Tuesday, November 16, 2010

Thought experiment, continued

This is a continuation of the thought experiment started yesterday.  Let's reduce
this down to just one car and the amount of solar panels you would need to produce
the electricity for that one car.  From my study of batteries, you can get about
3 miles per kilowatt hour with a battery.  This calculation is from my memory and
may not be entirely accurate nor up to date.  But this isn't exact, a round figure
will suffice for now.

The Chevy Volt has a 40 mile range for its battery pack.  Using the 3 mile per kwh
number, you would need about 13 kwh of power to drive with electricity.  Again, this
may not be exact, but it is using a figure from the real world in order to get a
rough estimate.

From this we can calculate how much solar panelling will be necessary in order to
produce the 13 kwh.  Using the web page from yesterday, we have 80 watts of continuous
power from a 1000 sq inch panel.  At 95% availability in geosynchronous orbit, the
panel will produce  (.95*80 watts * 24 hours)/ 1000 watt hours per kwh = 1.824 kwh.
Why 95% availability?  A solar panel in geosynchronous orbit at an altitude of
22,000 miles (approx) will be in the sun almost 95% of the time.  From this number,
we can now calculate the number of panels needed in order to produce the electricity
for 1 auto.  13 kwh/1.824 kwh per panel gives a little over 7 panels.

What would be the cost of these 7 panels?  This is about 49 square feet of panels.
From this webpage, we can calculate it (again a bit rough calc) it figures out
to be about 56 dollars per square foot.  So, 7*56 = 392 or about 400 dollars.

How long will it last?  Maybe it could last 30 years.  Perhaps that would be way
too optimistic.  Let's say 10 years.  Then the cost per year would be 400/10
which would be about 40 dollars per year.

Sounds pretty cheap, still.  What is the cost per kwh if you use electricity from
the wall plug?  At 10 cents per kwh, you and 13 kwh per day you get the following
.10*13*365= which is about 475 bucks per year.  The above solar panel configuration
would only have to be 1/10th as efficient in order to yield the same cost.

Is it worthwhile to launch from the moon?  You have to have the infrastructure in
place on the moon in order to make the solar panels and to put them into orbit.
That would take a considerable investment.  If you can get the costs down of
manufacturing the solar panels to anything near what it takes on earth, then what
you have left is the cost of launch.  How much would that be?  About 5% of what
an earth launch would cost.  Again, rough figures.  That would be about 10,000 *
5 percent or 500 dollars per kg.  You'll still have to do better than this because
7 solar panels would definitely weigh more than 1 kg.  The 10,000 number is based
upon shuttles which are expensive even for earth launches.

The launch problem still exists even on the moon, or at least in this example.
Can you do better than the 5% number?  Maybe.  It takes energy to launch from
anywhere.  Getting stuff off the moon would require an efficient way to use
energy for launches.  If you figure that out, you figure out this problem.

What about mass launchers?  It would only use energy and little to no fuel.  This
could be a start.  I pick this up again in a future post.

No comments:

Post a Comment