Thursday, January 27, 2011

Cost of solar power (3)

A colleague referred me to a recent interesting article: “Are Solar Thermal Power Plants Doomed” by Michael Kanellos and Brett Prior.  This appeared in a web publication from Greentech Media,, 18 October 2010.  The thrust of the article is simple – PV costs are coming down faster than solar thermal costs, therefore PV will come to dominate the marketplace.

Other factors cited by the authors in favour of PV are (1) PV plants are more modular and can be built smaller than concentrated solar thermal (CST) plants, and (2) concentrated PV (CPV) technologies help reduce overall costs of PV plants even further.  On the other hand, the authors acknowledge that CST plants have better storage prospects (thus giving despatchable utitlity-scale power) and have inherently less spiky output as clouds momentarily pass overhead.  At the moment, however, they say these advantages are not valued by project developers.

On-line discussion of the article was valuable.  Useful comments were made by
Farid Bensebaa, author of a recent publication “Solar Based Large Scale Power Plants: what is the Best Option?” in Progress in Photovoltaics: Research and Applications (Aug 2010).  Conclusions highlighted were:
·         capital cost of CSP and PV are about the same;
·         O&M cost of CST is at least ten times higher than PV;
·         current CST requires large amount of water (a disadvantage in desert areas);
·         the cost advantage of PV is smaller under very high solar irradiation;
·         energy storage is proven and cheaper in the case of CST when compared to PV.
Not everyone in the discussions agreed with the comment about O&M costs.  There was also knowledgeable discussion about the relative lifetimes of PV and CST plants.

What do I think?  Well, cheap storage and despatchability have to be very important eventually, even if project developers don’t value them highly today.  I also put value on the option of hybrid firing of CST plants with biomass or even fossil fuels.  So, I wouldn’t write off CST just yet, and I’m sure the proponents of CST in its various forms (tower, trough, dish, Fresnel arrays) would agree with me.  Until I have better and undisputed information, I’m going to stick with my assumption that all annual O&M costs are 3% of capital costs.

On to today’s examples – Olmedilla and Andasol 1.

Recall my standard assumptions used to analyse all projects:
·         there is no inflation,
·         taxation implications are neglected,
·         projects are funded entirely by debt,
·         all projects have the same interest rate (8%) and payback period (25 years), and
·         all projects have the same annual maintenance and operating costs (3% of the total project cost).
To those I should also add that government subsidies are also neglected.

Olmedilla (

This Spanish PV installation was completed in September 2008 and has 270,000 Si panels.  Other project details: site area 285 Ha, peak output 60 MW, annual output 87.5 GWhr, cost EUR 384 million. 

Cost per peak Watt     EUR 6.40 / Wp
LEC                            EUR 547 / MWhr

The components of the LEC are:
CAPEX           {0.094 × EUR 384×106} / {87×103 MWhr} = EUR 415 / MWhr
OPEX             {0.030 × EUR 384×106} / {87×103 MWhr} = EUR 132 / MWhr

Comment: even if the O&M costs were zero, the LEC is still very high.

This project is based at 1,100 m altitude in southern Spain and came online in March 2009.  Solar energy is collected by parabolic trough collectors coupled to molten salt storage that provides for 7.5 hours operation.  Other project details: site area 51 Ha, peak output 50 MW, annual output 180 GWhr, cost EUR 300 million.

Cost per peak Watt     EUR 6.00 / Wp
LEC                            EUR 205 / MWhr

The components of the LEC are:
CAPEX           {0.094 × EUR 300×106} / {180×103 MWhr} = EUR 157 / MWhr
OPEX             {0.030 × EUR 300×106} / {180×103 MWhr} = EUR 50 / MWhr

Friday, January 21, 2011

Research update

From web stats, I know I have a set of regular visitors to my other website at  Over the years I’ve corresponded with some of these visitors, and the message is they like to hear about my research progress.

So, here goes …

Since attending the Australian Solar Energy Conference in Canberra in early December, I’ve been working mainly on thermal storage.  A key realisation from Canberra was that solar thermal needs storage if it is to be competitive long-term with PV.  We all know that PV prices are coming down and PV efficiencies are going up.  Without storage, however, PV won’t keep the lights on after dark.  On the other hand, thermal storage is relatively cheap and also provides a boost to the capacity factor of solar thermal heat engines.  There will be a big opportunity for utility-scale despatchable solar power that only solar thermal can meet with today’s technology.

What are the storage possibilities for my evaporation engine as powered by passive heat collection under a transparent insulated canopy?  It seems to me the answer is – rather good!  Thermal energy can be stored by passing the air heated under the canopy through a bed of loosely packed rocks.  Provided the bed is large enough, both pressure and thermal losses will be small.  The stored heat can be used to generate power after dark or during cloudy periods, thereby increasing the capacity factor of the piston-in-cylinder evaporation engine.  Such storage would be cheap too!

I’ve estimated the pressure losses in the rock bed and I’ve started to build a simulation model for thermal charge and discharge.  A related issue is to develop a control strategy for the concept.  My ultimate goal is to develop software so I can simulate the annual performance of the canopy/storage/engine system at regular (15-minute?) intervals throughout an entire year.  I’d like to present that work at the next Australian Solar Energy Conference to be held in Sydney at the end of 2011.

Meanwhile, I’m planning to present the Wellington simulation results for a sloping canopy at the 2011 Solar World Congress, to be held in Kassel, Germany (29 August – 1 September).  I have already obtained results for the sloping canopy, and they are clearly better than for the horizontal canopy.  No surprises there!

For the next few weeks, another project (still confidential at this stage) has demanded my attention.  I expect to get back to the storage simulations in mid-late February.

Monday, January 10, 2011

Cost of solar power (2)

The New South Wales government today approved the construction of a solar PV power station at Nyngan, about 600 km inland.  A decision to build the power station now depends on the success of an application for federal government funding under the Solar Flagships program.  The announcement contained sufficient details to allow an estimate of the cost per peak Watt and Levelised Energy Cost under my standard assumptions.

Project details:  The developer is Infigen and the PV cells will be provided by Suntech.  According to today’s details, the cells will be mounted on a fixed frame, storage is not mentioned, the peak power is 100 MW, the site area is 200 Ha, and the project cost is AUD 300 million.  A precise figure for annual power output was not mentioned, however claimed savings on CO2 emissions are 150,000 t per year, from which I calculate the annual power output to be 187 GWhr based on an emissions intensity of 0.8 million tonnes CO2 per TWhr.

Recall my standard assumptions used to analyse all projects:
·         there is no inflation,
·         taxation implications are neglected,
·         projects are funded entirely by debt,
·         all projects have the same interest rate (8%) and payback period (25 years), and
·         all projects have the same annual maintenance and operating costs (3% of the total project cost).
To those I should also add that government subsidies are also neglected.

For the Nyngan project, we therefore have

Cost per peak Watt     AUD 3.00 / Wp
LEC                            AUD 199 / MWhr

The components of the LEC are:                   
CAPEX           {0.094 × AUD 300×106} / {187×103 MWhr} = AUD 151 / MWhr
OPEX             {0.030 × AUD 300×106} / {187×103 MWhr} = AUD 48 / MWhr

Infigen would cite different figures to the above, which are what you get under the standard assumptions.

Thursday, January 6, 2011

Cost of solar power (1)

Lots of very smart people work very hard to calculate the cost of electricity, no matter how it is generated – fossil fuels, nuclear fission, wind, geothermal, solar photovoltaics, solar thermal etc.  It’s an endeavour fraught with contentious assumptions.  For example … What is the estimated capital cost?  What is the capacity factor and output?  How is the project to be funded?  What is the level of government support?  What is the interest rate and how long will it take to pay back loans?  How is taxation calculated?  What level of profit is demanded by investors?  What is the present and expected cost of fuel?  What are the operating and maintenance costs?  How to cost externalities such as CO2 emitted or radioactive waste?

In this blog, I shall only look at solar power – both PV and solar thermal.  I’ll make a set of standard assumptions at the outset, namely
·         there is no inflation,
·         taxation implications are neglected,
·         projects are funded entirely by debt,
·         all projects have the same interest rate (8%) and payback period (25 years), and
·         all projects have the same annual maintenance and operating costs (3% of the total project cost). 
These assumptions will be applied consistently to all projects.  No currency conversions will be made.  The cost of capital is given by the rate of capital return.  To pay back a loan at 8% over 25 years, annual payments must be 9.4% of the funds borrowed.

Estimates will be given for two cost metrics – the power per peak Watt and the Levelised Electricity Cost (LEC).  To make these estimates, all that is now required is the capital cost of the project, the peak power output and the total power output per year.

Example (1) is my passive solar evaporation heat engine as described at and in a recent conference paper: N.G. Barton, “Annual Output of a New Solar Heat Engine”, Proc AuSES Conf, Canberra (2010).

Suppose we have a 1 km^2 horizontal canopy area, with output as per simulations conducted for Wellington, New South Wales.  The peak power output is estimated as 65 MW with annual output 74 GWhr.  Construction costs are estimated at AUD 25 / m^2 for the collectors (glass, land, frame, construction) and AUD 1,000 / kW for the evaporation engine and balance of plant including water treatment.  So the total capital cost for the 1 km^2 project would be AUD 90 million.

Cost per peak Watt     AUD 1.38 / Wp
LEC                            AUD 150 / MWhr

The components of the LEC are:                   
CAPEX           {0.094 × AUD 90×106} / {74×103 MWhr} = AUD 114 / MWhr
OPEX             {0.030 × AUD 90×106} / {74×103 MWhr} = AUD 36 / MWhr

Here, the power output has been considered in detail, but the manufacturing and operating costs are very preliminary.  To refine the estimates will require further R&D and construction of a prototype system.  As it happens, the LEC for a sloping canopy is significantly better than for the horizontal case, but I won’t present estimates for the sloping canopy until the results have been published.

Saturday, January 1, 2011

Welcome to my new blog

Now why should I start a blog, and what am I going to write about?  Well, the things I want to say typically relate to my philosophy of life and my resulting inventions.  By way of background, I have a PhD in applied mathematics from the University of Western Australia.  I lectured at the University of New South Wales from 1975 to 1981 and then had a number of different jobs in CSIRO Australia from 1981 to 2003.  I developed mathematical models for physical and industrial processes, I enthusiastically ran meetings on industrial applications of mathematics, I managed CSIRO’s applied mathematicians for a dozen years, I commercialised software, I did lots of work for professional mathematical associations and I directed a major international congress held in Sydney in 2003.

Over the years, I became concerned about humankind’s impact on planet Earth.  A long time ago, I accepted the science behind Anthropogenic Global Warming and I start to think particularly about how issues would play out in Australia.  This island continent is large, sparsely populated, prone to heat and drought, and occasionally floods.  Meanwhile, the world will need energy, and although we have sources in abundance (coal, natural gas, uranium, sunshine, wind, waves, geothermal), most of these are not good for the long-term health of the planet, whilst the others are difficult to harvest in a commercially viable way.

I found myself thinking more and more about inventions, eventually to the point where I decided to resign from CSIRO and follow my passion.  My thoughts focussed on power generation directly from sunlight, in a cheap and eco-friendly way of course.  In May 2004, I invented a new heat engine based on evaporative cooling of hot dry air at reduced pressure.  This seemed to offer good prospects for power generation from passive solar heat collection, so I worked hard to analyse the thermodynamic cycle as manifested in various ways – in a piston-in-cylinder engine, in continuous-flow, or using the Bernoulli effect. 

Associated inventions are a heat pump that operates on condensation in humid air at reduced pressure, and new schemes for desalination.  At the moment, they have been shelved whilst I work on the evaporation engine.

After completing and publishing the analysis of the thermodynamic cycle for my engine, I decided to build and test an experimental device, a project completed in 2008.  As the concept still looked promising, at least to this still-wet-behind-the-ears optimist, I started to look for investors to take things further.  Not easy, even if the Global Financial Crisis hadn’t occurred around the same time!  The only way forward was to work for a couple more years so as to answer the key questions that were always asked – how much power will it produce and what will it cost?

That led in 2010 to simulations of my heat engine, as powered by heat collection under a transparent insulated canopy, either horizontal or sloping.  I have also estimated how much my concept would cost to build and how the costs compare to other renewable technologies.  That brings me to the present.  From a technical standpoint, I’m now looking at thermal storage with my heat engine.  And, with an eye on commercialisation, I make comparisons on costs of power from various technologies.  Those are the things I plan to blog about – the technical work I’m doing, and cost comparisons.

More details on my inventions are available at my regular site,

I encourage input to this blog.   I want to learn from people who have concerns like me and scientific/engineering/commercial knowledge to share.