The efficiency of solar photovoltaics

BP provide data for the installed solar photovoltaic (PV) capacity in a number of countries together with the amount of solar electricity consumed [1]. This allows for the solar load factor to be calculated. If you take the installed capacity and multiply that by 24 hours and 365.25 days this provides a theoretical maximum electrical output against which the actual output can be compared. The results, which are in part surprising (spurious?) are shown in Figure 1 and Table 1.

Figure 1 The solar PV load factor in the UK is less than 9% and less than 10% in Germany. In the countries ranked from the UK to Australia, load factors appear to vary in a way consistent with latitude and sunshine. But the large jump to over 20% in Portugal looks suspicious and the figure of 30% in Spain appears to be wrong. At the other end of the spectrum, the figures in the USA and China appear to be spuriously low.

In detail, understanding the efficiency of a PV panel is not straight forward. In fact what I am addressing here is the load factor comparing the actual output to the theoretical maximum output.  It seems the panels are rated against sunshine falling directly (perpendicularly) onto the panel. So a 1 kW panel will only produce 1 kW if the Sun is shining directly on it. At higher latitudes this rarely / never happens since the efficiency at which the panels capture sunlight is influenced by the geometric alignment between the panel and the Sun, and this is continually shifting. On average, across an annual cycle, there is no Sun at all for half of the time (night time) and the maximum load a panel can have is therefore 50%. This may seem a terrible waste but it is in fact one of solar’s main strengths. The Sun shines and produces electricity during the day when electricity demand is highest. 

Beyond the diurnal cycle, latitude and cloudiness also impact the amount of solar energy reaching the surface. David MacKay [2, page 38 book, page 39 on line] has good graphics that explains the latitude effect (Figures 2 and 3). The latitude of southern England receives only 60% of the solar energy compared with the orthogonal alignment of Sun and Earth’s surface at the Equator (Figure 2). But the difference between England and Scotland is much less pronounced (Figure 3).

Figure 2 From David MacKay [2] illustrating how the surface at high latitudes receive less direct solar energy than the equator. Cambridge receives 60% of the solar energy received by Nairobi. The example is for midday in spring or autumn.

Figure 3 Whilst the difference between Nairobi and Cambridge is large, it is much smaller between London and Edinburgh, but this does not take into account difference in cloud cover [3].


Figure 1 and Table 1  plot the load factors of solar PV. In part the numbers make good sense but there are also some anomalies, that are difficult to rationalise as discussed below. The UK load factor is less than 9%. Given the maximum load is 50% (daylight) times 60% (latitude) = 30% max, a 9% measured load seems to make good sense. The reduction from 30 to 9% reflects the sub-optimal orientation of panels in the morning and evening combined with cloud cover.

Table 1 Data from BP [1] plotted in Figure 1.

The data show a progressive rise in load through Germany, Japan, Czech Republic, France Italy and Australia the latter country giving over 13% (Figure 1). This progression seems to make sense relative to geography (lower latitudes) and my perception of increased sunniness in these countries. But I am a bit surprised that the progression is not more pronounced. Perhaps the correct way to see this is that in relative terms the load in Australia is 50% higher than the UK. But then there is a massive jump to Portugal and Spain. In 2012, Spain appears to have a solar PV load factor of 30% which I believe is higher than the theoretical maximum. It is tempting to assume a data error but looking at the last four years (2009 to 2012) we see that the loads have progressively increased from 20 to 21 to 23 to 30%. There is no such progression in the numbers from Portugal that hover around 20%. Are the Spaniards making up numbers to justify their massive investment that is bankrupting the country?

If the numbers from Portugal are correct, they suggest that a 20% load is the maximum attainable for solar PV in a small sunny country with a Mediterranean climate. At the other end of the spectrum the USA and China, both of which possess large sunny territories at lower latitudes than Europe, have loads less than 7%. It is possible that dense air pollution in China is occluding sunlight but I do not really understand why these countries should be so much lower than Europe.


Solar load factors in temperate high latitude countries like the UK and Germany are less than 10%. In sunny lower latitude countries like Australia they are over 13% and perhaps as high as 20% in small Mediterranean lands like Portugal. If it costs €2,000 to install 1 kW of PV capacity [4] it will cost €20,000 to deliver 1kW of solar PV electricity equivalent 24/7 in northern Europe. That is a lot of money to pay up front to keep an electric fire burning!

[1] BP: Statistical Review of World Energy 2013
[2] Sustainable Energy Without The Hot Air
[3] The link between sunshine and temperature based on UK climate records since 1933
[4] Energy Saving Trust

This entry was posted in Energy, Political commentary and tagged , , , , . Bookmark the permalink.

57 Responses to The efficiency of solar photovoltaics

  1. Hi Euan,

    Timely topic. A couple of thoughts on the capacity factor for American PV. A lot of our generation is self-generation, but still grid tied. For example, my university, Caltech, has 2MW of PV. My cabin has 1kW. I believe these are counted for our national PV capacity, the GW. However, this generation is net metered, so it does not contribute to the TWh. Our statistics generally consider self-generation to be negative demand. This would skew the capacity factor calculation.

    Our cabin capacity is calculated by taking the capacity rating of a panel and multiplying by the number of panels. On the other hand, a large installation, like Caltech’s, will use the rating of the AC inverter, presumably a smaller number than the aggregate panel capacity. This would give a higher capacity factor.

    Some numbers. My capacity factor is 13%. Caltech’s is 16%. I believe the desert installations will be closer to 20%. I think a reasonable overall number for the US would be 15%.

    One other factor that you might consider. If the capacity is rising rapidly during the year, you would need to use a time-average capacity in your calculations.

    For Spain, you might see if some of their solar thermal generation has crept into the statistics.

    Overall, the solar pages are not BP’s best effort.


    • Euan Mearns says:

      Dave, thanks for that. I’m sure all these factors are at work. There needs to be some form of standardised reporting. It looks like 20% is the upper bound and I think many of the numbers from N Europe are likely reasonable. Writing this woke me up:

      If it costs €2,000 to install 1 kW of PV capacity [4] it will cost €20,000 to deliver 1kW of solar PV electricity equivalent 24/7 in northern Europe. That is a lot of money to pay up front to keep an electric fire burning!


      I need to get back to Nate Hagens since I recall he and Hannes Kunz having some big bruising posts on TOD about discounting rates and renewables. If you discount €20,000 for 20 years at 5% your “free” electricity in 20 years time is costing you a huge amount. Economics not my strong point though.

    • kakatoa says:


      I concur that lumping the entire US into one metric (without say an SD) is a bit sloppy.

      PG&E rolled out their SMART meter program for PV net meters about a year ago. They moved the accounting for Time of Use generation and usage from analog readings of a digital meter (which in my case is a E-7 net meter) to an algorithm based calculation from a data base on their servers. Prior to the smart meter roll out PG&E was capable of reporting the Net PV kWh sent to the grid by the different time bins for the billing rate (E-7 has two and E-6 has three time bins). With Smart meters they are now capturing different information. They are accounting for all the kWh that a PV system is sending into their grid- not just the net. They are incapable, at the moment, of measuring the instantaneous wattage that is used at a residence that the self- generation (PV system) provides.

      I had Smart Meter for a few months last year so I calculated nirvana as far as one way of looking at energy efficiency goes- my load that was balanced out by the PV system. I define my load balanced as my PV systems output being 100% used to provide energy to power the many processes that use electrical energy to accomplish tasks at our place- lighting, well pumping, heating some water, heating or cooling some structures, using the microwave oven, and the really big one drying our clothes in an electric dryer,,etc.. Depending on which data set I used, PG&E or Opower provided data from last April, my PV systems generation was perfectly balanced by usage for 414 kWh during the month using Opower data or 365 kWh using PG&E Chanel data (that IS not reported at the meter). My little system generated a total of 957 kWh that month (April, 2013). Depending on which input is used to calculate my load balanced metric my percentage PV generation that was load balanced by usage was either 38.1% or 43.4%.

      I assume the very smart folks at Telsa, and Solar City are aware of the load balancing role that the grid provides. How, and more importantly to Telsa is Who, makes $ from an updated system of accounting sounds like it will take some forensic accountants to figure out. The nuances of the billing codes used by CASIO is something that I have never really wanted to delve into, but there are a lot of ways to capture $ from the processes. It looks like Jack Ellis is aware of how the system is currently set up as his recent comments at this blog post- notes that there is a mind boggling number of ways (160+) to capture some of the value ($) in getting an watt of electrical energy to the location it is needed to do some work.

      From a technical perspective my 6.12 Kw (STS rated- ie DC in the LAB) or 5.22 Kw (CEC AC rated- which takes into account the real world, to a point, theoretical max output of my little PV system) just celebrated it’s 1 year anniversary from the date I opt’d out of the Smart Meter program. During that time frame it has generated 9288 kWh. I sent 1624 kWh to the grid at peak time (E-7 rate schedule). The total kWh that PG&E is likely reporting as our household usage for the year is 7137 kWh.(this value includes the 1624 kWh measured at peak times, alternatively they could report 7137-1624=5513kWh ).

      The capacity factor of PV is 0 for 12 hours a day- give or take. At this time of year my system hits close to its CEC theoretical maximum output for about 3 hours a day. My system has been degrading a bit for the last 8 years so it almost NEVER hits the original CEC theoretic max AC output value of 5220 instantaneous watts.

      • Graham Palmer says:

        One of the interesting questions is ascertaining exactly how much solar is being fed into the grid. In Australia, there was some use of gross feed-in tariffs with two circuits, which provided a measure of actual generation. However most systems are now on a net tariff with a single circuit, hence the information is lost at the meter and there is no measure of actual generation. Estimates have been made based on estimated installed capacity from solar rebates. Then using postcode data and a small number of well-maintained reference solar systems and/or weather data, an estimate can be made of actual generation. This provides a good first-order estimate but doesn’t allow for poor panel orientation, shading, dust build-up, inverter failure, diode failure etc. With a new fleet of solar PV, this is probably not a huge issue, but will warrant more research as systems age and warranties run out or can no longer be claimed on.

        • kakatoa says:


          The wonders of new technologies in the IT world sure make life different in 2014 than what it was like back in 2005/6 when I spec’d out my little PV system. I haven’t even thought about pulling out a slide rule in years. Starting about 2010/11 the majority of residential solar installed in CA occurred via lease agreements (the 30% federal tax goes to the lease holder).. In 2006 90+% of the residential PV was paid for by the owner of the home. With a lease agreement the resident pays the lease holder for the output of the system at a fixed price. Solar City, E. Musk’s firm, is a major installer of residential and commercial PV in the state. Hence they, any solar leased system owner, need to know how many kWh are generated from each system they own. PV system performance data is transmitted to servers via wifi in most cases (vs say a dedicated land line). They total up all their installed systems kWh for their own internal accounting and billing. I am not sure if they share this information with the regulating bodies, etc.. A wealth of performance data is potentially available..

          I have read one lease agreement, for a friend, and it had a clause in it that required the home owner to sign up for net metering and a Time of Use rate schedule if it was available from their service provider. Given this requirement, I assume the lease holder then can get some type of credit from the traditional service provider- say PG&E- for the kWh that are sent to the grid from the residential system they own. How all of this is accounted for I am not clear. Likely one of those codes that CASIO uses for energy provided to the grid.

          In CA, like Spain, we are trying to decide how to design the residential rates to encourage energy efficiency adaption while ensuring that costs for the different value chain parties are appropriately allocated based on the cost(s) to provide service. Many of our public based utilities, example being SMUD, have already had some minimum monthly fee allocated to residential PV owners to capture some of the fixed costs associated with generation, transmission, and distribution. Our Public Utilities Commission has a difficult job ahead as some of the assumptions about the fairness of our the current rate designs are being questioned. In case you’re interested in the particulars-

  2. Joe Public says:

    Thanks Euan. Another physics lesson which probably doesn’t get taught to secondary schools’ physics students.

  3. Willem Post says:

    US total solar (PV + CSP) energy generated in 2012 was 12,775 GWh, of which 4,327 GWh was by utilities.

    In 2013 a record MW of PV panels were installed, mostly in the last quarter, and utility energy was 9,253 GWh

    • Euan Mearns says:

      Thanks Willem, it looks like the installation number reported by BP is the Utilities and they are not reporting the privately owned installations, but are perhaps somehow counting all of the power produced. 12.8/4.3*6.9% = 20.3% ?

  4. Roger Andrews says:

    Since installation a year ago my 2.25kw solar system has operated at a load factor of 20.02% (4,021 kwh of generation in 372 days, to be precise). Nice to live at latitude 20N 🙂

    I’ve plotted this number, Dave Rutledges’ numbers and Euan’s numbers against latitude in the graphic below.

    • Hi Roger,

      You had a good idea to plot the chart. My guess is that a major part of the latitude effect is scattering over the longer paths. If I take a 100-W solar panel here in Los Angeles and aim it properly at the sun, I get 60W output. However, when I point the panel end on to the sun, I still get 15W. That is, a lot of light is scattered at our latitude.


      • Roger Andrews says:

        Dave: Solar panels respond to light in the visible and IR spectra so I guess scattering would give you some response even when your panel is end-on to the sun. 15W does seem a little high though. Smog amplification, maybe? 😉

      • Euan Mearns says:

        Dave, when you point the 100 W panel directly at the Sun in California you only get 60 W. Why is that? Is it because 40% of the light is scattered? It a kind of makes a mockery of the panel rating if it can never make its specification outside of the lab. My next post will be on Scottish solar (was originally part of this post but i hived it off) where we are just as likely to see panels facing N as S and I’ve been told it makes no difference, which is maybe true in a land where the Sun seldom shines. If you get 15% from scattered light * 50% for daylight * 60% for high latitude you zero in on a load of less than 5% for randomly orientated panels in Scotland which I believe is likely close to reality. If Pedro’s ERoEI of 2.4 is correct and based on S of Spain the ERoEI for Scottish solar will be one quarter of that, i.e. 0.6 meaning that they will never produce the energy used to create the devices, i.e. coal and diesel. More CO2 is produced before the device makes any energy than will be saved over the whole 30 year life cycle 🙁

        • Euan/Dave,

          There is a formal way to measure the panel specifications. for instance, for a 240 Wp moduel that the manufacturer says it has -0+5W tolerance, the measure has to be made under a irradiance of 850 W/m2, as measured by a calibrated device, with an Air Mass (AM) = 1.5 and with a temperature of 20º Celsius or below. This is somehow a trick some manufaturers use, because having an irradiance of 850 W/m2 occurs only in very snunny and clear sky days, usually in summer and usually, under these conditions, with an AM=1.5 (midday in our country in the best case) it is very rare that we get 20º Celsius. These conditions are for a hotter day and a hotted moment. But if all the conditions are given, the panel (brand new, of course, because it has a degradation pattern of about 1% average per year until year 10 or 20, depending on the manufacturer) must deliver the nominal power at the module output, in my case 240 W. Another measure is at the inverter output, that includes the inverter losses, usually bettween 3 and 6 percent.

        • Roger Andrews says:

          You guys have got me running experiments now.

          This morning, just before the sun began to shine directly on my 2.25 KW of panels, I got up to 100W of output from back scattered light. This, however, is less than 5% of rated capacity and it doesn’t go on for very long so it isn’t going to add much to total output.

          The most I’ve ever gotten out of my system is 2,032 W, or a little over 90% of rated capacity. Why not 100%? Probably a combination of factors. The air is never completely clear. The panels are never completely clean (dust, birds). I’m told that output decreases as panel temperature increases, and black objects left sitting in the sun around here get hot. But it seems that the installer allowed for all this because what he said I would get is pretty close to what I’m actually getting.

          Incidentally, solar radiation here peaks out at around 1,100 W/sq m at this time of the year.

  5. Glen Mcmillian says:

    It might be the case that in Germany the country has not made a big mistake in installing a lot of pv if one looks at this from the broadest possible perspective of establishing a domestic industry capable of earning a lot of foreign exchange thru exporting the goods and expertise and as a hedge against a market that might not always deliver oil and gas and coal due to depletion or war or acts of God.

    But it sure is hard to make the case for solar at ten percent given costs in the past isn’t it?

    Given future costs of oil and gas it might be that solar at todays prices is not outrageously expensive since it will produce fuel free for thirty years or longer. Tanks and fighter aircraft cost a lot too .

    Which option delivers more national security per EURO?

    There is another aspect to spending public money on things that are excessively expensive.
    that bears thinking about.

    In a modern western societies we ROUTINELY spend large sums on boondoggles. There is no real justification for most of the freeways and public buildings we build these days but we build them any way.

    There is virtually no long term benefit to be discerned from the money we spend on ethanol in the US for instance.

    That money would be far better spent on getting freight and cars off the highway and goods and people onto trains or city buses and street cars.

    There is this to be said for pv even in Germany. The money spent on it is not COMPLETELY wasted.

    I may eventually buy a modest pv system for my little farm just to be on the safe side in case of the worst coming to pass . I do not expect it to be a worthwhile investment compared to buying electricity from my utility but the marginal value of even a little electricity that cannot be counted on at any particular time will be huge in the event the grid goes down.Three or four kilowatt hours a day would be the difference between a primitive and a more or less modern life style for me – that is enough to keep a freezer cold and my well pump on and run a few light bulbs and the washer occasionally.My freezers are good ones and they will keep the contents rock hard for three days at least even in hot weather. I can get five hundred gallons of potable water out of my deep well with
    just one kilowatt hour of electricity.

    Now if I lived in Arizona or any part of the dry and sunny American southwest I think a pv system would turn out to be a good investment for almost anybody who uses a lot of juice and has plenty of room to ground mount it because the political prospects for building new nukes is slim and the price of gas and coal are going to continue to go up over the long haul.

    • Euan Mearns says:

      Glen, I think I just twigged that you are Farmer Mac – right? It is good to get these alternative perspectives. Is it better to buy a work of art for $20,000 or a solar PV array?

      but the marginal value of even a little electricity that cannot be counted on at any particular time will be huge in the event the grid goes down.

      The paradox here is that at present the greatest risk to the UK / European grid is the closure of base load capacity and the increase in renewables, especially wind. Better get a turbine incase your turbine causes the grid to crash 😉

      As you know, my preferred hedge against scarcity of FF and the security of their supply is nuclear.

      • Glen Mcmillian says:

        Yes I am OFM.

        I would like to comment here using that handle but I can’t figure out how to change it.

        I personally am scared silly by nukes given the long term implications of the safe storage of spent fuel in the event of social collapse for any reason.But I am scared about as bad or worse by the thought of NOT having them in the same situation.

        We are not at high risk of being invaded or blockaded or quarantined here in the US in terms of deliveries of coal and gas for good reasons (domestic supply, super duper sized military establishment, ocean on two borders and large friendly buffer countries on the other two) but in the case of your country a new fleet of nukes could be as important as the Royal Navy to the security of the country in the event of a war.

        The military capacity to keep sea borne coal and gas deliveries up and on schedule during a time of war simply do not exist and the capacity to build the ships needed doesn’t exist any more in our countries.

        One obsolete fighter plane equipped with obsolete air to ground rockets can sink a super tanker. They aren’t armored and they don’t have the size of crew and training needed on board to fight the sort of fires started by explosives.

        It is trivially easy to bomb a pipeline that is hundreds or thousands of kilometers long.

        I may be too pessimistic about the future prices of coal and gas in terms of constant money.

        I am equally pessimistic about the stability of all fiat currencies and believe that while massive deflation is a short term possibility massive inflation is a long term certainty barring an absolute collapse of the government in question.

        If you can permit and finance a new fleet of nukes now at a fixed low interest rate inflation alone virtually guarantees that making the payments on them will be easier than paying for imported gas and coal a decade or two down the road.

        The safe storage of spent fuel and decommissioning worn out plants is not a tough technical or financial problem compared to other options.

        The arguments about safe spent fuel storage are in my opinion mostly ill-informed or worse simply dishonest.

        Let us consider the case of a typical existing nuclear plant of the sort I have worked in the US.

        IF we were to have an INTELLECTUALLY HONEST discussion of the choices actually available to us given the constraints of money and so forth it would be ok to just remove every possible trace of fuel from an old nuke and seal the containment building and forget it.

        I am certain that in a few hundred to a few thousand years there will be some slow leakage evident.

        But compared the environmental damage we are doing to generate the same amount of power burning coal for instance – that slow leak is going to be a trivial problem in my estimation.

        Only an idiot would seriously entertain the notion of simply doing without nukes and coal.

        IF I were the all powerful dictator of this planet I my first priority after making sure of my own position would be to get cracking on a permanent war time footing on conservation and energy efficiency and reducing the birth rate to below replacement world wide as fast as my propaganda machine and secret police could manage the job.

        I have for often wondered why spent fuel could not be finely pulverized and buried in an old exhausted oil field using well field equipments such as the pumps used for tracking. If the oil stayed there safely for the tens of millions of years so would highly diluted spent fuel.

        Getting it out again might be possible but if the right chemicals were mixed in the particles might bind to the stone down below. At any rate it would be an extremely time consuming and expensive job to recover the fuel once dispersed- Consider how hard it would be to get back tracking sand or ceramic beads. And every year that passes would reduce the danger and the benefit ( to a terrorist ) of recovering the spent fuel.

        A government or major corporation might manage to recover enough of it to make a bomb or at least a dirty bomb but short of that – It couldn’t be done and the doing of it would be perfectly obvious in any case. Such an operation could not possibly be hidden because it would require many very large pieces of machinery for months at least.

        We could build a new fleet of reasonably safe nukes ; they would probably be safer by an order of magnitude than the ones we have now and the ones we have now are relatively safe considering the well documented dangers of coal and natural gas.

        Any body who has twenty grand to spend on art is a fool if he doesn’t have an implemented plan for his personal survival in the event of a Black Swan event ranging from a new emerging highly contagious fatal disease to WWIII.Such people usually have expensive locks and burglar alarms and lots of medical insurance and fire insurance etc.

        But they almost always fail to consider anything somebody is not actively selling to them.They buy the latest and safest heaviest cars because the industry of fear mongering convinces them to do so for instance.

        Part of my own personal security plan is to own some solar panels if I can ever put my hands on enough money. I am not broke but so far I have not gotten all the other bases covered that can be covered cheaper.

        I do have my own deep well and gravity fed spring water and my own sewage disposal and enough firewood in sheds to stay warm and cook if necessary for three years- and I have a chimney and my Mom’s well cared for wood fired kitchen range.I have a good supply of diesel fuel on hand and a couple of barrels of gasoline and three generators which I can keep running with fuel on hand for many months long enough each day to supply water from the well and keep the freezers cold.

        I am a real farmer born and bred and professionally trained and can grow enough food to manage without pesticides or fertilizers – if somebody doesn’t kill me in the act of stealing my proverbial last can of beans.

        I am well armed and will not hesitate to go proactive in defending myself and a few old and trusted friends and relatives in the event things go mad max- which would happen in the event of a large scale nuclear attack.

        I have a bomb shelter of sorts that is adequate for a few months -and if that is not long enough I am not going to last outside anyway.

        Maybe I am wrong but in my estimation a nuclear WWIII is considerably more likely than a catastrophic failure of any given nuclear power plant.I have relatives who are professional military people including medical officers and they agree emphatically on this point.The Pentagon estimated the possibility of a nuclear WWIII at two percent annually all thru the decades of the cold war.The chances are probably much less now but certainly not zero.

        None of these things have cost me an extra dime excepting putting some non perishable stuff in the shelter which consists of the underground basement of a large masonry barn we used to use for storage before we retired.THey are all the sort of things that pay their own way if you choose to live in a rural area on a farm.

        IF WWIII starts unexpectedly with nukes involved the chances of any body living in a big city who doesn’t just get in a car and go immediately surviving a day to a month are extremely slim to none in my estimation. There will be no grid, no food deliveries, no cops not looking out for themselves and their own families.

        The chances of long term survival out in the boonies are not much better of course unless you have friends in the boonies such as yours truly.

        Mine know what to bring- medicines, food, clothing, and weapons.Books.Survival skills.

        There are only six – carefully chosen- with the standing invitation.Three or four of them might actually manage to get here in the event of need to do so.

        I think we would have fifty percent chance of making long term radiation sickness excepted.A post WWIII world will be a very dangerous place indeed.

        Foolish fears on my part?

        Not half so ”foolish” in my estimation as worrying about any given nuke melting down.

  6. Roger Andrews says:

    “The Sun shines and produces electricity during the day when electricity demand is highest.”

    True, Euan, but the problem with solar energy isn’t matching daily demand. It’s matching annual demand, which in most parts of the developed world is 180 degrees out of sync with the solar cycle. Here’s the impact in the UK:

    Solar radiation is a good match to demand in July, but July isn’t when you need it. You need it in January, in particular during the 5-7 pm peak period, but the sun has set by then. You also need more solar power during the daytime because January electricity demand is about 15GW higher overall than July demand, but instead you get less than a quarter as much. (Note that the solar scale doesn’t imply any specific installed capacity.)

    The annual fluctuations could of course be smoothed out with pumped hydro, but the storage capacity needed would be humongous. To store say, an average of 5GW of surplus solar during the summer months for release in the winter could require as much as 20 TWH.

    And added to that you have the problem of intermittency on cloudy days …..

    I think that while solar power is good for some small-scale projects you can make a case that it’s potentially even more problematic than wind when it comes to grid-scale applications. You can’t even make the case that the sun will always be shining somewhere, unless of course you interconnect with New Zealand.

    • Euan Mearns says:

      Solar is certainly out of synch with the annual demand cycle in N Europe. Doesn’t do you much good N of the Arctic Circle 😉 But in wealthy, sunny climes that are running AC, there must surely be a better match to demand. The data I’ve seen from Germany means that they can more or less shut down their gas peaking plants in summer, but of course still need them in winter.

      • Roger Andrews says:

        I’ve been looking for wealthy, sunny AC-running climes that might give a better match between solar generation and demand, and the best example I’ve come across is the US, where the summer cooling peak exceeds the winter heating peak by about 20%. But when we plot the data there’s still a large imbalance between annual demand and annual solar output (calculated for latitude 40N, incidentally).

        There don’t seem to be any large countries in Europe where summer demand significantly exceeds winter demand. The closest is Italy.

        We also have to remember that winter electricity demand in many countries is suppressed by the use of natural gas (and oil in some parts of the US) rather than electricity for heating. In the green sustainable world of the future we wouldn’t of course use oil or gas for heating; we would use clean renewable electricity. The result, all other things being equal, would be a large overall increase in winter peak electricity demand which would make it even more difficult to integrate solar with the grid.

        • Graham Palmer says:

          In summer-peaking grids, PV can provide a generally good match but the problem is that it doesn’t provide a robust solution to the high costs of network augmentation. In recent years in Australia, there has been exhaustive analysis of this in order to inform feed-in tariff policy. The costs are mostly about distribution networks rather than generation. Peak demand in the capital cities on the hottest days often occurs in the late afternoon/early evening when people are arriving home and turning on their A/C, after solar has dropped. The PV is reducing the wholesale peak prices in the middle of the day, which in theory is good, but the distortionary effect has of course flow on effects and unintended consequences.

          At a local level, there may be instances where PV can delay or reduce network upgrades but not at a system-wide level. Many of these issues have been fortunately recognised and the states are pulling back on feed-in tariffs. But adding around 4 hours storage would allow PV to provide a meaningful role to the problem of increasing network costs, and it then simply becomes a matter of assessing the costs of storage versus alternatives (ie, transformer upgrades, demand management etc.)

          • Euan Mearns says:

            Graham, when the wind blows in Europe it suppresses the whole sale price of electricity – a good thing? Well the wind operators get their top notch ROC (renewable obligation certificate) price so they are not fussed about the price being dumped and consumers I believe have to pay that top notch price regardless of the wholesale price being dumped. I think it is the traditional utilities who lose, getting reduced market share and lower price for providing the invaluable load balancing service – at least I think that’s how it works. German utilities are certainly in trouble.

            Parasitic wind killing its host

  7. Syndroma says:

    Euan, at higher latitudes land area gets less sunshine, but you don’t have to place your panels parallel to the ground. If you place them at an angle corresponding to the latitude, you’ll get the same amount of light as on the equator. The shadows will be longer and you can place less panels on the same land area though.

    • Euan Mearns says:

      Syndroma, good point. I did think about this but not for too long. Would it be correct to say that on the Arctic Circle you could place your panels vertical to optimally catch 5 minutes of sunshine at Midday in mid winter? But the panels are then badly orientated for the rest of the year. The only solution is to have a tracking device….

      they claim a 45% uplift in efficiency – i.e. you will produce 45% more with this than without it. The other issue is the depth of atmosphere the light has to pass through, don’t know what impact this has on the relevant wavelengths.

      • Syndroma says:

        >on the Arctic Circle you could place your panels vertical to optimally catch 5 minutes of sunshine at Midday in mid winter?
        Yes, and at 46 degrees from vertical to catch 24 hours of sunshine in mid summer. It’s the same seasonal ±23 degrees variation in all lattitudes. But I guess it’s more important to track the daily motion of the sun. The efficient way to do it is to use something like an

        >The other issue is the depth of atmosphere the light has to pass through
        Good point.

  8. Joe Public says:

    “Figure 1 The solar PV load factor in the UK is less than 9% …………..the figure of 30% in Spain appears to be wrong.”

    Knowing nothing other than the Spain being at a ‘better’ latitude, and, the huge numbers of Brits who holiday / emigrate to Spain for their sunshine and relative lack of cloud/rain, is 3x the UK’s figure so far out??

    • Euan Mearns says:

      Joe, on basis of other comments here I think the theoretical maximum without a Sun tracking system is to be around 20% and so I think the Spanish number is wrong. If you look at the three prior years they are all just above 20% and that is where Portugal lies. I suspect the big jump between Portugal and the rest comes down to cloud cover – difference between Mediterranean and Temperate Regimes.

  9. Joe Public says:

    When the feed-in tariffs exceed the costs of using diesel generators to power arc lamps to power solar panels to harvest subsidies, man’s ingenuity knows no bounds.

    Courtesy Google translate:

    “Industry calls on the NEC to take appropriate measures against fraud solar energy
    Suspicions about the use of diesel generators to charge premiums
    The Photovoltaic Industry Association calls for ‘clean’ your image
    Last updated on Wednesday 14/04/2010 12:59
    Baltasar Montaño
    Madrid – . Secretary of State for Energy Pedro Marin, has sent a letter to the National Energy Commission (CNE ) in which he calls for the adoption of the necessary measures against the great fraud detected in the area of photovoltaics.

    There are solar plants in Castilla -La Mancha , Canarias and Andalucía mainly are producing energy all night even though the sun has gone down under , according to industry data released by THE WORLD.

    The Photovoltaic Industry Association ( Asif ) has joined the ” requirement ” of an investigation into the “alleged illegal actions ” in the photovoltaic activity and that ” we must identify the culprits to clean the image of the sector ” as has been alleged in a statement.

    According to government data , between November and January , in winter, the electrical system received 4,500 megawatt / hour produced by solar plants between midnight and seven o’clock , plus another 1,500 between 19.00 and 23.00.

    How are these megawatts generated ? Early indications , some developers might be using photovoltaic generators fueled with diesel oil to generate electricity , because the premiums are 436 euros a megawatt .

    “This is just the tip of the iceberg ,” says an expert. The CNE devoted 64% of all inspections conducted during 2009 to the photovoltaic sector. Specifically , of the 3,154 energy sector facilities inspected , 2,019 were for this branch .”

    [OK, I accept that is a 2010 story, but large subsidies do encourage the unscrupulous to be ‘creative’.]

    • Euan Mearns says:

      Joe, this is more than fascinating, keep digging.

      • Joe Public says:

        “A spokesperson for Spanish PV association APPA (Asociación de Productores de Energías Renovables) tells pv magazine that out of the 55,000 PV installations in the country, CNE suspected that around 9,000 had lied about when they were generating electricity.”

        “It was announced last July (i.e.2010) that a new decree “Fraude Fotovoltaico” was presented to the country’s Council of Ministers. Under it, around 800 megawatts of installed outdoor PV plants suspected of having swindled the feed-in tariff specified in the Real Decreto (RD) 661/2007 prior to September 28, 2008, were to be examined.” – 20th Mar 2011


        “The size of the subsidies paid annually, which amounted to about $68 billion between 1998 and 2013, had increased by 800% between 2005 and 2013.” [My bold]

        • Joe Public says:

          Euan – the Iberian subsidy-claimers would be interested ONLY, in the revenue earned. The Iberian authorities can account for the measured kWh received.

          Given the size of the subsidies available, the sceptic in me wonders if BP’s estimate of panels surface-area covering the peninsula is (woefully) underestimated, which in turn inflates the apparent load factor.

          • Euan Mearns says:

            Joe, I’m away from home and its a holiday, but…. looking at BP stat review, the Spanish solar PV loads have looked OK (compared with Portugal) up to 2011. In 2012 the Spanish solar PV capacity increased by +6.5% compared with 2011 but their solar power consumption increased by 36.8%. The installed capacity figures are specifically for PV while the consumption figures are not – simply solar. And so if they have added a large solar thermal electric plant and this is included in the consumption figures that could be an explanation. I’m not aware that this might of happened.

            It remains possible that the 2012 solar consumption figure is a simple data error – this happens all the time. But regardless of that, what you have turned up is very interesting. I have route ways into BP economics dept and can ask them to clarify the source and meaning of their numbers. I’m guessing the EIA may have similar records – Roger?


          • Joe Public says:

            @ Euan 9:23

            I don’t know whether this fact affects the figures, but I believe the Spanish are considering / have implemented a method of charging (no pun intended) self-producers for the benefit the self-producers receive by actually being connected to the grid.


          • Euan/Joe Public,

            I can not speak for the Iberian (Portugal + Spain), but in Spain the authorities do account, very accurately in digital sealed meters, for every kWh generated in every feed-in plant.

            I would reccomend, for a precise calculation, to forget the BP statistics and will take for Spain only that of the Comisión Nacional de los Mercados y la Competencia (CNMCC) whose link to the compressed file I have already passed.

            But in fact, you are right with respect to the underestimated installed power, leading to an inflation of the load factor.

            The fact is that the regulation was granting at the beginning (2004 till 2008-2009) the highest premium tariff only to plants below 100 kW. After many consults the CNE and the government estimated first that the 100 kW were referred to the output power of the inverter.

            Therefore, most of the scale utility designed multimegawatt plants based in the “farm” concept; that is, for instance, a 20 MW plant could be composed by 200 times 100 kW unitary plants, legally constituted under 200 different limited liability companies, which in its time were governed on top by an instrumental company i the hands of a Corporation. That was a legal “bending the rules” move and it was admitted that the 20 MW could have a single evacuation line and common land property and some other infrastructures (control room, etc.).

            Then, consequently, the next move was to install more peak power in modules. The fact that modern inverters may cut the output power to 100 kW, even if the take at the input 120 kW, made it possible (nota bene: this is good for business when tariffs are so high, but very bad for energy efficiency, because a lot of energy evaporates as heat in the inverter. Any inspection measuring more than 100 kW at the inverter output could have expelled the promoter from the feed in premium tariff system. But as they did not pass the upper limit, many promoters installed 105, 110 or even up to 130 kWp in modules in each 100 kW typical plant, to take advantage of premium tariff (more hours-peak in a year). Therefore, the losses by dust, in modules interconnections, in DC cabling and in the inverter itself, could permit the overpowering to have the inverter with many more hours-peak in a year.

            It was difficult to guess how much promoters had been exceeding the nominal power with the peak power at module levels.

            In our book we were conservative again and estimated an 8% of more power than the nominal registered by CNE (CNMC) at the inverter output as per the law. There were no public accurate data on this. In my experience with about 30 MW different plants, I had observed about 12% overpowering and used 8% to be in the safe side in front of the many solar PV supporters.

            Then we came to discover that we had been conservative. When the government entered into the deep economic and financial crisis and started to squeeze solar PV producers, one of the things they made in one of the many Royal Decrees downgrading previous granted rights, was to assign income to promoters only on the nominal value by means of an algorithm. Immediately, all the national the industry, and solar PV national associations reacted with logical anger, because they had been given previous securities about the considered power at the inverter output when they decided their investments. The losses that the industry claimed they were going to suffer with this backward legislation decision were 15% of their current income.

            So, now you have a couple of good data to know how much inflated the peak power is in Spain. And be sure that the EROI we considered is even lower in reality.

    • kakatoa says:

      Morning Joe

      Many of us who live in CA are hoping that the rest of the country follows our leadership in fighting global warming, climate change, climate disruption, etc.,, soon. It’s going to be a tad harder for local and statewide development folks to attract, or keep, much new business if we continue going it alone. It’s been a few years since I looked into the details of how CA is accounting for the costs of the RES. The link below is for the most recent report provided to the CA legislature by the CPUC. It can give you a feel for what it is costing ca to meet the RES. Some of the transmission and grid issues associated with adding in as much intermittent sources of generation is also noted.

      It’s a tad unfortunate that PG&E’s RES cost data is redacted for 2012, and 2013 in the data above as a lot of the new PV utility scale, must take contracts, projects started coming on line during these years. The rather large Concentrating Solar facility that PG&E is contacted to take 2/3rd of the output from didn’t start operation until early this year if memory serves me correctly. Our Energy Commission publishes Forecasts for the “Average Residential Electricity Rates” to assist folks in planning. It looks like the “Low Demand Scenario”, developed in 2011, is closer to matching the Actual Avg Residential rates for PG&E in 2014 as our actual AVG costs are a tad higher than this value.

      If one wanted to supply electrical energy to PG&E in the future this year’s contract details are noted below-

      The price a new generator will get paid for a kWh, with a new contract with PG&E, is dependent on Time of Delivery (TOD) factors that are denoted on page 94.

      • Roger Andrews says:

        I took a brief look at California’s plan to increase renewable energy’s share of electricity generation from 20% to 33% by 2020. Here are the results:

        * It will cut California’s CO2 emissions by ~5 million tons a year, representing a 1.4% cut in California’s total annual CO2 emissions, a 0.09% cut in annual US CO2 emissions and a 0.014% cut in annual global CO2 emissions.

        * It will result in the mean global surface temperature in 2100 being 0.0006C lower than it otherwise would have been, assuming a) a climate sensitivity of 3C, b) a doubling of atmospheric CO2 and c) that the IPCC has got it right.

        * It will cost an arm and a leg. (Expenditures are already in the tens of billions of dollars range and there are still six years to go.)

        Glad I’m not paying for it. 🙂

  10. Euan,

    Few words on this interesting subject

    1. Calculating the load factor on the BP statistical data has a big problem: they only give data of installed power in GW and generated energy in GWh at the end of each year. And usually for feed in installations, that are the ones better accounted. For the generated energy it is Ok, but for the installed power, they can have some gross mistakes, specially in countries were installations are ongoing throughout the year. This has to force you to make a gross average if more detailed data is not offered. That is why Charles Hall and myself choose Spain to calculate the EROI of solar PV systems between 2009 and 2011, because the period was long (3 year complete cycles) because the installed power growth had almost stopped due to the economic and financial crisis and also because the Spanish government offered very detailed month to month and region by region installed power and energy generated. The load factor was lower than that calculated here, by the way.

    2. No, we are not making up figures in Spain, just to justify the investments. These are real life figures, but they are not as good as you conclude from BP partial figures. And with respect to the Joe Public comment, the data he has presented with respect to people generating with diesel engines even by night, is from 2010. It is old and fake. That was the first marketed attempt of the government to discredit the solar PV promoters, when in fact, the fraud (see a specific chapter on fraud in our book “Spain’s Photovoltaic Revolution: Energy Return on Energy Investment –Springer 20. Prieto and Hall). The fact that a handful of idiot cheaters were using diesel generators to produce energy at premium rates and even forgot to disconnect them in the night or in cloudy days did not represent <0.1% of the installations of that time. That was severely cut by the regulatory bodies, because the measuring tools were very effective and detected these anomalies. There were thousands of inspections to the plants (I personally have passed a couple of them). The authors of this minimum cheating were expelled from the system, but this was a fabulous marketing argument for a government that had just noticed that they could probably not afford the payment of premium tariffs to discredit the solar PV promoters in front of a public until then very favourable to renewable energies. So, we were not so “creative” as some readers suggest. Were they found more anomalies was in the deadline fixed by the government to get the plants up and running (September 2008) to keep the highest premium tariffs. Then, many promoters, in knowing they will not have their plants fully operational in time (lack of modules was a key issue in 2008 in the world market), decided to start running the meters with few modules per each 100 kW plant. But this did not affect to the generated energy per installed MW, because the sealed digital meters with quarter of an hour registers and memory. And these promoters were also inspected and downgraded in the premium tariffs.

    3. As for the latitude effect, it is certainly existing, but it is not the only one contributing to more or less generation per installed power. Here, the main factor is the Air Mass (-AM- the more to the north, the more tilted the modules and the more mass of air to be crossed by solar rays), which is also considered in the sunrise and sunset of every installation. In fact testing measures in case of conflict use to clearly define the AM to verify if modules comply with the specs, by fixing this condition (i.e. AM= 1.5). Other factors are also taken into account when designing a plant. There are tropical countries with a lot of clouds. So, the important decision is sometimes the number of peak hour per year, as measured by pir/periheliometers for a number of years to be sure that it is a good place. Orientation and tilting are also important to optimize in fixed plants. One or two axis trackers in solar PV plants can help to increase the generation per installed MW, but on the other hand they are more costly and spent more energy to track. That is why big utility scale plants are more efficient than most of the rooftop mounted plants, because the later are generally placed on top of already existing roofs, many times following more the aesthetics than the proper tilting and orientation and maintenance is much more expensive when scattered over private roofs than in big accessible plants on the ground.

    4. Portugal is continental climate (Atlantic), rather than Mediterranean; that is, has more low pressure fronts than Spain and again, the irradiance depends not only on latitude, but on the regional weather (clouds, smog, fogs, etc.) and over all, temperature (the higher, the higher the losses) Even within Spain, we have places in the North-East generating more than others in the South East. One of the best places is a range of mountains in Western Andalucía, where the peak-hours in a year are optimum and temperature, due to altitude over the sea level is low enough and air is clean and absent of clouds and fog or haze. Deserts may have severe problems with sand storms and dust.

    5. As a summary conclusion, our data estimated that for the best irradiated country in Europe, in a study comprising 3 full cycle years, with data of about 4 GW of installed solar PV plants, stable without almost any growth, the EROI was in the verge of 2.4:1. Of course, we considered some energy input extended boundaries that had never been considered in previous EROI analysis, but that were all of them documented and considered SINE QUA NON the PV systems could not operate. Should we have considered all financial expenses in equivalent energy inputs and the complete labor associated to all the value chain in PV complete systems, probably the EROI could have been 1:! Or even lower. As Charles Hall said, this will not maintain a society as consumerist and energy intensity demanding as the one we have today at world level.

    • Euan Mearns says:

      Pedro, thanks very much for this detailed clarification, bbbut…

      In 2012 the Spanish solar PV capacity increased by +6.5% compared with 2011 but their solar power consumption increased by 36.8%.

      I think the installation time lag can account for under-genartion but not over-generation? But its late and I’ve been consuming bio-fuel again ;-(

      The big jump in consumed solar in Spain in 2012 requires explanation. It may simply be a BP typo?

      But the other comments you make about ERoEI are of course vital. This needs to be knitted together with Dave Rutledge’s comments about scattering of light at high latitude and my own estimation of solar PV efficiency in Scotland that is around 6% – where does that leave the ERoEI? I’ll have another post on this early next week where I hope folks with knowledge will chip in…

      PS – the literature view on ERoEI of solar PV is much, much higher than 2.4. MacKay says 4, but then again he claims 80 for wind. This is vital to know.

      • BP is very unaccurate in solar statistics and may be mixing solar PV and CSP. to get much more accurate information, please go to

        The unzip the document (latest report August 2013; we are still pending on the last and probably definitive Royal Decree to finally kill all the renewable energies in Spain) and go the label of Annual GWh_Mw-Inst and double check it by yourself in the solar FV (solar PV). These are official data from Comisión Nacional de Energia (CNE) at present Comisión Nacional de los Mercados y de la Competencia (CNMC).

        I hope this helps to clarify. We have about 60-70% of all CSP install power in the world. Here it is split and you can discriminate.

        • Roger Andrews says:


          Thank you for the link. I just used it to check the numbers I posted in the comment below, which are from the Red Eléctrica de España, and found a problem. The installed PV capacities for 2013 match:

          RES: 4,681 MW
          CNE: 4,641 MW

          But the generation totals don’t:

          RES: 8,397 GWh
          CNE: 5,924 GWh

          ¿Qué está pasando?

  11. Roger Andrews says:

    I’ve been pulling some solar pv numbers from sources other than BP (took a while) concentrating on the warm, sunny countries, all of which are at about the same latitude. Here are the load factors I came up with:

    Spain (2009): 19.6
    Spain (2010) 19.6
    Spain (2013) 20.5
    Portugal (2013) 18.0
    US (2013) 20.0
    Italy (2011) 9.5
    Italy (2012) 13.0

    The load factors for Spain, Portugal and the US now compare quite well and Italy becomes the odd man out.

  12. Yes, I agree it is vital to discern the solar PV or wind REAL LIFE EROI.

    The main difference in the Prieto/Hall methodology and previous conventional methodologies is that previous ones focused only in bottom-up approaches of given particular solar plants, with a theoretical given irradiance (of Southern Spain) and excluding many of the energy input societal costs that we analyzed, considered and included in our study, such as but not limited to accesses, foundations, canalizations and perimeter fences, evacuation lineas and rights of way, O&M full costs, module washing/cleaning, self-consumption, security and surveillance, transportation (even from China!), premature phase out of unamortized manufacturing and other equipment, , insurances, fairs, exhibitions, promotions, etc., administration expenses, municipality taxes, duties and levies, cost of land long term rent or ownership, circumstantial and indirect labor, agent representative or market agent, equipment stealing and vandalism. communications, remote control and management, preinscription, inscription and registration bonds and fees, electrical network/power lines restructuring, faulty modules/inverters, trackers, associated energy costs to injection of intermitent loads: network stabilization associated costs (i.e. combined cycle gas fired power plants) and we left out some important but comlex factors like associated energy costs to injection of intermitent loads: pump up costs and/or other massive storage systems, if applied or force majeure events (I have seen a complete utility scale plant flooded by a river). We also left out, as sensitivity analysis, the huge financial costs brought into energy equivalent and the bulk of labor costs.

    We have been discussing for months on this subject with other authors of previous conventional EROI analysis, more limited to energy input costs of the modules, inverters, trackers and metallic structures, that we gave for good.

    They finally acknolwedged, in general, that the above mentioned costs were REAL LIFE and SINE QUA NON energy input costs, but claimed that the official and previous methodology did not considered them. Some of them proposed that even if previous methodologies did not considered them (including the IEA methodology), if they are really energy inputs coming from our fossil fueled society, they should be included and considered in future EROI analyses. Now I live up to you to decide.

    The dire conclusion on our study is that conventional solar PV EROI studies considered about 1/3 of the total energy input costs and 2/3 of the energy input costs were societal energy inputs without which, the solar PV systems could not operate. In other words, even if solar modules would go to zero in price (and energy input costs), they could only increase abvout 1/3 the considered EROI. That is, that solar PV is absolutely underpinned in a fossil fueled society and would collapse or crash if this fossil fueled society collapse or crashes.

    • Euan Mearns says:

      Pedro, do you have a paper on your solar ERoEI work? If, so can you send me a copy please. If we have ERoEI of 2.4 in the S of Spain, I reckon Scotland must close to 0.6. The devices create more CO2 than they will ever save over their life before they produce any energy at all.

      • Euan,

        For sure England and Scotland will have an EROI well below 1:1 in solar PV. The latest program with premium tariffs in Great Britain is a complete craziness, IMHO.

        I believe now that we were extremely optimistic and conservative in our book about the Spanish case. Financial expenses and the bulk of labor were not considered, the later because some of the factors analyzed could have a pint of embodied labor and we tried to avoid duplication, but the energy costs of labor in all the long value chain were very important.

        Some information was given in a presentation at the State University of New York (Syracuse) in April 15th, 2011, as a prelude to the book. We considered there only two years cycle. The book added 2011 as a thrid year, without any significant changes. You can find the presentation at

  13. Glen Mcmillian says:

    These energy input and out put figures are sobering to say the least. An energy return on energy invested of only two or three to one is just not going to get the job done over the very long haul. Such a low rate of return indicates that pv cannot be a self supporting industry once fossil fuels are exhausted does it not?

    But the potential for improving the return is enormous by for instance utilizing large scale installations only and only in spots with a very good solar resource.Maintanence and construction costs are greatly reduced by such a strategy and many middle men and much work involving small systems can be eliminated.

    And there is the ever present fact that we routinely waste enormous amounts of energy without giving the waste any consideration at all.

    If for instance we expend a few tens of thousands of euros and countless man hours and many thousands of liters of kerosene ( jet fuel ) flying a professional sports team from one city to the next the opponents of renewable energy do not consider this energy wasted.

    But ten or twenty years from now nobody will remember much about the games that cost so much in energy and capital and talented manpower.

    If that capital and energy and manpower had been invested instead in a solar farm an EROEI of only three to one would look like a hell of a bargain.

    It has to be nukes or renewables or a combination of nukes and renewables or it means lights out sooner or later.

  14. Kit P says:

    I am going to do something that you will not hear someone from the solar industry. Brag about how my first commercial nuke plant is running 35 years later producing 10-15% more power than it was originally designed for and a plant life 25% longer than design.
    The fact is that solar does not work almost all the time. Making power on a nice sunny day is not a big achievement but keeping the power flowing on a bitter winter night requires lots of skill. Spain and California should stick to bragging about red wine exports.

  15. Phil Scanlon says:

    Hi Euan, the rating on a solar panel is not a theoretical maximum. It is a rating under standard conditions, which include taking into account the air mass the solar irradiation must pass through to reach the panel. From wikipedia’s entry on Air mass: “Solar panels do not generally operate under exactly one atmosphere’s thickness: if the sun is at an angle to the Earth’s surface the effective thickness will be greater… AM1.5 is useful to represent the overall yearly average for mid-latitudes.” Panels are rated (tested) by simulating AM1.5. Closer to the equator, where the AM value is less than 1.5, a solar panel can output more power than its rated power.

  16. Leo Smith says:

    I did attempt to if not get exactly TO the numbers correctly identify the SHAPE of the cost equations in calculating real levelised lifetime costs of various forms of electricity generation here:

    In essence there are various components to cost, and you have to be clear whether the thing you are costing is a randomly generated-when-it’s-there unit of electricity or a reliable available-when-you-want-it unit.

    My general point being that a unit of generated electricity that you cannot use is worthless.

    If you then take intermittency fully into account, and cost in the necessary technology to ameliorate its effects (you can’t get rid of them) and develop a model that includes the knock on costs of renewables, the picture for renewables looks a great deal worse than the popular headline figures of ‘solar generated so many TWh at such and such a cost’. In fact the cost of its FAILURE to generate on a regular basis has also to be added. And its propensity to generate far more on certain occasions than others, which reflects in high peak capacity links that are on average grossly under utilised.

    If a company were to announce that ‘our policy of taking on volunteer staff has resulted in them doing 30% of our work for free’ and failed to mention on the balance sheet ‘but we now have to have ten times the office space, because they don’t work very hard, and ten times the computer systems and cabling, to cater for them when they turn up, and we still have to keep our permanent staff on retainers for when they don’t and actually our costs have gone up 70%’ that company would be guilty of fraud and false accounting.

    However Renewable UK and government organizations are political lobbies and (like religions) are not required to adhere to the truth when stating their claims.

    Renewable energy on the nations balance sheet of electricity results in a very small profit in electrical terms and a huge cost reflected elsewhere into the system.

    It is even questionable as to whether it results in any reduction on CO2 emissions overall. IN a nation that has but little hydro to act as balance and all of that used to cover demand variations already.

    • Euan Mearns says:

      In my more recent post on Solar Scotland I conclude that Scottish solar over lifetime is always in energy debt. But here’s the rub. All of the CO2 emissions associated with creating the system are delivered to the atmosphere decades before the panels deliver the “CO2 free” power.

Comments are closed.