Can someone explain solar power pricing to me?

What I\’m looking for is a simple, handy, metric, which would tell me when solar power was actually cost effective.

Forget subsidies and carbon costs.

OK, so I see around the place that solar panels (panels note, not cells themselves) cost $1.40 or so per watt at present. (A$, US$, no real difference, call this £1 per watt).

There\’s on average, 12 hours of sunlight a day but cut this to 8 hours just because. 365 days in a year, so, 3,000 hours a year this 1 Watt cell, costing £1, is producing 1 Watt/hr.

So we\’re getting 3 kW/hr from our £1 cell over the course of the year.

1 kW/hr costs, from coal and the grid, maybe 10 pence.

So we get 30 pence worth of electricity from our £1 installation, we get that for 20 odd years, umm, hang on, I\’ve just shown that solar is hugely profitable right now.

So, what have I got wrong there?

So, clearly, I have got something wrong.

What I actually want to know is, what is the price for a solar panel which makes solar power comparable with coal/grid electricity?

$1 per watt? 10 cents per watt? What is that magic number that makes people want to run, without subsidy or consideration of carbon costs, solar power at least when the sun shines? So this means leaving out any battery systems, storage systems.

What capital cost for a solar panel makes it work?


Update: OK, so we\’ve established that I\’ve got the numbers completely wrong. OK, great, now, could someone tell me what the actual answer is?

What is the price per watt of solar panel at which solar PV becomes price competitive with coal \’leccie from the grid? That\’s actually what I want to know, after all.


16 thoughts on “Can someone explain solar power pricing to me?”

  1. Please Tim. Can you get the units right.

    kW/hr means to anyone with a grasp of anything KiloWatts PER Hour.

    You mean kWh, meaning KiloWatts TIMES Hours.

    Energy consumption is almost always quoted in kWh which is I’m certain what you mean, but it is such a basic error and displays such a basic to grasp the very basic maths that it really discredits an article when this mistake is made.

    Please correct throughout and then delete this comment.

  2. I can’t believe for even a millisecond that a 1 kWh installation would only cost £1000. If it does, I’ll take four please.

  3. I’d imagine your first problem is with the figure 1watt. That’ll be the theoretical output quoted by the manufacturer. Even at the latitudes we both live, on a clear midsummer day at noon, you won’t get anywhere near that.

    Secondly, where are you getting the £1/watt figure? Is this the cost to a PV farm of panel or installed cost to a user? Cheapest I can see for the latter is 3 times that for a feed-in installation (so there’d be no battery necessity).

    Thirdly, that 20 year figure looks very hopeful. The catalogues I’ve got do guarantee for 20 years but I’d love to look at the guarantee terms. I suspect that these things will de-rate over time like most anything else does so the actual power output over the service life’s going to be a descending curve. The shape of that curve is critical.

  4. £1 per watt = £1 000 per kW.

    3 000 kWh per year would then cost £1 000 or 33 pence per kWh.

    Amortised over 20 years (60 000 kWh) would be 1.67 pence per kWh.

    So far so good… but …

    An average domestic house needs a 14kVA supply from the grid, although this can vary depending on circumstances, so to provide for a similar maximum loading, 14kW, the solar cell cost would be £14 000.

    So to answer the question.

    This makes the final amortized unit cost 23.38 pence per kWh – thus nearly 2.5 times the cost of coal/gas..

    That also assumes all the solar electricity generated is used.

    Since the Sun shines most during the middle of the day, then the output would be highest at that time, yet in fact electricity requirement is at its lowest during the middle part of the day.

    Being generous, it is likely that no more than 50% would be used meaning the actual per kWh cost would double – thus about 46 pence per kWh.

    It is why solar power is so expensive unless it can be stored or sold to the grid at an inflated price.

  5. In amplification of that third point, it occurs that if the price/efficiency of PV continues to move in the current direction at present rates it won’t be long before the advantages of putting in the new stuff as against buggering about maintaining/repairing the old, bearing in mind there’s limits on the space available to mount the things even for utility companies, will be a factor.
    All in all, taking a ten year service life might be more realistic.

  6. “What is the price per watt of solar panel at which solar PV becomes price competitive with coal ‘leccie from the grid? That’s actually what I want to know, after all.”

    As the saying goes Tim, you need to get your head round what Pedant-General said.

    There is a distinction between power available (potential output) W or kW – units of measurement – and power delivered (actually consumed) Wh or kWh – units of consumption.

    If the grid charges 10 pence per kWh, then 1 W is sold for 0.01 pence for every hour it is consumed and the unit COST does not vary irrespective of how much us consumed, although the unit PRICE of consumption might vary depending on tariff.

    To compete with this, a solar cell must then provide 1 W for one hour at a cost lower than, 0.01 pence.

    That must mean a one watt solar cell must cost less than 0.01 pence.

    However because solar cells are a paid, fixed cost, the actual unit cost of production = the unit cost of consumption and will vary dependent on consumption.

    Household A which consumes say 45 000kWh over 20 years from the cells given in the example, will have a lower unit cost than a different household using only 30 000 kWh over 20 years.

    So to cover all eventualities the unit output cost of solar per cell must be lower than the output cost from the grid.

  7. Your cell in a year will run for 365 times 24 hours = 8,760
    If it delivered 1W all those hours, it would give you 8.76 KWH valued at 87.6p
    It won´t, because of day / night etc and because of clouds. I would guess 0.1W would be more realistic level of average output 24/7. Your one cell then makes you electricity valued at about 9p per year. Apply the discount rate of your choice to decide the capital equivalent of a flow of 9p per year. This gives you the break-even point on a back-of-envelope basis assuming nil ongoing costs and nil capital costs apart from the cell. To do it properly you need to make realistic assumptions about maintenance costs (and deduct them from the 9p/year) and ancilliary equipment costs (add them to the purchase price of the cell).


    Tim adds: OK, but at those numbers then £1 a W sounds like a good deal (ignoring maintenance and any other ancillary costs).

  8. Tim: “What is the price per watt of solar panel at which solar PV becomes price competitive with coal ‘leccie from the grid?”

    It has been established in previous comments that the cost of solar power has to include the cost of storage and/or transmission from places sunnier than northern Europe in order to maintain continuous supply. We know that battery development (energy per kilo, recharging rate) lags behind development of PV cells. So we’ll be able to generate electricity from solar at economic rates pretty soon, but we won’t be able to store it. Solar power will soon become an interesting niche source — if you only need electricity in the middle of the day it could be your primary source (eg solar powered air conditioning for offices).

    Remember as well that we need to recharge all of the electric cars over night when the sun isn’t shining. This is all obvious stuff.

    Essentially, Tim, you have asked a question for which there is no absolute answer at this time. Solar power, as it is utilised today, is not a continuous source and is not a direct replacement for traditional sources. It is something that you use “as well as”.

    How much does a business pay for electricity during working hours? 10p per kWh? If you can make electricity yourself for 8p or 9p per kWh, recovering the capital cost in five years, it would be worth opening a spreadsheet.

    8 hours per day multiplied by (your usage) N kWh multiplied by (nominal current cost of electricity) £0.10 multiplied by 5 years multiplied by capital cost (say 1.06^5). Add on the contingency cost of buying in power on cloudy days. That’s your budget for home brewed electricity for five years.

  9. A few misconceptions to clear away: First, residential solar is almost always interconnected to the grid. When it produces most strongly, excess power feeds the grid and net metering is used to credit the homeowner for the surplus generated in the middle of the day.

    Second, solar modules produce creditable amounts of energy even on cloudy days–up to 30% of rated capacity. It’s not a total loss on a typical British day.

    Thirdly, the cost of a solar module has been dropping for decades, and we are now at that embarrassing point where the actual module is only a quarter of the cost of an installed system. The inverter, which converts the DC output of the system to usable AC, costs about $2k in the U.S., depending on make and size of system–and large systems use multiple inverters. The panel, conduit, connection widgets, etc., plus the installation and interconnection, all add a hefty chunk. As more and more systems are now financed via third party leases, the cost of finance has become a very relevant percentage of total costs.

    Grid parity for residential solar is a fluid concept, and highly dependent on government support at this time. Think of bubbles on a map. The bubbles are of different size depending on insolation, utility rates and government support, as well as the actual cost of solar. All of those factors change frequently, if not regularly. The solar industry has done quite well at lowering the cost of solar installations. Utility rates have had the unfortunate tendency to rise. Both of those factors increase the size of the bubbles. Insolation fluctuates slightly with climate patterns, but is more or less a given. Government support has been fickle in most areas.

    In the U.S., if government support is maintained at current levels and if utility rates increase in line with the recent past, those bubbles will expand to include about half the residential population by about 2015.

  10. There is a nice cafe near me that has it’s sizeable entire roof area covered with cells and a lovely power meter display on the wall. About 10 years old I think.
    I have never seen it produce more than 60% capacity even on the brightest summer days. Cloudy day in Spring and they couldn’t run 100w bulb so less than 1%

  11. Solar power will soon become an interesting niche source — if you only need electricity in the middle of the day it could be your primary source (eg solar powered air conditioning for offices).

    Yes, if you’re taking a UK-centric view. On the other hand, for the parts of the world which are warm, summer daytimes are peak electricity demand times.

    Remember as well that we need to recharge all of the electric cars over night when the sun isn’t shining.

    Why? Most people with cars leave them parked during most of the day, whether at work, at the train station, or at home. If the grid were to switch to solar PV, then electricity prices could be varied to incentivise daytime recharging (and, by extension, to incentivise car park owners to include recharging in their services).

    This would make life more expensive for the small minority of car drivers who are sales reps / estate agents etc and do use their cars all day and would need to charge them at night, but that’s supply and demand for you.

  12. Interesting returning to this post to see how you’re getting on with it, which apart from Al @ 12 seems to be essentially nowhere. Your figures are meaningless. Especially as mostly you’re discussing domestic installations.
    Show you why. I’ll quote you for installed cost for one watt PV, your house: £257.00 plus VAT (that’s 1 PV panel from Maplins,£5, good deal it actually rated fro 1.5W & 10m of 2 core – rest of it’s me playing silly buggers up a ladder). If you’ve a 10,000 m2 warehouse with a steel panel systems roof you want to cover we can do a whole different deal.
    What you need is a typical installed cost. (We’ll presume you’ll want a roof installation because few people have enough spare land they want to clutter it with panels.) Run of the mill UK 3 bed semi, plain pitched roof, you’ll get 12 1.5×0.65m (1 m2)panels up there. The catalogue I’m looking at is a year old & the panels are rated 120W per so in very round figures 1.5KW for the array. From what I see company’s are quoting £4500 up.
    Of course, distinguished company like this, we may be talking more capacious properties but that doesn’t necessarily imply more usable roof area. Architects don’t get their jollies on big houses with plain pitched roofs. They look boring so they stick all manner of hips & gullies on. Likelihood you’ll have trouble even getting a dozen panels to fit.
    OK, now we’ve some sensible figures to work with.
    Grid price 10p kW/h? 8h x 1.5kW x 365 = 4380kW/h or £438 at grid rates. Wow. That’s payback in the eleventh year.
    Now the hitches.
    The 120W figure is pure & unadulterated bollocks. It’s the maximum current peak the unit’s rated for, not the produced power.
    Let’s start folding in some practical problems. The UK’s at 50degN+ so not exactly the equator.
    Not all roofs orientate due south & few are at optimum pitch so unless you fancy tripling
    the installation costs with complicated firrings you’ll need to do simple trig to work out the divergence from optimum radiation intersection.
    Cable routes from the panels to the inverter can be quite lengthy. Low DC voltages result in high amperages. [email protected]=59A. That’s either some serious cable or high current losses.
    Go crawl around on a roof for half an hour & you’ll come down filthy. That’s dust, birdshit, pollution deposits, algal growth…An uncleaned roof window can reduce light transmission by 25% so what does it do to PV panels? (Unless you’ve a good head for heights you may have to pay to have them cleaned periodically. Reckon on a couple of hundred quid a time.)
    Put that lot together & it might explain why in my experience solar generally seems to produce about 50% of its claimed output.
    Revise our figures & we get 750Wh from the typical installation. Trouble with 750W is that it’s not much use for anything. Wouldn’t boil a kettle or power the microwave. Just about cover the fridge/freezer & a few bits & bobs. Of course if you’re feeding it into the grid & getting 40p/KWh feed in tariff that’s a different matter but read Tom Fuller’s post.
    “…if government support is maintained at current levels and if utility rates increase in line with the recent past, those bubbles will expand to include about half the residential population by about 2015.”
    No they won’t. If the feed-in tariff is passed on to the energy consumer as at present, at some point around 2014 the Minister for Energy is likely to get himself lynched by the folk who haven’t got a roof to stick solar panels on & who’ve been watching their electricity bills triple. In other words the low income groups, inner city dwellers & people whose streets inconveniently run north/south. (If it’s still Chris Huhne, every cloud has a silver lining.)
    All in all, I can’t see how domestic PV is ever going to be viable unless the efficiency of the panels rises by several fold. Reduction in the cost of the actual photovoltaic materials isn’t going to make much difference because you’ve still the fixed costs of the rest of the panel assembly & installation.
    There’s another thing. The installation costs will rise. Currently PV’s going on buildings where it’s easy to install. Generally detached 1 & 2 storey. The further into the housing stock you want it to penetrate the harder the installation. Nobody in their right mind is going to try tackling a Victorian terrace off of a ladder, health & safety rules aside. Scaffolding one could set you back £2000.

  13. In general, the domestic solar thing is a bit of an irrelevant sideshow, for more or less the same reasons we don’t all have diesel generators in our sheds.

    Unless you’ve got either a dedicated out-of-town site or a large industrial/commercial building, domestic solar is only ever going to be window-dressing, probably marginally viable for a fair amount of new-build and major renovation (since there’s a dude on the roof with a latter and a dude downstairs hooking up the wires already), and almost entirely pointless for most retrofit.

    The important thing here is to not let the fact that domestic PV is fairly meh obscure the fact that large-scale PV is going to become viable without subsidy in large swathes of the world within this decade. Which is good news for everyone, green or not, except for those in the hydrocarbon and nuclear industries…

  14. Pingback: Trying again on solar power pricing

Leave a Reply

Your email address will not be published. Required fields are marked *