The PV scalability problem

The PV Scalability Problem

by Dave Kimble at www.peakoil.org.au

Australia's electricity consumption is growing at 2.55% per year.
Suppose you want to build enough solar PV flat panels
to provide for that 2.55% of Australia's annual electricity consumption in 1 year.
How much would it cost ?
How much energy would it take ?

BP's Statistical Review of World Energy (2006) shows
Australia's electricity consumption for 2005 was 243 TW.h
and the trend is for an increase of 6.2 TW.h per year, (2.55%)
so I forecast that in 2008 we will use 262 TW.h
That amount of production (in 2008) is equivalent to 33.2 GW of generating capacity,
assuming operating at 90% load factor (typical for coal).

2.55% of that amount is 6.68 TW.h per year
or 3.03 GW operating at 25% load factor
What area of solar panels will we need to capture that much energy ?

With the panels set facing north and tilted to the latitude angle
(the most efficient fixed position)
locations in Australia will vary from 5.5 to 6.5 KiloWatt.hours / m² / day
averaged over a whole year,
see http://www.peakoil.org.au/australia.insolation.xls

So let us assume 6 KW.h / m² / day , or 25% load factor averaged over the year
and assuming a thermodynamic efficiency of 14% ( for PV flat panels)
and sunlight at 1,000 W/m²
that will require 21.7 million square metres of collector
and at current market prices,
see http://solaronlineaust.yahoostore.com.au
the most cost effective panels cost AU$1,145/m² (March 2007).
So to meet the expected growth in consumption with PV panels would cost AU$24.7 billion per year.

And the price of solar panels is RISING,
not falling as the spruikers will tell you.
This is because the PV industry is now outgrowing the supply of
scrap silicon from the electronics industry.
From here on, the PV industry will have to build their own new factories
and make their own silicon,
without being able to share some of the infrastructure costs
with the lucrative silicon chip business.

Assuming the ERoEI of the system is 3 and the lifetime is 25 years,
as reported in the ISA Report as part of the recent nuclear review,
"Life Cycle Energy Balance and Greenhouse Gas Emissions of Nuclear Energy in Australia"
http://www.pmc.gov.au/umpner/docs/commissioned/ISA_report.pdf [2.75 MB] ,
[ archived here ]
then it takes 25 / 3 = 8.3 years' worth of generating to repay the energy expended in the manufacturing process.
so 8.3 x 6.68 TW.h = 55.3 TW.h is needed to build 2.55% of Australia's consumption in solar PV.
This is equivalent to 7.0 GW at 90% load,
or SEVEN VERY BIG POWER STATIONS' WORTH.

Now it does not makes sense to use 21.1% of the nation's energy
to bring on-line 2.55% of consumption as solar.
It would have made sense in 8.3 years' time,
when the panel has paid off its energy debt,
and is then producing pure energy profit,
but there is too much of a barrier at the beginning to get started.

And could we use the output from the first year's solar production to fuel the next year's production ?
No, its not enough by a factor of 8.3 .

And of course the above calculations don't allow anything for the energy and economic costs of
control equipment, wiring, transport of materials, labour to build, labour to maintain, management, sales, advertising, insurance.

The above demonstrates that it is impossible to grow a power generating industry quickly
unless its ERoEI is at least 5, and maybe more.

The sad fact is that greenies, including Bob Brown, Mark Diesendorf, and Ian Lowe
are seriously over-selling solar energy.
It can never scale up as well as they would like it to because of this initial energy barrier,
and they are as guilty as the oil companies telling us that oil has a long-term rosy future
or that coal will become clean and green.
They don't want to hear the bad news about solar
because it is firmly part of their policy platform,
and it would be embarrassing to have to change.

The Federal Government has made a big fuss about building
"the biggest solar generating station in the world" - a paltry 154 MW.
But the annual increase in electricity usage is equivalent to 708 MW per year.
So this project would provide for less than three months' worth of electricity growth
and can have absolutely no impact on replacing fossil fuels.

The Federal Government then mandates compact fluorescent lamps,
said to save us 800,000 tonnes of greenhouse gas emissions,
( Sydney Morning Herald 19/2/2007 )
So emissions will be "slashed" by a whopping 0.14 percent! Wow.
The savings from using compact fluoros will be swallowed up within less than 3 weeks at current rates of growth.
It cannot possibly prevent any fossil fuels being burnt.

Dave Kimble