrapolate it.
Look at how your own life has changed in 40 years, and look
for technologies capable of evolve towards that level of


(*) PS: To make a long posting even longer, there WILL be many
fools who forget to charge their batteries in their Chevy Volts,
and run out of juice in inconvenient places.  Those heavy cars
will be difficult to tow.  So imagine building a small truck
(Smart phone dispatched) with a small crane and a few 300 pound
jump charger battery boxes in back.

The charger-truck driver goes to the stalled EV car and driver,
collects $200 (remember, the folks that buy these have lots of
money), lowers a charger unit and connects it to the EV
charging plug.  Come back a few hours later, retrieve the unit,
and take it back to the central charging station.  GPS and
cameras and lojack on the deployable charger units, of course,
so you can track down and jail thieves and vandals. 

Service 15 cars a day, and gross $1M/year.  A lot easier than
operating a tow truck.  That is an example of a business which
can scale from tiny to large as quickly (or slowly) as the plug
electric car business does, with far fewer risks.   Few gold
miners got rich, but the fellow who sold them denim pants did.

Think outside the box.  There will be many such opportunities.

Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs

More information about the PLUG-talk mailing list