Wave/Geothermal - Energy Return on Investment (EROI) (Part 6 of 6)

This is the final piece of a series on Energy Return on Investment from Professor Charles Hall's EROI Workshop at SUNY. Today's papers outline the energy technologies of wave and geothermal power, concluding a 5 part series that has looked at Why EROI Matters, Natural Gas and Imported Oil, Tar Sands and Shale Oil, Nuclear Power, and Passive Solar, Photovoltaic, Wind, and Hydro-electric. Previously, Professor Hall also wrote the thought provoking, At $100 Oil, What Can the Scientist Say to the Investor. Forget not about the simple 'balloon graph' below of EROI x Scale for fossil and renewable energy sources that this project is attempting to update with the help of theoildrum.com readership.


by Charles A. S. Hall

Most of the energy sources that we use or might use are dependent directly or indirectly upon the sun. This includes wave energy which is derived from wind (e.g. the sun). Nuclear, geothermal and tidal energies are different in that they depend upon nuclear decay within the Earth or Earth’s materials or, in the case of tidal, the processes of celestial motions. The advantage of these energies are that they are truly immense. The main disadvantages are that they are, with a few exceptions, dilute and hence very difficult to extract energy from. Another issue is that for some forms (e.g. heat from the ground) high quality energy (electricity) must be invested to extract low quality energy (heat), which can be a losing proposition even if the direct EROIs are positive. These issues for many situations imply generally low EROIs and hence low profitability. On the other hand some hot steam procedures in very favorable sites have high EROI and generate high quality electricity via investment of general engineering and materials, which implies lower quality investment energy. So unless these most favorable circumstances can be applied more generally or better methods are derived it is likely that development will be quite slow. On the other hand if and as EROIs from other fuels continue to decline they might be increasingly attractive. Tidal energies are likewise potentially enormous but there are few operational plants and we have not examined them. Daniel Halloran summarizes here such information as he could find on EROIs of various geothermal and wave energies. They are interesting but remain more as potential than realized energy and appear unlikely to effect our energy situation significantly for decades, if ever. As usual we seek your critiques and, especially, other hard literature that we missed.



Daniel Halloran SUNY-Syracuse

Definition: Geothermal energy is the heat within the earth, which can be “mined” by extracting hot water or steam, either to run a turbine for the generation of electricity or for direct use of the heat itself (Brown and Garnish 2004; Dickson and Fanelli 2005).

Resource Base

Theoretical: The heat content of the earth has been estimated to be about 13 trillion EJ (Dickson and Fanelli 2005). That heat comes from radioactive decay inside the Earth. Obviously, most of this is not practical to exploit.

The 2000 World Energy Assessment estimates that “140 million EJ per year” could theoretically be tapped within a depth of 5 kilometres”, with 5,000 EJ/yr being economical within 50 years (UNDP 2000). A recent MIT study estimated a stored thermal energy of 14 million EJ between 3 and 10 kilometers (Tester et al 2006). This energy could be tapped with enhanced geothermal systems (EGS), also known as Hot Dry Rock (HDR), which exploits the heat available at greater depths in the absence of groundwater.
Geopressured-geothermal systems could theoretically provide thermal energy from hot brine, mechanical energy from highly pressured fluid, and chemical energy from confined methane. The Gulf Coast of the United States has an estimated stored thermal energy of 11,600 EJ in geopressured sedimentary basins (John et al. 2006).

There is not a consensus in the literature regarding resource base estimates.

Actual: World-wide capacity for direct use of geothermal heat is about 16-17,000 MWt and world-wide installed capacity of geothermal electricity generation is about 9,000 MWe. Currently, the only places being exploited for geothermal power generation are places where hydrothermal resources exist. In a hydrothermal resource, heat is transferred to groundwater at depths penetrable by drilling technology. No power is generated commercially using HDR. The world leader in geothermal electricity is the United States with a capacity of over 2800 MWe, which accounts for 0.36% of U.S. electricity production (GEA 2007). Growth of geothermal power capacity worldwide has slowed from 9% per year in 1997 (EERE 1997) to 2.5% per year in 2004 (Dickson and Fanelli 2005).

Geothermal heat pumps, which extract heat from the normally “warm” shallow soils or their water, have grown to over a million units world-wide, led by the U.S. with 600,000 (Lund et al 2004), accounting for most of the four-fold increase in direct use capacity between 1992 and 2000 (Brown and Garnish 2004). While the heat pump industry has continued to grow, total geothermal direct use has slowed to 6 or 7% growth (Bronicki and Lax 2004). Total use of geothermal energy world-wide was an estimated 2 EJ in 2000 (Sawin 2004). Geothermal heat is regional in availability. Countries such as Iceland, Japan, the Philippines, Costa Rica, and the United States, have successfully exploited the shallow geothermal energy available at plate boundaries (Huttrer 2001). Most of the terrestrial Earth does not have those conditions.

Although in theory ground heat is indefinitely renewable there is concern about the sustainability of geothermal systems. Technically, geothermal resources are not renewable, because heat is always removed faster than it is replenished by the heat source (Brown and Garnish 2004; Lee 2004). The most important US site is The Geysers in California which has shown signs of cooling with heavy use. Nevertheless, geothermal energy sources are constant and require no storage other than the earth.

Technology: The general technology is that of steam turbine power generation, with rare “dry-steam” reservoirs (vapor-dominated) being the ideal type of resource. Because most resources are not dry steam, technological improvements are necessary for the geothermal industry to continue to grow (Brown and Garnish 2004), possibly including improvements in enhanced geothermal systems.


The EROI for electricity generation from hydrothermal resources has been reported by a handful of researchers with a range of 2.0 to 13.0 (Table 1). Some conceptual EROI values have been calculated for HDR ranging from 1.9 to 39.0, and for geopressured systems with a range of 2.9 to 17.6. The ranges represent the lack of a unified methodology for EROI analysis and disagreements about system boundaries, quality-correction, and future expectations. No EROI values of geothermal direct use were found. Because they exploit and use lower-temperature resources rather than electricity generation, and are more universally applied, it is probably safe to assume higher EROI values for most direct use applications.


In addition to geography and technology, high capital cost and low fossil fuel costs are major limiting factors for geothermal development, especially for HDR and geopressured systems which are still in the developmental phases. A kilowatt-hour of electricity generated at The Geysers, the largest field in California, sells for 3-3.5¢, and many other plants are economically competitive at about 9¢ (MDEQ 2007). Economic feasibility could be potentially improved in the U.S. with an extension of the Production Tax Credit (Gawell 2007) and with cascading geothermal systems, which use lower temperature waste fluids in succession for secondary applications (Lee 2004).

Environmental and Social Impacts

Positives: Reduced emissions and low land area compared to fossil fuel plants, employment benefits, decreased dependence on foreign energy for countries rich in geothermal resources (EERE, No Date).

Negatives: Small danger of air, water, thermal, and noise pollution, erosion and solid waste buildup. Subsidence, hydrothermal eruptions, aesthetic disruptions, local or indigenous objection, and changes of surface manifestations are rare and site-specific. There is also a controversial possibility of induced seismicity.

Prospects: The limited hydrothermal resources are unlikely to become a silver bullet solution to meet increasing global energy needs but could continue to be important regionally. If HDR were to become economically feasible, much larger, less-depletable geothermal resources would be opened up worldwide, potentially increasing EROI, geographic relevance, and long-term sustainability of geothermal power, with an estimated increase in production of a factor of ten or more (Tester 2006). Geothermal heat pumps already seem to be generating net thermal energy on small scales and are nearly limitless geographically.

Table 1. Geothermal Power EROI


Bronicki, L., Lax, M., 2004. Geothermal energy, In: World Energy Council, 2004. Survey of Energy Resources, Elsevier Inc. http://www.worldenergy.org/wec-geis/publications/reports/ser04/overview.asp

Brown, G., Garnish, J., 2004. Geothermal energy, In: Boyle, G. (Ed.), 2004. Renewable Energy, Power for a Sustainable Future, Oxford University Press.

Carson and Underhill, 1976. [I don’t see this referenced in this summary paper. It refers to a paper referenced in Herendeen and Plant for which I found no other info. See Table 2 of my larger paper--DH]

Cleveland, C.J., Costanza, R., Hall, C.A.S., Kaufmann, R. 1984. Energy and the U.S. economy: a biophysical perspective. Science 225, 890-897.

Dickson, M.H., Fanelli, M. (Ed.), 2005. Geothermal Energy, Utilization and Technology. Earthscan, London. 205 pp.

Geothermal Energy Association (GEA) website, 2007. http://www.geo-energy.org.

Gilliland, M.W., 1975. Energy analysis and public policy. Science 189(4208), 1051-1056.

Halloran, 2007. Unpublished [this refers to my own EROI calculations, which are shown on Table 3 of my larger paper.]

Herendeen, R.A., Plant, R.L., 1981. Energy analysis of four geothermal technologies. Energy 6, 73-82.

Huttrer, G.W., 2001. The status of world geothermal power generation 1995-2000. Geothermics 30(2001), 1-27.

Icerman, L., 1980. Net energy production history of the geysers geothermal project. Energy 5, 29-33.

International Geothermal Association (IGA), 2001. Report of the IGA to the UN commission on sustainable development, session 9 (CSD-9), New York, April 2001. http://iga.igg.cnr.it/geo/geoenergy.php

Lee, K.C., 2004. Geothermal power generation, In: Cleveland, C.J. (Ed.), 2004. Encyclopedia of Energy, Elsevier Inc.

Lund, J., Sanner, B., Rybach, L., Curtis,R., Hellström, G., 2004. Geothermal (Ground Source) Heat Pumps – A World Overview. http://geoheat.oit.edu/bulletin/bull25-3/art1.pdf

Muffler, L.J.P (Ed.), 1979. United States Geological Survey Circular 790: Assessment of Geothermal Resources of the United States – 1978.

Murphy, H., Drake, R., Tester, J., Zyvoloski, G., 1985. Economics of a conceptual 75 MW hot dry rock geothermal electric power-station. Geothermics 14 (2-3), 459-474.

Odum, H. T., C. Kylstra, J. Alexander, N. Sipe, P. Lem, M. Brown, S. Brown, M. Kemp, M. Sell, W. Mitsch, E. DeBellevue, T. Ballentine, T. Fontaine, S. Bayley, J. Zucchetto, R. Costanza, G. Gardner, T. Dolan, A. March, W. Boynton, M. Gilliland, and D. Young. 1976. Net Energy Analysis of Alternatives for the United States. pp. 254-304. In: Middle and Long-Term Energy Policies and Alternatives. 94th Congress 2nd Session Committee Print. Prepared for the Subcommittee on Energy and Power of the Committee on Interstate and Foreign Commerce of the U.S. House of Representatives. Serial No. 94-63. U.S. Gov. Printing Office.

Petty, S., Porro, G, 2006. Updated U.S. Geothermal Supply Characterization. Conference Paper NREL/CP-640-41073, March 2007. Available online at http://www.osti.gov/bridge.

Sanyal, S.K., Morrow, J.W., Butler, S.J., Robertson-Tait, A., 2007. Cost of Electricity from Enhanced Geothermal Systems. Proceedings, 32nd Workshop on Geothermal Reservoir Engineering.

Tester, J.W., Anderson, B.J., Batchelor, A.S., Blackwell, D.D., DiPippo, R., Drake, E.M., Garnish, J., Livesay, B., Moore, M.C., Nichols, K., Petty, S., Toksoz, M.N., Veatch, R.W., 2006. The Future of Geothermal Energy – Impact of Enhanced Geothermal Systems (EGS) on the United States in the 21st Century. An assessment by an MIT-led interdisciplinary panel. Available online at http://geothermal.inel.gov.

United Nations Development Programme (UNDP), 2000. World Energy Assessment.


WAVE ENERGY: Potential, EROI, and Social and Environmental Impacts

Daniel Halloran, SUNY-ESF, Syracuse NY


Wave energy is solar energy concentrated by the wind (Thorpe 2004), which can be converted into mechanical energy for the generation of electricity. Wave energy has been a part of the renewable energy discussion since the 1970s (Duckers 2004), but has yet to materialize as a viable option for large-scale power generation. It can have a high power density (storm waves up to 1700 kW per meter of wave crest length) depending on the speed, duration, and fetch (unimpeded distance over water) of the wind (Duckers 2004). Because of this high power density, and more recently, its relatively low environmental impact, wave energy research and development continues in many countries, most notably the UK, Japan, Norway, and Portugal (Thorpe 2004).


During the energy shocks of the 1970s, wave energy research was mainly government funded and academic in nature. After tapering off for a few decades, it is beginning to reemerge, this time led by small engineering companies (Thorpe 2004).


The wave energy resource base is highly regional in potential. Estimates of the global potential vary widely for reasons that we do not know (Table 1) but many agree that there is a significant exploitable energy source in the waves, especially between 30 and 60 degrees latitude. Wave intensity at these latitudes is greatest in the winter, corresponding with the seasonal peak in electricity demand (Lemonis 2004). Tropical regions have some potential because of the prevailing steady trade winds, but power densities are generally not as high (Thorpe 2004). As with most “environmental” energies a large problem is that the supplies cannot be counted upon, but is at least partially intermittent.

Table 1. Global Resource Base Estimates: Wave Power

Regional differences in the direction, duration, and fetch of wind across the ocean combined with differences in ocean depths near shore cause certain areas to have greater wave power densities than others. Greater wave power densities (or levels) are more feasibly converted into useful mechanical energy. The illustration below shows the semi-latitudinal pattern of wave power density.

Presently, there is very little wave power being generated anywhere in the world. Recently, a Pelamis off-shore unit developed by the Scottish firm Ocean Power Delivery (OPD) was deployed off the coast of Portugal with a grid-connected capacity of 2.25 MW (Power tech 2007). This installment approximately doubled the previous worldwide capacity of about 2 MW that had existed in demonstration projects (Table 1).

There is significant wave potential in the Northeastern Pacific, and the State of Oregon has begun looking into options for exploiting it using technology developed at Oregon State University (Profita 2007). The United States is well behind Europe and Japan, despite estimates that the U.S. wave potential is twice that of Japan and nearly five times that of Great Britain (OEC 2006).


There are over 1000 patented techniques under development for converting wave energy into mechanical energy that can be used to generate electricity (Lemonis 2004). The challenge is the precise engineering required to enable a turbine or other moving part to move relative to the central structure (Duckers 2004). Because the physics of waves varies geographically and temporally, many technological solutions have been proposed, tested, and in few cases, implemented. They are generally classified according to their relative distances from shore.


Net energy analysis of wave energy appears to be non-existent. One study (Banjeree et al 2006) reports life cycle emissions of 21.67 g CO2 per kWh and energy payback time of just over one year for the Pelamis off shore device. Therefore, with an expected lifetime of 15 years per device, the Pelamis could be a sustainable net energy producer with an EROI of nearly 15:1. It is not known how much this would be reduced by including maintenance and other costs. This analysis does not account for the small scale of wave energy production and the inability to demonstrate significant commercial production to date.


There is optimism in the field about wave energy becoming economical in the near future (Duckers 2004; Margolis 2007), although the capital costs are very large. A kWh of wave energy costs about 20 to 30 cents to generate, but one expert compared the wave energy cost to that of wind 20 years ago (Profita 2007). Wind is now down to 4 to 6 cents per kWh. Reportedly, the best technology in the UK is producing at an average of 7.5 cents (OEC 2006). Although waves are more predictable than wind, the variability of the wind causes wave power stations to run at relatively low capacity factors, perhaps around 40% (Duckers 2004), compared to 95% or higher for geothermal energy, for example. This threatens the ability of production to pay back high costs. However, the capacity factor for wind power systems is lower than for wave energy, and the wind industry has been able to reduce costs significantly. In addition, waves are much more dense than moving air, meaning smaller turbines can generate the same amount of electricity (Profita 2007; Lemonis 2004). Smaller turbines should imply smaller cost. However, a major economic consideration is durability and plant lifetime, which may be greatest at near-shore pressure plants such as OWCs with fewer moving parts and less susceptibility to storm damage. In addition salt water implies a very difficult corrosion environment.

Overall there is little experience with wave energy and although the EROI appears moderately favorable the lack of experience and the irregular nature of the resource appears to have resulted in very little research. We have heard that there was one system built in Portugal that was destroyed by a storm but we cannot find a reference even from our Portuguese colleagues.


Positives: Little to no chemical pollution during operation and little to no land use (Lemonis 2004). These devices would have very low greenhouse gas emissions estimated at 11g of CO2 per kWh for near-shore schemes (Duckers 2004), and 21.67g per kWh for the off-shore Pelamis device (Banjeree et al 2006). This compares to a release of about xx KG of CO2 per kWh for coal-fired electricity production.

Negatives: These devices require very high construction costs. From a net energy perspective, the energy required to build the infrastructure may outweigh the small amount of electricity wave projects are capable of producing in the short term. Sever storms have dashed the hopes of some earlier projects, probably before serious energy has been returned. They may also alter coastlines by changing energetic patterns of waves (Lane 2007 may generate various environmental impacts, most of which are unknown. Other potential impacts, such as disruption of marine habitat and fish migration patterns, and sedimentation, are generally agreed to be minimal, but important considerations on an individual project basis.


Noise pollution is usually low (Duckers 2004), but could be a problem in some situations (Thorpe 2004). There has been some concern about aesthetics (Lane 2007) and disruption of fishing, shipping, and boating (Lane 2007). These impacts would occur in both construction and operation.


Wave energy has yet to be demonstrated as a possibility for large-scale commercial power generation. However, with the rising costs of fossil fuels and increasing environmental concerns, a competitive wave industry, if developed, could be one of the most environmentally benign of the renewables. The most practical application for wave energy in the short to medium term could be on small, remote islands without easy access to fossil fuel shipments or the need for long transmission lines. The potential for these sorts of small but locally important projects seems highest in the UK, where wave power density is high and much of the research is centered. Ocean Power Delivery, the Scottish company that provided the 2.25 MW installation in Portugal, is planning a 3 MW project in Orkney, the small island systems off the north coast of Scotland (OPD 2007). There has also been research into potential uses for wave energy other than electricity, most notably desalinization and hydrogen generation.


Banjeree, S., Duckers, L.J., Blanchard, R., Choudhury, B.K., 2006. Life cycle analysis of selected solar and wave energy systems. Advances in Energy Research 2006. http://www.ese.iitb.ac.in/aer2006_files/papers/142.pdf

Duckers, L., 2004. Wave energy, In: Boyle, G. (Ed.), Renewable Energy: Power for a Sustainable Future. Oxford University Press.

European Thematic Network on Wave Energy (ETNWE), 2002. Wave energy utilization in Europe, current status and perspectives. Centre for Renewable Energy Sources.

European Wave Energy Network (EWEN), No Date. History of Wave Energy. http://www.wave-energy.net/Schools/History.htm

International Panel on Climate Change (IPCC), 2007. Working Group III "pre-copy edit" to the 4th Report, Climate Change 2007: Mitigation of Climate Change. http://www.mnp.nl/ipcc/pages_media/AR4-chapters.html

Margolis, J., 2007. Wave farms show energy potential. BBC news.com, March 2, 2007. http://news.bbc.co.uk/2/hi/technology/6410839.stm

Lane, N., 2007. Issues affecting tidal, wave, and in-stream generation projects. CRS Report for Congress, May 2, 2007.

Lemonis, 2004. Wave and tidal energy conversion, In: Cleveland, C.J. (Ed.), 2004. Encyclopedia of Energy, Elsevier Inc.

Ocean Energy Council (OEC) Website, 2006. Wave Energy FAQ. http://oceanenergycouncil.com/faqwave.html

Ocean Power Delivery (OPD) Ltd Wesbsite, 2007. UK’s first wave project announced. http://www.oceanpd.com/default.html

Oxley, R., 2006. An overview of marine renewables in the UK: a synopsis of michael hay’s presentation. Ibis 148(s1), 203-205.

Power Technology, the Website for the Power Industry, 2007. Pelamis, World’s First Commercial Wave Energy Project, Agucadoura , Portugal. http://www.power-technology.com/projects/pelamis/

Thorpe, 2004. Wave energy, In: World Energy Council, 2004. Survey of Energy Resources. http://www.worldenergy.org/wec-geis/publications/reports/ser/overview.asp

United Nations Development Programme (UNDP), 2000. World Energy Assessment.

Wave-energy.net, 2007. History of Wave Energy. http://www.wave-energy.net/Schools/History.htm

For wave power the Wavehub project is very important, as it will test several different technologies and provide some real insight into costs:
UK plugs into Wave Hub | Cleantech.com
Several different methods of power generation will be built there, and run their power ashore together.

Another widely available and relatively energy dense resource is ocean and tidal currents, which are attempting to be tapped using turbines:
Gulf Stream's Tidal Energy Could Provide Up to a Third of Florida's Power (TreeHugger)

Further to the discussion of ground source heat pumps, it should be noted that CO2 air heat pumps are now able to operate to very low temperature levels, and are vastly more cost effective than ground source:
"Eco Cute" CO2 Heat Pump Water Heaters

If considered as part of a total heating package in conjunction with electricity sources, multiplying the heat value of the electricity by between a factor of 2.5 for existing buildings and up to 4 for new builds greatly raises the energy efficiency of that portion of the eletricity used for space heating, so polar PV, wind, nuclear and coal and gas would all operate at rather higher EROEI.

Another wave power demo plant:
PG&E is building a roughly 2MW wave power plant in northern California. Within a few years I think we will have a significant amount of data on the economics.

Is polar PV something new? :P

Is is not new, but they are expanding it greatly at the Antartic base, as it works very effectively in the summer when it is staffed and supplies power around the clock - the low temperatures help, too! :-)

Is gravitational collapse still considered an original source of geothermal energy?


Ground sourced heat pumps that use shallow buried pipes over a fairly large area (usually a cheaper option than deep vertical pipes if the land is available) are not geothermal devices. They cool the surface of the ground slightly and this tips the equilibrium between absorbed and re-radiated energy at the surface. The energy gain over the electrical input is thus a form of solar energy. The mean solar input per unit area of ground is about 10,000 times more than the geothermal energy coming up except in exceptional areas (like Iceland). Vertical pipes are solar dominated for the first 15 metres or so. Thereafter geothermal energy travelling radially in the surrounding area becomes the dominant source

The continuing flow of Geothermal energy has its origin in radioactive decay as the article states. Gravitational collapse stopped long ago and the heat generated by it would have long since leaked away.

I would dispute the air sourced heat pumps are a better option then ground sources ones if you have the ground to install one. Any advance in working fluids, thermodynamic cycles and heat exchangers will apply equally to both types. Even a slight drop in the external temperature makes a big hit on the coefficient of performance. The link given says they have achieved a COP of 3. This will be under the best conditions. It would be unlikely to achieve better than 2.3 over the winter. I have instrumented my ground source system and it achieved a COP of better than 4 for nearly all the winter.

For an annual 6000kWh heat input to the house that is an annual saving of about 1100kWh of electricity. The cost of the pipes and excavation for my system was about £1500. Subtract from that the cost of the air exchange unit and take a guess at how much electricity will rise over the next few years and it does not take that long to get your money back. In the UK where there is a a government grant of about £1500 available for ground sourced systems but not air sourced ones and there is no contest. In addition you do not have an ugly air exchange unit stuck on your house.

Wow! I have never seen a figure remotely near to £1500 for the pipes and excavation!
I am not disputing it, but have you a link to your supplier?

A rough estimate I got at a recent home show in Canada was about $30,000 for an average suburban house. Another estimate from a few years ago was $20,000. Needless to say, I'm concentrating on insulation first.

I was quoted €6500 euros for a complete ground source heat pump system here in Germany, if installed as part of a new house build [1] so I could imagine laying the piping being around 1500GBP. Fertighaus's are largely made of chipboard and vast quantities of expanded polystyrene. The walls are typically around 40cm thick and filled mostly with insulation. 0.1 -> 0.14 W/m^2K heat flow.

The crucial point being that it is part of a new build, much more expensive to retrofit to an existing property. Plus it comes down to supply and demand, if you're the only one in your state trying to install an HP system it's going to be pricey.

You may well find it's cheaper to knock an old property down and put something like a fertighaus up in it's place rather than trying to heat and maintain a traditional property over the next 25 years.

[1] Fertighaus: http://www.fertighaus.de/

Sounds like you could do this with ThermaSave products.

Small quibble - the heat from the original gravitational collapse may have leaked away but the heat from continued gravitational pull (collapse, except that things are fairly dense liquids or solids) continues in conjunction with radioactive decay. In fact, this mechanism has been presented as one possible theory of terrestrial heat generation. As the paper referenced notes, there are aspects of this theory that are more consistent with observed statistical values than the fully radiogenic theory. Further, work published in the journal "Nature" also calls into question the radiogenic theory.

Thus, I would state that it is far from settled that radiogenic sources are the only or even primary source of terrestrial geothermal heat.

I agree with Nick Rouse above.

Ground source heat pumps are only sold as 'geothermal' because it sounds good. Basically the heat comes from the air by making the ground a bit colder than it would have been otherwise.

Air source heat pumps have a problem of the heat exchanger becoming coated with ice in damp climates like the UK. This means that their midwinter performance deteriorates rather seriously and at worst they may have to revert to plain resistance heating. There was interest in them in the UK in the 1970s but, as I understand it, the electricity industry dumped the air source technology when they realised that it would not reduce the peak winter load on the grid.

Ground source heat pumps get round this by using the thermal mass of the ground to coast through midwinter cold periods without performance dropping too much.

You have to be careful in assigning an EROI to heat pumps. Most are electrically operated, so they are really part of the electricity distribution network rather than an actual energy source. A true EROI calculation needs to start with the EROI of the fuel used to generate the electricity. Their advantage is that they can compensate for the efficiency loss at the power station. If 3 units of coal go in to make 1 unit of electricity, then the heat pump with a COP of 3 can convert this back to 3 units of heat.

The same applies to engine-driven heat pumps. The Festival Hall in London was originally designed to use diesel driven heat pumps sucking heat from the Thames. I'd guess at a diesel to heat COP of around 1.5 for this, so the overall EROI would be the EROI of the diesel fuel multiplied by 1.5 less a bit for the energy used to make the diesel heat pump.

Finally can I ask again -

Where is the EROI of building insulation as a potential generator of negawatts (i.e energy not used)? Surely this is just as valid as positive energy flows in the bubble diagram?

I reckon it has potential EROIs in the range 10-100 and potential equivalent negawatt flows in the US of the order of 5 quads or more.


Bob, I referred to ground source heat pumps as geothermal because that is the commonly accepted term, not out of ignorance as to their source of heat.

On the issue of icing etc, that may well have been a problem in the 70's, as might inadequate performance, but it seems clear that the latest Japanese designs have overcome both, and have very good performance.
They are extensively used in the island of Hokkaido, where the winters are harsh and humidity is often high.

Not only is the source of geothermal heat the energy generated by the decay of uranium and thorium atoms within the earth, but that decay produces radioisotopes which pose certain radiation dangers that might be associated with the use of geothermal power. I became aware of this while investigating radiation problems associated with the extraction of natural gas from the Barnett Shale of North Texas. The extraction of heated steam and water from subsurfaces sources will inevitably bring radioisotopes including radium and radon to the surface. These isotopes in tern have the potential for contaminating geothermal electrical generating systems, as well as releasing significant amounts of radioactive gas into the atmosphere. Thus the use of geothermal energy sources pose a hazard to human health, it the associated release of radioisotopes is not contained and cleaned up.

Secondly, in practical terms earth source heat pumps are far more expensive to install than air source heat pumps. They are also potentially very expensive to repair. A COP of over 6 is possible with air source heat pumps, and air source heat pumps capable of operating in Canada have been designed. Air source heat pumps are far more practical replacements for gas furnaces in the American South, because of their lower price, and because Air Conditioning contractors would already be familiar with many air source heat pump features. An air source heat pump is basically an air conditioner that can reverse its heating and cooling cycle.

Because they tend to level summer-winter electrical demand, electrical generating companies might well subsidize the installation of air source heat pumps. The Government could also offer tax incentives for switching from natural gas to ASHPs.

I would dispute the air sourced heat pumps are a better option then ground sources ones if you have the ground to install one. Any advance in working fluids, thermodynamic cycles and heat exchangers will apply equally to both types. Even a slight drop in the external temperature makes a big hit on the coefficient of performance. The link given says they have achieved a COP of 3. This will be under the best conditions. It would be unlikely to achieve better than 2.3 over the winter. I have instrumented my ground source system and it achieved a COP of better than 4 for nearly all the winter.

Hi Nick,

I keep detailed and I like to think fairly accurate records of my heat pump's operation and its seasonal COP averages between 2.4 and 2.5 over the course of our long, cold and surprisingly damp winters (colder than Buffalo, NY and about as damp as the Pacific North West!). I've crossed checked these numbers with my fuel oil consumption and daily meter readings and everything seems to mesh well. My heat pump is an older model that falls below current federal standards; its HSPF is 7.2 and the new minimum is 7.7. High end air source units such as those offered by Fujitsu, with HSPF ratings as high as 11.0, are 1.5 times more energy efficient than my own and generate far more heat at the low-end of the temperature band and continue to operate down to -15C and I'm told, anecdotally, as low -20C. The COP in this case is as high as 3.2, which is not all that far off from that of a typical GSHP, particularly if heating and cooling loads are not well balanced and taking into consideration additional fan and pump related losses.

I paid $2,100.00 CDN for my ductless unit, installed -- a GSHP would easily cost ten to fifteen times as much (likely more since my home has no existing duct work) and would at best save me another couple hundred dollars on my utility costs. As it stand now, my home's space heating costs are about $650.00 a year (electric + backup oil) and a Fujitsu high efficiency model could knock another $150.00 off that. I'm trying hard to understand how a GSHP can offer better overall value in moderately cold climates like mine (well, that is if you consider Buffalo, NY moderately cold) and for the life of me I'm just not getting it. And I'm not saying this to be argumentative -- I genuinely want to know if they are truly cost effective and offer superior value when compared to the other alternatives.


The heat of formation of the Earth is considered a contributor to the heat flux from the Earth. Radiogenic heat is usually considered to be the larger contributor these days. Tidal heating also contributes.


Just look at the graph.

I love the prominence of the *C*O*A*L* big red spot hanging up high and in center.

Look at it. It's HUGE. It's Efficient. It's abundant.
It's here in the west. In US, Canada, Australia, Germany, Poland.

Please, try to find on that picture Gasahol, biodiesel, and Tar sands. This is not trick, it's there!

How much more does it have to hurt before US politicians stop fooling around with regulations on corn ethanol and Carbon Sequestration and run for this proven, cheap and abundant and All-American resource?

You might not like it, but you have to admit that in short term, there is no substitute, because we do not have 30-50 years do develop and scale up wind, solar, biomass and thermonuclear power Generation and (don't forget!) Storage.

I have trouble being roused into patriotic reveries over Coal, which is poisoning our air and water before, during and after it is used for maintaining our expectation of this energy-fat diet.

I don't deny that it is the most convenient source to turn towards.. but this thinking is dangerously short term as we look at Atmospheric Carbon, Mercury and other pollutants in the waterways as they become constantly more important for fish and other wildlife health (and their relationship to food supply)

You're right I don't like it, and yet I do not admit that there is no substitute. We've known about the substitutes for 30 years or more. The fact that we've basically ignored them so far doesn't suddenly make coal a reasonable option.

With any luck, we'll have enough unemployed Auto and Airline workers to finally initiate a new WPA equivalent.. and then we'd see how fast a PV/Wind/Conservation/Transit(etc,etc..) Buildout could really be effected.

Thanks for the patriotic remark, but that misses the point. This becomes a fight for survival, there is no time to be spoiled.

My point is that without coal, we do not have enough energy to replace 50 years of infrastructure before oil hits prohibitive $200/barrel. No matter how many workers you have, if you have no affordable energy, you have no steel, no pipes, no wires, no transportation and you cannot achieve anything.

Yes we did ignore substitutes too long.
Yes, oil is energy of a far better quality than coal.
But quality of energy from coal is still far better than biomass, wind, direct solar, and most of the rest.
As far as clean, I am burning coal and it far cleaner than wood. Good coal is smokeless. Biomass never is.

I am very enthusiastic over Underground Coal Gasification. That is even much cleaner, consumes nearly all coal underground and leaving all the ash and trash right in place. Clean syngas is then taken to surface for processing.
It is much less wasteful, more energy efficient, way cleaner, safer for miners, and the resulting syngas is ready for synthetic diesel or gasoline.

Seems we have a Coal industry lobbyist onboard...

Awesome! He can hang with the nuclear lobbyists and that guy who's always promoting the little carriages hanging from cables.

Well done. That was funny!

Seems we have a Coal industry lobbyist onboard...

Awsome. If this serves as the only remaining "argument", then that is a solid acknowledgement that the bare facts are undisputable. Thanks for contributing this important scientifc argument.

Be forewarned, that I am also Nuclear lobbyist, Polywell Fusion lobbyist, External Combustion Quasiturbine lobbyist and Everything That Makes Sense lobbyist. And I do all that lobbying for free!

On the other hand, I would not lobby for Carbon Sequestration of Corn Ethanol if you paid me a million bucks, because I think that it is gravely immoral to put lives of millions of people in misery and diminish their chances for survival.

On a second thought, I cannot take credit for such an efficient lobbying for Compassion with poor and basic Common Sense.

The full Coal lobbying credit goes to Nate Hagens who posted the graph above and Cutler Cleveland and C. Hall of US EIA who supplied the graph and underlying data.

If you had been following along, you'd know that the number for coal is mouth-of-the-mine and thus can't be directly compared with other sources on the graph.


China is already stumbling with king coal and they effectively have no restrictions. We have been exploiting coal for a very long time. Sure King coal will make something of a comeback as oil peaks but the transition from oil -> coal is not favorable even given electricity as a energy carrier. A move back to coal for the US does not result in a expanding economy we are not china. Without economic expansion our current way of life is dead.

If the situation is a declining or at best steady state economy with energy say taking 20-30% or more of economic output your dealing with a radically different economic landscape and set of economic forces vs today.

Put it this way simply to cover the liquid fuel needs for our global economy into say 2020 or so means we need coal to supply about 50mbd at least of liquid fuels not to mention expanded electrical needs. And this is assuming very conservative declines in oil production say 1-2%. Your free to consider almost any resource and even given anemic growth of 2% a year off our current base the requirements by 2020 are mind boggling. And this does not even include potential growth in the Middle East or Africa just India and China.

So its impossible to consider the business as usual scenario as valid and its more important to focus on smoothing the transition to a renewable society. Not that coal does not have a large role as a intermediate source of power during a transition but only as part of a bootstrap to renewable. CTL etc will simply eliminate our chances of getting out of this one without serious hardship. The window if you will of opportunity is rapidly closing the longer we ignore the basic resource/economic problems we face.

For some uses liquid fuels are very difficult to replace, for instance in some heavy equipment like agricultural machinery.
For others electricity can be substituted, and where it can is often enormously more efficient, for instance the entire car fleet in the US if run on electricity and batteries would only consume some 75GW or so of power.

The use of air source heat pumps alone could increase the efficiency of space heating with electricity by a factor of between 2.5 and 4, whilst residential solar thermal could provide most hot water needs.

Once we start getting on with it then a lot of substitutions can be made which greatly reduce the power requirements.

"For some uses liquid fuels are very difficult to replace, for instance in some heavy equipment like agricultural machinery."

It just takes a little creativity. For instance, see:


For more efficiency, the next step is a bigger battery and a plug...

Developing batteries capable of running massive pieces of mining equipment or giant combines for extended periods of time strikes me as non trivial problem. Also in the case of mining we will be faced with the necessity of running power lines to every remote mining site which I assume will jack up expenses, although these expenses would be offset in some degree by decreased expenses in the oil industry. Also I do not anticipate either ocean going vessels or airplanes running in a hybrid configuration any time soon.

I understand that in the case of mining much electrical equipment is used anyway.
You could power it either with nuclear battery technology or solar or wind.
Obviously if you used the renewables option then storage would have to be provided, so you could run the mine 24/7.

In discussions here agricultural machinery seemed to be a lot tougher, as it has got to be able to move around.
A rather rambling discussion came up with a couple of ideas such as a power truck to replenish the equipment, zinc-oxide batteries or making an exception by using biofuels, but nothing worked too well.

"Developing batteries capable of running massive pieces of mining equipment or giant combines for extended periods of time strikes me as non trivial problem. "

Mining is, as Davemart notes, often electrified already. Underground operations don't like fuel...

Combines are harder. You might have to have overhead lines, or swappable batteries. Either would be infinitely better than any alternative with draft animals, as we occasionally see proposed. I expect that we'll see hybrid machinery gradually expand their batteries and start plugging in, as batteries get cheaper.

"I do not anticipate either ocean going vessels or airplanes running in a hybrid configuration any time soon."

Planes are hard, due to their weight limits.

Water shipping is much, much easier. In fact, container vessels could easily run mostly on solar and wind, due to the very low power to surface ratios of these huge boats. Large batteries could be carried for the remainder, to be recharged at frequent port stops, as used to be done with coal. Or, the ships could just slow down - a speed reduction of 25% reduces power consumption by 50%.

If this is so easy, why don't we do it already? Because bunker fuel has been so cheap. Now, even at PV's currently high price points it would be cheaper than bunker fuel for propelling ships.

Now, even at PV's currently high price points it would be cheaper than bunker fuel for propelling ships.

Excellent news Nick!
I don't suppose you have any links?

In fact, container vessels could easily run mostly on solar and wind, due to the very low power to surface ratios of these huge boats.... Now, even at PV's currently high price points it would be cheaper than bunker fuel for propelling ships.

Are you absolutely sure about that?

Suppose a container ship is 350 meters long and 50 meters wide, with a deck area of 17,500 m² (the biggest container ship in the world is somewhat smaller).  This is covered with PV at 18% efficiency.  At high noon at the equator, the deck receives 17,500 kW of sunlight and generates 3.15 megawatts of power; averaged over the whole day, this is perhaps 630 kW.

The biggest container-ship engine generates more than 80 megawatts.  You might be able to power such a ship with wind, but not with solar.

That 80MW is maximum output - cruising power would be much less. I was interested to find that it was put into service in September 2006 aboard the Emma Mærsk, which is a hybrid electric design!

I'll have to see if I can re-find my calculations - I estimated that PV could provide 20% of power, and wind 30%. That would be enough to provide all power, if they just slowed down by 50%. If not, they'd need some batteries and perhaps some supplemental bio-diesel (especially for emergencies. The Emma Mærsk can carry 156K metric tons, which would allow a pretty large battery.

I think just a little creativity would make it easy to deal with fuel issues. For instance, you could place PV on outriggers to expand available surface area; a lot of light is reflected from the sea, so you could put PV on vertical surfaces - perhaps even something that looked like a traditional sail!

"we do not have 30-50 years do develop and scale up wind, solar"

Wind supplied 20% of new US electrical generation in 2007. It could supply 100% in 5 years, and start replacing coal after that. Solar is about 10 years behind wind.

we do not have 30-50 years do develop and scale up wind, solar

what is your study that says that? how do you claim to know that? what if those don't have to scale? what if we have massive conservation? what if we have some new and easy way to get energy?

He's quoting another poster. Look upthread.


Any alternative will need material, work and energy. Where will you get the energy needed to build renewable energy solutions, when the energy is scarce and too expensive?

You could not, if it were not for coal. Coal is both cheap and abundant. And it will stay so for another 30 to 50 years.

After Peak Coal, Coal will not be cheap.

However, as the quality of coal produced will be declining continuously the world coal energy peak is projected to come around 2025. It is also important to note that ‘‘peak coal exports’’ should come even earlier, as lower-energy-density coals are not worth transporting long distances.


Cheers to wind! I like wind - except when it does not blow and I need hot shower.


Please look at the graph again. It shows less than 1% windmill energy. (perhaps the disclaimer "new generation" is what makes it 20%) It takes a lot of energy to make the wind turbine. It is TONS of steel, which needs - guess what - tens of tons of COAL to produce. That's why it shows EROI of about 20, while using coal directly has EROI between 40 and 80. Hence my statement that it takes 30-50 years. To replace all our elecricity with wind, you will have to mine 10% more coal for about 50 years to get the energy for manufacturing all that network.
Now, turbine is only the cheap part of the solution. How will you get electricity when wind does not want to blow?

Batteries ;-)? Or thousands hydrostations charging dams and then releasing it?
Or perhaps backup COAL station?

Either way, you will have to spend lots of energy fighting environmentalists.
They managed to get the Massachusetts Martha's Vineyard windmill field plans trashed.

Or take it another way. Suppose all 100% of the 2% of new US energy generation will be wind turbines, it will take 100%/(2%/year) = 50 years to replace the all power with wind. So even with 50 years I was too optimistic.

To clarify, this calculation assumes a rough estimate, that our 50 year old energy network is replacing 2% of the old energy Genaration with new one each year. I bet that my estimate is +/-1% accurate.

At best deviation of 3% of energy network replaced, we still need 33.333 years for total windmill replacement.

At the moment the main renewables resource is wind power, which does indeed need back-up, not at 100% but in increasing quantities as penetration increases.
However, although solar power is not economic at the moment I have recently become convinced that solar power will shortly (by 2012-15) be viable in many areas of the US, providing that you play to it's strengths and use it in hot regions where peak demand is mostly when the sun is up and cooling is the main problem.
To avoid the extra costs associated with rooftop installation then Nanosolar may have a good approach:
Nanosolar Blog » Municipal Solar Power Plants

Please note the multiple advantages of this approach, with the scale big enough to be easily serviced, readily accessible as it is ground mounted, and not needing extensive transmission lines, or indeed steppers ass the power comes out at 20volts ready for local use.

Advances by multiple companies and technologies seem to show that in this sort of time frame we could hope for a cost of around $1/watt.

For residential and commercial use amorphous silicon or thin film technologies can be built into building materials, and it si a lot easier taking thee tiles with you when you move than uninstalling a mounted system.

Peak capacity of the US grid is around 1TW, and the average load is about 460GW, of which baseload might be around 300GW(my estimate)

For this base-load you would need around 200 of the latest 1.6GW reactors, or a twin reactor on each present site.
At $9bn per reactor that would be around £1.8trn, over a realistic 30 year build that would be $60bn per year, a fraction of oil import costs.

That would only take care of electricity suppply, so what about natural gas etc, supplies of which look like being tight and expensive?

For the space heating part of requirements air-source heat pumps could multiply the effectiveness of eh energy input by a factor of 2.5 for old properties, up to 4 for new build, and tighter insulation standards could further reduce usage by large amounts.
Residential solar thermal could provide in excess of 50% of hot water almost everywhere.

I would see around 30 years as being the sort of time period to very substantially, although not completely, alter the US supply and use of energy landscape.

So why go this route rather than the cheaper coal option?
I would clarify that I do not see coal disappearing overnight, but the reason is basically because I have never seen an ICC project I think is realistic.

Since I am a non-specialist, I take the climate projections of the IPCC as the best guide we have, and so would try to keep below the recommended limits for the lowest case scenario.

Part of the job appears to have been done for us, as there simply does not seem to be as much oil as they project, and I can foresee steeply rising coal prices in many areas of the world, although not the US without carbon charging.

The only thing on the horizon that I can see that might change the equation is in-situ combustion, but it is simply too early to evaluate the potential or carbon sequestration possibilities of that.

Anyway, I would be interested in your critique of these ideas.
Shoot me down! :-)_

You are looking for a constructive, scientificaly based critique?
1. The link is dead but http://www.nanosolar.com/blog3/ works
2. You must be a nanosolar industry lobyist ;-)

I think that these ideas are great!
Silicon photovoltaic shingles might be a bit expensive, but surely will be more durable than asphalt shingles.

And as a added benefit, in the winter, you can run them in reverse to prevent ice dams!

The you cannot achieve this goal without increasing coal energy.
1. How many square meters of photovoltaics replace our energy needs?
2. Do we have resources to convert US to phitivoltaics within 5 years?
3. How much coal is needed to unlock that much Si from SiO2 and for
other manufacturing processes?
4. How do we store solar energy for nighttime and how much it costs?

So it is not Photovoltaics or Coal. It is Photovoltaics AND coal.

And do not worry too much about what IPCC says. At the rate their temperature increase guestimattes are decreasing, they will be negative by 2020. And then they will realize that mankind could not generate enough CO2 to revert the catastrophic global cooling.

Worry about CO2 sequestration fools such as signatories Clean Energy Act bill, because they can kill all the energy input to realize our photovoltaic plan by making coal energy too expensive.

I share your sympathies with On Site Combustion. Very effective!

Thanks for the reply.

I don't go in much for 'Grand Plans' and hence am not talking about converting totally to PV in 5 years or anything like it - that is one of my reasons for favouring nuclear for base-load, that we know that it can be done, and we know how to do it, since the French are doing it at present.

That is the reason that is my solid proposal, but I would also not discount geothermal for base-load in some regions, and possibly solar thermal, and would advocate vigorous research into those areas, but we are not far along enough with them to rely on them as yet.

For similar reasons I would at present discount solar PV for baseload, and in northern areas with poor winter sunshine.
But things can change!

If you use nuclear or geothermal for baseload, you don't have to fiddle with storing solar for nightime, and that is what makes it cost effective.

The remaining peak load market in the US is vast though, and exploiting that should further reduce costs.

We should conserve coal burn though, and not just because of GW, but because globally we may not have as much as we think:


I agree with your words.
1. We should be thrifty even with coal and buld the post-coal infrastructure,
2. backed-up with nuclear as a baseload.

3. The third best candidate is likely tidal energy, barrage tidal twin basin dams with hydrostations. (Predictable, abudnant, high EROI)

4. geothermal (But not for all uses)
5. wave (Good desity and constancy)
6. wind direct (turbines) (Fair density, needs storage)
7. Algae biomass (Low desnity Generation, high density storage)
8. Wood growing and gasification
9. thermal solar
10. photovoltais, sadly the last and least.

A lot of those have good potential, and need further support and research, but are at an early stage of development and so can't be properly assessed.

I'd disagree about solar PV.
At the moment it is very expensive, but audited companies like First Solar have now said that they can do cells for not much more than $1/watt.
Prices are not currently so low, but that is partly because silicon has been in short supply.
Those bottlenecks are now being rapidly addressed, and by 2012-15 in the right configuration in hot areas it should become competitive with natural gas, perhaps earlier with the rapid massive increases in NG prices.

There are enormous advantages to this, as it does not use water in dry areas like Arizona, and can be put right near where you want to use it so you don't have to transmit it far, and matches peak use in hot areas well so you don't need to store it, and you are not competing against cheap base rate power but expensive peaking power.

Anyway, I sure hope it works out as that is what I mainly base my projection that massive low-carbon supplies of electricity can be made available fairly soon in the US.

I like the cleanness of PV so much that I would love to give it higher marks. But to be honest, I cannot, not even for $0.10/W . There are few reasons:

1. Why is most abundant earth metal in short supply? Because Silicon is commercially prepared by the reaction of high-purity silica with wood, charcoal, and coal, in an electric arc furnace using carbon electrodes. At temperatures over 1900 °C, that yields only 98% pure Si. More repeated reheating is necessary for purification. Lots of COAL! Until the Cambridge FFC process is developed, Silicon will stay in short supply and expensive.

2. PV electricity is not the same quality as power grid electricity.

I power grid, the generator is freewheeling, until some consumers flip the switch. Then the Generator starts feeling some resistance and has to work for power. Esentially, you brought piece of the gas turbine home.

In contrast, PV electricity is use it or lose it source. As long as sun shines, you do not need lighting and house is heated directly. Rain, night or winter comes and as sun becomes more scarce, so is your PV electricity but so much more is it needed. PV is technology that delivers energy only when you don't need it. Sure you can buy synchronized convertor and send excess to power grid, taking it later. But on large scale, if everyone did the same, outage would come every Rain, night and winter.

What comes to your mind then for backup, Coal?
So PV is at best supplementary source.

Why is most abundant earth metal in short supply? Because Silicon is commercially prepared by the reaction of high-purity silica with wood, charcoal, and coal, in an electric arc furnace using carbon electrodes.

You forgot:

Metallurgical silicon is very abundant and cheap; the only reason PV silicon has been scarce and expensive is because it has been using the castoffs from the semiconductor industry, with its much more stringent purity requirements (10 ppb impurities instead of 1 ppm).  Now that PV's volume is set to exceed the semiconductor industry's, a dedicated supply of much cheaper material is viable and is being developed.

PV electricity is use it or lose it source.... PV is technology that delivers energy only when you don't need it.

You mean, only during the day when demand peaks, especially on sunny summer days when the air conditioners are cranking?

You're giving in to rhetorical excess, but I'll let you in on a little secret:  PV plays well with a number of technologies which exploit off-peak energy supplies (like ice-storage A/C) and displace oil with electricity (EV's and PHEV's).  The solutions for the variability of wind work even better for PV.

You mean, only during the day when demand peaks, especially on sunny summer days when the air conditioners are cranking?

You're giving in to rhetorical excess, but I'll let you in on a little secret: PV plays well with a number of technologies which exploit off-peak energy supplies (like ice-storage A/C)

These statements seem to contradict each other. If the solar resource matches well with the demand for AC then ice-storage is not needed. Ice-storage would better for wind which blows during the night time. However, since winters are windier than summers this option does not seem like a great match either.

Here are some comments about the cost of rapidly ramping up renewable generation capacity. Electricity costs for renewable energy sources are generally calculated by dividing the total costs by the total kWh produced over the lifetime of the renewable generator. These costs correspond to those that would be incurred in a steady state system in which the generation capacity is fixed and the new generators installed every year are merely replacing worn out generators which have reached the end of their life. During a period of capacity growth in which the total installed renewable capacity is increasing the costs would be much higher than those commonly quoted.

Consider, for example, the case where we are installing a fixed amount of new capacity every year. I will give the installation in units of the energy which those installations would produce in one year's time. I am going to neglect interest costs and operation and maintenance costs. For specificity I will refer to PV panels as the devices being installed. I define the following variables:

E0 = amount of new PV capacity installed each year

L = expected life time of PV panels.

After n years the total installed capacity would be:

EC = n×E0

Ignoring operation and maintenance costs the total cost of the PV panels during these n years will be proportional to the total installed capacity EC.

Next I will compute the gross output of energy from all of the installed panels during these n years of growth. For now I am ignoring the issue of the energy expenditures required to manufacture and install the panels. I am simply going to calculate the gross energy output. A subtlety in this calculation arises because if we install PV panels continuously throughout the year then not all of the panels will produce a full year's output during the initial year of installation. It is not hard to show that under the assumption of uniform installation rates the total energy provided by the all of the new PV panels in the initial year of their installation is ½E0. In each of the succeeding years the panels will produce E0 units of energy. The gross energy output over the time period of n years is given by:

EG = n×E0 + (n-1)×E0 + ... + E0 - ½×n×E0


EG = E0×[(1+2+...+n) - ½×n]

Since 1+2+...+n = ½×n×(n+1) we find:

EG = ½×E0×n×n

We now ask the question: What would be the quantity of PV installations required to maintain a steady state system (i.e. a system with a fixed generating capacity in which the only new installations made are to replace worn out panels)? Clearly the steady state capacity required to produce EG units of energy over n years is given by:

ES = EG/n = ½×E0×n

If the lifetime of the panels is L years then a fraction 1/L of the PV panels has to be replaced every year. After n years the total fractional replacements is n/L. Therefore the total required replacement capacity is given by:

ESR = ES×n/L = ½×n×n×E0/L

Since both the growing system and the steady state system produce the same gross energy output over n years the cost ratio (for gross energy production) of the growing system to the steady state system is given by the ratio of the total required installations for each system:

Cost Ratio = EC/ESR = (n×E0)/(½×n×n×E0/L) = 2×L/n

To give a specific example suppose L=30 years and n=10 years. Then the cost ratio of the growing system to the steady state system is 6.0. The cost penalty of growth is substantial. Note that this result is independent of energy payback time (i.e. the time required for the PV panel to produce as much energy as was consumed during the manufacturing and installation process). It does not matter whether the energy payback time is 6 years or 1 month. If you are depending on long lifetimes for low costs then a substantial cost penalty has to be paid during the initial ramp up of generating capacity.

The chart attached below shows average long term cost of gross energy production as a fraction of the steady state cost in the case that constant yearly installations are maintained for a long period of time after which the system is allowed to decay. This graph is done for the specific case where the PV panel lifetime is thirty years. During the thirty years of the ramp to a steady state capacity we are paying the full cost of steady state system, but since only a fraction of the full capacity has been stalled we get a lot less energy for our money. In the first year, for example, we pay the full price of a steady state system but get only 1/60 of the energy. The average cost drops fairly rapidly with time, but after 10 years the average cost is 6 times the steady state cost, and even after 20 years the average cost is still 3 times the steady state cost. After 30 years of constant yearly installations our PV system has reached full capacity and further installations merely replace worn out panels. At this point our yearly cost of energy production is steady, but the cost deficit of the thirty year growth ramp is never made up unless at some time in the future we allow the system to decay away. In this case we get 30 years of free but diminishing energy which makes up for the deficit of the first 30 years and finally brings out long term average cost down to the steady state yearly value.

These ideas can be extended to take of account net energy production (i.e. Gross energy output - minus energy input of PV manufacture).

In order to account for net energy I make use of the energy payback time

Y = Number of years required for the energy output of a PV panel to equal the energy consumed during manufacturing.

In the above example, since we installed n×E0 units of capacity during n years we must have consumed an amount of energy equal to Y×n×E0. If we subtract this number from the gross energy output we will get the net energy output. We then need to determine the amount of new installations that would be required in a steady state system providing the same amount of net energy. If we then take the ratio of the required installations in each of the two cases we find the cost ratio for net energy production. I won't bore anyone with the arithmetic but the result is:

Cost Ratio = (L-Y)/[(n/2)-Y]

If Y=0 this cost ratio becomes equal to 2L/n which is the same as that given above when we neglected energy expenditures in the manufacturing and installation process. To take a specific example suppose L=30 years, n=10 years, and Y=4 years. Then then cost ratio = 26/1 = 26. This number huge. It will rapidly drop of course. For example after 20 years the ratio will be 4.28. Notice that the denominator of the above expression is a negative number until n>8. This fact does not mean that the costs are negative. It means that they are infinite. Through the end of the eighth year the energy spent installing new panels is greater than the cumulative energy output of all of the new installations. Even if you installed 100 Terrawatt hours of capacity per year you would not get any net energy for the first eight years of growth. Long energy payback times represent a serious barrier to aggressive growth.

These same ideas can be extended to the case where the amount of new capacity installed in each year is larger than the capacity installed in the previous year by a fixed percentage. The excess costs during the growth ramp are even higher in this case as is not surprising since new installations are biased toward more recent years from which less total energy production has been obtained.

Of course the excess costs of growth are transient. Once we stop growing and only replace worn out generation capacity the average yearly costs drop to the long term steady state value. I naturally assume that all regular readers of TOD understand that we must abandon composite growth as the unvarying goal of our economic activity. However, just in case some innocent person who has not thought seriously about these matters has stumbled onto this discussion thread, I mention this fact in passing for his or her edification.

Much more important is how growth in installations affects the average payback time of the overall system. The steady accumulation you models leaves many older factories contributing substantially to the system while more rapid growth would have newer factories providing most of the system. These will include factories that use energy more efficiently and reduce the overall cost.


I am assuming that PV costs will not drop exponentially to zero forever but will eventually asymptote to a finite value. If this is not the case there is no point in having any discussion whatsoever about energy technology. We should quit our wasting time posting on the Oil Drum, invest all our spare change in PV companies, and sit back and relax while we wait for the age of solar abundance to arrive.

Either I have time (to write here on TOD) or I have money (to invest in PV, for instance).
Both?? Just show me the man..

Well, PV is cheaper than oil now and we expect it to be cheaper than coal around 2015. It will drop further beyond that by perhaps another factor of two or three. It seems to me that we get there sooner the faster we grow the industry since a portion of the savings comes from scale. The issue you raise is a concern but the main thing really is to have most of the overall system be cheap. The current exponential growth tends to help with that.


I agree that within the context of the current economic system growth in PV sales is stimulating fundamental technology improvements and manufacturing efficiency improvements. However, if we go into a global recession in the next few years we will have to find a different method of stimulating the necessary technology development other than the desire of money to make money. In the long run we are going to have to learn how to make wealth preserving investments which require different economic and social institutions than wealth increasing investments.

Interestingly, one of the legacies of the great depression is soil and water conservation swales that are still working today. PV has a similar long life and might be a response to recession.


Chris - this would be a great guest post topic, if you should find the time.

Of course the excess costs of growth are transient. Once we stop growing and only replace worn out generation capacity the average yearly costs drop to the long term steady state value. I naturally assume that all regular readers of TOD understand that we must abandon composite growth as the unvarying goal of our economic activity.

This seems like a non-sequitur to me from the preceding argument.

Sure you take time to get either your investment back or show net energy, but for quite a while the net energy input doesn't much matter, as we have plenty of it, just not in a form we want, as coal.

By the time the volume of solar panels reaches a level to have any significant effect on energy consumption, or production for that matter, then it should be possible to turn them out for many purposes at a price at which they can be amortised on normal commercial terms - ie they amortise themselves over 7-8 years, not their expected lifetimes.

So long as the costs of the solar power are cheaper than that which they replace, both in terms of money and energy, then you are 'retiring' an existing system, and so benefiting from the savings you mention in that connection.

After amortisation, of course, you are getting very cheap power indeed - that is the reason a lot of power from nuclear stations is so cheap, they are already amortised.

If costs for low-carbon alternatives remain higher than coal and oil, then you are in a less advantageous position, and you would have to pay in money and energy terms a premium both for the replacement and any additional capacity.

Really, all this analysis is far too complex.

So long as costs in money and energy terms are reasonable, then there is little difficulty.

In the same way, whether an economy should expand by increasing energy use is dependent on how much value you get per unit of power.
That is likely to increase if we are successful in conservation efforts, but Jevon's paradox kicking in doesn't greatly matter as long as emissions are down substantially.

Sure you take time to get either your investment back or show net energy, but for quite a while the net energy input doesn't much matter, as we have plenty of it, just not in a form we want, as coal.

It matters for the short term 'health' of the economy as that quality is usually measured in our society. If the global economy goes into a recession in the next year or so, I assure you that making huge capital investments for the better part of a decade without any return will be a big deal.

Really, all this analysis is far too complex.

I did not realize that addition and division were complex forms of analysis. All I am doing is adding up the total dollars and the total energy output and then dividing these two quantities.

So long as costs in money and energy terms are reasonable, then there is little difficulty.

This statement begs the question I am raising. During the growth ramp the cost will be much higher than during the steady state, so that my analysis is highly relevant to the "so long as" condition that you mention. It is true that in the OECD nations are we have a lot of room to compensate for the cost of a renewable energy buildup. As you mention we could ramp down coal mining, and oil and gas exploration and production. We could also ramp down the production of many wasteful goods and services such as SUVs, jet skis, monster houses etc., not to mention improving efficiency. However, the underdeveloped world which has much lower energy use and much lower standards of living does not have this luxury. If the vision you have of the future is for the underdeveloped world growing to catch up to us using renewable energy sources then then this sort of analysis is relevant to the real growth rates that can be achieved.

There are some good points there, Roger, and sorry if I sounded too dismissive of the points you raised in the previous post.

As regards to the developing world, they are certainly in for a tough time.

It would seem to me likely that they will operate at a time lag behind the developed world, IOW the costs of solar development and so on will have been bourne in the west and they will go for solar etc at a later date.

In China and India energy supply is likely to be basically a matter of burning more coal for a while, together with residential solar thermal power.

In places like Africa, many places do not have supplies at present, and if they do it is often in the form of vastly expensive portable diesel generators.

In the absence of power grids, then distributed solar will surely be cheaper.

Systems will not be specified to Western levels, but will use tiny panels powering a LED - enough to read by, and study, but completely unacceptable by western standards, surely cheaper than it would have been to build a power grid though.

I really don't know whether all this will be enough to pull the poorest through.
I am more gloomy about this than at any time since the 60's

Does anyone know about the economics of geothermal when one has an active volcano close at hand?

It would seem like as good a case as any. The "big island" in Hawaii currently gets 17% of electrical power from geothermal and 76% from burning oil. Their sustainability plan looks at various options for expanding electrical power. These options include wind, pumped hydro storage, and solar, in addition to geothermal. I was surprised that geothermal does not play a more prominent role in their future plans.

One thing that surprised me is that they do not seem to be looking at replacing a good part of oil-based electrical power with geothermal. I wonder if part of the problem is grid related. If the power for the big island were all on one closely tied grid, then clearly they could use more base power than the 17% from geothermal. The discussion in the report seems to indicate that if they were to expand the plant, it would exceed base usage, so adjustments would need to be made to provide variable power from the plant.

I imagine the other issue is cost. If geothermal were really cheap, it might make sense to start stringing transmission wires to the rest of the island, to distribute the power. If it is expensive, it may look cheaper to continue burning oil.

Here you go, Gail:

And here is the site it is from:
Geothermal — Department of Business, Economic Development & Tourism

Via the Maximum Power Principle, organisms and ecosystems that maximize 'power' have had adaptive advantages. In human systems, I would define this as:

MPP= EROI X Scale X Flow Rate X Transformity (Quality)

EROI is important but only one piece of the puzzle. As Charlie attempts to articulate in the above 'balloon graph', scale (magnitude of resource) is also important. Flow rate is also essential - if we have trillions of barrels of unconventional fuel underground, but can only get it out at 2 million bpd, then it doesn't make up for the depleting 'power' we have historically been receiving from oil and gas. Finally, quality is very important. Electricity currently is of higher quality than oil, but if we have liquid fuel shortages, that will invert. Our society currently is heavily dependent on the quality and scale and flow rate of liquid fuels. So alternatives have to answer all 3 of these, as well as account for environmental externalities(which we would have to address with or without oil depletion). So as we've been discussing, alt energy sources may have something to contribute but might lack the necessary trifecta (e.g. the potatoes in my yard have 30:1 EROI, much higher than oil, but their scale is limited).

Alternatively, we could change (as quickly as feasible) the definition of 'transformity' (quality) for our culture - more towards electricity that can be generated from pure renewables and away from machinery and systems needing liquid fuels. e.g. meet maximum power halfway.

Thanks for your efforts in trying to expand the literature on this piece of the puzzle, Charlie.


Tghe graph shows scale of present use, but it does not show potential scale. Color coding that shows which souces can grow in the next fifty years are which must shrink might be a help. The lines indicate the progression in in time for the current color coding I think. I notice that the chart is not current in tar sands. I don't see how a fuel shortage can invert energy quality. Can you explain this?



I don't see how a fuel shortage can invert energy quality. Can you explain this?

Chris, we essentially use energy for heat, electricity and transportation. Electricity is its highest use in the current economy because it allows myriad tools, toys and electronics to just be plugged in. Historically electricity has been about triple the energy quality as oil. But it is also subsidized by oil - e.g. we will not be able to effectively upgrade/scale electrical infrastructure without relatively cheap and abundant liquid fuels. Energy quality is really what is valued by a culture/civilization. Thermally a barrel of oil would have been much higher quality to a 1850 Sioux warrior than a horse, but his culture would have had no use for the oil. Horses in that sense had more energy quality for that culture. To me, quality means 'most limiting input' required by its users. If we do have liquid fuel shortages, the price, desirability and quality of gasoline/diesel will invert with electricity.

Well, one can make liquid fuels from electricity and air so I guess as a product of electricity you might count the quality higher or the same. But, if you got more transportation from the same amount electricity, I'd think that the quality of the electricity would still be higher. The only place where I see a big issue of this sort is in aviation: http://mdsolar.blogspot.com/2007/12/jet-fuel.html


Depends if the efficiency loss (which would be huge) from the primary energy source would be larger than the quality disparity. I expect it would. Clearly is in the case of hydrogen..e.g. for a hydrogen fueled prototype plane: roughly 70% of primary energy is lost in making the hydrogen and a further 50% of the remaining energy may be lost in the fuel cell and a further 10% lost in the electric motor. This would result in only 14% efficiency of the primary energy source.

Yes. There is sufficient space heating demand that one can make use of the process heat to cover aviation fuel use at essentially 100% conversion efficiency since all the energy is used, but going to a larger scale might run out of uses for the heat. Based on the current heating oil futures price, I'm guessing the heating season will open at a delivered price above $4.20/gallon. This probably makes making fuel from wind power and coheating less expensive than using heating oil for heat. Jet fuel is currently about $3.85/gallon.


Hi Chris,

In parts of Ontario, heating oil has already hit $1.36 per litre which translates to be $5.15 a gallon before taxes. At this price, it's cheaper to heat with electric resistance if your cost per kWh falls below $0.15 in the case of newer systems (85% AFUE) and upwards of $0.18 for those with older models (70% AFUE).

I'm corresponding with someone in north-west Montana who consumes roughly 1,600 gallons of fuel oil a year and his last quote was $4.24 a gallon (when his house was built in 2002, he tells me he was paying just $0.95 a gallon). Electric resistance at $0.0716 per kWh is the equivalent of oil at $2.39. However, after crunching the numbers a little further, he decided to install two high-efficiency Fujitsu ductless heat pumps (a 12RLQ and 15RLQ) which will now heat his home for the equivalent of $0.80 a gallon -- less than 1/5th of what he's paying now!


A 2002 furnace should have some years left. Too bad about that. I found that resistance heating (including my indoor growing) saved in two ways. It cost less per BTU than oil and I was heating less of the house. The heat pump you mentioned, or the 9000 BTU unit, only takes a 15 Amp breaker. Might be a do it yourself kind of thing if it goes in on some wires you can dedicate.


Hi Chris,

These units have inverter drives, so they're all 240-volt. My Friedrich 14,000 BTU/hr is a 115-volt/15-amp model and as I recall draws 1,230-watts in heating mode and 1,290-watts in cooling. From what I gather, the newer 115-volt designs top out at 8.0 or maybe 8.5 HSPF, so to reach 10.0 or 11.0 requires that you go to 240-volt/dual pole.

In these parts, I can pick up a Fujitsu 12RLQ for about $1,800.00 (wholesale) and slip a buddy of mine who is a qualified HVAC tech another $400.00 to install it. I'm still thinking of moving my current system downstairs to better serve the lower level and popping a new 12RLQ in its place. That would eliminate virtually all the remaining fuel oil related to space heating and the electric water heater I'm installing this weekend should take care of the rest.


Theoretical: The heat content of the earth has been estimated to be about 13 trillion EJ (Dickson and Fanelli 2005). That heat comes from radioactive decay inside the Earth. Obviously, most of this is not practical to exploit.

And perhaps not a good long-term plan if one can't get off the planet.

Mars is thought to at one time had an atmosphere - but as the core cooled, the magnitism that 'pushes back' against the sun shrank and allowed the sun to 'whisk away' the atmosphere.

Taking the heat of the core and placing it in the atmosphere strikes me as a way to raise the atmosphere temp and cool the core. Both have 'known' issues in the future.

Don't worry, Eric, at any rate we might be able to deplete the earth's heat resources, wee would run them low well after the billion years we have until the sun expands enough to swallow the earth.
That should keep up pretty warm, I would have thought! :-)

Don't worry, Eric, at any rate we might be able to deplete the earth's heat resources,

Show the math.

I can't be bothered, Eric - it was meant to be a fairly light hearted comment, but if you are interested you can run it yourself as the figures are readily available, together with the rate at which radioactive decay replenishes the heat of the earth.

If we are talking about geological timescales, the biggest problem will be land erosion. Eventually we will all have wet feet

hi eric,

even if we have an infinite source of energy to use , Arthur C clark pointed out that the waste heat would be the final issue - there is only so much a planet can radiate away ! ( a differnt kind of global warming that it appears we will never find out about ....... )

in the mean time we do out best with carbon di-oxide and methane...



there is only so much a planet can radiate away !

Hence a concern over cheap and easy fusion or the 'lets beam power into the atmosphere from orbiting space stations'.

"we" might all want to get comfortable with trying to keep the equation of photons in = heat radiated out. (if only there was a good model for that equation)

The orbiting space station can easily beam in the night and shade during the day to maintain overall photon ballance.

Thermonuclear is another matter. However, it would be a very simple calculation to find out what the stedady state temperatire increase would be:

dT = (ConsumedPower-(SolarIrradiance*Albedo))/(SolarIrradiance*Albedo))^0.25

Anyone has the figures to substitute into the calculation?

Solar Irradiance is something like 1350W/(m^2)*Area, Area = somethig like Pi*Radius^2, Radius =~ 6000km+. ConsumedPower = ConsumedEnergy/time (expressed in W = J/s)

As far as cheaper, abundant electricity is concerned, Kite Gen seems to have a really good idea. Wind power from relatively constant and powerful upper altitude winds is a pretty good idea. I think these could be scaled up relatively easily though, the company is just having trouble getting capital.


The largest plant they could make would be only 2000m wide with a 5 gW power output, which is like 5 nuclear power plants. What do you guys think?

I completely agree, and find the fact that it has not been funded properly a tragedy.
It should be pointed out that there is similar research on the Laddermill concept going on at a University in Holland,

Makani wind power have backing from Google:
Makani Power, Inc.

Magenn Air Rotor System Finally Floats : TreeHugger
Magenn is testing:

Sigh, I couldn't tell from their site but they have either given up or are building a small concept power plant that will be finished in 2 years. The Kite Gen idea seems the most put together out of these other High Altitude Wind concepts.

I'm not quite sure who you mean by 'they' - assuming it is Makani, they are extremely secretive as they don't need publicity, getting all their funds from Google.
Here is the last known data form Pete Lynn, one of the founders:

Read down to the comments for the technical info.

From this it appears that they are working on a wing which carries it's generators, as opposed to one which has the generator on the ground.

Magenn actually sounds furthest along to me, as they are extensively testing their prototype.
I'm not too keen on using helium as the lifting agent though, as it is in short supply and needed for more urgent purposes.

You may also be interested in this, which is being used to provide additional propulsion for ships:

This means that many of the basic problems of deploying, furling and tacking the kites to get maximum power have basically been solved, or at least their use on ships will show whether any issues remain.

Perhaps engineers here can correct me, but it sounds to me as based on this you could certainly generate energy from a similar system.

Perhaps Skysail themselves will branch out, although they seem to be a typically conservative middle-sized German business, so they may just concentrate on their present area of expertise.
I would have thought in that case it would be worthwhile for others to buy it's expertise for use in non-marine applications.

Based on the relative inefficiency of the mechanical power transmission, I think gyromills may be a better bet.  You're also not dragging a bunch of kites through altitudes where there's little or no wind, like the laddermill does.

I would hope that at some point, the NASA budget will go to zero and we start some equivalent agency for renewable energy research and development. Adding to the budget of that agency would be (well, alright, SHOULD be) the majority of the United States Defense Department budget.

A lot of DARPA money is going to solar and battery R&D.

The military has always been acutely aware of their fuel dependence, but they've recently realized that they can do something about it, and they're pretty excited.

I have seen where they are working with two companies to make jet fuel from non-oil sources. One uses coal, and I think the other uses something else (natural gas?) to make it.

Yes, I think that's the Air Force.

I suspect the Navy is working on synthetic jet fuel using aircraft carrier surplus nuclear power. Guantanamo has a wind turbine to replace diesel!

A lot of DOD projects...

geothermal plants are going to be huge. I think we will also see geothermal from oil wells. that would be an interesting EROEI calculation. an oil well that produces electricity and oil and then eventually just electricity.

Enhanced Geothermal Systems
According to the Geothermal Energy Association (2007), EGS currently generate 0.371% of the electric power in the U.S. New projects in 12 states will double this amount to total nearly 6,000 Megawatts of power in several years. A 2007 study by MIT, The Future of Geothermal Energy, concludes that by 2050 the U.S. could increase this amount by 100,000 MW, thus generating about 10% of the nation’s electric power. Hence, EGS power will not be available for making significant reductions in the use of oil and natural gas in producing electric power for at least several decades.

Ocean Energy
Ocean energy includes wave energy, tidal energy, ocean thermal energy conversion (OTEC), and wind energy off shore. These four technologies provide electric power. Wave, tidal, and wind are functional technologies. But due to siting requirements they are limited by their numbers; and will not generate much power. OTEC is in research and development. I estimate that combined ocean energy power generated for the U.S. is no more than 2,000 Megawatts. In the next 10 years this figure could be tripled. Hence, ocean energy will increase power generation slightly, and it will not replace a significant amount of oil and natural gas that is used in producing electric power for at least several decades.

The introduction to the key post says,
"This is the final piece of a series on Energy Return on Investment from Professor Charles Hall's EROI Workshop at SUNY. Today's papers outline the energy technologies of wave and geothermal power, concluding a 5 part series that has looked at Why EROI Matters, Natural Gas and Imported Oil, Tar Sands and Shale Oil, Nuclear Power, and Passive Solar, Photovoltaic, Wind, and Hydro-electric."

So there is to be no discussion of one of the most promising of alternative energy options, Concentrating Solar Power (CSP) or concentrating mirror solar?

This is disappointing, given the history of EXCELLENT articles that TOD has posted on this option. Recent developments in Concentrating Solar Power using different materials and engineering solutions are opening the possibility of capturing solar power with far less investment of materials:

The most impressive point of the above technology using mylar mirrors is that the imput of material could be very small compared to the potential power output. Well, actually the most impressive thing about the CoolEarthSolar company is that when discussing possible investors, they say no thanks, we are not seeking investment at this time! How rare is that in the alternative energy business?

Some posters on TOD have pointed out that an alternative energy would be truly "renewable" when the energy used to manufacture the alternative energy infrastructure could be provided by the alternative energy itself. I have seen proposed CSP systems that claim a quarter of a million horsepower from one square mile if placed in the desert areas of the world. It is hard to understand how this cannot be promising and worthy of discussion if we are to truly include all viable options in the discussion of EROEI.


CoolEarth looks promising. Still in pilot and prototype phase. I wonder what is left to "work out". I could see a City perhaps installing their own power plant using this technology should it become feasible. I wonder what the costs would be to build a plant. And...how does it scale.

CoolEarth is definitely on the right path - IF it's collection method of the focused light (which I don't understand) is scalable enough.

It would need one heck of a control system.
If you get a breeze then how do you keep it in focus?
At those concentrations if you miss you fry the equipment almost instantly.

Ah, come on, Dave, you can do better than that! look at the diagram. The focal point is inside the balloon, meaning "in a closed system". Wind might move the system (balloon), but not in relation to the focal point..

OK, I'll admit that my comment/question above was also irrelevant, because it is focused onto a small PV-cell. I still wonder if that'll work as wished.

Cheers Dom

I'm not sure whether shaking the balloon would not still throw it off - pretty fine tolerances, I would have thought.
Still, we are not dependent on any one approach working out for solar power - there are lots to compete with each other, and the end result seems sure to result in a major contribution to power needs - I just wish people wouldn't get bees in their bonnets and try to use them in winter where it ain't sunny, on the grounds that they are a 'good thing'.
Under those circumstances they are a waste of money.

Thank you and all who performed this great work. Do you have the entire series available in PDF format? I suppose I could cut/paste, but if you do have it available in PDF or another printer friendly format that would be fantastic. Thanks again. It was a great education for me and I certainly look at energy options in a new way.