Clean Power SolarCity Credit Suisse

Published on July 3rd, 2016 | by James Ayre

45

SolarCity PV Modules Have Useful Lives Of 35 Years Or Longer

July 3rd, 2016 by  

The solar photovoltaic modules used by SolarCity in its residential and commercial installations have useful lives of 35 years or longer, going by new data and information recently presented by the company.

The new findings are the result of independent, third-party, accelerated stress testing of SolarCity’s solar photovoltaic (PV) modules performed by the world-renowned laboratory DNV GL.

SolarCity Credit Suisse

The accelerated stress testing that was performed by DNV GL went beyond the IEC 61215 industry standard, reportedly. The testing found that “power degradation of modules supplied to SolarCity by its key external suppliers is as much as 35% lower than for a comparable industry-wide selection of non-SolarCity modules, which are typically warranted for 25 years.”

The press release that was used to announce the findings continues: “The greater longevity of modules fabricated for SolarCity is due to SolarCity’s stringent and industry-leading Total Quality Program, which was adopted in early 2014 and is based on similar programs in the automotive industry. SolarCity requires its third-party suppliers to maintain effective quality assurance programs and refined manufacturing processes, and to demonstrate reliable product and manufacturing quality. Rigorous tests performed by a qualified third-party lab need to be passed on an on-going basis.”

It’ll be interesting to see how the company’s US-manufactured solar PV modules perform over the long term, after its under-construction manufacturing plant in New York State is up and running. The company has in the recent past achieved some impressive new solar conversion efficiencies, as I’m sure many of you know.

Those wanting to take a closer look at the full report can find it here.


Check out our new 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , ,


About the Author

's background is predominantly in geopolitics and history, but he has an obsessive interest in pretty much everything. After an early life spent in the Imperial Free City of Dortmund, James followed the river Ruhr to Cofbuokheim, where he attended the University of Astnide. And where he also briefly considered entering the coal mining business. He currently writes for a living, on a broad variety of subjects, ranging from science, to politics, to military history, to renewable energy. You can follow his work on Google+.



  • Epicurus

    What’s the best online calculator or downloadable spreadsheet for calculating the financial benefits of a solar panel investment? It can get complicated if you include all the variables. Thanks!

    • vensonata .

      Try “PV Watts”. Everything is built in to the calculator. A simple mortgage calculator can be added for financing.

    • Calamity_Jean

      Here’s the link to the site vensonata suggested: http://pvwatts.nrel.gov/

      • Epicurus

        Thanks.

  • Mike Dill

    James had a great answer for me earlier, so I now have a question: I live in the hot part of the southwest (Las Vegas). I think my cells will die from heat and UV radiation before getting water induced failures. Are there ‘special’ panels for different places?

    • Epicurus

      It’s ironic that the places with the most sun like the Southwest desert also have the most heat which is not good for the solar panels.

    • JamesWimberley

      All I can put in here is that First Solar claim better performance in seriously hot conditions (Middle East) for their CFTE modules. However, they have not cornered the market thereor in India or Iran, so the advantage may be quite small.

      • eveee

        The advantage is output, not longevity. Silicon output falls with temperature. CdTe doesn’t as much.

  • JamesWimberley

    SolarCity’s claims for its own panels are pretty credible; companies can’t lie to investors. As usual, the comparison is phoney, to an unspecified “Brand X”. Of course you can find cheap stuff in any market, taking your chances on the quality. This may actually be a sound strategy for utility developers, who can monitor performance and replace failing panels; householders should pay more for the assurances that come with a leading brand name.

    The useful comparison would be with other leading brands – SunPower, Canadian Solar, Jingko, Trina, Panasonic and so on. I would ve very surprised if there were demonstrable differences between them.

    • eveee

      Some of the studies of earlier solar cells indicated such , but newer Chinese cells are lookin better, I recall.

  • Bryan

    Most of SolarCity’s modules were made in China. I find it hard to believe that Chinese made solar modules will last 35 years. The Chinese manufacturers are notorious for substituting high quality backsheet material, encapsulations material, and diodes for lower quality Chinese brands. Not to mention cell grades and the potential for micro-cracks in the cell’s structure. 35 years? I don’t think so.

    • JamesWimberley

      Cites please. Are these reports for first-tier manufacturers, or second-and third-tier ones? Unsubstantiated rumours smell “smear campaign” to me.

  • Brunel

    Can DNV GL test Powerwalls to determine cycle life.

    • Ronald Brakels

      DNV GL is testing GCL’s E-KwBe. I would be confident the very large majority of Powerwalls will make it to the end of their warranty period, provided the conditions of the warranty are met. But third party testing is always welcome. I don’t know if anything has been done along those lines.

      • Brunel

        Everyone wants to know the cost of storing electrons in Powerwalls vs LG vs Samsung vs Aquion.

      • vensonata .

        Something interesting has happened to the warranty itself apparently. It has changed recently, I have heard, through second hand sources. The cycle degradation description seems to have been eliminated. This all needs further confirmation, I haven’t been able to find the new warranty on line. This may be to do with the Powerwall 2, which will be displayed this month, probably at the gigafactory opening. All in all though, I see an avalanche of battery subsidies and wide scale distribution is about to happen in Australia very soon.

        • Ronald Brakels

          I’ll be interested to see if there are any changes for Powerwalls currently being sold.

    • Matt

      Go back and watch the video by battery guru on testing. Most is not very go, they have created a new way of doing it and get amazing results. But short version of video is that lots of very fast charge/discharge cycles don’t give you very real results, you need to do extra work. Oh and he comments on TESLA batteries as an a side, and TESLA has at least one of his disciples.

      • vensonata .

        The most interesting part of Jeff Dahn’s presentation was the graph on the lithium oxide cells from Medtronic. They have been tested in real time over an 8 year period. They fall to 80% capacity after about 1200 cycles. But then, something astonishing happens. Degradation almost stops. From 80% down to 60% takes 20,000 cycles. Then Dahn says “and Tesla vehicle batteries are better than that”. So there is a new wrench thrown into battery storage discussions. It makes the lifespan of the vehicle batteries…re-purposed for home storage, so long, as to reduce the cycle cost to a few cents.
        This crudely made video and hand illustrated curves are not all that inspiring of confidence, however Jeff Dahn’s qualifications are impeccable in mainstream science and engineering circles. Not enough has been made of this data.

        • Bob_Wallace

          I view a 200 mile EV range as a ‘threshold of usability’ for long distance driving. I suspect we’ll move to a higher range as battery prices and weights decrease.

          1200 cycles to 80% would mean a 300 mile range EV would still have a 240 mile range at 240,000 miles. I doubt we’ll build our cars to have a longer than quarter mile lifespan. If that battery is capable of 20,000 cycles then it could do once a day grid cycling for over 50 years after it’s pulled from the EV.

          It seems that we’ve got the mass storage problem solved. We just need to wait until the first generation of EVs reach the end of their useful life and then move their batteries to the grid.

          It may be 20 years before we need large scale storage. By then we could be seeing a large number of used EV batteries starting to appear in the recycling yards.

          • Bob_Wallace

            Some rough math. Assume battery pack drops to $100/kWh. That’s a fairly safe assumption. Finance at 5% over 20 years. During the payoff period if the battery is cycled once a day the price to finance and pay off the battery is $0.01/kWh.

            Based on the Medtronic data the battery would, after payoff, continue to do daily cycles for another 30+ years. Zero battery cost. (Inverter, storage space, costs would have to be covered.)

            Price has to be adjusted up a bit for capacity loss. Assuming 80% capacity through the payoff (20) years the cost rises to $0.0125/kWh.

          • eveee

            Battery cycle life is non linear with depth of discharge.
            https://c1cleantechnicacom-wpengine.netdna-ssl.com/files/2016/05/Cycle-Life-vs-Depth-of-Discharge.png
            At a typical daily usage of 30 miles on a 270 mile 85 kWhr tesla battery pack, the cycle life and total mileage soars. The worst scenario is repeated 100% discharge. Calendar life is a greater limitation because it’s proportional to temp. That’s why the 85kwhr pack is warranted to 8 years unlimited mileage. The calculation is conservative as long as it’s kept to less than 15 years and temperatures are kept moderate. Jeff Dahn will be helping Tesla improve that shortly.

          • Bob_Wallace

            Take a look at the Medtronic graph. This is 20,000 full discharges. My eyeball analysis says average capacity over the 20k cycles is about 70%.

            At an average 70% capacity and a 20k cycle lifespan the cost would be $0.014/kWh.

            In his presentation Dahn says that Tesla has moved on to a better battery.

          • eveee

            Yes, thanks. This graph matches Tesla owner experience that shows a fast drop off and then a slow down. The chemistry is different in this graph. It’s the laptop LiCo type, not the NCA used for the Model 3, but they behave similarly in this case. Another wrinkle. For storage apps, the degeneration to 70% is less of an issue, so more cycle life is available. With these advances, calendar life is more limiting than cycle life.

          • Bob_Wallace

            What’s the best take on calendar life for current lithium-ion batteries?

            The cells used for the Medtronic graph are still around 80% at 1,000 cycles. For an EV with a 250+ mile range that means well over 200,000 miles before range drops to 80% of new.

            I don’t think we’ll be looking at 15 year old cars as what we expect to use for a lot of long distance driving. Those cars (because their bodies and interiors are looking tired) are likely to be sold on to people who can live with a 150+ mile range. A 250 to 300 mile drive day will be reasonable and a 500 mile drive day possible, just slow due to the number of charge stops necessary.
            —-

            For grid storage I can’t see batteries undergoing more than 2x cycles per day. One night wind -> early day and one day solar -> evening cycle at most. At two cycles a day 20k would mean a 27 year lifetime and the charted batteries aren’t dropping below 60% very rapidly.

            At 30 years of two cycles a day a $100/kWh battery costs less than half a penny per cycle. That’s incredibly cheap. (Financing and battery charger/inverter costs not included.)

          • eveee

            As I recall, calendar life depends on temperature and electrolyte oxidation. I would hesitate to generalize without environmental operating conditions, but Dahns video indicates it may be possible for 15 year calendar life today, given non stressful conditions

  • Mike Dill

    My panels only have a twenty-five year warranty to better than 80% output. If they do not suddenly fail, I should be getting at least 50% in fifty-seven years. It all depends on the definition of ‘useful life’.

    • Bob_Wallace

      According to the NREL you should be getting at least 80% after 50 years. Panels manufactured post 2000 should lose between 0.1% and 0.4% per year.
      You could be getting as much as 95% after 50 years if you live in a “gentle” climate. Little wind and snow loading. Low UV exposure.

      • OneHundredbyFifty

        About 15 years ago I was at meeting at one of the, then, major mfgs. After the meeting we were given a tour of the facility and then there was some unstructured time afterwards. In an off the record discussion with one of the engineers, I asked how long they really expected the modules to last. He said that they were confident in 30 years but thought that 50 was more likely. He said that they did not market them with those numbers because they believed people would think they were being dishonest and it would look bad – since, of course, it would take 50 years for them to prove their point.

    • JamesWimberley

      Doesn’t this depend on the failure of encapsulation rather than the degradation of the cells? The latter is quite smooth, the former catastrophic, as with car parts. The trouble with estimating longevity IIRC is that encapsulants have changed much more than the cells themselves over recent decades, so there is no direct experience of their lifetimes, only accelerated tests in labs.

      • Matt

        Yes this is the hard part and the one that say time will tell. The accelerated test give insight. With an announcement like this you would have more faith if they had also bumped their warranty up by 5-10 years.

      • Bob_Wallace

        All the long term studies of panels report very low panel replacement. In a 20 year study only one out of 288 panels needed to be replaced.

        There was another study that I’ve lost the link for. They reported a slightly higher failure rate.

        Failures in the latter study were due to delamination and terminal corrosion. I would expect those problems are well behind us. Edge sealants have received a lot of research.

        My 17 year old panels have junction boxes where it is fairly easy for moisture to enter. When wiring the panels one has to open the box and make internal connections. If not all the seals are correctly placed then there’s going to be moisture penetration.

        Newer panels have sealed junction boxes with the connecting wires installed. The only corrosion opportunity would be at the plug. And the plugs look well designed to hold up for decades.

        • Epicurus

          Any failures in 17 years?

          Thanks for being an early adopter, Impressive.

          • Bob_Wallace

            No panel failure. But some installer failure.

            One morning I was standing at the window, drinking my coffee and enjoying the newly fallen snow. Eventually it dawned on me that there was only one bump in the snow created by a rack of panels and not two as there should have been.

            Turns out that one rack leg had become unbolted (lack of lock washer or failure to tighten?) and the wind had ripped the rack off its footings and destroyed six of the eight panels.

            My bad. Now the racks are double nutted with lock nuts and tightness is checked before the winds start up in the fall.

      • eveee

        Partly. There is surface erosion and aging. Some panels break. And some connections fail.

    • Right. I think a lot of people incorrectly assume that if a solar panel is warrantied for only 25 years, then that’s it. But look at a car warranty – like 5 years. But cars last a lot longer than 5 years. Just like an old solar panel, an old car is not optimal, but that doesn’t mean it’s not working just fine.

      • eveee

        Susan – you should explain that to the PowerWall critics who think it’s performance equals warranty then base economic calculations on that. You don’t see Lazards figure LCOE on warranty or Edmunds figure cost of ownership on warranty. It’s superfluous for economics except in case of rare early failure for an individual
        Owner. It’s useless for figuring average economics of average owner because the statistical data is unavailable.

        • I didn’t know about that. But yes, the same mistake is clearly being made in that case!

  • Epicurus

    A year or two ago I saw Fox News favorite Stephen Moore, alleged economist, claim that solar panels would only last 10 years and therefore would never make back their cost. Idiot or liar, one of the two.

    • Bob_Wallace

      Is there a “Both of the above” box?

      • Epicurus

        This jerk appears on CNN and other stations, and he (like others) is never challenged by the interviewer about the stupid, false things he says. This is the principal failure of the American mainstream media and the reason the citizenry is uninformed and often misled.

        • Karl the brewer

          And the mainstream media of other countries as well.

          • Epicurus

            That’s depressing. I thought the Europeans were brighter and better educated than we are. For example, I find The Guardian to be superior to any U.S. newspaper in terms of in-depth, truthful reporting.

            Even the European right-wingers seem much more intelligent than ours. None of them deny anthropogenic climate change, do they?

Back to Top ↑