Solar Panel County

Via Tom Nelson and Schlotzville, comes this from our modern day Savonarola, Al Gore:

Gore mentioned a few statistics that drove home the notion that we actually have the capability to be oil free with existing technology. If, he said, we were to build on a 90 mile x 90 mile tract of land in the Southwest a field of solar panels, we would have enough electricity to power the entire United States. So, why don’t we build it? What is stopping us?

Its kind of cool to think about – I always get excited about man-made structures you can see from space.  When I think about this, my mind keeps jumping to the Sunflower fields in Larry Niven’s Ringworld.

So, is this feasible?  Well, I was suspicious, since I live in one of the best solar sites in the world (Phoenix) and could not even come close to making solar pay on my house, even with 50% subsidies. 

First, is it enough power?  Well, its turns out the answer is "sortof."  I looked around at solar panels, and decided to assume a 200 watt panel that was 13 sq ft and cost $900.  Actually, you can’t quite get that panel today.  You can get a 200 watt panel that is that cheap, but bigger, or you can have one that is that small and more expensive.  But you will see soon that it does not matter.  I assumed a third of the 8100 sq. miles would be dead space between the panels, roads, transformers, access paths, etc.  I assumed you put the installation in the best solar sites in the southwest, which yield on average about 6 peak-sun-hour-equivalents a day.  I assumed a 20% loss in conversions and transformers.

So 8100 sq miles  x  2/3  x  200 watt/12sq ft  x  6 hours x 365 days x 80%  (with necessary unit conversions thrown in) yields 4.08 billion Megawatt-Hours of electricity, which is about exactly our current US generating capacity.  (Way to go!  Al got a number right!)

I say sortof for the following reason:  This does not cover elimination of fossil fuels in the transportation sector.  And it does not address the problem of how you store this power at night, which of course is a catastrophic problem for the idea.

Al doesn’t know what is stopping us.  Well, other than the storage problem, one thing might be the cost.  Using the assumptions above, and assuming that installation costs (with land acquisition, transformers, inverters, roads, mounting, installation, etc) is as much again as the panel costs themselves, the total installation would cost just under $21 trillion dollars.  This is orders of magnitude more than a nuclear program of the same size would cost.  And presupposes the environmentalists would let you cover 5 million acres of desert with metal and silicon. 

Postscript:  Al Gore thinks its the oil companies at fault (of course):

Well, he gave one possible answer – the oil companies. Apparently, according to Gore, the oil companies drive up prices reducing supply and then depress them in a telling pattern. As soon as the political will swells to a light boil, the companies reduce prices/increase supply. And we, really the pols that be, fall for it all the time and the political will it is vanquished

LOL.  Environmentalists have one card to play – its the oil companies fault! – and they are going to play it every chance they can.  Of course, the boom-bust patterns in oil are characteristic of nearly every other commodity out there, which therefore presupposes that if oil prices are the result of manipulation, then every other commodity must be as well since their prices demonstrate the same patterns.  We see these patterns in commodities that politicians have never even heard of and in which they have never thought to exercise their "political will."  (political will in this context defined as use of government force against a segment of the populace).

A reasonable person might suppose that the surge in prices followed by a drop a number of years later is better explained by the time delay in increasing oil production after oil prices spike. In many ways, Al’s theory is simply delusional.  If your friend started trying to tell you, in all seriousness, that every action Microsoft takes is actually aimed at thwarting him personally, you would think him insane.  But this is effectively Gore’s argument, showing the immensity of the politician’s ego.  Oil prices move not because of supply and demand, but because of us politicians.  Every tick up and down is carefully managed to thwart us brave Congressmen!

I had a long post here on why conspiracy and manipulation can’t possibly drive oil prices but for the shortest possible periods.

Update:  Here is my spreadsheet if anyone thinks I made an error in the numbs.  Download solar.xls

  • I posted some questions at the Tom Nelson site that I think go beyond some of the stuff mentioned here.

    I’d enjoy reading comments about that.


  • Captain Obviousness

    And the $21 trillion doesn’t account for the maintenance costs – not just the cost of keeping the equipment in good order, but the cost of paying for the small army of workers it would require.

    However, a giant solar array in the middle of the desert would be an excellent place for an official GISS temperature station. Heehee.

  • It’s always interesting when numerate people look at this stuff. Engineers rock (okay, so I’m one too).

    Just to make sure my facts were still current, I did a little quantitative looking at current alternative electricity generation alternatives.

    For solar, I looked at home on-grid applications – the one’s lots of folks propose. This is a solar system with no storage, where you can put power into the grid when you have more than you need (and get paid for it), and pull power from the grid when you don’t.

    20 years ago, or so, you could do this at about break-even, but only because of enormous subsidies including regulations forcing the utilities to pay you retail for the power you put into the system, whether they need it or not.

    So… throw out the stupid incentives, because that’s what they are – things to misdirect the market.

    Solar power proponents say it becomes competitive $1/peak-watt of solar panel cost. Some companies claim to be close to that.

    Nope. It’s $1/p-w system cost, as far as I can tell.

    Today, wholesale costs of solar panels are around $5/peak-watt (1/2 of what it was some years back). Then you have to install them. Then you have to provide the inverters to turn that into usable power.

    Inverters cost $1/watt – the required system cost.

    So the solar panels and all other things have to be free!

    Now, I think inverters could be built in high quantity for $.30-$.50.

    But ultimately the panels still need to be free.

    Then you need to store the electricity…

    And as you point out, the real issue is that we need energy storage that is mobile – so we can use it in cars. Energy density is how much energy you can store at a given weight (technically, the units below are MJ/kg):

    Gasoline: 46.9

    Lithium Ion Battery w/ nanowires: 2.7

    Lead Acid Battery (car battery today): .11
    Ultra-Capacitor: .02

    So a pound of gasoline stores 17 TIMES as much energy as a pound of the most advanced battery, and 400 times as much as lead-acid batteries.

    Wikipedia reference

  • Just for fun, I took your spreadsheet and used a cost of $1/peak watt. Result: $4.54 trillion.

    The same power requires (if I didn’t screw up the math) about 372 122MW nuclear reactors (r about 124 nuclear plants equivalent to Palo Verde (Arizona).

  • kuhnkat

    Actually, you should look into the newer solar mirror farms. We have one in the Mojave that is producing at a reasonable output. By using a heat sink the power from the daytime can go past light hours. As mentioned, you still need backup for low output days/weeks and nights!!

    Al “the inconvenient moron” doesn’t consider the DISTRIBUTION issues either. Here in Cali we are still trying to deal with the issue of not being able to backup from the north to the south with our power output due to lack of transmission facilities!! Imagine all the trunks needed to be built to distribute power from ONE area to the rest of the continent!! How about how easy it would be to TURN OUT THE LIGHTS by hitting ONE power plant!!

    How about what the enviros would say when they realise 8100 SQUARE MILES OF PRISTINE DESERT ARE GOING TO BE COVERED!!!!

    Remember Kennedy and the wind farm off the coast in his sailing area!!

    Actually, with the mirrors and fuel cell and battery systems for mobile it wouldn’t be that bad of an idea. Except for the total cost of conversion of most of the country!!!!! Slowly will work just fine thank you!!

  • Raven

    You need to factor in the lifetime of the equipment as well. Solar cell efficiency drops as they age and will likely need to be replaced every 30 years or so. Assume an optimistic lifecycle of 50 years and you will find that $21 trillion capital cost works out to 400 billion per year – forever!

    It is also worth noting that despite billions spent in research fuel cell cars are still rediculously expensive. For that reason we cannot bet that the price for solar panels will decrease significantly over time.

  • Luis Dias

    Well, I can sympathize with your skepticism, and of course, it is the “market’s” fault, not big oil, for the lack of solar power massive investment. But I agree not with your conclusions, as solar power is definitely the most accelerating energy technology, is only ten years behind wind power and has a lot more potential. We have to be more foresighted to this. Some people in this comments talk about 1$/w total system cost and how only that golden value will get solar a big boost, but they aren’t being reality based.

    1 – Solar power is accelerating like hell, is the fastest growing energy economy in the world. Of course, it is tiny, so any acceleration doesn’t mean that much in the beginning. Think of exponentials and you get my idea.

    2 – Coal and Gas are reaching all time highs in price, due to stressing oil supply conditions and the commodity bubble. You can argue that it will pop, but given the very tensed world set of energy situation, I doubt it. They are just following oil, up up up. Soon, that 1$/w will be far surpassed by the coal industry itself.

    3 – New solar technologies are arriving that promise incredible results. PV printing, concentrated solar power, balloon coolearth systems, biosolar, etc., etc. These things just keep on appearing. The reason is that while in the nineties oil was at its lowest values, destroying any alternative markets, oil has surged the past seven years now, giving a boost to this kind of investments.

    It may well be needed 5, 7 years until these techniques get matured and mass produced. But anyone that doubts that solar power will be the energy of the future is deluded. Fossil Fuel price will be higher and higher, due to the simple fact that these are finite reservoirs, and one day we will simply not be able to grow its production by the amount we need, rising its prices. Arguably that is starting to happen.

    What are the limits of solar power?

  • IMHO, solar is attractive and it may become a viable source sometime in the future. Currently, it is not for the reasons discussed in other posts. My biggest concern is that solar, as it is envisioned, may be damaging to the environment. Concentration of solar panels into mega-farms in what we consider to be throw-away topography (deserts) could lead to problems for the following reasons;

    1. by capturing incoming solar energy, before it hits the ground, we may reduce overall temperatures in the landscape covered by the farm. I can’t see any way around this.

    2. reducing temperatures in these desert areas will, ultimately, alter the bio-diversity of the local environment.

    3. reducing temperatures on such a scale may also change weather patterns such that a cooler landscape allows more cloud cover, inuitively, this reduces the efficiency of the solar farm’s output.

    Transporting the energy to where it is needed and then converting it back to heat will cause an increase in temperatures in those places where the energy is used. While it does not cause global warming, since the net effect is zero, it does change the way weather works by disrupting energy economies (natural or environmental).

    As for the solar concentrator concept, it may provide a few more hours of energy output per day as Luis states. I’m unsure about the environemntal effects of operating what amounts to an open-air blast furnace day in day out for an extended period of time.

    Do I have any facts to back up these observations? No. Having done a quick search of the literature, I’m not sure that anyone has done any detailed studies to disabuse me of my thesis either. I’m prepared to change my mind.

    I think that we are barking up the wrong tree with wind and solar if they are being pursued as end point technologies. What we need is a new, clean, source of energy. While nuclear is an alternative, there is so much fear in the minds of the public that it becomes unusable. We don’t have it and I’m not convinced that anyone knows what it might be. Lubos Motl and the string guys might, but they aren’t sharing any details if they do.

    For now, solar and wind are niche technologies, they may remain that way, who knows. Even if you get the cost per kw hour down to below coal, you still have to have conventional thermal-electric plants idling, below peak efficiency, as a backup. So you still burn the coal or natural gas. How does that help?

    Should we pursue these technologies? In my mind, yes. Its still considerably cheaper to mount a solar panel to run your cottage on a remote lake than it is to have your local power company run 15 or 20 Km of wire. Should we expect them to be a near-term solution to conventional technologies, no, not for the forseeable future.

    So, I may be deluded as Luis posits, but it is always wise to look before you leap. The unintended consequences may outway the gains.

  • Rob

    Concentrating your entire power supply for the country in one area is a national security risk.
    Imagine how easy it would be for an enemy to cripple the entire country by hitting one target. Ouch!

  • Luis Dias


    I never intended for you to make a leap without looking. Nor am I making a case of the fact that solar will get out of our energy hole situation no-time, of course that in the mid-term future I am not seeing that happening. I see nuclear power getting a boom, gas and coal booming a little, and oil probably descending. Wind and solar are still a niche like you said. What I also said is that we have to be more foresighted. And in that case, both fossil fuels and nuclear power don’t really compute well in the longer road, because FFs are finite, and nukes are rather damned expensive and slow to start-up (almost a decade!). In contrast, solar fields are very fast to produce and place, and can be modular.

    Wind has a problem of NIMBY (Not In My BackYard), but not so with solar. People could put solar roof-tops and discard desert altogether, that’s a possibility. But I also don’t see that happening, for centralized stations are often more economical.

    Having said that, I find your arguments quite staggering. You claim enormous disastrous environmental repercussions, without considering the simple fact that we’ve been building cities in the desert for a long time now, that we’ve been altering vegetation (and therefore Albedo) of a great percentage of the planet. This kind of preoccupation is similar to those greentards that claim that anything mankind makes is wrong. Just because. I don’t adhere to that.

    “Should we expect them to be a near-term solution to conventional technologies, no, not for the forseeable future.”

    Well, some technologies are already promising a 5cent/KWh solar power, and that before 2010. It’s a bit stretched, and they can be wronged, but the curve is steepening for solar so much that they are already talking about such values. For me, it is a matter of time until solar gets a big boost. I’d guess 2015 we’re gonna be surprised at the investment figures for solar. Is that inside a “forseeable future”? If you think you can foresee things, I’d say yes. But I kinda like to be surprised.

  • Stop trying to solve the problem for the whole country or the whole world. Think locally and regionally. These conversions can be done incrementally, a bit at a time. But if you look at the huge total costs of converting the entire US or the entire world, forget about it. Nobody can solve all that at one time.

  • morganovich

    here’s another issue not raised in the above discussion that will make the idea impractical and is likely as big a deal as the inability to store the power:


    consider how much power would be lost to the lines and transformers so in sending electricity from arizona to new york. if half of it got there, it would be an amazing accomplishment. a more likely number is significantly lower.

    add to this that the power is only generated at peak levels for a few hours and at all for about half the day, and solar is just not going to be able to be a major part of baseline power until the ability to store massive amounts of power and transmit it with low loss is created. at present, it is, at best, a regional peak power provider. none of the big solar farms can yet compete with the existing grid sans sunsidy.

  • What would be the effect of changing a 90 mile by 90 mile region from .4 albedo to 0?

  • There are currently a number of large, operational solar power plants around the world. The third largest of which is in Nevada, USA, commissioned in 2007. It, like so many others around the world uses the Heliostat design. There are a number in the Mojave Desert that have, I am assured, a generating capacity in excess of 350MW at peak.

    Why don’t these get a mention?

  • it does not address the problem of how you store this power at night, which of course is a catastrophic problem for the idea.

    Use solar thermal, store the heat in a big pile of rocks.

    Solar thermal, done on a really large scale, is at present cheaper than solar cells.

    Nuclear, however, is still cheaper.

  • Doug

    “Sunsidy.” I like it!

    Has anyone ever run the numbers on a Tesla Motors electric car? I must admit to being a little perplexed by it all. The facts as I know them are that the car has 6831 lithium-ion cells, and that Tesla claims that the running cost is a scant 1-2 cents per mile, and it has a 250 mile range. I’m trying to do the math (out loud). Tell me where I’m wrong.

    A 250 mile range at 2 cents per mile means that the car consumes $5 worth of electricity per charge. At $.10/KWH, that means that 50 KWH is “pumped” into the cells for every 250 miles driven. A later comment in the article confirms this number: the charging source is a 220VAC/70A outlet, and charging takes 3.5 hours. That works out to 54 KWH. Close enough.

    A standard internal combustion engine with acceleration like the Tesla affords might get 15 MPG. Thus, to drive it 250 miles between fillups would require 250/15 = 16.7 gallons of gas. At $4/gallon, this would set me back $67/fill up.

    $5 per fill up versus $67 fill up?

    In a mathematical sense, the Tesla machine claims to have better than 10x the efficiency of a gasoline-powered car. Martin Eberhard of Tesla claims that a gallon of gasoline spinning a generator could ideally power a car for 110 miles. If this number is correct, then $4/gallon gas would translates to 3.6 cents/mile for an electric, but about 4/15 = 27 cents/mile for an internal combustion car. What has Eberhad discovered that countless others before him didn’t?

    If it were this easy, electric cars would have overtaken the internal combustion engine decades ago. Who’s jiving who here? If it’s this good, might a suitable energy storage reservoir for a solar farm be a bank of spinning flywheels coupled to Eberhard’s motors? (Yeah, I know, I can just imagine acres of spinning wheels…)

  • morganovich


    an electric motor is far more efficient than an internal combustion engine. THAT is not the problem with the electric cars.

    here’s what you are leaving out of your calculation (and i happen to know this precisely as i recently had dinner with a friend from college who works for tesla):

    the batteries in the TR cost $25,000 and are difficult to produce. you need to add this into the equation. there is a reason this is a $90,000 car. the engine is quite cheap and the size of a big watermelon, but the batteries are expensive and heavy.

    the back breaker is that they are good for about 200 charges. so, assuming you get the full 250 miles out of the car each run, you get 50,000 miles and the battery is shot.

    this implies about 50 cents a mile just in battery amortization. even at 15 miles a gallon, this implies a $7.50 cent/gallon gasoline price just to cover the battery (and you haven’t paid to dispose of it yet, and that can be nasty)

    that is the problem with electric cars. the batteries wear out. the batteries in the TR are essentially a large number of the same cells you have in a laptop battery linked cleverly with software to manage them. ever try to charge a 3 year old laptop battery and take it on a plane?

    assume only 200 miles/charge (still generous as you don’t want to get stranded and won;t always run it all the way down before wanting a long trip) and a 20 MPG car and the pump price equivalent of battery degradation is $12.50/gal.

    and you haven’t even paid for the electricity yet.

    so i’m going to stick with the stuff from stuttgart for the time being…

  • Doug

    Thank you, Morganovich. Your numbers definitely compute: $25k for 6831 batteries is about $3.66/cell, which sounds about right. However, they have to be matched in their characteristics. I believe your $25k figure could be low, but it’s one I can work with.

    Yes, I know a great deal about using Li-Ion batteries, working with them on a daily basis. In my work, we deliver small amounts of power, so the battery life is more like 500 charge cycles. However, I suspect that in a high-power automobile environment, with temp extremes and much higher discharge rates, the life could be about half this. Which also brings up another interesting point: these batteries can only be charged in a narrow temperature range, usually 0-50 degrees C (-32 to 122 F). Imagine being unable to charge because it’s too hot or too cold!

    I also thought about the charging. Imagine explaining to your local power company that you need to up the power line capacity into your house to provide 220V * 70A = 15.4 KW of power delivery for a few hours. You can’t do this in a standard residence without some serious changes at the power pole — I can see the need to rewire the entire neighborhood to deliver such power.

    Current and voltage into the battery must be limited by electronics to avoid explosions. The electronics to convert the power line voltage into a DC one at these wattages is no trivial exercise either, driving the cost even higher.

    Let me then follow up this question with another: California, where I live, refuses to build any more power generation facilities. The NIMBY complex is intense here. Consequently, when I charge my Tesla, I have to buy that power from Arizona or Nevada. If I need 50 KWH delivered to my Tesla, how much power must be generated at the power plant to cover for the power losses in the cross-state power lines? I’ve never heard how inefficient it is to move energy across the power lines. Anyone know that one?

    (Sorry if I’m a bit off-topic, but I am SO tired of hearing about “the realities of alternative energy sources,” including electric cars, when it’s clear that the various analyses are never complete. And no, we are NOT running out of oil; we are simply doing such a great job of restricting supplies in the name of wacko environmentalism that it appears that way.)

  • morganovich


    i live in CA as well, so i feel you. the great gray davis blowout of the whole rainy day fund for a few months of subsidized power was one of the great acts of modern idiocy.

    power loss, even at very high voltages, is about 30% as a minimum figure for a “long haul” like AZ to CA assuming you just run wires and don’t have too many transforms. we are still using the same basic power line tech that has been around for 80 years. i’ve seen a couple of companies purport to have a better solution using nanotech or what have you, but they have all turned out to be either smoke and mirrors, or much too expensive.

    but again, the killer is not the cost of the power going into the car, it’s the cost of storing and releasing it.

    i’m going to be very interested to see how the pruis owners feel when they get about 80k on their cars and their batteries die. they may wind up a bit less enamored of them…

    the other back breaker on electric cars is charging time. are people really willing to put up with 4-8 hour charge times? (and that’s from full 220) i have seen some new battery technologies that allow a quick charge (35kwh into a car in about 8 minutes or a full cell phone charge in 20 seconds) but they are wildly expensive ($70,000 for a car) and not yet producible in quantity. and, as you rightly point out, whose house can provide that kind of current? how many home builders are even qualified to work with that kind of power?

    i have long said it will be very simple to tell when an alt energy becomes more efficient than existing systems because it will spread like wildfire. people are pretty good at figuring this stuff out.

    the one boon of high oil prices is that lots of people are working on alternatives. eventually, somebody will figure something out and the market will take care of getting the idea propagated when it comes so long as we can keep the government out of the way. clearly, they have a less than less than stellar track record for mandating innovation and a nasty habit of trying to get the cost curves to cross by making existing power more expensive…

  • morganovish,

    If the other problems with batteries (cost, weight, temp range) went away, the charging could be handled by a battery exchange service. You drive to the charge station, where a machine removes your battery and puts in a charged one. Voila – charge time problem gone.

    To bad that the other if’s are so huge, but they are. Electrochemical batteries just don’t hold enough energy/kg to do the job. The best have a tiny fraction of the energy density of gasoline. So unless you can make up for the rest in the weight of the IC engine, transmission, etc, you still have a problem.

  • braddles

    Don’t forget that you are going to need much more than the supposed 8100 sq mi if you are going to use the collectors efficiently. To get the maximum 6hours equiv sunshine on each panel, they will have to be spaced out so that each one does not cast a shadow on its neighbours in the morning and afternoon. This spacing will have to be big if you want good performance in early morning or late afternoon.

  • braddles

    BTW, transmission losses are a bit of a killer for long distance electricity (5-10% per 100 miles I think). If it wasn’t, nearly all coal-fired power stations would be located next to coal mines. With remotely-located coal mines (Wyoming) it is more efficient to put the coal on trains and carry it to distant power stations than to transmit as electricity.

  • See ultra-capacitors.

    Likely what you will see in the future is a combustion engine with a battery and an ultra capacitor or two. My guess would be diesel. I think stanford has gotten over 40% efficiency out of diesel engines. We will pobably see the transmission, transfer, and drive shafts replaced with a generator, wires, and an electric motor to drive each wheel (people are also doing interesting things with permanent magnets in motors and generators).

    Reason you will see multiple power storage in conjunction is that most have a weak spot where they can’t deliver the power needed (you just can’t accelerate when needed), a second source is needed to cover an gaps in performance.

  • morganovich


    i don’t think battery exchange will ever work. it’s just a totally unfeasible business model.


    even if batteries get really cheap, you’re still talking about a couple thousand dollars for a car sized battery. for the sake of discussion, let’s assume they get 80% cheaper to $5000.

    a recharge station would need enough batteries to fill all the cars that come in in a day. consider the inventory implications. you’re still looking at millions of dollars of inventory that degrades as it goes along. the up front costs of starting a charge station would make them pretty uncommon relative to a gas station. further, unlike gasoline, batteries are easy to steal. slippage would become a big issue.

    and this assumes that everyone uses the same battery. cars have different form factors etc. the batteries in the tesla are spread all over. this makes sense from an even weight distribution standpoint. so quick change outs are unlikely and stations may need to carry several kinds of batteries which makes the whole biz model worse. moving a several hundred pound (at best could easily be 1000’s) cannot be “self serve” either. it will take automation and a degree of standardization that seems unlikely.

    batteries are big, fragile, and heavy as well. swapping them out will be impractical. high speed charge is the way to go, but that is much more expensive still.

  • joshv

    Re: Battery charging issues

    Why can’t we make modular batteries that are swapped out in a semi-automated fashion at a filling station? You’d have a battery composed of smaller, identical cells – maybe 25-50 of them depending on range. Filling stations would purchase and charge the individual cells, and recycle worn out cells.

    To “refuel”, you’d pull in to the station, your spent cells would be offloaded, freshly charged cells would be loaded, and you’d be fully recharged in about 5 minutes. Granted the mechanics of automating this would be tricky, but it could be done. You’d have the added benefit of providing for some sort of emergency battery eject if one of the cells decides to make like a laptop battery and ignite.

    You would not have to purchase your battery, the filling stations would purchase the cells, and amortize the cost over the life of the battery, charging an added fee for every “fillup”.

    The filling stations will charge the cells using renewable sources, or perhaps off-peak grid power.

  • Luis,

    You said that I “claim enormous disastrous environmental repercussions, without considering the simple fact that we’ve been building cities in the desert for a long time now, that we’ve been altering vegetation (and therefore Albedo) of a great percentage of the planet. This kind of preoccupation is similar to those greentards that claim that anything mankind makes is wrong. Just because. I don’t adhere to that.”

    I made no such claim, I simply said that it could happen. Like you, I don’t adhere to the thesis that everything we do is bad. I am realistic enough to recognize that some of the things we do are wrong, it is those things that we must avoid. Likewise, we should try to avoid creating situations where things go terribly wrong.

    Yes we have been building cities in the desert for a long time, that doesn’t make it the right thing to do. Note that I didn’t say it was the wrong thing to do either. Las Vegas comes to mind, but I don’t think that we have come close to altering the vegetation of “a great percentage of the planet.” Deserts don’t account for that much landscape.

    As for my “preoccupation” what is the sense of creating more damage than one has to to solve a problem? Look before you leap, if there is no problem, now or in the forseeable future, then go for it. If we are not cognizant of the possible negative consequences, the “greentards” will have a better arguement against the plan. Don’t give them any more ammunition than they already have.

    Solar, if it has a future, is a local solution. Put the collectors closer to the end users.

  • joshv

    @morganovich – replying to you because john posted an idea similar to my own.

    “Even if batteries get really cheap, you’re still talking about a couple thousand dollars for a car sized battery. for the sake of discussion, let’s assume they get 80% cheaper to $5000.

    a recharge station would need enough batteries to fill all the cars that come in in a day. consider the inventory implications. you’re still looking at millions of dollars of inventory that degrades as it goes along. the up front costs of starting a charge station would make them pretty uncommon relative to a gas station. further, unlike gasoline, batteries are easy to steal. slippage would become a big issue.”

    The 1000lbs of batteries required for a car would not be easy to steal. The batteries would of course incorporate some sort of “DRM” so that you yourself could not charge them. There would of course be pirate chargers, but I don’t think they would be widespread enough to be much of an issue.

    As for the cost – that’s what loans are for. Perhaps the government could get in on it at first, but once the concept proves itself you’d have no problem finding someone to loan you money for batteries. Either the batteries make money, or you sell them – it would be nearly impossible to lose money.

    Also, your inventory does not “degrade” it’s continually swapped out with inventory from other stations. The entire circulating inventory would quickly reach an equilibrium distribution. Assuming the batteries refuse to be charged past some charge cycle limit, filling station operators would be force to regularly buy new modules to keep the overall inventory fresh.

    “and this assumes that everyone uses the same battery. cars have different form factors etc. the batteries in the tesla are spread all over. this makes sense from an even weight distribution standpoint. so quick change outs are unlikely and stations may need to carry several kinds of batteries which makes the whole biz model worse.”

    Simply solved with a modular battery composed of 25-50 identical, standardized cells. If a manufacturer wants to put the batteries in multiple locations, well that would be an inconvenience for swapping out batteries, but for most car designs I don’t imagine creating a single battery pod with an external hatch would be that big of a deal.

    “moving a several hundred pound (at best could easily be 1000’s) cannot be “self serve” either. it will take automation and a degree of standardization that seems unlikely.”

    As they are modular, each cell would weigh on the order of 20lbs or so. One could imagine any number of standardized positions for the battery compartment that would allow easy loading/unloading with a semi-automated device. You might need an attendant to operate it, but that does not add that much extra cost,

    “batteries are big, fragile, and heavy as well. swapping them out will be impractical. high speed charge is the way to go, but that is much more expensive still.”

    I don’t even want to think about the changes required to the power network to support ten cars at one station fully charging a Tesla sized battery in 5-10 minutes, let alone the risk of operating such equipment near customers. With a removable battery, changes might still be required, but the charging of the stations battery inventory could be spread out over an entire 24 hour period – most likely though concentrated on off-peak hours when the power is very cheap.

  • SunSword

    I have a different set of calculations. I base these on the costs and output of a relatively new (went online in July of last year) solar power facility called Nevada Solar One. Cost: $220 – $250 million. Output: 64MW. Land used: 400 acres. Therefore:
    90×90 miles = 8100 square miles.
    1 square mile = 640 acres. Therefore:
    8100 x 640 = 5,184,000 acres.
    5,184,000/400 = 12,960. Factor in your 2/3 and we get 8640. Let’s round that up to 8,650. That’s how many plants we build in our 90 by 90 mile footprint.
    8,650 x 64 MW = 553,600 MW. Cost would be (using the low number in the range of $220M) 8,650 x 220M = 1,903,000 million, or 1,900 billion, or 1.9 trillion dollars.

    So — my calculation shows about 1/7 of current US generating capacity, but at < $2T. So more efficient than your calculation. That said, all comments related to limitations to peak solar efficiency and cost of high transmission lines are all valid. The Nevada Solar One does heat salt to enable some power generation at night but I lack details on efficiency and amount of power generated by this technique.

  • Samuel Pickwick

    The storage problem kills the idea stone dead. You simply can’t store that much power for use during the night in batteries. Suppose we use those super batteries mentioned by John Moore that store 2.7MJ/kg. The US energy usage averages at about 3.3 TW which is 2.8×10^17 Joules in a day. That’s the ballpark figure of what we need to store, so we need about 10^11 kg of batteries!
    Nuclear power is the only answer.

  • morganovich


    you’re not thinking about this like a business. no one is going to open a battery exchange. the up front costs are massive just to get all the batteries and to keep them up to grade.

    by degrade, i don’t mean all move away to other stations. by degrade i mean they will stop holding a charge and need to be replaced. this is VERY expensive. see my comment above on gas price equivalent of battery amortization.

    the batteries in the tesla weigh about 2000 pounds. let’s assume some development can cut that in half (double energy density). 1000 pounds of batteries to be moved for each car? 50 20 pound units? go to the gym and move 50 25 lb plates from one place to another then back again. excited to do that every time you refuel? oh, and do it very carefully, as all would need to be seated and connected properly and secured so they do not move or shift, especially in a crash where the last thing you want is 50 20 pound projectiles… some would break, connections into the car would break, this is a very difficult task. it’s not like putting a bunch of batteries in series into a flashlight. the power management systems are very sophisticated. there is not a system anywhere near existence that could make 50 modular batteries work together like this and discharge evenly. the software to manage the power discharge in the tesla roadster was the big breakthrough. it would never work with batteries of varying quality (number of uses etc).

    and 20 pound batteries are incredibly stealable. DRM would never work. one, it will get cracked. two, who is to stop me from just putting it in my car and driving to a station to charge it there? are you proposing to link every station and have them compare battery serial numbers in real time as they unload them to make sure they aren’t accepting bad ones? the logistics of that are mind boggling. 50 times every fueling?

    this leads to another issue. if i show up with a nearly worn out cell, will the charger want to take it? don’t they have lots of incentive to just pass it on? the same moral hazard applies to consumers, who, as they don’t own the batteries, will have no incentive to take care of them.

    further, using small cells will increase weight and space, not reduce it. so you are likely loading a ton of batteries. 4000 pounds moved per fill up. who takes liability if you drop one or worse damage the contact points in a car? and who is going to load this stuff? and take it in back to be charged etc. the wear and tear from this would far exceed any benefit.

    it also prevents people from adopting new, better tech. why would anyone buy new and improved batteries when they are just going to give them up in 250 miles? why would any station offer them when they are just going to go off down the road? maybe there could be a couple of grades, but then you are just jacking up the inventory problem to new levels.

  • Doug

    God, I love this readership — it is SO refreshing compared to the mainstream press, where all we hear is the lunacy of how CO2 is “the primary driver of global warming.” At least here we get a good, healthy dose of back-of-the-envelope calculations. Engineers and scientists having a good, honest exchange of ideas, and most of the time it’s damned civil. Thanks for the forum, Warren. I love it!

    I did a quick calculation for the fudge factor required to optimally place a bank of solar panels that track the sun’s angle. In 24 hours, the sun strikes a 360 degree arc. In a 6-hour period, the sun will move +/-360 degrees * 3/24 = +/- 45 degrees from a vertical (normal) line, which each panel has to track. If you do the math, you find that a panel with dimensions X must be placed SQRT(2) * X to avoid shading any of its neighbors, side-to-side. The other axis distance is a function of the normal inclination of the sun on an annual basis. That calculation is regionally dependent, but it would not be too out of line to assume a maximum tilt of 45 degrees, which yields the same fudge factor. Thus, a panel with an area of X x X needs an area of 2*X^2. So back to our 8100 square miles of panels exercise: you need to double the amount of panel real estate (which still does not count for road access and the like).

    Good lord, Morganovich: 35 KWH in 8 minutes? That’s 263 kW/minute. Surely you mean 35 KW for 8 minutes, although that translates into “only” 4.7 KWH. 263 KWH would be 1200 amps @ 220VAC. You’d take down the entire power grid with each charge. This would surely be one very large extension cord required to connect to your garage outlet! Me: “Why are the lights in the house going dim? Do I smell ozone?” Wife: “Not to worry. Morganovich just got home and is recharging his car. Dinner will be delayed until he’s done. That is, unless Jones on the other side also decides to do the same…”

    All of this just goes to show that it’s hard to beat gasoline for energy density. Ditto for nuclear. Anything else is simply pie in the sky, and will be so for decades to come.

  • morganovich


    no, i actually mean 35kwh.

    that is about the power you need to drive a car 250 miles.

    here’s the tech:

    the problem is that they cost $75k each.

    do they work as advertised? i don’t know.

    can an infrastructure be put together to charge them this quickly, maybe. i’m not really sure.

    but i have to believe that if these batteries could be made affordable (a massive if) that solving the charging station problem would be manageable through some determined engineering.

    i seriously doubt you could charge that quickly at home with 220v power.

  • morganovich

    oh, correction:

    you get 130 miles on 35kwh, 250 on 70.

    they are getting sub 10 minute charge times (told me 8 when i saw them at a conference) using a 480v charger.

    no idea what the 70kwh version costs…

  • Doug

    How would the existing power grid support such massive amounts of energy? California invokes these “spare the air” days whenever the weather gets hot, trying to keep the power use down to some manageable level so that we don’t suffer rolling power outages. (Such is California’s eco-nut wisdom for not building any more power plants in the state.) Now add a million electric cars on the grid, sucking additional bursts of gigawatts of power. Yeah, right.

    These numbers verify my past suspicions: a state filled with all-electric cars, as California thinks would be a great thing, would be one whose power grid would immediately collapse. A hybrid vehicle works only because it off-loads the power transfer from the power grid to the gasoline tank of the car, taking advantage of the massive energy storage capabilities of gasoline.

    There’s no way this all-electric car stuff is for the masses. It won’t happen in a century, much less decades. Gasoline is a far better method of energy storage and transfer, by at least an order of magnitude.

  • burnside

    Regarding storage technologies and the discussion of electrically-powered vehicles, MIT’s Technology Review reports some recent and relevant design developments today at their website:

  • Using higher voltage speeds up charging and improves battery life.

    You don’t need to swap batteries, you just need a bunch of electricy and the right transformers.

    But batteries that you can both charge quickly at stations on the go and charge slowly at home would be a bit more difficult to engineer.

  • Doug

    Aaron: you don’t know what you’re talking about. Higher *power* might be used to speed up charging, but with the power and time levels that already exist, such increases are to the *detriment* of a battery. At best a Li-Ion battery can be charged at 0.7 times its C-rating. Beyond that, you boil the chemicals in the battery into useless goo.

    I have no idea what you mean with your second sentence, and I don’t think you do, either. Do I go down to Home Depot and buy “a bunch of electricity” in bulk? I put a transformer …. where? And …. why? And this eliminates the need to swap batteries …. why?

    Sentence #3 is also gibberish.

    Here’s the battery you’re dealing with in a Tesla, which is probably typical of other EVs, that you think you can alter or adapt:

    Not shown, I believe, are the special cooling cage to keep the batteries at 35 degrees C (for reliability) and the charge system. Tell me how you would build your ideal system. I’m dying to hear your plan.

  • Mrs Veronica frankling

    Dear Sir,
    I will like to purchased Solar Panel in your store please am waiting for your reply as soon as possible………..

    Mrs Frankling