Re: "The Only Thing They Learn" -- does anyone else find it irritating?

John Schilling <schillin@xxxxxxxxxxxxx> writes:

On Wed, 18 Jun 2008 18:16:05 -0400, lathamr@xxxxxxxxxx (Richard D. Latham)
wrote:

John Schilling <schillin@xxxxxxxxxxxxx> writes:

On Wed, 18 Jun 2008 07:09:12 GMT, Gene <gene@xxxxxxxxxxxxx> wrote:

John Schilling <schillin@xxxxxxxxxxxxx> wrote in
news:0m8h54trrku1hf0bg124vkvkfa1llus1p9@xxxxxxx:

The reality is, even if we had retro-futuristic too-cheap-to-meter
nuclear power, you'd *still* be driving a car that burns $4/gallon
gasoline. Two completely different markets, now and for years to
come.

If energy is too cheap to meter, what is the cost associated with production
of hydrogen from water or methanol from air? It seems to me the price could
easily undercut those kinds of gas prices.

If by "easily" you mean "after tens of gigabucks and tens of years of
investment", sure.

I think you are wildly over-estimating the capital formation
difficulties. I mean, yeah, $10 billion sounds like a lot of money,
but it probably less than 1% of what the various investment entities
are going to end up writing off, by the time we have the current
housing bubble behind us.

Yes, but "various investment entities", plural. And really, that
one comes down to millions of eagerly optimistic amateurs each
dealing with a few hundred thousand dollars at a time.

Oil refineries and the like, require tens of billions of dollars
all at once, from a single "investment entity". If you think
that's easy, or even really plausible, I recommend you give it
a try.



You keep waving your arms real hard, saying "this is really
outrageously expensive, trust me".

I'm trying to get some sense of the actual numbers.

If it is ( given the free electricity you already threw in <smile>)
economically feasible, Exxon will the first folks in line to build a
plant, Shell will be about 20 seconds behind them, and _they_ can spend
$10 billion a year out of their current cash flow.

ObSF: There are Just three things the uS of A is still a world leaders
at, pizza delivery, microcode, and capital formation :-)

Heck, the US Air Force burns a shitpot full of Jet-A every year. You
don't think _they_ can come up with $10 billion of capital ?

BTW, you mentioned that "oil refineries and the like require billions
of dollars all at once". Hint: those things already exist, so I think
we have an existence proof about the ability of companies to allocate
capital in those amounts, with payback periods measured in decades, in
the industry in question.

And this isn't "invention required" stuff, like fusion is.

Who said anything about invention being required?


This is just bog-standard chemistry that we've known how to do,
and done, on a commercial scale, in war conditions.

Yes, by way of many billions of dollars of investment and many
years of effort.


I'm questioning the "years of effort". The Germans produced 28 million
gallons a year in the 1940s. It didn't take them 10 years to tool
up. And even if it had, we didn't throw the recipe away.

What I haven't been able to find out is "given 1940s technology, what
did it cost per gallon ?". I'm sort of assuming that price help
establish a lower bound on what we could do it for today.

Secondly, I don't see there being massive economies of scale
associated with this sort of processing. A 1/100 scale plant ought to
be (to a first order) 1/100th the capital expense.

I suppose the right way to get a reasonable estimate is to go pull
some 10Qs etc. for Dow Chemical. They build large chemical plants all
the time. (They've probably got the plans in a file drawer somewhere,
waiting on the price of oil to exceed X for Y years ).

Banging around on google, I see that UIG <www.uig.com> sells turnkey
liquid CO2 plants in the 100 to 250 tons / day range. Haven't been
able to find a price, for either the plant or industrial quantities of
CO2 and liquid hydrogen. I'd be shocked to find out they're radically
more expense than LOX, and that goes for pennies per pound, right ?

This is a persistently baffling blind spot in the SF community -
the belief that once an invention has been invented, and the
necessary raw materials identified, the problem is solved and
the part where stuff actually gets built can be handwaved off
into exercise-for-the-student territory.

I suppose this could be a way of suspending disbelief for all
those lone-genius-transforms-the-world! plots, cutting out the
boring parts in the middle. But the real world doesn't work
that way, and IMO it's been a long time since SF where it works
that way has been really credible.

If Dr. Morbius shows up tomorrow with plans for the Krell Machine
(now with Id-supression safety interlocks!), it's going to be a
long, long, *long* time before dreams start coming true.


But almost nobody who is investing big money for the long haul, seems
to think that gasoline is going to stay at $4/gallon for more than a
year or two. And if it drops back down to $1.50/gallon, well, you'll
be paying $500/gallon for all the red ink your accountants will need
to explain that shiny new hydrogen plant in the back lot.

Now we're off into a "peak-oil" discussion. I will note that, right
now, there seem to be a lot of people investing pretty big money in
the bet that gas prices are going to go _up_ from $4 a gallon.

Yes. *Next year*.

And I suspect they're wrong even about that, but it's far from a
sure thing either way. Predicting short-term market fluctuations
is mostly a fool's game.

It is _already_ over $5 a gallon at the pump in portions of California.

But there's a difference between buying gasoline at $4/gallon[1]
today because you think you can sell it for $5/gallon next year,
and buying a synfuel plant that will only be seriously profitable
if gasoline *stays* at $5/gallon for the next *twenty* years.


My personal only slightly informed position is that OPEC (Saudi
Arabia, primarily) has shot their wad, meaning that I don't think
they _can_ raise production enough to satisfy demand and drive prices
back to the 2004 level. I think China and India are just growing too
fast to permit that outcome.

The Chinese and Indian economies have grown by about 40% since 2004.
That's impressive, but the Chinese and Indian economies combined
only make up about 10% of the global economy[2]. So all their
economic growth since 2004, will have resulted in a ~4% increase
in the global demand for oil over that period.


This statement conflicts a graph produced by the US Energy Information
Administration, which showed world-wide usage continuing it's upward
trend right thru 2007, with US consumption leveling off in circa 2004,
but an accelleration in usage growth in China whch effectively
maintained the long term growth trend line.

( I apologise, but I was sure I'd seen it over at www.oildrum.com, but
I can't find it right now, if you don't beleive me, I'll go root it
out The web-site I saw it on credited the EIA, and that was the
first time I'd ever heard of that organization ).

Actual world oil production has increased by more than 10% over
the same period.


Wrong. This reference I did still have lying around.

"Total liquids" has increased, but crude production appears to
have plateaued in 2005. Here's the url to the latest US EIA Oil
Production numbers via the Oil Drum at
<http://www.theoildrum.com/node/2300> .

Excerpting from the Executive Summary:

1. All Liquids: the peak is still July 2006 at 85.47 mbpd, the
year to date average production in 2006 (11 months) is 84.59
mbpd, up 0.01 mbpd from 2005.
2. Crude Oil + NGL: the peak date remains May 2005 at 82.08 mbpd,
the year to date average production for 2006 (11 months) is
81.40 mbpd, down 0.03 mbpd from 2005 (11 months).
3. Crude Oil + Condensate: the peak date remains May 2005 at 74.15
mbpd, the year to date average production for 2006 (11 months)
is 73.48 mbpd, down 0.09 mbpd from 2005 (11 months).
4. NGPL: the peak date remains February 2005 at 8.05 mbpd, the year
to date average production for 2006 (11 months) is 7.92 mbpd,
up 0.06 mbpd from 2005 (11 months).


Furthermore, even if world oil production hadn't increased at all,
the short-term price elasticity of oil is in the 0.1 to 0.2 range.
That means a 4% increase in global demand, only translates into a
20-40% oil price increase.


This is one of those issues where you really, really need to have
done the math to understand what's going on.


Even if you do the math, it's important that you know what the
underlying inputs are.

Your arguments would be more persuasive with more figures (with
references, since we've already seen that you'll assert facts not in
evidence) , and less bluster.

For the nonce, I'm not interest (nor capable) of debating "peak
oil". I'm just trying to figure out a plausible upper bound for the
cost of gasoline, assuming we have to synthetize it out of the
atmosphere, adding energy as required to drive the chemical reactions.

But you've consistently refused to do the math. You continue to just
wave your arms real hard, and provide more proof by assertion.

Assume free electricity. You already positied that.

We need ( it appears ) liquid CO2 in industrial quantites. My best
guess is that costs something on the close order of 10c a gallon. LOX
is "pennies per pound" or so they claim over in sci.space.

( I've looked, and haven't yet found a quote for CO2 or LH in
industrial quantities ... you happen to have any pointers ? )

We need hydrogen (probably liquid, just from volumetric
considerations) in industrial quantities. I'm (wild ass) guessing
something very close to $1 a gallon.

So, a $1.10 or so per gallon for feedstock + depreciation + labor
etc. + profit ought to be the upper bound for gasoline costs, in the
somewhat distant future.

What I don't know how to estimate to within even 1 order of magnitude
is the capital construction cost for the plant.

I'm guessing your estimate for chemical processing plants construction
costs may be on the very high side.

If you have figures for the current costs for the various feedstocks,
I'd like to hear them.

--
#include <disclaimer.std> /* I don't speak for IBM ... */
/* Heck, I don't even speak for myself */
/* Don't believe me ? Ask my wife :-) */
Richard D. Latham lathamr@xxxxxxxxxx
.