It appears to me that Philadelphia gets year-round-average insolation
of 175 watts, at least 170 per square meter. Compare that to the 1 KW
per square meter that I have some impression that solar cell arrays are
That sounds to me like 34-47 dollars per watt in Philadelphia, if the
panels are laid horizontally.
I am guesstimating that by making them facing 40 degrees south of zenith
(towards "high noon" sun on days of the equinoxes), an improvement of 40%
of the 30.5% high-noon-equinox improvement would be achieved on
year-round-average. Maybe tilt them to face 37-38 degrees south of zenith
because the sun is up and out more in the spring and summer than in the
fall and winter. Maybe tilt them to face 2-3 degrees north of the
celestial equator and 7.5 degrees east of "high noon meridian" to take
advantage of the fact that on average during daytime there are more clouds
after noon than before. (I am aware of "morning fog / foggy low cloud"
exceptions to this "general rule"). At this rate, I expect about a 15%
improvement by optimizing how the solar panels are aimed. Without further
optimization to motorize them and get them to track the sun.
So, I see good opportunity of a Philadelphian to get the cost down to
$9.50-$1 per watt.
Considering that a Philadelphian pays nowadays about 14.5 cents per KWH
of electricity at "residential rate", excluding the surcharge for
consumption past 750 KWH (I hope I got that right, too lazy at this
moment to dig out a recent electric bill) per month during a defined air
At this rate, even a Philadelphian who wears nice cool summer dresses in
the house and uses fans rather than air conditioners and tolerates
subtropical to often-worse-hot weather well and frets more about winter
than summer and who lives in a rowhouse (good for stealing some climate
control from any more-climate-control-needy next-door neighbors) would
have a $9.50-41-per-watt solar panel paying for itself in 41 / .000145
hours, which works out to...
About 23-32 years, assuming inflation in electricity cost is the same as
inflation of cost of cost of whatever else a homemaker has to pay. If one
does all maintenance required and in the likely event the solar panel and
the likely-associated DC-AC inverter (at maybe 90% efficiency) keeps on
truckin', then it's 25-35 years to pay for itself, and 50-100 years to
double the invested money.
Suppose sun tracking with at least one motor is deployed, at consumption
of 5% of the output of the solar panel. At that rate, I have liking for:
Philadelphia above atmosphere year-round-average appears to me to
achieve about 315-320 watts per square meter of insolation, while at
ground level appears to me to achieve 170-175 watts per square meter.
That makes me think "at this rate" that 1 KW per square meter of direct
sunlight is degraded to 530-555 watts per square meter of a solar panel
that tracks the sun, before loss from electrical power consumption of
devices to track the sun and before correction for maybe 40% of the time
there is significant cloud cover that at least mostly destroys
optimization of sun-tracking (as a "representative figure").
At this rate, I have a liking for 95% of 60% of the way from 175 to 320
watts per square meter to be divided by 175 watts per square meter...
42% improvement of rate of return, down to 70.3% as much time to get
money back to you - as in about 16-22.5 years to pay for itself, or about
32-45 years to double your money should you have no expense at maintaining
the system that long.
(If the system requires rechargeable batteries, plan on additional
expense because such bateries have a high rate of expiring in less than 32
or even 16 years.)
If you double your inflation-adjusted money in 16.6667 years, then the
annual rate of return is 3% above the effectively-actual inflation rate.
I seem to think that the total rate of returns of major stock index
funds, especially "total stock market" ones, have done better from 1970 to
now or would have if they were in operation according to their stated
rules should they have been in existence in 1970 - I would guess likely
even in UK, let alone USA! For that matter, fair chance even from the
roughly-1970 high to the 2009 low! (second-worst 39 years of USA stock
market - behind the 39 year stretch starting at or a bit before the 1929
high. I seem to think that stretch had USA total stock market outpacing
"official inflation" by 4-5% with reinvestment of dividends, minus the
~.3% annual expense ratio that a good index fund like Vanguard "Index
Total" should have and that Vanguard achieves.)
- Don Klipstein (firstname.lastname@example.org)
On Tue, 01 Jun 2010 17:18:57 -0700
Probably from reading the specs and knowing about the typical
efficiency of PV cells. The output rating of PV cells is usually quoted
under "full sun" conditions of around 1000 W per sq. metre.
That's the electrical output of that panel, which given the normal
efficiency of panels like that (10-15%) means an insolation of around 1000 W
per sq. meter. Actually the specs for that panel claim 13.1% efficiency so
slightly over 1000 W per sq. metre is required to achieve that output.
Not so - those were pretty accurate calculations.
Of course if you want efficient use of solar energy then solar
thermal is the way to go - it's not too hard to get 70-80% of the
insolation energy available as usable heat.
Steve O'Hara-Smith | Directable Mirror Arrays
C:>WIN | A better way to focus the sun
The computer obeys and wins. | licences available see
You lose and Bill collects. | http://www.sohara.org/
On Wed, 2 Jun 2010 09:38:28 +0100, Ahem A Rivet's Shot
If Klipstein mounts one of the modules I referenced above in full sun
in Philly on a cool day and measures the output, he'll conclude that
it costs out at ~$ per Watt, not the $0-$0 he managed to arrive at.
To avoid starting with worst case PV costs he could google "best price
PV". And he could skip even more GIGO by using HOMER or some such.
http://homerenergy.com/ Which would prevent erroneous assumptions such
as his 5% of production for tracking. Seriously? Try ~20Whrs per day,
which on a 1000W array in Philly might net out to ~.5% minus for the
motor, but >20% plus overall. PV economics aren't great, especially if
one is willing to ignore the unbilled-cost of grid energy and the
unsustainability of the billed cost. So those who seek to be negative
about the economics really don't need to exaggerate.
On 02/06/2010 16:35, wmbjkREMOVE@citlink.net wrote:
That is their price per peak output per watt installed and seems
unusually low. $/W is still about the going rate and some are closer to
$/W where you paying a premium for higher efficiency.
But unless you can arrange continuous sunlight the average output over
the year allowing for clouds and including diffuse light is something
like 1/8 to 1/10 of peak installed capacity. So his $0-40/W delivered
for use is basically in the right ballpark in the long term.
Operating at peak efficiency with a clear sky and normal incidence
sunlight then the array can achieve peak performance, but the rest of
the time it does not by a long way. And obviously at night it is idle.
I think you just have to be clear about what measure you are using.
The PV array link you pointed at is the cheapest I have seen on offer -
has anyone here obtained one? Or are they vapourware?