Remote solar PV vs small nuclear reactor – electricity cost comparison

It is often claimed that small-scale renewable energy, such as solar photovoltaic panel arrays, will fill an important future energy niche by providing much-needed electricity to developing nations and other remote regions (such as the outback of Australia). That’s a seemingly reasonable argument, but how do the numbers stack up? Below, Gene Preston (SCGI member) provides some easy-to-follow calculations (currency is in US dollars/cents). The results might surprise many:


A friend of mine at the University of Texas and I were talking about his desire to develop a presentation for educators in Africa to use in estimating energy costs. He has just 3 hours for his presentation. He wants the teachers to be able to do the economics calculations themselves. I suggested he narrow down the discussion to just a comparison of solar versus small scale nuclear. Here’s what I came up with:

Solar – Lets go low tech with fixed solar panels. The cost is $8/watt (W) and runs about 14% of the time (its capacity factor). You will need energy storage, which costs $1/W + $.4/Wh (that is, 40c per watt hour**).

Lets say that we develop a solar system to serve a 5 kW peak load with an average load of 1 kW. The daily energy demand will be 24 kWh and peak load is 5 kW. This could be a few houses or a small school with some PCs. To produce the average amount of energy needed will require 1/.14 = 7.14 kW, so lets say 8 kW just to put in a little extra energy production factor. The 8 kW will cost $8/W (for 8000 W) = $64000. The energy storage system will cost $1/W (5000) = $5000 for the electronics and switchgear plus $.4/Wh (24000) = $9600 for one day’s energy usage. I would double this and install two days of storage just to be safe, which would cost $19,200.

Therefore the cost of the 5 kW peak demand solar system is:

$64000 for the panels (only half this cost is the PV array)

$5000 for the storage system electronics

$19200 for the batteries (2 days storage)


$88000 for the entire system. (see what I mean about this being a rich person’s energy source?)

Let’s calculate the cents per kWh energy cost. Assume a loan at 6% annual interest rate to pay for it. Assume the system has a 20 year life.

A = PW [(i)(1+i)^n] / [(1+i)^n-1] where A is the annual payment, PW is the present cost of the system, i is the interest rate of .06, and n=20 years.

Then A = (88000)(.06)(1.06)^20 / [1.06^20 – 1] = .08718 (88000) = $7672.24 annually.

The energy produced annually is 24 kWh/day (365 days/y) = 8760 kWh. The cost per kWh = 767,224 cents per year / 8760 kWh per year = 87.6 cents per kWh. (first wow — that’s expensive!)

What about if we instead generated this energy from small nuclear reactors? First, some examples/references:

The Pebble Bed Modular Reactor would have been in South Africa but there is local opposition

This is an interesting discussion of micro reactors, especially the Russian Navy’s design

This information paper from the World Nuclear Association shows the huge number of small-scale reactor technologies being considered

Here is an IEEE paper on small nuclear (2, 5, 10 and 20 MW reactors)

The objective of many of the above references is to get the nuclear power cost down to about 10 cents per kWh. Suppose we could buy into nuclear at $5000 per kW (that’s the estimated cost of the Babock & Wilcox small nuclear plant [called ‘mPower’], for a 125 MWe plant). The 1 kW of nuclear power portion of the small plant would run all the time so one kW would have an average energy based on the calculations for the solar plant. All we have to do is replace the $88000 with $5000 in the previous “A =” calculation.

Therefore the small nuclear program energy cost is .08718 (5000) (100) / 8760 = 5 cents per kWh. (second wow — that’s low cost!)

However we will need some peaking power to get 5 kW peak load. We can use the battery storage system to get the peaking power. We only need 4 kW since we will have the 1 kW nuclear running all the time. Also, the energy storage need only be about 4 hours at the most at 4 kW (conservatively). The peaking power using nuclear energy is $4000 for electronics + .4 (16000) for batteries = $10,400. Note that the peaking power system costs twice as much as the base load nuclear generation. The total cost is about $15,000 and the energy cost is about 15 cents per kWh.

This small nuclear + peaking system is only about 18% the cost of the solar + storage system.

This is an example of how anyone can, fairly easily, go through the economics calculations for solar and nuclear. Such an exercise would probably an eye opener for them, and dispel the myth that solar is ‘free energy’ or even a cheap source of power. But how are they to afford any type of power plant if they do not have industries that need power and produce income for them?


**This estimate is based on presentations given by Xtreme Power Inc.

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Furl


  1. Barry, My cost calculations for the B&W reactor suggests a somewhat lower cost, somewhere around $3.50 per W.

    The B&W cost is almost certainly higher than the cost of the Indian small (220 MW) PHWR. Recent PHWR costs have been running about $1.40 to $1.50 per W overnight costs. The Indians figure electrical charges of from 4 to 5 cents per kW hour, from these small reactors. Large scale serial production would probably bring indian manufacturing costs down further. Of course Indian reactors are built with Indian labor, compensated at Indian pay rates.


  2. Yes Charles, I agree it will probably come in lower (they did say less than $5/W) — I think Gene is erring on the side of conservatism with respect to nuclear power costs.

    An unraised issue, of course, is that in some instances even 2 MW produced from a very small reactor will be far too much energy (for a village, for instance). What do they do in that instance? Perhaps a smaller reactor again is possible (in the kW range), or otherwise, they’ll either have no power or grin and bear the cost of the solar PV (who will subsidise them?), or, get transmission lines out to them (but what kind of LCOE would that result in??). Else, rural Africa and other regions remote from a robust and low-cost grid are going to take many, many more decades before they pull out of their low-energy state.


  3. Barry, if you build small Indian PHWR at well chosen locations in Africa, they will come. Cheap electric power will draw people from villages, into urban centers, whether low cost electricity will support industrialization. Better to move the people to the Grid, than the grid to the people. That will increase access to services, education, and the tools for economic development.


  4. You’all missing the bigger picture! Such solar implementation are NOT grid connected. They run *exactly* that 14 hours a day….actually it’s more like 4 hours for the peak/name plate capacity. Most to NOT have batteries and those batteries only last about 5 years anyway then they, their lead and acid, get tossed in the Savannah someplace. Africa, BTW, is a major used-battery dumping ground.

    So realistically, solar systems are “light-switch” compatible (not refrigerator, no lights when you need it, no power on demand). Even with batteries you get those dust storms and very cloudy tropical periods with weeks of cloud cover.

    The smallest of reactors being discussed this year…and what a year it is!…are ALL grid compatible. In fact, they are great in *starting grid infrastructure*.

    You take a 50MWe reactor. You build a grid that can serve…at least 100,000 – 200,000 people. You are asking yourself…WAIT! 50MWs isn’t going to serve this many people. You would be oh-so-wrong. This is because initially, that is for years, instead of 200 to 1000 KWhrs/month like we all do in Perth and Dallas, the average use, per month, might be about 25KWs…for that ONE light the average non-electrical-non-connected-to-the-grid-but-now-connected-to-the-grid-family might have. So *over years* they start adding some important appliances: a *small* affordable TV set, a small, say, 12 cu ft. refrigerator, and, maybe, one or two lamps.

    This is who grids are build. This region of 100k/200k people spend *most of their investment* on such a low usage distribution grid around the new nuclear reactor. Assuming the economy gets better, people’s living standards rise *some* because of access to regular electrical power, you start investing more money into this small grid to buff it up as people start using more power. A larger refrigerator, maybe an electric hot-plate. Middle class elements might go for AC here and there.

    Now two things start to happen. Based on rising expectations…the load increases and:

    1. We can add another reactor module and
    2. We start interconnecting the grid to perhaps larger reactors around the main capital city to develop a truly national grid that can back each region up.

    The solar cells some well meaning European or American NGO donated years before? Well, they are broken down to use on kid’s remote controlled toy cars and other worthwhile distractions.



  5. You should check solar panel prices at PVinsights, Solar panel price is around US$ 2.0 /Watt now. This off grid system should cost less now.

    [BNC Ed: The price of $8/W matches that charged for a system in Australia — I just had a 1 kW system installed for a retail price of $11,500, which, based on the US to AUD exchange rate, is a little more than $8/W. This includes panel cost, installation, inverter and so on, so without an inverter, the costs would be lower]


  6. The PV panel in the image above looks to have a maximum output of about 50w. Via a lead acid battery it could power a 12v trickle pump and I have something like it on my garden shed. However mini NPPs should suit an already grid connected settlement where the residents aspire to 10-30 kwh/day lifestyles. For example Botswana wants 3 GW of new coal fired generation

    Our own Penny Wong says that developing nations should be allowed to increase their per capita CO2 but that would be a recipe for disaster. Perhaps rich but nuclear shy nations like Australia should help pay for mini NPPs in other countries like Timor Leste. That may help them out of poverty (for a stable population) while not adding to global CO2. Better still let’s help our current export coal customers switch to nuclear. At last Australia would be helping not hindering the mitigation effort.


  7. I’m not really satisfied with Gene Preston’s assumptions e.g. “Lets say that we develop a solar system to serve a 5 kW peak load with an average load of 1 kW. ”

    So I wanted to work through the numbers.

    I googled ‘peak sun hours for Sydney’ and got an annual value of just over 5 kWh/m^2 from This website also claimed that the average Sydney family uses 5000 kWh per year of electricity or (5000/8760)*24 = 13.7 kWh in one day.

    If this figure for a Sydney family is correct then lets assume a non urban African household uses less electricity and so can we say that we can cut that figure by, say, 3 so its 4.6 kWh a day.

    Now Gene says a few houses, OK, can I assume three households so we are back up to our figure of 13.7 kWh and lets round it up to 14 kWh for simplicity.

    If we take 5 Peak sun hours (I’ve just googled peak sun hours Africa and got this then we will need 14kWh/5kWh/m^2 or 2.8 m^2 at 1 kW so if we assume an efficiency of 15% (in between present PV technologies) then we need 1/.15 = 6.7 m^2 of panels.

    Lets round the 2.8 kW of panels up to three kW for simplicity. Obviously this is where I would disagree with Gene.

    Gene’s figure of $ 8 A Watt gives us 8 X 3000 = $24,000.

    Now we’ve got to go for storage for peaking. Is 5 kW a realistic peak for three households? We certainly need storage for nighttime etc. So would a better expression of storage be in kWh as a battery can supply 5 kW for a certain amount of time and 1 kW for a longer amount of time. What can we assume? So I’m going to follow Gene’s lead and use one days storage of 14 kWh and then double it to give two days storage so 28 kWh of storage. Googling around (sorry) for costs of storage using lead acid batteries I’ve come across a number of websites and am choosing electropedia from Chester in the UK for no particular reason. They quote $150/kWh and that gives us 28 X 150 = $4200, obviously different from Gene’s so I’m going to be pessimistic and double it to $8,400 and will round it up to $9,000 for simplicity. I will assume they need to be relaced once in the 20 year lifetime so its $18,000.

    So perhaps we should add 1/3 more PV to charge the batteries (i.e. $8,000 worth).

    I’ll keep the electronics the same.

    So my costs for the PV option are:

    Panels: $32,000

    Batteries: $ 18,000

    Electronics $5,000 (only because I haven’t had time to go into this)

    Note neither my nor Gene’s costing have added in construction costs unless Genes cost for the panels is a delivered cost.

    So my total cost is:$ 55,000 over 20 years though PV does better and I would expect nuclear to to better.

    So using Gene’s formula:

    Then A = (55,000)(.06)(1.06)^20 / [1.06^20 – 1] = .08718 (55,000) = $4794.9 annually or 54.74 cents a kWh.

    Now that doesn’t need the cost of grid connection because gene’s figures of the nuc plants CAPEX didn’t say whether it included grid connections costs so I’m going to assume it doesn’t.

    If we add the cost of grid connection into a centralised generator and again I’ve googled (sorry, sorry but its all I’ve got time to do) the website Practical Action ( on their grid connection page quoting a 2000 World Bank study gave the cost of rural grid connection as $8,000 to $10,000 per km in non difficult terrain (up to $22,000 in difficult terrain).

    If we assume an average distance of 5 kms and the lower cost thats $40,000 per three house holds

    So back to the costs of our nuclear plant

    Then we have operational costs for the nuclear plant and here I am going to assume its largely auotmated and so say three people per operational shift across three shifts plus leave, illness etc so thats 12, then a small engineering dept of say 3 then thats 15 and I’ll add 5 more for back office etc so I will assume the plant has 20 people on ist pay roll @ $50,000 each or $2,000,000 per year or $40,000,000 for the 20 year lifetime. I’m excluding maintenance costs, land leasing costs and so on. The $40,000,000 is equivalent to 6.4% of the $5000 or $320.

    So we have:

    $ CAPEX $5,000

    Op costs $320

    I didn’t really understand Gene’s battery costs for the nuclear installation so what I’m going to do is use what I did but halve it (and I will assume that the nuclear plant has a 99% availability and when its down other plants can cope along with the batery systems)so thats….

    Battery $9,000

    Grid $40,000

    Thats a total of


    Plugging that in to Gene’s

    0.08718 X 54320 X 100/8760 = 54.06 cents per kWh.

    Thats as opposed to 55c per kWh for PV. Believe me I haven’t worked backwards from this answer, I have only worked forwards choosing assumptions as I go. Its a big suprise to me.

    But thats one of the killers for centralised systems….. bloody grid connection!!

    And I’m assuming the $,5000 costs for the nuc doesn’t include grid connection.

    So what we need is those even smaller nucs so we cam eliminate more and more of the grid costs…….. that is if we are prepared to swallow the idea that a grid connected system may not be the in the best interests of rural development.

    I am happy to have my assumptions challenged, ripped into and my maths is notoriously bad. And the data I found if anyone can find better data especially on the grid connection costs that would be helpful, but I was pretty conservative on that i.e. 5 kms distance.

    There are some other things that could be factored in apart from the use of small nucs to lower grid costs. PV costs are coming down and $1/watt is being talked about for a 2012/13 time frame. Then there is the cost of money. Will the entire output of Gene’s 125 MWe plant be connected to every customer from day one, otherwise that foregone revenue will have to be paid for? I also haven’t factored in maintenance costs for the PV cells apart from battery replacement. We could add @ 6% to the cost of the PV in that case.

    I’m doing this because I like both nuclear and renewables and think the debate about the two is based on faulty outlooks.


  8. Barry,
    this article completely ignores the reality that the first 10W is much more valuable(ie recharging mobile phones) than the next 100W( one or two lights or refrigeration) etc and much more valuable than the next 1000W.
    Even in Australia households are not using more than a few kW, what would a village do with 220MW(say a 1MW/household)?

    Nuclear power’s advantages are in providing large amounts of power close to major demand( ie large cities). Solar PV advantages are the low unit costs of supplying small amounts of power to individual applications, especially in remote homes or villages.

    Next you will be suggesting farmers pump bore water with small nuclear power plants instead of expensive inefficient wind turbines!

    If we could get reactors down to a few 100KW’s, what about using small reactors to power electric cars and trucks instead of those expensive lithium batteries?
    This may sound familiar to readers of “Popular Mechanics”(1960’s publications)


  9. Barry, this article completely ignores the reality that the first 10W is much more valuable(ie recharging mobile phones) than the next 100W( one or two lights or refrigeration) etc and much more valuable than the next 1000W…

    Good point.

    If we could get reactors down to a few 100KW’s, what about using small reactors to power electric cars and trucks instead of those expensive lithium batteries?

    There is little difficulty in getting them down to 1 watt, or 1 milliwatt, but for a typical accelerating-car power demand of ~200 thermal kW, the neutron and gamma ray muffler doesn’t get below roughly 500 tonnes. 1000 tonnes if it’s water.

    That is why I want many nuclear-powered cars to get their nuclear power from one shared stationary boron deoxidation plant.

    (How fire can be domesticated)


  10. Barry and Neil,

    I believe Gene Preston missed one important point in his analysis. He says:

    The cost is $8/watt (W) and runs about 14% of the time (its capacity factor).

    I presume the 14% Capacity Factor is the annual average. What is the capacity factor on the worst days, and for a sequence of overcast days? At Queanbeyan, NSW (Latitude 35 South and 150 km from the coast), the capacity factor for the worst days is 0.75% for 1 day, 1.56% for 3 days, 4.33% for 5 days and 9.4% for 90 days of winter. (The annual average is 13.7%; so similar to Gene Preston’s site)

    To keep the calculation simple I’ll assume the worst case capacity factor is 1.4% for 2 days. So, either the Afican village’s solar PV system needs 10 times as much storage as Gene assumed, or the village must be prepared to go without power for days at a time. Once the villagers have seen electricity they will not be happy with power outages. So in reality, we will have to provide sufficient storage capacity to meet demand continuously, even through sustained periods of overcast conditions.

    Let’s recalculate the cost. Assuming the capacity factor is 1.4%, not 14%, over the worst two days, we will need 10 times the generating capacity of solar panels. So the cost of the panels is 10 times what Gene Preston calculated. The cost of the panels is $640,000 not $64,000.
    So the total system cost, with 2 days storage, is: $664,000 not $88,000.

    Gene Preston said:

    (see what I mean about this being a rich person’s energy source?)

    He is certainly correct about that. But it is even worse than he told the African school teachers. About 7 times worse!


  11. The onboard fuel problem will become critical for many vehicles unsuited to batteries such as aircraft, long haul trucks and broadacre farm machines. To keep them running on existing cheap reliable engines I think hydrogenated synfuels (methane, methanol, DME) are the answer. They can combine nuclear hydrogen with organic carbon which is looped within the biosphere. Since the calorific energy return on food is 0.1 and the human metabolism is say 150w we need 1500w per person just to keep the food system going.

    That’s where I see a need for mini NPPs; to replace oil with hydrogenated synfuels close to where demand will be. When Peak Oil starts to bite places like Africa will be the first to miss out on fuel due to low purchasing power.


  12. You can make nuclear reactors quite small. Both the United States and the Soviet Union launched fission reactors into space before the nuclear test ban treaty. Here is a picture of a technician loading fuel rods into the core of one such unit and you can see that the core isn’t much bigger than the canister on an Electrolux.

    I don’t think it would be cost effective for running a village. I think a solar thermal system running a Stirling Generator would be better. If you want, you can hook it up to a flywheel system to provide power at night.


  13. It seems strange to go with a small scale concept example (on just fixed solar panels.

    For a 5000W system (24 kWh per day), I would probably use fewer panels, smaller battery storage and install a small wind generator as well. This increases the chance of something feeding into the system more often. Of course there are constraints on sunshine and wind availability with some places being better than others. Where our farm is at Mount Pleasant we have heaps of wind and/or lots of sunshine most of the time so at this location when we add to the 25 year old solar panels that are still producing energy, I will be doing this with a small wind turbine rather than more panels.

    I love the panels because there is very little maintenance required but the new small new small wind turbines also seem very efficient and light to manage (We used to have a 32V Dunlite mechanism feeding batteries and a modern 32V inverter but I got sick of keeping the moving parts going).

    I also get annoyed at having to change batteries rather than being able to have a standard casing that an be re-conditioned with new plates and reprocessed electrolyte.

    I think that maintenance and ease of maintenance is a critical issue for small village scale systems.

    Tim K


  14. I’m also not so sure about the cost comparisons. Much of the labour would be done locally by people paid at local rates.

    PV might be best for internal lighting, community solar cookers for cooking and heating water, an anaerobic digester for producing the methane needed to run power after dark, genuine windmills for pumping water, including for a micro-pumped storage unit.

    Much if not all of this could be maintained by local labour.


  15. Peter Lang,
    I now have my hard drive up and working again. I did see your post about Tantangara/Blowering. It would seem to me that having one tunnel(2.8GW) would be a less expensive way of using the 500GWh as long term(5-10day) storage, and adding more turbines to reservoirs requiring shorter tunnels or pipelines for short term pumped storage.

    Your statement that wind power cannot be used for pumped hydro seems to be wrong as the present 18 wind farms connected to the NEM grid often produce 1000MW during off-peak periods when existing pumped hydro is in operation. There is a good surplus of spinning reserve to accommodate 1000MW, and at least the 5 min output doesn’t change very much over 30mins-1h. It looks like about 100MW spinning reserve would accommodate 1000MW of wind power. Since we have >6GW of hydro turbines in addition to the 1GW used in pumped storage we should be able to handle the minute by minute fluctuations of up to 60GW of wind capacity.

    Longreach solar output would be more relevant to most of Africa( within tropics) than Queanbeyan. The point I was trying to make is that very small amounts of power to keep a mobile charged or a refrigerator cold is much more valuable($/W) than large energy uses( cooking, hot water, air-conditioning) that would require grid power or very substantial battery back-up.


  16. When the argument revolves around on-grid and off-grid you are basically arguing between subsistence electrical energy and the availability of power to advance the standard of living.

    Prior to the Rural Electrification Programs of the 1930’s many farms had small electrical systems to power some lighting and perhaps a radio for a few hours a night. Often these were collections of batteries charged by homemade windmills. My father-in-law, as a teenager, built such a system, as did several others in his area, but nobody refused to connect when the lines came through. The advantages were just so overwhelming, that they all scrapped their turbines and batteries at the first opertunity.

    What pushing for solar/wind over grid is doing is essentially condemning vast areas to poverty, with no hope of climbing out.


  17. For the life of me I just can’t imagine village scale nuclear systems… and if it were cheaper to do a larger (but sitll small) generator and grid connect all these small places to it… then how come that isn’t being done now at least with coal or other fossil fuels…

    I struggle to imagine that these places are being given small poxy renewable systems that confine them to a future of poverty (not my opinion ust going with the flow of other comments) instead of simply whipping in a state of the art grid connected power system, giving them a 1st world power supply at a fraction of the cost of a system that only gives enough energy to boil a kettle when there are 3 or more sunny days in a row…


  18. Hi Neil, (#18)

    Thanks for the further comments. You said:

    It would seem to me that having one tunnel(2.8GW) would be a less expensive way of using the 500GWh as long term(5-10day) storage, and adding more turbines to reservoirs requiring shorter tunnels or pipelines for short term pumped storage.

    I think you would need to estimate the cost of the schemes you are proposing yourself to get a handle on their cost. The calculations I have done indicate the tunnels are 50% of the cost of the project. A colleague who is totally immersed in designing and building such schemes says I have under-estimated by a factor of two. He says there is no way the Tantangara-Blowering project could be economic. I do agree with him. And it is the best of the four options we considered. Take a look at the Wivenhoe project here and compare the tunnel length of that beautiful site with the 53 km tunnel for Tantangara-Blowering and for Eucumbene-Talbingo.

    I posted on another thread a list of the major components that would be needed for the Tumut 3 expansion. You might want to use this as a prompt for costing. I don’t think you will ever really be convinced until you have had a go at the costings yourself.

    You said:

    Your statement that wind power cannot be used for pumped hydro seems to be wrong as the present 18 wind farms connected to the NEM grid often produce 1000MW during off-peak periods when existing pumped hydro is in operation.

    That is a misunderstanding.

    Wind and hydro (not pumped hydro) work beautifully together. That is the case in Scandinavia and Brazil. When western Denmark’s wind power generates electricity it saves water from having to be used to generate power in Norway and Seweden. Similarly in Brazil. It would also work like this in Canada.

    But those countries have plentiful water inflows for their hydro generation. In those countries, hydro generates beseload power. Any wind power generation saves water to produce power when the wind isn’t blowing. But this is not pumped hydro.

    We do not have hugh water inflows like those countries. The Snowy had a capacity factor of just 14% in 2007. We need all the water we can store just to balance the grid and for emergency use (when a power station goes down).

    As I explained before, pumped hydro needs steady, cheap power. The power needs to be bought at about 25% of the price it will be sold for. This can be achieved when the power is available from cheap, reliable, constant baseload generators such as coal or nuclear plants. It would not be viable to pump with high cost electricity. Wind farms need to earn $90 to $140/MWh to be viable. That is high cost electricity. The owner has to earn that income for the wind power station to be viable. So wind and pumped storage would not be economic in Australia.

    Wivenhoe, near Brisbane, has two pump and generators. The pumps pump for about 7 hours, from around 11 pm to 6 am. Once started they run for 6 to 7 hours, pumping continuously and drawing steady current, other than minor changes to control the frequency on the grid. Wind power is totally unsuitable for providing the steady power required by the pumps. Look at the aerial photo here to see what a beautifully economic site this is.


  19. You all do know that there are “village size” nuclear power plants that have been designed and built, do you not? One that I am very familiar with is AECL’s SLOWPOKE series.

    SLOWPOKE is an acronym for Safe Low-Power Kritical Experiment (the “K” is justified — it is the traditional symbol for “criticality” in the field of reactor physics). SLOWPOKE is a low-power, compact-core reactor technology developed by AECL in the late 1960’s. There are five SLOWPOKE-2 reactors operating in Canadian universities, rated at 20 kW and supporting nuclear education and/or neutron research. The SLOWPOKE-2 core is only about 22 cm diameter by 22 cm high, and sits in a pool of regular (“light”) water, 2.5 m diameter by 6 m deep, which provides cooling via natural convection. In addition to passive cooling, the reactor has a high degree of “inherent safety”; that is, it can regulate itself through passive, natural means, such as the chain reaction slowing down if the water heats up or forms bubbles. These characteristics are so dominant, in fact, that the SLOWPOKE-2 reactor is licensed to operate unattended overnight (but monitored remotely).

    In the early 1980’s AECL developed a higher-power variant of this technology, called SLOWPOKE-3, which could act as a district heating source for remote communities, and can be hooked to a Rankine-cycle engine to crank out 600 kW of electricity.

    Naturaly these were designed for the Canadian North but they were also designed to serve small communities with a passively safe reactor that could be safely operated with a minimum of training. Of course fuel changes and other more complex procedures would be done by visiting experts, but nevertheless these reactor were expected to power a village over the Winter without outside intervention.

    Obviously district heating is not needed in Africa and other tropical regions, but this type of reactor could have provided good, clean and reliable power had the program continued. As it was the cost of oil at that time was such that SLOWPOKE’s were not considered cost effective at the time.


  20. There are several references to the B&W $5/W. Here is one reference
    The new NRG 2700 MW plant at the South Texas Project site has had its estimate creep upward until now it is also about $5/W. I’m sure if we had a few good installation experiences the cost would be lower.

    I wasn’t envisioning a grid connected nuclear plant either. However, there would have to be a local distribution system to collect together enough loads in an area to make a small nuclear plant feasible. Who would run a small nuclear plant? How would it be made safe from attack from terrorists. Who will run the small nuclear plant? The education level needed seems to be at odds with what the local population is capable of. This applies to a village in Africa as well as my neighborhood wanting to build and run a small nuclear plant here in the US. The same impediments exist here for small nuclear as they do in Africa.

    With regard to installing solar without storage, forget it. Even solar on a sailboat has batteries for storage. A stand alone solar system without some storage is useless. I was simply proposing a minimal battery system capable of supplying power to a school. You will have the students using many PCs in their classes (this is a way to skip the need for books) and batteries are necessary to store enough energy so the school can function without interruptions during classes and at night when they have time to do their homework. The PCs I am thinking about are of the $100 variety – see This PC program has been highly successful but it needs power to run it.

    With regard to the comment about solar cells costing $2/W, which is true. However when you add all the supporting hardware and electronics and labor to install the panels, the $8/W is a more realistic estimate. If you are installing a large system, such as a centralized tracking array, you can get the total cost down to about $6/W. This observation about the difference between solar panel cost and the total solar system cost is interesting because it shows that even if solar panels went to 0 $/W there are still other costs that make the overall system rather expensive. Think of the solar cells as being the engine in a car and the solar panels as being the engine and transmission. There is still the cost of the wheels and body of the car, the insurance, and other costs involved in owning a car. These other costs must also be reduced to make solar more economical. Unfortunately the other costs have been creeping upward cancelling out the cost reductions of the solar panels themselves.

    The 5 kW system I chose was just an example. Each installation would need to be sized for the load to be served. The comments about not having enough energy on the “worst” days is a good point. If the solar system ran a school, and there were too many cloudy days, the student would just have to not do school those days. Thus this points out a fundamental problem with relying on solar power and batteries. Its not a reliable source of power. Concerning the Sterling Engine, fine, go for it, but the economic calculations are about the same and the reliability problem is the same. For areas with wind, they are fortunate, and the wind economics calculations are similar to solar, except wind power is usually more economical than solar – if the wind blows enough. I would think that pumped hydro would be a nice storage system for anyone who can install such a system. Its definitely better than a battery storage system. But how many of us can install pumped hydro? Very few.

    Thanks for the reference on the small nuclear generator. We need more experience with small nuclear facilities. Can I get one to power my electric car?


  21. As I said in my previous post my maths is notoriously bad so checking over it this morning I realised I made the mistake of doubling operational costs for the nuc plant so that comes down to $160 a year.

    I also realised the cost of the bloody-grid-connection (!!) could come down if instead of giving it a 20 year payback it was given a 40 year payback by assuming a grid only needs renewing after 40 years.

    So those changes give our nuc costs as:

    Nuc $5000

    Ops $120

    Batteries $9000

    or $14120 which is 0.08718 X 14120 = $1230.98 per year

    Treating the grid connection separately over 40 years,

    A = (40000)(.06)(1.06)^40 / [1.06^40 – 1] = $2658.46 per year

    total is $3889.44 so putting this per kWh gives us 44.4 c a kWh for the nuc which opens up the divide betwen PV and nuc (bloody grid connection!). However where PV will be breathing down the neck of the nuc is the decreasing costs in PV panels and its too early in the morning here for me to back that up with some linked numbers.

    BTW I didn’t understand how people here arrived at their ‘capacity’ factors for PV. It would be useful if someone could explain how those were arrived at.


  22. There is a reason that small villages in Africa don’t have electricity. In relative terms (ie relative to income) both the solar option and the nuclear option are expensive. If the villagers want electricity then the best option is to move to a town that is grid connected.

    Speaking of electricity in Africa the following youtube discuss an important paradox about electicity in parts of Africa. In this case rules and price are the issue. But in this case the price isn’t too high, it’s too low.


  23. Eugene,

    your comment

    “I was simply proposing a minimal battery system capable of supplying power to a school. You will have the students using many PCs in their classes (this is a way to skip the need for books) and batteries are necessary to store enough energy so the school can function without interruptions during classes and at night when they have time to do their homework.”

    I think demonstrates where we all perhaps bring our own outlooks and presuppositions into discussions like development. I would question why use PCs over books? Books are still the best random access device ever invented and are self powering where ever there is light to read by. I think in many situations they remain superior to PCs (I know, I know, I’m an engineer and this sort of talk is supposedly heresy for an engineer). I’m not saying give away the PCs but use them for what they are perhaps best for rather than just a blanket application.


  24. Thanks Peter,

    I see where you are coming from and I am trying to relate your analysis on the
    Queanbeyan Solar Farm to peak sun hours and I think the difference is that your capacity factor arises from a historical operational analysis of a site while peak sun hours takes in insolation available at a site. I would didn’t look through all of your report but did it investigate reasons as to why the O/P levels of the solar farm were what they were? I ask that because there could be other engineering factors particular to that installation.

    Just to reiterate my position I am a fan of both nucs and renewables. I think they arguments over which is best si a zero sum game, what we need to be asking is what we want to do with energy.


  25. Jeremy, capacity factor solar PV in Massachusetts (USA) was 12 to 15%, Arizona 19%, average for all US was 14%.

    Click to access RERL_Fact_Sheet_2a_Capacity_Factor.pdf

    It would be higher in desert regions and lower in temperate and wet or wet-dry tropics where cloud cover in certain months can be a persistent issue. I think ~15% would be a reasonable approximation for Africa.


  26. Thanks Barry, I’m not convinced that ‘capacity factor’ is the best way to understand how a system such as a solar is designed to optimise against factors such as annual insolation. As I said in my previous post it doesn’t take into account non solar factors in the design and operation of Peter Langs example fo the Queanbeyan Solar Farm. I think capacity is something that has come about with coal and gas fired plants and may not suit a smaller system where you may be taking end use into account. Thats not a good explanation on my part but its not an excuse.

    Just thinking about it further. I think you could use both PSH and capacity as part of the analysis for a PV plant, but I am not convinced that capacity gives the overall picture.


  27. Interesting stuff. If we’re talking specifically about the developing world, though, the analysis somewhat misses the point.

    I haven’t had the chance to fact check the figures but the basic result is not very surprising for those of us who accept the low power yield of techosolar vs high energy density of fissiles, and can envisage small scale, relatively low-cost nuclear technologies maturing in the not-too distant future.

    But the comparison doesn’t reflect the reality of what’s happening on the ground.

    Rather than several fully wired-up houses and PC-equipped schools, the electricity requirement in an African village typically starts with a few lightbulbs, somewhere to charge mobile phones, a radio or TV set (usually only one/village), maybe a water pump, maybe a vaccine refrigerator for the clinic. This is a long, long way from a village-wide grid with assumed 24h supply, peak-demand management etc, anything close to what we take for granted in the developed world. If there is a clinic the fridge becomes the only real 24h essential for cold chain purposes. Expensive or otherwise, renewables are being used in this way now- the following examples are those I’ve personally witnessed without any apparent NGO/development aid sponsorship (there are usually obvious signs and stickers on those projects!):

    *micro-hydro in north of Vietnam for village lighting
    *PV setups in Mongolia providing satellite TV for nomadic tribesmen
    *PV/battery setup for clinic fridge in Peruvian Amazon
    *wind and/or PV setups for water pumps- many localities

    Gene’s proposed nuclear setup is basically a western-style grid in miniature. Perhaps it’s something for these communities to aspire to in the future. More likely, this sort of plant would be used first in remote regions of developed countries, or by armed forces or western aid agencies in temporary camps/field hospitals etc, and in time for rural regions in rapidly developing middle-income countries. This assumes all the technological and political/ideological hurdles are overcome. In the meantime, I think renewables will be used out of necessity in the developing world for a quite a while yet.

    Gene’s last point is the most telling- without development these communities’ electricity requirements will remain low, and without income they will not be able to afford anything but the most basic low-yield power supply anyway, certainly not nukes of any size. I’d welcome the development of small scale nukes to be ready when (hopefully) these economies start to develop, so that there’s a mature low-CO2 emitting technology in place for them to use, but the current reality is a choice between fossil fuels and micro-scale renewables.


  28. How will solar energy in Africa support 24/7 365 usage? I don’t see it.

    Several things are going on there. One, there is a huge hydro development scheme going on in Congo…the single largest hydro development in the world, larger that 3 Gorges dam.

    Secondly, several African nations are *beginning* to explore their nuclear options: Egypt, Algeria, Nigeria and Uganda. These are all big, larger LWRs.

    Thirdly, S. Africa’s PBMR was saved by the intervention of Dr. Chu’s US Dept. of Energy that wants to partner with the S. Africans over deployment of their PBMR.

    Africa will only be saved from further resource depletion, wars and civil wars, by development. The more semi-developed areas of West Africa would be ideal for a regional grid plan around a few larger LWRs. Building out from the Congo River hydro project for local grids would also be appropriate. There are no plans to do this now which is a major problem.



  29. I think we should all be careful pronouncing on what is ‘good’ for Africa. Years ago, as a young, bright and bushy tailed engineer, I won a scholarship to a course held in the US on space systems and management – very high tech as you can imagine. There were people from ll over the world on this course including three guys from Kenya, two engineers and the third a young physics graduate who taught in the boondocks. One day we were given a special guest lecture by Peter Glaser the inventor of the solar power satellite concept. He talked about the SPS (very exciting stuff!) and during it mentioned that Africa must industrialise. In the Q&A session I put my hand up and asked why Africa should industrialise. It was an innocent question but Glaser roundly slapped me down declaring,” Africa must industrialise”! I remember it because I was very embarrassed at being slapped down so roundly and publicly. At the end as I was slinking out trying not to be noticed the young Kenyan physicist caught up with and told me I was right to ask such a question saying, gently but emphatically, “why should Western countries expect us in Africa to do the same as them, it may not be good for us”.

    Thats always stayed with me and not just in regard to development but I always ask why we are doing things. Itbwas the best piece of engineering advice I’ve ever received.


  30. Just some perspective on Nigeria, which as David Walters points out is working on opening their grid to nuclear power plants. The country is 25% larger than Texas with 150+ million inhabitants (compared to Texas’ 25 million). Lagos is home to nearly 20 million.

    The church I attend paid for the construction of a health clinic in Jos, Nigeria (Faith Alive Clinic) and the medical doctor/director recently spoke to our congregation. He sleeps only 4 hours per day and sees thousands of patients each month. Humbling compared to what we do here. Would love to place a NPP in that city to assure the clinic (and all the inhabitants) have reliable electricity without the need for back-up generators.


  31. Barry #30, you said:

    It would be higher in desert regions and lower in temperate and wet or wet-dry tropics where cloud cover in certain months can be a persistent issue. I think ~15% would be a reasonable approximation for Africa.

    This statement reinforces the belief, for those who do not understand, that the average capacity factor is relevant for this analysis. The annual average capacity factor is only relevant if we have 1 year of energy storage. It is the minimum capacity factor that must be used. That is, the minimum capacity factor over the period for which we have energy storage. If we have just two days of energy storage, as per Gene Preston’s analysis, then the capacity factor to use in the analyses is the minimum capacity factor that could occur over two days.

    Jeremy (#29),

    There were no problems with the output from the Queanbeyan solar farm over the two years of detailed records. The low capacity factors recorded in winter by the Queanbeyan Solar Farm are corroborated by other solar farms over similar periods in winter. For example, Mudgee: Refer to Table 1 and the minimum output for June. We find periods of low energy output, hence low capacity factor, on all PV records. You need continuous records at 5 minute intervals, or at least 30 minute intervals, over a period of at least two years.

    The important point to recognise is that the design must be based on the minimum capacity factor, not the annual average capacity factor.


  32. The question of variable capacity factor will be paramount in asking whether a proposed wind farm ‘offsets’ the emissions from the giant Wonthaggi desal plant
    Rough calculations suggest the plant will consume over 100 MW of energy and perhaps much more depending on what output and technology is planned. If the ‘offset’ wind farm has 33% CF that suggests 300 MW nameplate would cover it if the energy could be stored or the process is interruptible. But sometimes that wind farm will be becalmed while water flow is needed. Then places like Hazelwood power station will be spewing well over a 100 tonnes per second of CO2 to cover that demand.

    Another ruse I expect is that the ‘offset’ wind farm won’t be built in time so they will declare that an installation built years ago will carry that mantle. However when or if the five year ETS holiday for brown coal ends then the average Melbourne water and power bill could be breathtaking.


  33. John … by all means use the word spew in relation to Hazelwood, and coal plants generally. Personally, given that the matter is has passed the plant’s digestive tract and is now composed entirely of what the plant deems waste I see it as more like effluent, so “defecate” sounds more apt, anthropomorphically speaking. (There you go DV8 …)

    Mostly though, one wopuld expect that a desal plant’s operations would be interruptible, since its output — potable water — is capable of being stored and would require significant inventories. It being costly to white start plants, you pretty much have to keep the plant going during the night so the marginal cost is really only the difference between desal and the reserve. Since a large chunk of the demand for Hazelwood power is aluminium smelting, as long as that’s there, it’s the cost above that that is marginal.

    I’d happily close the aluminium smelters running of these old coal plants of course. They attract a significant cross subsidy as things stand and there are far more GHG-friendly places in the world than Victoria to do smelting — Iceland for example. Of course, if the aluminium plants want to buy hydro or some other low footprint energy, then fine, let them do that. The sunk costs might make it worthwhile switching.


  34. Fran #40,

    you said:

    Of course, if the aluminium plants want to buy hydro or some other low footprint energy, then fine, let them do that. The sunk costs might make it worthwhile switching.

    Where are we going to get the hydro in Australia? Snowy had a capacity factor of 14% in 2007 due to lack of water inflows. It is not much better in wet periods. We need all of it an more for balancing the grid, for emergency use, and to hep provide peak power. We simply do not have the water nor the topography suitable for hydro. The idea of capturing energy from town water supplies is simply not viable. There is nowhere near sufficient hydraulic head or storage. The potential is miniscule, the cost of such energy enormous.

    What other low footprint generation options are there? Certainly wind and solar are not low footprint options. They are relatively high footprint options in alsmost every respect when compared with nuclear energy.

    Several bloggers on these threads have made comments about “sunk costs”. What do you mean by your comment? Do you mean that the investors (eg our supperannuation funds) who have invested in the coal power stations should be told sorry, we’re not going to pay you a return on your investment. Fran, if you were approaching retirement age you might be concerned about the effect on your superannuation. Some bloggers here make statements about sunk costs in a way that suggests they do not understand investment and return. I hope you are not one of them.


  35. Where are we going to get the hydro in Australia?

    Ever heard of the Ord River?

    There were plans for a tidal plant off WA (near Derby) as well. And of course if they could get nuclear through, I’d support that.

    Conceivably though, they could run the plant from waste biomass — it’s not as if there is a shortage.

    Several bloggers on these threads have made comments about “sunk costs”. What do you mean by your comment? Do you mean that the investors (eg our supperannuation funds) who have invested in the coal power stations should be told sorry, we’re not going to pay you a return on your investment.

    What I mean is that if the subsidy aluminium smelting gets were eliminated — during the early years of this century about 250k per job apparently, the sunk cost losses of shifting aluminium smelting someplace else would probably discourage moving offshore. So eliminating the subsidy would be a good thing.

    Fran, if you were approaching retirement age you might be concerned about the effect on your superannuation.

    I’m 51, so I don’t know if I meet that standard, but I do hope that my super fund doesn’t have assets tied up in filthying up the planet, and if it does, then I’m in favour of that being remedied as soon as possible regardless of the cost to my super. If fixing the problem would entail measures that would cause the value of my super to drop to zero and stay there until first I then my younger son drew our respective last breaths, then so be it, as far as I’m concerned. There is no economy without a healthy biosphere and I want no part of harming my fellow humans, regardless of the “benefits” to me and my heirs.


  36. AFAIK processes requiring critical pressure (reverse osmosis) and temperature (aluminium smelting) are not easily interruptible. Of course producers could maintain a large battery bank (ie a UPS) for the required shutdown power but that would cost more than the 1-3c per kwh they feel they are entitled to.

    I don’t know about Ord hydro but Neil Howes’ idea of a trans-Nullarbor HVDC could make sense for another reason. That is because most of the population lives in SE Australia but in future most of the gas reserves and hence cheap peaking power will be in the NW. Perhaps add some hydro to that. That would create a national grid, not separate E and W grids. Then with that new cable tack on a NPP/flash desal plant at Ceduna to power (700 MW) and water (200 ML/d) Olympic Dam and surrounds, with surplus electricity going to the national grid. Somebody who knows costs should do the sums.


  37. Fran #42,

    I suspect most of your response is tongue in cheek, or you are a stirrer (probably both).

    The tidal schemes and the hydro potential of northern Australia have been investigated about as frequently as the petrol companies have been for collusion. They are completely uneconomic. I am sure you realise that.

    Regarding investing, we can’t just take assets from investors because we change our mind. If we want to stop coal fired generation, we have to pay out the investors at fair price. If we do not compensate fairly, their are consequences of sovereign risk. What will we do next? I suspect you do agree with that, even if you cannot say it on this blog site for ideological reasons?

    Talking about subsidies, and $250k per job, I agree with you. We should not subsidise anything. But that goes for the car industry, the textiles industry and especially for the renewable energy industries too!!!!

    If we send aluminium smelting off shore, it will be smelted by dirtier electricity than we are producing. So the planet is worse off and Australia’s economy is worse off.

    Fran, if we waste money – slow the economy – on fantasies to suit the crazies we will be less able to take the most appropriate actions. I suspect you really do understand that too.

    What we need to work on is convincing the majority of the population that, if they want to cut GHG emissions, there are two options: 1) cheap nuclear or 2) expensive nuclear.

    Let’s get on with educating the population and stop all the nonsense. The voting public is not going to buy into a high-cost renewable energy dream. If we do not get low-cost nuclear, then cutting GHG emissions is going to take a very long time.


  38. Barry,
    Looking at that picture reminds me that Australians are currently installing a solar panel, one 12v battery and one light in rural homes in East Timor and are charging the householders 10 USD each. They think that this is their value and trade them for another $10 article. They currently seem to all use candles. Not even lamp glasses which will magnify a candle considerably, enough to read by.
    This seems to indicate that good lights aren’t highly valued.


  39. RE: Fran #42, Peter #45

    Fran loves to suggest ‘solutions’ that are implausible or outrageous, and then gently brings us back to Earth with a reassurance of what is possible. I consider this to be subtle trolling, and believe Fran is actually quite disingenuous. I think someone else in another thread called Fran a troll. I increasingly agree.


  40. Peter Lang

    1. Aluminium smelting is probably not cost-competititve in Australia and if a prioper price on CO2 (ie about $100 per tonne) reveals that, so be it.
    2. see this link:

    What it shows that the most energy efficient place to produce Aluminium is Africa. Since nearly every serious aluminium exporting country uses less coal than we do to produce it and/or has lower per capita emission that we do, the fugitive emissions risk is zero. In Iceland, they have near zero emissions for stationary energy. If aluminium can survive without subsidies and with this price so be it

    3. I’m in favour of nuclear (see my posts at LP and elsewhere), but I’m not dog in the manger about it. If the choice is 30 years of coal, then nukes or 15 years of coal phase out in favour of gas/biomass and then nukes, then I take the latter.

    4. I favour reconfiguring our cities to radically reduce energy demand per person so that we can live with a sharply increased cost per unit of energy. We need to greatly increase densities to achieve this. When each of us demands a lot less dwelling space and less private transport then the fact that energy might cost three times what it does now won’t matter that much.

    If we can process water locally then we can also cut water supply costs too and maybe avoid those big desal plants.


  41. Mr Siadora

    If you think my claims are implausible, you should outline the substance of the claim. Merely calling me a disingenuous troll is … trolling … tu quoque

    Perhaps you hope to provoke a flame war … that too would be trolling.

    Make a pertinent claim on topic or hold your piece.


  42. Fran, well, I think your post #48 was a perfect example. Evidence is in your actions, and someone only needs to cast back over your many previous comments to see this illustrated many times over.

    As I said, you’re quite subtle, but your motive for delay and careful admonition is plain enough to me, and I’m sure many others reading this blog. I’m certainly not going to bother to take the time to engage in a battle of wits with you. No, I’ll simply point out to others what I am carefully observing (as mostly a lurker and only occasional commenter on brave new climate), and let others work it out for themselves (or not, as the case may be).


  43. Peter,

    Regarding your post .37

    Thankyou for the link to Muriel Watt et al’s paper, its an interesting paper.

    My question regarding thre performance of Queeneyan was whether it has been optimised for operation as you find in Muriel’s paper where she asks similar questions of technical design and performance for Mudgee and the Olympic village. I asked that question both as an engineer and because I have been in meetings with Country Energy on other matters when the Solar Farm has been raised and there have been mutterings about the system but being outside the scope of the meeting I wasn’t able to delve into the comments.

    Regarding your reference to Mudgee in Muriel’s paper you said Table 1 but I think you meant Figure 3 if we take your usage of capacity factor (if I have understood it) then there are a number of days where the O/P falls below 3 kWh so I would take the approach of what was driving the sizing of the system e.g. how much of the load is it meant to meet and then decide what storage might be appropriate.

    With your reference to Gene Preston using two days storage don’t forget he plucked those figures and the load out of the air to illustrate the point he was making. That’s fine but we must consider that and if you have read my critique of him you will see that I used those but arrived at a different result because I considered tha energy available.

    Yes, capacity factor is relevant but its one factor in design, amongst a number of relevant things, but you seem to be divorcing it from what someone is trying to trying to do, i.e. what is the load and its characteristics, what is the energy available and what sort of system will bring the two together. Its interesting that Muriel Watt et al’s paper doesn’t make use of it and also the analysis she and her colleagues made of the Olympic village PV system is very interesting in demonstrating how to regard the use of PV in a grid system.

    I think if capacity factor is taken on its own wrt to systems such as PV we can end up misleading ourselves.


  44. To Esteban Siadora

    I’m sorry but I think you are wrong about Fran Barlow. I have read her entries on this and other blogs and she ask bloody good and useful questions showing that she thinks through things.

    Tough if you don’t like her thinking!


  45. No, I don’t think I am wrong but you are quite right that I don’t like her thinking.

    On other blogs, she harps on about nuclear. Here, she harps on about other solutions and cautions about nuclear. Always, it is exercise caution, take what we can, or offer the listening crowd something they don’t really want, as a compromise. Like Lawrence said on another thread (I found it after much searching):

    “And I still think FB is a troll, and also TerjeP (say Troll). They talk a lot, and I don’t think they add anything; infact I think it’s like listening to Saruman – a bad choice… You talk too much, about stuff that has nothing to do with energy, and I think it’s a deliberate distraction.”

    TerjeP, I don’t put much credence in your objection, as like Lawrence, I also consider you in a similar light. I do not think this way of Jeremey C., I simply think he is mistaken about Fran.

    Look at her post #48 as one (of many) examples:

    Point 1 = She wants a carbon price, but then carefully reinforces the negative view among those who oppose a carbon tax that, hinting subtly it will destroy jobs that this will indeed occur.

    Point 2 = More on this: the jobs got Africa or Iceland, but all couched in terms of this being better for the planet. By now the anti-carbon price people are foaming.

    Point 3 = She’s in favour of nuclear (of course Fran), but will settle for gas because nuclear in 15 years isn’t really a decent prospect after all, is it? So gas isn’t really so bad, is it?

    Point 4 = An appeal to the power-down doomer crowd, once again hinting (carefully) at the negative aspects of energy efficiency to those who oppose a drop in their standard of living or take a libertarian viewpoint. This way, she enhances the bad vibes of a couple of crowds – the anti-carbon people and the anti-power down people. But of course she does it all in such a nice, earnest way that no one could surely disagree could they?

    Or, if they do disagree, there’s always a gradual shift away from coal to gas.

    Divide and conquer. She’s smart, I’ll give her that, and ultimately, her ways are totally destructive to positive and constructive thought. We’ll, I’ll have none of it, and I hope other people on this blog are not hoodwinked either.


  46. Jeremy C #52,

    I believe the questions you are asking are answered in the paper. Could you follow it through first. If there is a serious error in the paper, I’d be pleased to find it and fix it. So far there have been 445 comments, so it has been fairly well chewed-over so far.

    No, I meant Table 1. Look at the June, Min. That is the minimum daily output for the worst day in June. If you calculate the capacity factor you’ll find it is similar to the minimum CF recorded at Queanbeyan.

    How much difference will optimising the system make? It needs to be a factor of 20 to be cheaper than nuclear, and that is if we could have 30 days of centralised energy storage (which is totally impracticable). With on-site chemical storage, we’d need to reduce the cost of the solar system by a factor of 40. The point of the papers, and what came out in the discussions on that thread, is that there is no point in talking about improvements of 10%, 20%, 50% when we are looking for a factor of 20 or more. There is simply no point in wasting time exploring the solar option when faced with these sorts of differences in the cost.


  47. Peter,

    Thanks for the clarification re table 1 and Figure 3 in the Watt et al paper. However,Table 1, I think, reinforces what I said in my previous post.

    I didn’t say there are errors in your paper I said that overall capacity factor is one part of what needs to be considered. Nothing I have read of yours has convinced me otherwise.

    In regard to the post/replies on I haven’t read through them. Apologies, its just due to availability of time. This means I cannot comment, at this point in time, on the the costs numbers you put in your last post of 10, 20, 40 and so on.

    I will read it and see if you are corect in the use of and emphasis you place on capacity factor.

    But…. in regard to you bringing up hydro etc going back to Gene Preston’s posting on rural electrification in Africa it seems he negelected to factor in the cost of grid connection (I am happy to be proved wrong). This is an important cost and I attempted to include it in my posted critique. If I am half way correct it pushed up the cost of a centralised generating system to being comparable with solar based on Preston’s parameters/assumptions.


  48. If you want to know what will happen to small scale nuclear reactors in impoverished villages, Eli suggests you consider the “pipline tapping” that is endemic in the Niger delta, and the occasional murderous explosions. That is not a good idea. You are also competing high tech solar against high tech nuclear. More realistically the solar/wind sources in small villages will be low tech, serviced by the local jackleg mechanic, at best, simple wind generators with a few high tech parts, or some sort of mirror heating a fluid with thermal electric generation. The first few watts are the most valuable as was pointed out in the discussion.


  49. Eli,
    Africa’s own vision is not based on “jack-left” mechanics. It’s based on a modern high–-tech nuclear infrastructure, scaled way down for the current low load usage. This means developing, with a lot of outside help, a nuclear infrastructure that includes an understanding of where they want to go, educational and training resources, elevating math and other pre-engineering in high schools, developing security, etc.

    No one is going to be *tapping* into a nuclear plant, only the distribution network if there is a grid.

    The idea, again, is to start small, and slowly develop small grids, then in country regional grids, then national grids, then a continental grid.

    It’s called development.



  50. Barlow appears to have missed that Al production consumes huge amounts of carbon for electrodes that get converted to CO2. In the US most electricity for Al production is hydro, but it still is a large CO2 producer.


  51. David,

    Do you have a reference for the nuclear vision of development in Africa? I think you might find there are a number of visions for development in Africa. Eli’s comment on ‘jack-left’ is about what is appropriate for communities, something that takes account of local economies and local desires and situations.

    There are many, many, many examples across the 20th century of inappropriate technological ‘solutions’ being forced on various groups. Haven’t us so-called high-tech types learnt that lesson yet?

    (Eli, you bring up people stealing stuff in rural situations. I am more worried about that here, where I’m currently living, south east London).


  52. @ Eli, 61:

    Yes, there is some direct GHG production from the aluminium smelter itself. But it’s extremely small compared to the GHG output from the generation of the electricity required, if the conventional coal-based electricity generation “mix” is used. Pretty much every metallurgical and chemical process has some small amount of GHG emission… they’re not the significant sources of the majority of the emissions, and they’re much harder to replace, so focusing on the energy systems needs to take a higher priority.

    It’s really unavoidable in the developed world… so much of our modern engineering (eg. aircraft) depends on aluminium. Before the electrolytic process of aluminium extraction was discovered, aluminium was a metal more precious and exotic than gold… and nobody has ever found a better process.


  53. @63 Jeremy,
    I garner this from a lot of sources including national plans that are published. The WNA for example had general nuclear infrastructure plans for most countries in the world. Many countries file overall development plans with the World Bank/IMF and UN that I’ve look at years ago.

    I’m not trying foster any technology on to unwilling recipients. It is generally the policy of *every* underdeveloped country to become *at least* a developing country. As such, there are a lot of plans out there. But all have a common ‘something’: the need for cheap, abundant and reliable energy.

    Starting with that one makes *proposals*. Some reject this altogether; many idolize the “nobility of poverty” and low energy usage and consumption. Like most in Africa who if given half a chance, would like *more* energy, not less. Thus my POV is in line with the majority, I believe, of Africans and other countries in Asia and Latin America who want to increase, *vastly* their energy production and per capita energy consumption.

    No country I’m aware of, except S. Africa which has thought this through, along with the Indians, has actually grabbed onto the small-reactor, small-grid concept…yet. I propose this simply as way to implement, extremely incrementally, and cheaply, the adoption of nuclear energy and, as a seed-kernal for developing national grids with distributive atomic energy.

    As some of you know, I’m not, generally, a ‘small reactor’ proponent. I like the big ones. But that’s because my focus has been on countries like the US and China that can use “big ones”. Africa has a whole different type of menu of problems that I think the sub-400MW market is preferable for.



  54. The idea of “ending” all CO2 emissions is really quite silly, and irrelevant. We need to focus on the *spewing* CO2 industries and practices, not focus on every wood burning stove in a house or an industry that outputs incremental CO2. There are thousands of industries that do this: wood pulping, fertilizer plants, chemical refineries, concrete manufacturing, aluminum, iron ore, copper smelters, etc ad nausium. So what? They are NOT the issue. Together they don’t add up to a single digit % of what the coal industry alone put outs. Focus folks, focus.



  55. Peter,

    Regarding capacity factor, in your paper, Solar Realities’ you define capacity factor as “Capacity Factor is the actual energy produced over a period divided by the total energy that would have been produced if the generator had run at full power throughout the period.” What period or periods did you use for capacity factor on the Queanbeyan Solar Farm? Was it 24 hour days or daylight hours and does it factor in aspects such as tracking vs non-tracking? As well did you consider aspects such as the efficiency of the inverters (under different conditions) used on the installation?


  56. Jeremy C (#57)

    Factor of 20 to 40!!

    Transmission costs here:

    Alternative to transmission is on site energy storage. Refer to post #12 above and to

    Note, I agree there is a role for remote, off-grid power supplies. But it is very high cost, and the poor coutries cannot afford it. A cheaper and better option in most cases is to electrify the larger population centres and facilitiate people moving to these for improved health, education, work opportunities and personal fulfilment. All options will take time. The least cost alternative overall is the one that will win out. Pushing high cost alternatives is bad policy.


  57. Jeremy C #67,

    Jeremy, all this is covered in the papers and has been discussed in previous posts on the respective threads. I don’t want to go over all that again, and especially not if you haven’t read the four papers and the references cited in the papers.


  58. Mr Siadora

    Finally some substantive claims …

    On other blogs, she harps on about nuclear. Here, she harps on about other solutions and cautions about nuclear.

    That’s simply wrong. I don’t “caution” about nuclear anywhere. In every place where I post, I alwaysadvocate resort to nuclear power, and have been the subject of abuse for doing so.

    Always, it is exercise caution, take what we can, or offer the listening crowd something they don’t really want, as a compromise.

    Actually, it is nearly the opposite. Where people reject nuclear power but assert an interest in lowering GHG emissions, I merely point out the implications of trying a non-nuclear course with suggestions that would work, albeit at a greater cost than with nuclear in the mix. Here in Australia, it is politically improbable that nuclear power will be used in the next 15 years, and so although I continue to advocate resort to it I don’t wish to do the very thing you suggest — force people to choose between what for them are unpalatable choices, coal or nuclear. If in Australia, you make these the only choices, coal will win every argument. That is precisely why the advocates of coal demand that if the greens are serious about GHGs, they support nuclear power. It’s an effective wedge which leads to massive subsidy of “clean coal”.

    Coal is not clean and never will be at a cost that would make it viable.

    TerjeP, I don’t put much credence in your objection, as like Lawrence, I also consider you in a similar light.

    Now you are calling someone else a troll. I don’t agree with Terje. He’s a supporter of American “libertarianism” and I am a leftist. That doesn’t make him a troll though, IMO.

    Point 1 = She wants a carbon price, but then carefully reinforces the negative view among those who oppose a carbon tax that, hinting subtly it will destroy jobs that this will indeed occur.

    There’s no question that the aim of carbon dioxide pricing is to reconfigure industry and human activity away from reliance on dumping industrial effluent into the biosphere. This will destroy some jobs, demand others and improve the life chances of almost everyone under 50. There’s also no question that it will tend to preserve those fossil resources that would be hard to replace at the moment, thus preserving the jobs of those in those industries. Neither coal nor oil will last forever and at current projected usage will not be available at close to the currenbt real cost by 2100. We need coal for steel and oil for polymers and yet we are burning these for fuel.

    Point 2 = More on this: the jobs got Africa or Iceland, but all couched in terms of this being better for the planet. By now the anti-carbon price people are foaming.

    The anti-carbon pricing people are always foaming. Peter’s point was about fugitive emissions, since a ton of CO2 released anywhere does the same damage to ecosystem services as anywhere else. My point was that in almost all places, the CO2 footprint of aluminium production wouyld be less than it is here, and that since the jobs here are being subsidised at 250k each, even if production went to Africa, we would be better off. At that price, you could pay people to lie on the beach — and frankly, at 250k, what laid off worker wouldn’t like that?

    Point 3 = She’s in favour of nuclear (of course Fran), but will settle for gas because nuclear in 15 years isn’t really a decent prospect after all, is it? So gas isn’t really so bad, is it?

    Be honest. What I’m really doing is accepting the lesser evil. In most countries of the world, nuclear power is a part of the mix. Australia is a sideshow in this respect. While I’d prefer we accepted nuclear, I still prefer gas to coal.

    When you walk into a restaurant, you choose from what is on the menu or at any rate what is on offer. Perhaps you’d prefer something else, but if it is not available, you accept the next best thing. Perhaps you think the prices are too high, but at that point you either walk out or agree to pay. That’s how life works.

    Point 4 = An appeal to the power-down doomer crowd, once again hinting (carefully) at the negative aspects of energy efficiency to those who oppose a drop in their standard of living or take a libertarian viewpoint. This way, she enhances the bad vibes of a couple of crowds – the anti-carbon people and the anti-power down people. But of course she does it all in such a nice, earnest way that no one could surely disagree could they?

    This is simply nonsense. It is clear that the current configuration of cities is unsustainable and doesn’t serve people’s needs, quite apart from the CO2 footprint. Sitting for hours each week in cars driving to work paying too much for housing and working overtime to do it are not good things. Changing the configuration makese sense not only in energy efficiency terms, but also in human terms. We get better quality of life. Given that in Australia, we are not going to get nuclear power anytime soon, it also means cleaner skies and cleaner water and less demand for desal, massive ocean outfalls etc.

    Divide and conquer. She’s smart, I’ll give her that, and ultimately, her ways are totally destructive to positive and constructive thought. We’ll, I’ll have none of it, and I hope other people on this blog are not hoodwinked either.

    It’s nothing to do with divide and conquer — at least not on my side of the argument. I do want to divide the other side, i.e. the pro-business-as-usual side by pointing out that their slanders of mainstream science also harm humanity in other ways and that coal and crude oil is not essential at the current scale to producing useable energy.

    You are attributing to me attitudes that have nothing to do with my advocacy. There are far better grounds for me calling you a troll than the reverse.


  59. Moving to cities has been going on for some time in Africa and it does not (usually) work out. Maybe better let each African country and region work out their own solutions, hmmm?

    I disagree David that you can generalise in this way about urban life in Africa. While I agree that Africans (and indeed any viable community) ought to be allowed scope to ‘work out their own solutions’ it’s unclear in this context what that would mean.

    Certainly it is clear that if Africans, amongst others, are to have access to the kinds of benefits we westerners take as the starting point for dignified existence, then urban existence will be the scaffold for this leap forward. It’s far easier to bring people to existing infrastructure and the associated resources than the other way around.


  60. Kind of interesting that so many people presume to speak for Africa, which I should remind you all is a continent, not a country, and very diverse in technology and culture.

    Anyone arguing for a particular solution is liable to make the alternatives seem really expensive. Here is a different calculation for solar that comes out much cheaper. No doubt you could questions some of the assumptions; your batteries for example are 4 times the price. But the assumptions for nuclear here are very favourable. Nothing is allowed for building a grid over potentially inhospitable terrain (I presume we are talking about regions with no existing infrastructure, otherwise we would not need batteries).

    Solar PV offers the potential to build up access to electricity on a small scale without a grid, and gradually extend access until a grid is built up. In parts of Africa that lack infrastructure, a good example to compare is the roll-out of cell phones, which also do not need a wired grid to get started. Anyone working the numbers like call charges based on developed world costing would think cell phones are the technology of the rich, and we ought to be pushing wired phones to third world countries.

    Another factor neglected in this calculation is that there are extremely strong incentives to reduce the price of batteries, a big factor in the pricing here.


  61. Now you are calling someone else a troll. I don’t agree with Terje. He’s a supporter of American “libertarianism” and I am a leftist. That doesn’t make him a troll though, IMO.

    American libertarianism is a nice idea but I’m mostly interested in Australian libertarianism.


  62. American libertarianism is a nice idea but I’m mostly interested in Australian libertarianism.

    Until about 1988, based on all the Australian libertarians I’d met and heard of, I thought all libertarians were lifestyle anarchists. I was actually quite shocked when someone acquainted me with proof of the possibility of being a pro-capitalist libertarian and pointed out that John Singleton had called himself the oxymoronic anarchocapitalist.

    Then again, I attributed this to Singleton importing Americans’
    eccentric taxonomies. They regularly call Democrats “socialists”.

    Australian libertarian? You’re probably better off saying “Randian” or “Hayekian” or “Norquistian” …


  63. Fran, et al, I never mentioned urbanization. Urbanization in Africa, Asia and Latin America is a function of the decay of the free enterprise system in under developed countries as nations have adapted Structural Adjustment Programs that smashed subsidies to farmers, health care etc. The whole socialist revolution movement in Latin America is a *direct* response to neo-liberal, usually US supported, SAPs.

    I’m not talking about urbanization. It exists now and I’m not, for one, passing judgment on it good or bad, it exists, period, as part of the result I describe above. I’m saying is that countries want to develop. Does anyone here dispute this?

    Solar does not make it. Is it “better than nothing”? Yes, of course. Is it a solution? No, it’s charity, at best. I’ve seen the systems developed for Africa, specifically, and they are woefully inadequate. At best the provide some form of power for low wattage, single bulbs with a small TV for a few hours a day. The batteries become huge waste issues and often do not stand up to tropical heat.

    I went to a fair last year in California for these systems. Nice shinny systems. No one was adequate to run a home with a 3 cu ft refrigerator. No one had more than a days storage of power. In the tropic, cloud cover can last for weeks when it rains.

    Much of Africa, even in urban areas, relies on charcoal, propane, wood etc. The first latter stand as ‘renewables’ I might add. Such ‘renewables’ are destroying what’s left of Africa’s tropic. Those in government and at the universities, almost totally, want to end reliance of on the trees-into-charcoal, savannah-into-cow dung forms of energy. They want 24/7 electricity.

    What most countries there do is this: they build diesel generators. Basically the same kind that run trains in the West. The do this in parts of the US as well, i might add. Like in Maui, parts of Alaska, etc. Lot’s diesel and very expensive, of course, but you *can* build a small grid around it. Most of Africa’s major cities run on this sort of power. It is, oddly, very expandable but it places the country in a dependent relationship with western diesel suppliers. You might see why this has disadvantages.

    Countries with hydro try to exploit that and of course, this is a good way to go if you consider the short term environmental damage it does but it’s reliable and, unlike Australia, it doesn’t appear to be ‘going away’, yet, under the impact of climate change. But countries are also interested in nuclear and that’s where the choices they make are very, very important. S. Africa and Esckom, it’s major utility, is trying to market the smaller PBMR for exactly this reason, to build grids, slowly, outwardly, and then attack the issue continentaly.

    If countries in Europe took the ‘me alone’ approach, David B. Denmark would have NO wind industry. It’s silly and ridiculous for countries in regions not work together to solve problems that are intrinsically international in solutions.

    What I hope they do is investigate the possible usage of the small IFR and LFTR regime.



  64. Get somebody who has been there to tell you about Nairobi turd missles!. Sorry, but many parts of cities in Africa have no infrastructure to speak of.

    David B., jokes aside that is absolutely correct. But do you think the residents want to live like that? Do you think the sprawling slums of Nirobi, Lagos and Soweto are somehow a question of “lifestyle” choices? This is a form of barbarism, plain and simple.

    What they need IS urban infrastructure: sewers, schools, grids, services, etc. This takes massive gov’t intervention and a *national* vision of pulling one’s country out of misery. I believe nuclear energy can play a vital role in doing this.


  65. Australian libertarian? You’re probably better off saying “Randian” or “Hayekian” or “Norquistian”

    None of those alternatives really fit in a way that is useful or even accurate. You could call me a social democrat who believes in a very small role for the government but that would simply confuse people. Labels suck but they are useful and libertarian is the best fit that I’ve been able to find. Classical Liberal would also be a pretty good fit.


  66. I’m not sure why/if you’re disagreeing with me David, though it sounds as if you’re trying to …

    Marx didn’t come to the phrase the idiocy of rural life by accident. While urban life is not a sufficient condition for large numbers of people to live adequately and have some prospects of dignified existence, it’s a starting point, especially if per capita resources are a little scarce. Sustainable rural life only works when highly arable land in a mild climate can be occupied by comparatively few people without contest from other humans, and then only if they are prepared to work very long hours and live modestly. In a world composed of 6.8 billion people almost all of whom would like to live no less well than Europeans, if not Americans that’s never going to work.

    I’m for people, including rural Africans of course, devising solutions that work for them, but I’m also cognizant of the fact that there’s no dignity in having to flog off your children to child traffickers (or having them taken at gunpoint) or living in ignorance and squalour. I don’t accept that respecting local wishes allows the treatment of women as breeders so the head of a clan can live OK when he’s 70 is OK. I don’t see that as any kind of respect for humanity. The genuinely uplifting story of the young Kamkwamba boy from Malawi with cleverly scavenging the materiel for his windturbines posted above, was inspiring precisely because he had done this despite the fact that like most of his village, his family was unable to pay the measly $80 for basic education. So ignorant were they that they thought his windturbines may have brought the drought. So uplifting, but sad. Here was a character the poet Gray might have been talking about in his famous Elegy Written in a Country Churchyard

    Full many a gem of purest ray serene
    The dark unfathom’d caves of ocean bear:
    Full many a flower is born to blush unseen,
    And waste its sweetness on the desert air.

    Some village-Hampden, that with dauntless breast
    The little tyrant of his fields withstood,
    Some mute inglorious Milton here may rest,
    Some Cromwell, guiltless of his country’s blood.

    The point is that if we want the most marginalised of the planet to have the kind of shot at dignity we have, ensuring that they mostly get to live in well-designed energy and resource-efficient cities with good services supported by suitably cornucopian economic hinterlands is a starting point.


  67. Re Fran Barlow
    After much thought I think I agree with Esteban Siadora – Fran is a VERY subtle troll!

    Her last remark @ 42 in which she proclaimed that she wouldn’t worry if she lost all her super to “sunk” or “stranded” assets, as long as we stopped spewing filth into the atmosphere, was very cleverly aimed at worrying those on the blog dependent on their super (like me)now, or worrying about their future balances. Just the people the Coalition and big business want to frighten! Who are you working for Fran?
    The only other explanation for her tactical remarks could be that she is one of those contrary people who always has to have the last word and loves to play the Devil’s Advocate. However, in her case, it seems more devil than advocate!


  68. Her last remark @ 42 in which she proclaimed that she wouldn’t worry if she lost all her super to “sunk” or “stranded” assets, as long as we stopped spewing filth into the atmosphere, was very cleverly aimed at worrying those on the blog dependent on their super (like me)now, or worrying about their future balances.

    Hardly. I was simply responding to Peter Lang’s challenge @41:

    Do you mean that the investors (eg our supperannuation funds) who have invested in the coal power stations should be told sorry, we’re not going to pay you a return on your investment. Fran, if you were approaching retirement age you might be concerned about the effect on your superannuation.

    I was declaring in the most graphic terms that I was not. It does not occur to me that this would happen because long before the assets were worthless, or even significantly diminished, restructuring would take place with the losses spread over a longer period or the super funds themselves would leave. But seriously Perps, would you be happy knowing that your super fund was helping destroy the life chances of your children and theirs? IMO, there’s no economy without the biosphere. I want to keep both healthy and the latter equitable, but if you ask me which I’m more ready to risk, it’s the latter. The latter we can rebuild, but the former we cannot and the collapse of that would hurt more people than a decline in the economy. Simple Benthamite logic.

    Who are you working for Fran?(sic)

    The NSW Department of Education and Training. For whom are you working?

    Pathetic, really.


  69. Here’s another reality check. This site costs a typical home system as needing 2kW solar panels, batteries with 20KWh capacity, inverter, charge protector and installation kit at US$22,000-25,000 plus installation. The same site costs connecting to the grid at $15,000 to $50,000 per mile. Even if you convert this to per km, it’s still quite an impost on the basic price. And that assumes you have a national grid to connect to; in a really remote location, you won’t have that.


  70. Sorry Fran – I didn’t realise we were sitting for an English language exam. If you know anything about linguistics you will appreciate that language is ever changing and that outdated expressions such as “For whom are you workin?” are an anachronism. Still – another example of your penchant for putting people (and their ideas) down. Perhaps, therefore, my second assessment of your input is correct and you are just a “know-it-all”, who always has to have the last word, and not a troll.
    If you have read any of my previous posts you would know that I am seriously concerned for the future of my grandchildren and actually I already have my super invested in ethical products. It puzzles me that you would mention,let alone “declare in the most graphic terms” the fact that assets may be “sunk” if not to cause angst for some people reading your posts. Otherwise what is the point of saying that?
    I suggest you concentrate on supporting the obvious winner(nuclear power) in the battle to save the planet as we know it. I am a retired librarian.


  71. I agree with Perps, Esteban and Lawrence, I much prefer people to be open and honest about their intentions and motivations. Now it may be that we have all misjudged you Fran, but the concern they express is certainly one that has also struck me recently. So, if you have been misread, I apologise — but at the very least you should take it as an indication that your commenting ‘style’ is easily misconstrued.


  72. Fran @ 85 said:
    “It does not occur to me that this would happen because long before the assets were worthless, or even significantly diminished,restructuring would take place with the losses spread over a longer period or the super funds themselves would leave.”

    EXACTLY! So, being aware of that fact, why did you not say so in response to Peter Lang’s question? Instead you chose to answer in “the most graphic terms” designed, I inferred, to frighten people.
    Typical troll behaviour!


  73. God you blokes ganging up on Fran are pathetic. What’s this troll nonsense? Can’t you see her style of reply is just like yours? So she bites! Get over it. She contributes to the general knowledge and makes others think. It looks a little to me as if she trod on some Coal or Aluminium industry toes, or maybe some cornucopian toes, and the best anyone can do is get personal and offensive.

    I can’t see any troll-like post here, not even the sycophantic #47.


  74. SG @90 said:
    “She contributes to the general knowledge and makes others think.”

    Fran(everyone’s opinion is valuable)Barlow preaches delay and spreads doubt. Fran is pro nuclear,with reservations. Fran now thinks we should wait for thorium reactors, nothing wrong with that really, they’re just further down the track. I feel sure Fran will gravitate to fusion before long. What happens whilst we ponder and doubt? Why,we keep on burning fossil fuel of course. Ask yourself, who benefits?


  75. As it now seems there’s a consensus including the blogsite host that I am trolling and at best contribute nothing of value here, I will withdraw.

    I do apologise if I offended anyone genuinely concerned about the issues on this site or “frightened the horses”.

    Best wishes on your project Barry. I do hope that one way or another, Australia and the world can rapidly decarbonise.

    Best …


  76. This is my last comment on this matter, as it’s not what this thread is about, but needs to be said.

    Fran, if you want to withdraw, that’s fine, there’s plenty more internet out there and my blog is but one tiny corner. But it is not my intention, in agreeing with those people, to drive you away. No commenter can disrupt the logical train of the blog, and dissension is the grist of interesting comment and I welcome that!

    No, my intention was to be constructive — to point out to you that I can understand where their feeling comes from, and hence you may wish to consider how your comments, in the way they are often framed, can be misconstrued. Look, sometimes Fran, it’s worth being ambitious rather than conciliatory, else you risk giving excuses to those who would take your compromises and use them to their own advantages (just look at the way the gas industry uses wind, and coal uses CCS). It may very well be an unintended consequence of your method, in which case I apologise. But you can see the perceptions that people like Siadora, Perps and Lawrence have taken away, right or wrong. The fact that at least 3 commenters have said this means there is a perception issue, like it or not. Perhaps it is because you are so well known as a regular poster on BNC that they are willing to say this, rather than just letting it pass through to the keeper.

    Anyway, up to you — I’m happy to have you around, but I stick to my view that there is a strange impression given in many of your comments that makes one confused as to intention, deliberate or accidental. Aged Husk’s last comment summarised my disquiet as well as any.

    My comments on this matter are henceforth closed.


  77. Peter re your post #71.

    I’m sorry but I think you misunderstood my post or didn’t perhaps read my previous posts. As I quoted from both your Solar Realities Paper and the Muriel Watt paper you linked I think that lets you know I took you seriously by reading them.

    If I haven’t understood what you wrote then thats a different matter (rather than not bothering to read your stuff) and then it would great if you could help out this dumb student.

    So may I ask again, what approach do you use for a time period for capacity factor. Have I missed how you do this in your papers/reports if so, my apologies? Is it a daily figure, monthly, yearly etc? If so how do you relate it to the amount of resource available. For example if you set it on a daily basis then if you use PSH then the capacity factor is 100%, if daylight hours then is PSH/no of hours, if a whole day then PSH/24 and so on and that would relate to. Or is different and, say, dependent on average montly insolation as set out in Table 1 of the Watt et al paper we have both already referred to?

    I was a little surpised that in referring to that table in the Watt paper you seemed to take little account of solar insolation data with your use of coarse BOM maps in your papers. As well as that table can I suggest a better source such as the NASA Surface meteorology and Solar Energy website at Or have I missed something?


  78. Jeremy C,

    I’ll make a short response here. However, if you have more questions, could you please ask on the appropriate thread so we can keep the dialogue together for the benefit of all readers who want to follow the discussion.

    The Capacity Factor is over the period defined. It can be 1, 3, 5, 10, 20, 30, 60, 90, 365 days. A day is 24 hours. The capacity factor is determined from the average output (kW) of the solar farm over the period divided by the capacity of the solar farm (55 kW). The average output is from the actual output measurements (recorded every 30 minutes for 2 years) not estimates calculated from insolation.

    I am not sure what you are referring to with regard to the BOM course insolation data. If more questions, please ask on the appropriate thread.


  79. For household-scale or small-village-scale power production, generators burning gasoline or petroleum-derived diesel fuel are what works now. At those scales neither nuclear reactors nor local collectors of renewable energies are ever likely to compete; not even if the operators of the diesel generators were to begin to be required to pay for the harm the diesel fuel does before, during, and after its combustion.

    What would compete is small internal combustion-engined generators that were fed a clean safe storable fuel centrally produced on a multi-gigawatt scale, perhaps with the aid of a solar dish tens of km in diameter supported by unspecified means in the thin air above North Africa, perhaps alternatively with the aid of a compact ~20-GW(t) collection of seaside nuclear reactors.

    Between now and when those big fuel plants go into service there will be a period when boron is too expensive, because its production facilities are too small-scale, to help the third world; during this time first-world drivers will carry the ball. By paying more for more, they will allow the production scale to increase and the costs to come down.

    — G.R.L. Cowan (‘How fire can be domesticated’)


  80. Peter,

    I disagree due to the subject of this thread, and that you did bring up capacity factor on here.

    But anyway, thankyou for setting out how you calculate capacity factor. From what you have posted and reading your papers I think your method is of limited use in assessing the usefulness of a PV installation. If you think it is worth it we can pursue it on the other threads but that may limit other people from contributing if they are interested.


  81. Marion Brook @ 94 said:
    “Damn it Fran, don’t be scared away by a handful of lurkers”
    I resent that comment – I doubt Marion is unaware that I have been commenting on this blog since its inception and Barry Brook can hardly be called a lurker on his own blog! Marion is entitled to her own opinion, just as I am, but calling us lurkers, when that is patently untrue, only serves to weaken her case. And that is my last comment on the subject!


  82. Well, I for one do not think Fran is trolling. I knew her posts over the *years* and I’ve seen here POV change in this time. Fran is a supporter of industrial level bio-fuels and she is a very articulate one at that. I know her questions and comments are contratrian but no obnoxiously so. We actually need MORE people like Fran in the Green and envriro movement, not less and she should be encouraged.

    @Fran, no, I wasn’t commenting or responding to your posts but to others on Africa.



  83. Which capacity factor you use (minimum 24 hour in a year or annual average or other) will in tandem with the storage system selected determine the system reliability. So one way to sort this out is to add reliability to the system specification. For instance can we tolerate outages 10% of the time, 1% of the time, 0.1% of the time etc. I’d suggest that except for things like refrigeration of critical products the power probably doesn’t need to be super reliable. And you can quarantine some of the energy storage specifically for refrigeration such that it gets a more reliable supply relative to other loads. And for PCs and the like you might be able to improvise some form of alarm that indicates when a power outage is imminent.

    Obviously this is still a simplification because there are other things that also effect reliability.


  84. Wiki states that the term ‘troll’ is often used as an ad hominem strategy to discredit an opposing position by attacking its proponent.

    This is exactly how a handful of needledicks here have gone about their nasty business of trying to discredit Fran who has been a strong advocate of Barry’s cause on other online forums. Way to go Barry Brook. Now that I have seen in full the type of wankers associated with 4th gen, I won’t be advocating it elsewhere either.


  85. David Walters (81) — Maybe Egypt and South Africa have the expertise to run nuclear power plants. Doubt that Malawi nor most of the rest of the 53 countries in Africa do.

    I approve of two village scale projects in India; both Jatropha oil based. One has a refinery (quite simple) to produce diesel. The other has a diesel engine modified to burn the Jatropha oil directly. Both projects produce a modest amount of electricity each evening, aobut 4 hours worth.

    The best part of both schemes is that all the villagers participate in growing a small number of Jatropha plants, taking the seed pods to the press and resturning with the pressed oilcake for other uses, such a cooking heat. In both cases, although organized slightly differently, there is an essentially co-operative effort. I think that important.

    And I wasn’t joking about the conditions in the Nairobi sums nor do I suppose that the people there much care for the quality of their lives. I was objecting to commenters here who seemed to think as the World Bank did, that they knew what was best for various African peoples. I’d rather, as much as possible, for the Africans to do what they need to do for themselves, with as little of the PL 480 and World Bank and IMF corrupted assistance as they possibly can.


  86. David @107, I don’t object to villiage scale projects. But remember what we talked about? We are talking about a country, er, countries that have, say, 20 million people living in ONE city. Not a village. The whole dislocation of African society as well as much of Asian and Latin American society is one toward rapid, and massive, urbanization. Growing fuel from plants isn’t going to cut it for a country of 200 million people, or 40 million or any number.

    What you advocate is also the NGO sponsored “telling people what to do”. This is how most solar ends up being in these small villages: western NGO’s coming with gifts and saying “here, try this”. I do not object per se, but it’s not a development model.

    A *small* country like Malawi, one of the smallest in Africa, is simply not going to develop beyond a village standard of living thinking it’s going to produce all the energy it’s ever going to need. Far from it. Regional *African* designed energy grid development is being discussed. From bringing power up from S. Africa to the large Congo hydro projects. Small plots of land are only, also, going to keep these people oppressed and repressed with no *hope* for a better life. Or it will drive them to emigrate to other African countries and join the millions of poor living in the slums around Johannesburg and other African cities.

    I’m all for making whatever it takes to make Africans control more of their own lives. I don’t see producing diesel as one of those but in the meantime, if it works for a few thousand, I don’t oppose it. I’m *suggesting* there are better ways and many Africans know this.

    BTW…despite the huge political problems in Sudan (Dafur, underdevelopment, civil war), if you look at what THEY are doing, actually proposing and staring to build, you will be amazed at the very high tech and progressive development of that region in terms of infrastructure. They are NOT looking a small village economies to solve their problems. I don’t believe the World Bank and other malevolent forces have anything to do with it, but I’m not 100% sure.



  87. Now that I have seen in full the type of wankers associated with 4th gen, I won’t be advocating it elsewhere either.

    The merits of 4th generation nuclear is not related to the sexual self gratification undertaken by proponents. You are attempting to discrediting an idea by attacking the proponents. That is an ad hominem which seems to be precisely what you were complaining about in the first instance. Strange behaviour indeed.


  88. Christ you all spend a lot of time dicking each other over. Who gives a damn what a commenters motives are, deal with what they say, not why they are saying it.

    I got an idea. Why don’t you all start sprinkling the thread with some content on the subject at hand?


  89. Salient Green # 106

    Gen 4 supporters come in all shapes and sizes. A little on line spat between a handful of people is hardly reason to abandon a technology that could well be instrumental in saving the entire population of the earth.


  90. #110 terje, you know damn well I meant ‘wanker’ in the sense of idiot, fool, self absorbed person and the ‘idea’ of 4th gen is being discredited by those wankers, not me fella.

    Marion# 112, you are right of course and I will not abandon my view that the technology could save lives as long as it is not used to maintain our present high consumption, high population and economic growth lifestyle because that would lead to a greater loss of lives further down the track.

    I will not be advocating 4th gen in other forums however as I believe that certain BAU elementals have attached themselves to this blog in the form of auschlecken and I don’t want to be associated with them.


  91. @Salient Green:

    You won’t be supporting gen IV nuclear technology? Of course you won’t! You’re an anti-human, anti-technology advocate of depopulation and energy starvation. Your pretence of not supporting a revolutionary, viable, economical, game-changing, carbon-free power production system on account of being offended by the style of posting of a few people here is as transparent as vacuum.


  92. I have read through the comments on this thread, and note that it has strayed off original topic. As many threads do on many sites.

    As I see it, there is no ONE source of energy that is the key to the future. All have their places. The disadvantages associated with each one are all being separately worked on and solved, at whatever rate for each. Thus,

    a. solar cost effectiveness is improving;
    b. battery storage (eg lithium ion for electric cars) is improving;
    c. it looks like thorium reactors and IFR offer a way out of the high level nuclear waste problem associated with most established reactors;
    d. there are other promising technologies, such as wave and tidal power.

    I do not see any ONE of these as being THE answer to energy problems worldwide, but assume that coal-fired generation has to be phased out.

    On the rural property I live on, the water supply is from a very reliable bore with a mains-powered submersible pump. A 22,000 volt line runs reasonably close to our other bore, which stopped feeding water into our farm system when its windmill became uneconomic to maintain (and later got blown to smithereens in what I can only describe as a tornado.) The last quote I obtained (about 20 years ago) said that 2 poles would be needed to get the power to the bore at A$10,000 per pole. I have neither had the need nor the courage to ask what it would cost today, and that is before installing any pump, wiring and electronics.

    A big increase in electricity cost we would have difficulty passing on to our customers. Most of our consumption is for pumping water for livestock. We would have to destock and perhaps eliminate some of the wild animals (mainly kangaroos) that drink at our troughs as well if electricity costs were to go through the roof. So a major interest of mine has been in non-mains-powered water pumping systems.

    In Barry’s original cost-comparison calculations of solar vs nuclear, I did not notice any reactor decommissioning costs as being taken into account. (This is one area where accounting tends to get hazy for the existing nuclear industry.)

    I do not think it wise to assume that promising but yet unproven teachnology will solve the problems of disposal of the (relatively) small masses of high-level waste or of the (relatively) large masses of low-level waste, and so both for the present should be factored in at present real cost.


  93. Ian, regarding decommissioning:

    Utilities currently play 0.1-0.2c/kWh for decomm.

    For a 1 GW reactor, this would result in a yearly deposit of $8-16 million (assuming 92% capacity factor) and an accumulated fund (at 5% interest) after 60 years of operation of $3-6 billion (more if the interest rate is higher).

    Current cost estimates for complete shutdown and decommissioning for light and heavy water reactors, are in the range of $200-600/kW (or up to $600 million for a 1 GW reactor). For gas-graphite reactors, the cost is higher, around $2,500/kW, but there are few of these outside of the UK. The EBR-II decomm (a sodium-cooled fast reactor) was a spectacular success.

    So, currently, the utilities in the US are covering their decomm costs at least 3-6 times over. It’s a(nother) non-issue that is hyped by antis (not saying you are, just generalising) and yet has no substance.


  94. Barry, I’ve read of these insanely high costs to decom the Magnox reactors. And also the British failure to finance the decom.I have some questions.

    Why ARE the costs so high?
    What is about these reactors that make them so expensive, is it the graphit?
    Will all graphite moderated reactors have such high decom costs?


  95. David, regarding Magnox decomm costs. Magnox structures were generally bulkier and more heavily irradiated than PWRs. Sellafield is apparently so high because it was originally a facility for breeding Pu for the UK atomic weapons project. This involved large amounts of neutron bombardment of materials. More detail here:

    WNA says the following:
    “For gas-cooled reactors the costs were much higher due to the greater amount of radioactive materials involved, reaching $2600/kWe for some UK Magnox reactors.”

    A lot more detail here:
    “Decommissioning Nuclear Power Plants: Policies, Strategies and Costs” (OECD)

    You can view most of the book on Google books.


  96. Surely to make grand statements about ‘only nuclear’ meeting our power needs for the future one also has to investigate the incremental but important advances in other corresponding technologies. So Gen3 and Gen4 can ‘eat’ old waste… OK, I’m a fan for that purpose alone.

    But nuclear as our one-stop shop climate solution, because all the others are too expensive, especially when backup is required to smooth supply to the grid? Really? This list seems happy to count the blessings that advanced Gen4 reactors will bring, but unable to also anticipate other advances in other technologies.

    What if we didn’t need a super-grid anymore to ‘smooth supply’ from wind power or solar farms? What if giant, CHEAP batteries became possible, that could almost make State grids irrelevant?

    It’s possible. Pump the cost of a large state wide grid into these batteries, and various renewables may even become cost competetive. Local power from local cities into local battery banks on a more localised grid becomes a more secure prospect for a simplified and more locally rewarding grid.


  97. Eclipse Now,

    Just for you, I did this quick analysis. It is from the latest ABARE reeport (released yesterday) on the major electrcity generation projects under construction or with applications in the piper line

    Wind farms are running at between $2.5/W and $3/W.

    Multiply by 3 to get equivalent energy to nuclear or coal = $7.5 to $9/W (say $8/W)

    Add $1 for gas to back up for the wind (not including capacity credit) = $1/W

    Add $1W for transmission = $1/W

    Total cost for wind power with same availability as nuclear and coal = $10/W

    Solar thermal in ACT for $6.41/W.

    Capacity factor in winter assume 10%

    Solar with equivalent energy to nuclear or coal (multiply by 9) = $57.69/W

    Gas back up = $1/W

    Transmission = $1/W

    Total = $60/W

    Thank you John Stanhope; that is why my rates and electricity prices are skyrocketing – and we don’t even have the CPRS yet :)


  98. I should have made that last post clearer. Wind capital cost is about twice that of nuclear to get the similar quality of energy supply. To get wind down to being equivalent you need to halve the cost of the wind farms, halve the cost of transmission (you cannot get rid of it) and provide batteries for half the cost of the fossil fuel back up – ie about $0.5/W and near $0//Wh.

    What price do you have for storage in $/W and $/Wh? (actual prices not researchers’ optimistic dreams)


  99. What price do you have for Gen4 nuclear power plants? (Actual Cost not researchers’ optimistic dreams)

    I don’t care that I don’t know the cost of this new battery gizmo, as I was just making the point that you all tend to allow glowing reports about Gen4 future development but not really allow for incremental advances in boring non-nuclear technologies.

    Anyway I have repeatedly explained that a LARGE share of the battery storage will be… what was it again? Oh yeah, FREE!!!

    Utilities will NOT pay for installing it, neither will wind farms. A certain very generous electric car system setting up in Canberra in 2012 will demonstrate this new technology and business plan.

    And of course, instead of taking the 1 cent / km coal powered energy they are taking the 2 cents / km wind powered energy because it STILL beats oil at an equivalent price / km working out about $0.80 cents a litre.

    Oh no… we’re all going to be bankrupted by Better Place using wind not nuclear! ;-)

    The cents / km are only marginally affected by the electricity price.

    The majority of it is the battery technology and battery-swap business plan. But of course, as I already spelt out, the UTILITIES do not have to pay this… us car users do… at $0.80 cents a litre equivalent energy cost. Know anywhere you can fill up for that today?


  100. Pingback: Key concepts for reliable, small-scale low-carbon energy grids « BraveNewClimate

  101. What Barry Brook failed to take into account about the nuclear option is the enrichment process the get the uranium and the 250,000 years that the spent fuel is not only extremely dangerous but is impressively expensive to maintain, man and protect (add that little number in there and suddenly nuclear fission is not as viable). There’s also no CFCs put into the atmosphere with solar cells. The solar option is also made from silicon, one of the most common elements on the planet! Solar cells also have NO moving parts, won’t ever pollute and are portable!


  102. Matt -Check again

    Many industries produce hazardous waste. The nuclear industry has developed technology that will ensure its hazardous waste can be managed appropriately so as to cause no risk to future generations.

    In fact, the radioactivity of nuclear wastes naturally decays progressively and has a finite radiotoxic lifetime. The radioactivity of high-level wastes decays to the level of an equivalent amount of original mined uranium ore in about 1,000 years. Its hazard then depends on how concentrated it is. Compare this to other industrial wastes (e.g. heavy metals such as cadmium and mercury), which remain hazardous indefinitely.

    Most nuclear wastes produced are hazardous, due to their radioactivity, for only a few tens of years and are routinely disposed in near-surface disposal facilities.

    If a typical in a First World nation got all of the energy they used in their lifetime (~70yrs) from nuclear fission, the total high-level waste ball would be the size of an orange.


  103. Hi Matt,
    it is precisely the IFR’s ability to burn all that horrible, long lived nuclear waste, reduce it to one tenth the mass, and leave behind a radioactive waste so ‘hot’ that it burns itself out in about 500 to 1000 years that has swung me around from being anti-nuclear to at least being open to the idea. In the USA alone the waste is sitting there in what, hundreds of thousands of tons of steel containers? We need to reduce that stockpile of nasty stuff by burning it. Then we store 1/10th the mass for 1000 years underground.

    So the only way to do it is the variety of new generation nuclear plants being considered, and the main thing we are waiting on now is the economics of the right way to produce them in mass volumes off a production line.

    So I’m for nuclear power, IF baseload renewables really don’t prove economically competitive in the long run. (About which I remain agnostic over the next 10 years as there are so many developments on all electricity generation fronts).

    Remember, with peak oil in about 2014, there are far worse scenarios concerning *lack* of energy that will impact the entire globe.

    EG: * many international airlines bankrupting and flight increasingly belong to only the uber-rich and governments
    * international tourism taking a huge hit
    * massive inflation in the price of EVERYTHING because everything involves production with oil or transport by oil
    * the American petrodollar collapsing
    * Greater Depression
    * Carter Doctrine creating the possibility of REAL oil wars that dwarf Iraq and Afghanistan

    So while trying to produce baseload power with renewables, we’ll also be forced to move much of mining, agriculture, transport and construction systems across to electricity… we can’t really afford to have our electricity supplies in doubt.

    (On a more morbid note: another Chernobyl might even be good for the local environment, as look at all the wildlife that has come back around there now that us humans have abandoned the place. But I’d hate to lose Sydney… one of the most beautiful cities and harbours in the world.)


  104. Eclipse – Two things: You have to make a distinction between ‘used fuel’ and ‘nuclear waste’ in particular in reference to one of the nuclear weapon Powers. The bulk of the real hazardous stuff, not just because it is just radiotoxic but also because it is chemotoxic and highly reactive, (and often liquid) is from nuclear weapons production. The antinuclear movement has always loved to confound the difference in there propaganda. Spent fuel per se, is relatively benign, once its initial cooling is complete.

    The second thing is that spent fuel can be burned now, in current reactors – if it is reprocessed – something that is done in France, for example. Gen IV is not required for that. The leftover is a small mass of isotopes that may have other uses once there is enough of it to make it worthwhile to develop.


  105. “The leftover is a small mass of isotopes that may have other uses once there is enough of it to make it worthwhile to develop.”

    Like? I’m interested in what you think *might* be possible with the final waste product.


  106. Like this.

    I asked GRL Cowan at the top of that thread for info on the elemental composition of the waste. Then I tried to estimate the value in the materials. Like I say in the comment, its a very dodgy calculation. If anyone has better information I’d be interested.


  107. Eclipse,

    Well neptunium, is a precursor in plutonium-238 production which is needed for RTGs for powering deep-space missions; americium is used for smoke detectors, and could be used in some designs of so-called ‘atomic batteries’; curium sources are used for alpha particle X-ray spectrometers; iodine-129, technetium-99, caesium-135 and strontium-90 are all used as ether medical or industrial isotopes.

    In Japan, platinum group metals are also targeted, for commercial recovery from spent fuel reprocessing.


Comments are closed.