[ExI] another excellent meaty post from robert kennedy p.e.

[spike]

WOW thanks Robert, very interesting post.  Note to ExI, for some reason,
Robert's ISP is apparently fighting with the ExI server, and I don't know
how to make those things get along with each other.  Here's Robert's second
post:

 

 

 

Spike, while you're working the mailserver issue, please re-post this brief
reply to several responses.

 

WHY THE FOCUS ON OIL (per Rafal)?

Petroleum just by itself is the largest single industry on Earth, about 10
cents on the dollar of everything humans make or do.  Today, petroleum is
also the greatest source of primary energy for the human race, having
eclipsed coal a few years ago.  (Btw, natural gas is catching up fast, and I
think already exclipsed coal also in this country last year.)

 

With my quick & dirty estimate last night, I showed the trend of the EROI
figure of merit in petroleum is rapidly sliding in a worrisome direction, in
a short timescale as Extropians would view time.  Since petroleum is our
greatest source of primary energy, this developing situation should concern
everybody.

 

ELECTRICITY.

Total nameplate capacity of the U.S. electricity grid is ~1 TW; electric
generating capacity averaged over the year is just half that,

~0.5 TW.  In other words, the duty cycle of the average prime mover is ~50%.
(Surprised?)

 

For the human race as a whole, multiply these figures by about 4.  

(Electricity is the most useful form of energy there is, but most people are
not as well wired as we are.)  So, humanity's electric grid right now is ~4
TW peak, or >2 TW steady-state.

 

Re: "one quarter of the land area" of CONUS (per Spike).

CONUS contains 2 billion acres.  1 acre of land can host a quarter-megawatt,
+/-, depending on where it is and which solar tech you're using.  So figure
2 million acres = half a terawatt to equal the steady-staate grid.  But,
since solar only works when the sun is shining, obviously, the duty cycle is
only ~20%.  Presupposing a decent method to store electricity storage comes
along (one does not exist yet), you'd need five times that much land to meet
nominal average demand.  So figure 2 million acres X 5 = 10 million acres to
actually feed the grid round the clock.  That's half a percent of what we've
got in the Lower 48.  More than that much is already covered with roofs, not
to mention roads and parking lots.  (Roughly 100 million acres is
"developed" in some way right now with some permanent

improvement.)  So, without covering up a single blade of grass or other
greenspace, I figure there's enough space on rooftops that exist right now
and happen to be oriented in a useful direction (southeast thru southwest
azumith) to host much more than that 1 terawatt of nameplate national
capacity.  You could easily double that again just putting solar carports
over every parking lot (and then you'd get to park in the shade while
charging your car).

 

(As an exercise, try comparing the gross revenue per square foot from PV on
your roof or parking lot with the annual property tax per square foot.
You'll be amazed.)

 

ALL PRIMARY ENERGY.

This naturally leads to the question, how much land do we need to provide
all energy needs via exclusively solar power, including storage?

 

Well, about as much sunlight falls on the lit face of earth (land &

sea) in one hour as the entire primary energy consumption of the human race
in one year.  (Both figures roughly 500 quads.  Since this is the  

Extropian list, you can say that a quad is very nearly an exajoule.   

Heh.)  So raw sunlight::primary energy is a ~10,000-to-1 ratio. That is a
lot of headroom.

 

Once transmission loss, embedded energy, and parasitic or "house loads" are
factored in, thermal processes for generating electricity are roughly 30%
efficient at converting the raw energy of the fuel into a useful form.
(This is improving rapidly in some sub-sectors.  

It may surprise you to learn that the electricity sector is the  

cleanest and most decarbonized of the three major end-use sectors.   

All forms of solar are still only a tiny fraction of U.S. electricity
supply, but the electricity it produces goes straight into the grid without
further conversion,  Furthermore, the sunlight is not counted against
primary energy input, just like rainfall is not counted against hydro dams,
nor wind against windmills.  Seems unfair, but that's how the accounting
works, due to how primary energy is

defined.)  So to get Americans' primary power input to the electricity
sector, simply triple the steady-state grid load, you'll be close enough.
Solar's penetration right now won't affect this number.  That is, 0.5 TW x 3
= 1.5 TW.

 

For most people in the developing and developed world, the primary energy
partition amongst electricity/ transport/ raw heat is roughly
one-third/one-third/one-third per sector.  So to get total steady-state
primary power for a country, triple the number again, i.e. 3 X the primary
power input.  Note that 3 x 3 is about an order of magnitude.

 

So now you have a quick means for a horseback estimate of a reasonably
developed country's primary power demand to run everything they've

got: multiply their known steady-state electricity demand by an order of
magnitude, and you'll be within 10% of the true figure.

 

(As I said, Americans are a little more wired than most, 40+% not 33%.  

  So, multiplying by ~2.5 instead of 3 to get American primary power for
everything results in 1.5 x 2.5 = ~4 TW instead of 4.5 TW.)

 

So the average primary power to run the whole world, expressed in SI units,
is something like 4 to 5 times 4 to 4.5, or 16 to 22 terawatts.

 

Remember I said we'd need 10 million acres (including storage) to run a
half-terawatt grid round the clock here in CONUS, and that total American
primary power is 4 terawatts.

 

We would not need 80 million acres to run everything in America, however,
since the sunlight falling on PV modules does not count against primary
energy consumption the way chemical fuels mined from the ground does.  I
estimate that 30 million acres would enough, certainly no more than 50.
What's interesting about that latter figure is, that's the amount of land
covered by asphalt just for America's 4 million miles of paved highways.  A
similar amount is available in the medians and rights-of-way.

 

For the world as a whole then, 100 million acres (just 5% of CONUS) to

200 million acres (amount of active good farmland cultivated in CONUS) would
do the job.

 

And remember with things like solar carports that this power production need
not exclude other land uses.  There are substantial benefits re: energy
security and governance with locating supply close to loads.

 

PS. And now I know what BOTEC means.

 

--

Robert G Kennedy III, PE

www.ultimax.com

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20121223/9f9ddb6f/attachment.html>