No Script, No Fear, All Opinion
RSS icon Home icon
  • EEVblog #632 – Solar Roadways Are BULLSHIT!

    Posted on June 20th, 2014 EEVblog 48 comments


    Dave debunks the Solar Roadways project that generated over $2.1M on Indiegogo, as well as almost another million dollars in US government funding.
    In this video Dave does the solar generation calculations to see if this system is viable from an energy generation viewpoint, and also how much power the LED’s might typically consume.
    What are the losses in a typical practical system like this?

    Datasheets & Links:
    Sunpower E18 Solar Panels
    Dave’s solar system data
    John Deere 3038E tractor
    Philips Luxeon Rebel LED
    Thunderf00t’s videos HERE, HERE, HERE, and HERE

    Forum HERE

    Be Sociable, Share!
    • http://www.skybeam.ch/ Rainer Meier

      Thanks for the video; it’s just a perfect summary of my arguments when I get occasionally asked about my opinion on such a thing. So now I can just point them to your video.

      Just a note to the signal LEDs: I think the problem of wasting most energy about sending light straight up to space could be “solved” by quite cheap lenses which spread the light in driving direction which would increase efficiency of signaling a lot. Though still pretty ridiculous efficiency compared to simply passive paint. But painting the panels would reduce efficiency again. Just like mud, dirt or even simple shadow.

      The second thing I would be worried about is safety. Your calculations took standard glass panels into account. Well, glass is very slippery to walk or drive on. But with rough surface you get some grip but lose a lot of solar energy which is absorbed by the surface. Moreover visibility of LEDs would drop again and cleaning cost would increase as well

      Rather than just talking about how much bullshit the current design is I tend to think about how this could really be turned into a success even with technologies not yet available. Though in this case I think we would have a hard time. Thing which come to my mind:
      – More efficient solar panels
      – Use e-ink style display (passive) rather than active LED (expensive, might affect solar efficiency)

      – Need of surface which provides grip but does not accumulate any dust, dirt or even tire rubber ;)
      – Efficient and cheap storage of power output

      Rather than using visible marks on the panels it might be more efficient to provide positional sensor signals and project the road signalization directly to the car HUD. But this would require to exchange all the cars and would you disallow any “traditional” car to drive on solar roadways just because they can’t see the signs? It would pretty much turn into a second road system to be maintained independently from the existing road network (road, solar roadways, train) but this is impractical especially in cities.

      Anyway, there’s a lot of ideas but certainly the current plans are impractical and inefficient but hey, perhaps some people will be able to earn some money from the hype and walk away as rich people.

      • Karl Tanner

        Dave,
        You got so mad you didn’t end the blog with you usual EEVBlog smoke and sound effect signature!!
        I think this is on the same level as the “Thorium” Nuclear reactor technology thing going on a while back with Google and the ex-NASA guy (don’t remember his name now)… As where in principle you initially think “good idea that”… However, the practicalities soon dawn on you and soon realize that it’s not-that-easy.

        Solar and wind farms are an absolute eye-sore, and the amount of area needed to generate a decent amount of electricity as well just doesn’t justify the use of them.
        Near to my parents they have installed acres of windmill generators, which most of the time are at a stand-still and just pointless. I’m afraid in the wrong use solar is just as pointless if you want to run something big i.e a 8.5Kw shower or an electric hob or even become completely detached from the grid. Great for offsetting your power usage. If you generate i.e xx kW during the day and use x kW from the grid at night… quids in. (I don’t know the numbers off the top of my head)
        I’m not dissing solar, I’m all for it, I hope to install it myself when I can afford it. But if anything its OUR BEHAVIORS we have to change! Our usage of appliances.. AND the manufacturer must start making gadgets and gizmos work not on A, or mA, but uA or pA…
        More efficient panels, yep, another idea, but I believe there is only so far you can take that.
        My Dad used to work for the mighty CEGB back in the day, a child of the fission area, and he says back in the day they were looking for alternatives, 30 years on and this is as far as we got? It’s our usage, we have to look at. Micro-generation is the way forward, but you use UV heating, Solar and wind to charge batteries and inverter tech for generating AC power. (Batteries, another mind-field)

        And finally back to the roadways, the amount of traffic will cause the panels to be shaded most of the productive times that the panels could be generating.
        I would really like to see them crack it… But, erm… I’m sitting with Dave on this one.
        Regards,
        Karl

        • Simon Ludborzs

          Thorium Reactor Debunking? Any links – keen to take a look at that.

          On the surface I like the sound of Thorium Reactors but as a sceptic with an open mind I’m interested in both sides of this argument.

          Just like Solar Freakin’ Roadways – Happy to look at both sides of the argument but with roadways, it just doesn’t add up….

      • tlhIngan

        Actually, most LED lights on roadways are angled towards the driver’s eyes. Almost no system in use today shoots the light “straight up” where it’s wasted. It’s always angled.

        That’s current existing technology today, too. using those standard bumpy reflectors. If you’re pointing the LEDs straight up, you’re doing it wrong.

        • Ancel Bhagwandeen

          I think the inventor got a little ahead of himself where the technology isn’t ready for the application. Perhaps solar roadway medians or even shoulders would be reasonable given the lack of wear and tear and no traffic shadow. No wasted power/cost running LEDS, passive reflectors are just fine as all cars have lights. That way, including some piezoelectric vibration energy generators could also augment the energy yields and provide value in traffic data that can distinguish vehicles by weight & speed. Further such vibration data becomes valuable in bridges and overpasses as early warning for structural problems. It’s all about adding value when innovating. Not hopeful claims.

    • Matt

      I gave them some of my money… Give the thing a chance… I don’t know where you read that this was production ready? Also, I do think you raise some valid issues, but they just need to be worked on…

      I’m really surprised, and to be honest a bit disappointed, by your conservative stand on this… That’s not this way of thinking that put people on the moon and brought other paradigm changing innovations.

      • http://www.skybeam.ch/ Rainer Meier

        In fact I did and having a look at http://youtu.be/0c0QLaVLhTs he’s just right too. Why would you carry around a pound of alcohol powder to a trip just to get alcohol drinks at the campfire while you could also just carry half of the weight in its liquid form. Or even roughly 10% of its volume (imagine you would have to carry about 1l of poweder to carry the same amount as in a 0.1l bottle of pure liquid alcohol.

        Well, this is off topic, but he’s certainly not just a troll.

        Back on topic: I would love to see something like solar roadways but with current technology it’s just that far off so it’s clear to anybody it can’t be achieved at this point. Moreover it’s totally inefficient by its design as solar roadways would waste energy even if nobody is there to see it. In such situations passive technologies like simple paint is much more effective. Sure you can claim you can use sensors to disable signalization when there’s nobody around but this sensors would have to be reliable and they increase cost a lot (again).
        Anyway if you can afford the money to be lost on such project then I won’t object. It’s your money. But by myself I am too concerned about these guys who admit their selves not having completed any study and collecting $2M for basic research.

        I am just curious how they will get away. People spent money on it and they expect some results. Perhaps they will update the web page every now and then and claim they are still in research until everybody lost interest.

      • http://www.eevblog.com/ Dave Jones

        The big blanket statement at the top of their web page that says:
        “Our technology works. Now it’s time to gear up for manufacturing.”

        You have to defend this because you invested money and don’t want to look like a fool. That’s understandable, and normal human psychology.

        But your investment in this is clouding your technical judgement.

        This isn’t something that it going to get solved by “working on it”. There are deep rooted fundamental issues why it can’t work. And where even if it did work as claimed by some miracle of technical innovation in dozens of areas, the return on investment means no one in their right mind would ever touch it.

        My stand is not conservative, I simply ran the numbers, and very generously gave this project every opportunity to prove itself on the power generation numbers. But it even failed that generosity.

        it’s not my fault the numbers didn’t, and won’t ever add up. If they did, I would be a backer and supporter.

        All ideas do not have an equal chance of success, this one has so many fundamental problems on so many levels, like dozens of different show-stoppers. It’s just ludicrous to even consider this viable.
        But if they, or you can show me the data that proves it’s viable then I will happily change my mind and support it. Please, go on, I encourage you.
        But please don’t simply come at debunkers with rhetoric, this is a technical issues, and it’s all about the technology and the data. You prove things based on facts, evidence, and empirical data.

        Carl Sagan said it best:
        http://quotehd.com/imagequotes/authors2/carl-sagan-scientist-quote-but-the-fact-that-some-geniuses-were.jpg

        • Matt

          Yes, pay to check rather than raise really, give the thing a chance to go further…

          I think it’s a case of shooting for the sun and reach the moon, if you see what I mean… Not all of it will probably make sense in the end (the road heating jumping to my mind), but some aspect of the project may be worthwhile in isolation…

          As I said some of the points you make are definitely valid (and I wouldn’t dare to lecture you on anything electronics related, under penalty of getting my ass handed over to me!). I do think that you’re a bit quick to dismiss on the basis of optics and structural engineering claims, which surely require more study. I sort of can see how that the lack of credibility on the electronics would make you doubt the rest though…

          Yes the claims on their page are a bit enthusiastic, but go raise some money saying: “it probably doesn’t quite add up, but we’d like to have a go…” (well that could work with me, but I’m special, I give money to everyone who seem to make a genuine attempt a pulling something off, including you! :)

          Have a good one!

          • http://www.eevblog.com/ Dave Jones

            I’m not sure you get it. The entire concept of solar roads is a pointless waste of money if it can’t generate enough power to pay for itself. There is no point researching if it’s a suitable road material or not, and all the countless problems involved with that. The entire concept is flawed to the point of stupidity. If they are unwilling to admit this, and focus the project on something that is practical then it’s just a big waste of money of your money.
            If this wasn’t about using them as roads, and being more sensible and upfront about their claims, then I’d probably be backing the project. I’m all for solar power and renewable energy, but this is just absurd.

            • Matt

              No, I’m not sure you get it (to reuse y our formula)…

              Maybe the solar side of thing doesn’t add up (yet?)… but there may be some interesting other aspects worthwhile of consideration: intelligent/connected roads, dynamic road markings (what about a mix of LED and ebook kind of technology? this would likely require a fraction of the power you use in your calculation, but hey I’m no expert in any of these fields and can debunk your debunking in 20s).

              Anyway, all the incidental side effects would be interesting to evaluate and dig into.

              You know what: it’s possible I made an error of judgement… but that isn’t the last one I made, which was obviously hoping to engage constructively with someone as obtuse as you! :)

              • http://www.eevblog.com/ Dave Jones

                Please go ahead and debunk my debunking. But please, make it detailed and include real world facts and numbers.

                Start by show how these tiles will even get close to paying for themselves. I honestly encourage you to do this.

              • http://www.eevblog.com/ Dave Jones

                Still working on your rebuttal video?
                That’s ok, take your time..

    • hdavis

      I definitely agree with you on this. With the exception of maybe some niche applications, there is no way this will become mainstream. I had to facepalm when they said melt snow. If the tiles are covered with snow, how do they get the light to melt the snow! Even if you heated them while it was snowing, it is not very sunny during a blizzard!

      But I do disagree with the LEDs. While they probably couldn’t be seen, couldn’t they angle the LEDs so the light is aimed at the driver? I still don’t like them since they would add so much light pollution.

      In Chicago, the Willis Tower (formerly known as the Sears Tower) has a skydeck where you can stand on three layers of 1/2 inch thick glass rated to hold 5 tons to look at the city below and still it cracked. True this is fat Americans standing on it, but it does not bode well for glass hexagons being flexed thousands of times per day by cars and trucks with large temperature swings.
      http://www.nbcchicago.com/news/local/chicago-willis-tower-sky-deck-ledge-crack-261079001.html

      Unless they came up with “transparent aluminum” that glass is going to crack.

      • tlhIngan

        The glass that cracked on the Sears Tower was designed that way. It’s a multilayer glass laminate with the top surface (the one people walk on) as a protective layer to protect the structural glass underneath it.

        And guess what? Someone either had elephant feet, or a pebble in their sole and it cracked the protective layer, like it was supposed to. It still had the full load carrying capacity, its just the top protective layer got damaged

        And you always design things that way because otherwise the structural element would undergo much more wear.

      • Zachary Heilig

        “Transparent Aluminium” already exists, has for a very long time. It’s known as sapphire glass — sapphires are aluminium oxide and some impurities. It has uses in chemistry where there is heat, pressure, and a need to view the results. The iPhone 6 is supposed to have the front glass made from sapphire glass.

    • Vliegendehuiskat

      In fact, the leds only have to be on, for about 300 meters in front of a driving car. If a car is past them, they can be switched off again. This is not a complex algorithm and can easily be implemented in even the smallest and/or cheapest micro controllers!
      Combine this with other solutions and you can drive down the power consumption of the line markings spectacularly.

      But as Dave already noted, the economical return on investment will be way to small to be profitable, even if you combine these solutions.

    • flop

      http://www.swarco.com/futurit-en/Products/Signals/LED-Road-Marker/LANELIGHT-MLK-150-bi-directional
      LED-markers do exist, but the leds are facing the driver (although they only really do their job in tunnels or at night…)

    • penguin42

      Quick nitpick: You shouldn’t just subtract the losses like -20 -10 -10 = -40%, you need to multiply them like 80%*90%*90% = 64.8% (ie -35.2% loss). Doesn’t change the outcome of your analysis but it was bothering me.

      • http://www.eevblog.com/ Dave Jones

        This is not an exact math problem, they are ballpark figures of a complex system were the stackup losses aren’t exactly known. Perfectly fine to just add losses in a case like this.

        • penguin42

          The fact that you got close to the right answer was complete luck. Even if you’re only estimating, subtracting losses is absolutely incorrect. Here’s an example: say you have a system with 4 components, each component has 25% loss. If you subtracted the losses, you’d say you had 100% loss in your system — but the real answer is about 70% loss! That’s a huge difference, even for “estimating.”

          • achim1989

            In my opinion, this depends on the reference value of the loss. If you say that you have a 100W generated by the system and then you say: from this 100W 25% are lost because of the surface structure and another 25% are lost caused by the angle then you have a loss of 50% in total because both percentages refer to the 100W. It’s just a matter of phrasing it right…

            • http://www.eevblog.com/ Dave Jones

              And that’s precisely the context in which I made these estimations,

              • penguin42

                I’m really confused as to why you would intentionally chose an invalid mathematical context to make your estimations, especially while in the process of trying to “debunk” someone else’s invalid claims.

                • http://www.eevblog.com/ Dave Jones

                  Because it’s easier for the audience to follow along, it doesn’t matter squat in the end, and is fairly common for such back-of-the-envelope calculations. You don’t dick around with exact values, that’s the essence of ballpark calculations.

                  • penguin42

                    This has nothing to do with “ballpark” vs “exact.” As I said earlier, the only reason your technique came close was pure luck. It could have easily gone wildly off. It’s honestly appalling to me that you’re continuing to be so obstinate about this. An intellectually honest person would admit their mistake and move on. I thought your purpose was to try to provide good information to your audience. Now I’m wondering.

                    • achim1989

                      Dave could also have used absolute example values for his calculation. It wouldn’t have changed the result. More important is the fact that Dave is an experienced engineer who also owns a solar system and therefore knows his figures and is able to interpolate a likely outcome for this solar road project.

                      Btw, an intellectual honest person would haved realized by now that both sides have their point and would not discuss this any further.

                    • http://www.eevblog.com/ Dave Jones

                      Oh for goodness sake. It was NOT luck. These are not random numbers being added up. I knew very well what I was doing and what the outcome would be. If you want me to admit I was wrong, then fine, I was wrong, I didn’t do it the mathematically correct way. I could have done it better. But I CHOSE to do it that way. If you don’t like it, fine, you don’t have to, point taken, but please, stop being so anal about it.
                      I’m just trying to explain why I did it the way I did it, but it seems that will never be good enough for you.
                      Sigh.

            • penguin42

              Except that’s confusing and wrong. Say you had a 100W ideal power, 50% lost from surface, 50% lost from the angle, therefore 0 power gets through? There is no system in which that description would make any sense. System efficiency is treated in a very specific way in engineering for very good reasons. Hand-waiving numbers can get you into the dangerous area of your estimates not being valid anymore. Why not just use the math right? It’s not that hard.

    • Clee

      Thanks for putting together your video. But I have a few issues including penguin42’s nick pick and the points that LEDs can be aimed directly or by reflectors and/or diffusers) towards oncoming drivers instead of pointing up to the sky.

      The 18.5% efficient Sunpower E18 is not “the world’s best”. SolarRoadways seemed to call it “conservative”.
      Sunpower sells a 21.5% efficient modules in their X21 series.
      http://www.stellarsolar.net/downloads/SPR-x21-335-345.pdf

      This 21.5% efficiency already accounts for the transmission losses of the glass as well as the area lost to the frame. The cells themselves are over 24% efficient.
      http://www.pv-tech.org/news/sunpower_lifts_lid_on_new_record_21.5_efficient_x_series_solar_modules

      It would probably be more accurate to base your numbers on the cell efficiency and not the panel efficiency since SolarRoadways isn’t including the glass and frame from panels in their tiles, otherwise you are double counting those losses.

      • http://www.eevblog.com/ Dave Jones

        It’s still not going to work out, not even close. My numbers were already conservative enough to cover any of this.

      • Elandril

        Just to add: these figures only account for the transmission losses of the normal glass protection, which will definitely not withstand any car driving over it (and I’m not even talking about trucks with XX tons). The losses from a material (of sufficient thickness) that can support such weights will be enormous. If I’d venture a guess, the transmission coefficient is probably going to be less than 50%. So you can immediately kiss away 50% of the incoming solar power.

        • http://www.eevblog.com/ Dave Jones

          At least, yes. I was very generous in my calculations!

    • Elandril

      Thank you dave for this video to point out some of the most basic flaws of the whole idea. Honestly, I’m bewildered that the inventors even dare to call themselves engineers (and with a degree no less)!
      I also would like to point out one other flaw that is mostly overseen in the engineering discussions: the non-availability of the raw materials in the required amounts. Take for example the silver for the contacts: the silver alone needed for those panels would be unobtainable – and using any cheaper/more available material for the contacts would just reduce their effectiveness even more. And it goes on with the other components of a solar cell.
      All in all this idea has so many flaws that it can only be called ridiculous.

    • Chris

      I havent looked into the solar roads much, but I didnt need to in order to know they just didnt seem right.(nope didnt watch the video, dont need to) The very first problem that comes to mind is simply how dirty roads are. They are covered with car fluids, exhaust residue, sand/dirt, rubber from tires, etc etc. This doesnt even mention the problem that when it comes to winter weather, they use sand and some people use chains.
      Even if you tried keeping them clean – which you actually couldnt without losing any practicality; not to mention the fact that the surface would end up far from polished after only a short time.
      If they want solar power, why would they not just put panels on the top of the buildings?

    • Michael K.

      Thank you for saying it out loud and clear! It should be easier to define pi equals 4. ;-) Unbelievable that they earned 3M$ with this huge solar garden light project. Yes, $33/m2/year as an unreachable best case scenario – that’s all you have to point out. LOL.
      BTW: How much energy do you have to invest to produce all the glass, solar cells, copper, PCBs, CPUs, LEDs, …

    • Julio Rodriguez

      What you fail to consider is that even with high intensity sunlight, you might still generate almost nothing during periods of high traffic because every vehicle will cast a SHADOW
      (That’s right, they are not transparent!) on the road surface below.

    • Jonathan

      I’m a little late to the conversation, but one of the loss factors I’m surprised wasn’t mentioned is the thermal effect on the solar cells. Embedded in a roadway is not an ideal thermal situation not just for the LEDs, but also I’d add another 10% loss impact to the panel generation due to operating at the high end of the temperature range simply due to the inability to dissipate heat. Unlike a roof or rack installation, the panels wouldn’t have convection cooling due to air contact on the back. Plus heat additions due to vehicle braking, friction, exhaust, etc. Terrible, terrible idea!

    • Bearman

      Great high level analysis. I loved it.

      You should do another one regarding electric cars vs. gasoline cars.

      Discuss the efficiency of burning gas under the hood in an engine directly as compared to generating power at a power plant, distributing it to your home, sending it through a battery charger, losses in battery charging chemistry, losses in motor drive circuitry and electric motor losses.

      Then also compare the cost per energy density of gasoline vs ,coal, oil and natural gas fuels involved in the power plant generation. I think you will end up with a similar result. Electric cars are not the answer yet either.

      Not to mention the fact that who is going to buy a used electric car (say 7 to 10 years down the road) when the first thing they will have to do is plop down possibly $8000 plus for a new battery for it to have a reliable vehicle that isn’t worth the cost you will need to put into it.

      BTW Anyone hear any news about all the electric vehicles that were all the rage a few years ago!?!?!?!?

      B

    • Bruce W

      I have lot of respect for Dave and love your great videos. But
      unfortunatelly in this one, you just sound like a pesimistic old man. A
      lot of technology today we would not have if there wasn’t an idea, come
      up by someone and was pusuit, even though technology was not viable at
      the time.
      In my opinion the project is great. Without a massive
      investments from the government for the energy isssues (look at the cost
      of CERN), the whole world engineers are in this great debate, regarding
      the technology and it’s viability. Sure it’s not perfect, sure the
      current technology is limited, but it is a great start and viable
      product for early adopters. And you don’t need much – the shear wolume,
      would make a great difference in the electrical power production.
      And finally, I do hate the solar panels on th roofs. They just ruin the house looks.

      • http://www.eevblog.com/ Dave Jones

        Ok, please prove that my numbers are wrong.
        Please prove that this technology is in the least bit viable. Please show me the numbers.
        It doesn’t matter what you think of the technology, if you can’t back it up with evidence and numbers, then your argument doesn’t have a leg to stand on.

        This is NOT technology or research investment, they claim their technology works and they are taking it into production. They are massively wrong and they have mislead the investors.

    • http://engineerllc.org Vernon

      Perhaps they have mislead the investors. However…

      “Heavier-than-air flying machines are impossible.” — Lord Kelvin, President, Royal Society, 1895

      “The aeroplane will never fly.” — Lord Haldane, Minister of War, Britain, 1907

      “I think there is a world market for maybe five computers.” — Thomas Watson (1874-1956), Chairman of IBM, 1943

      “Computers in the future may weigh no more than 1.5 tons.” — Popular Mechanics, forecasting the relentless march of science, 1949

      “There is no reason anyone would want a computer in their home.” — Ken Olson, president, chairman and founder of Digital Equipment Corp., 1977

      “640K ought to be enough for anybody.” — Bill Gates (1955-), in 1981

      • Michael

        Did you not read the quote from Carl Sagan?

        Your quotes are irrelevant as they are either people making statment based on the current deamnd of technology at the time, or were people making statments based on flawed asumptions and lack of knowledge.

        The Solar Roadways people have stated the technology is ready to go. Clearly it is not only not ready, but is currently not remotely viable. Until, and unless, we see some quantum leaps in solar panel efficiency and large reductions in the manufacturing cost of both panels and the ‘harder than asphalt glass’ it is unlikely to ever be viable.

        The project leaders haven’t even made a proper attempt to address the arguments and reasoned deductions by Dave et al. Instead they continue with their hand waving generalisations, flawed examples and emotional blackmail. I sincerely hope I’m wrong, but these are the typical tools of a scam artist rather than a genuine engineering project.

        • http://engineerllc.org Vernon

          @Michael,

          No, just trying to inject a little levity in the thread :) My perspective with the quotes was somthing more along the line of how one thing in common with many innovations is a reluctance to heed the advice of influential Naysayers.

          • http://www.eevblog.com EEVblog

            My video wasn’t “advice”, it was some basic ballpark feasibility calculations, of which this project failed miserably. Not to mention all the offer issues I could also do ballpark calcs on of which I am sure it will fail as well.
            Facts have a nasty inconvenient habit of being true whether or not you believe in them.
            These are the basic calcs that a guy with a masters degree in EE should have been able to do 5 years ago when he started this project. But he obviously didn’t do these, or did them wrong, and has gone off on a silly delusional crusade and duped a whole lot of people out of their money for a project with no hope of succeeding as advertised.

            Comparing this project to other like you have done is complete folly.

            • http://engineerllc.org Vernon

              Complete folly? Yes I do that from time to time.

    • http://engineerllc.org Vernon

      But Wait There’s More!

      Now if you want to talk about blatant scams…

      I got a call this morning from “PC Speedy”!

      If you get a call from “PC Speedy” or anything similar, here is an example of a “Go away quickly” script (didn’t have time to play with ‘em). They get discouraged fairly fast when there doesn’t appear to be a good mark on the line…

    • Jessica

      Using your final numbers, prior to the installation of LEDs, and the surface area of our driveway, if paved with solar panels, it would produce 17337.5kwh/yr. According to the US Energy Information Administration, the average US home uses 903kwh/mo x 12 = 10836kwh excess per year x .24/kwh = $2600.64/yr. Average Colorado electricity usage was actually closer to 7500kwh/yr or 625kwh/mo. So even allowing for cloudy days, or snow, or rain, dust and dirt, that $2600/yr is probably about right. 30% federal tax credit on installation, state and city rebates will cut installation costs. The $2600 is above and beyond what we would have payed for electricity, over the course of 5-10 years, though they give an estimate of 20 years, and would not replace panels with completely new panels, but retrofit and replace any broken parts within a panel and replace it, so the big cost is all at once, plus rising electric prices, how does this not work again?