Clean Power

Published on October 9th, 2015 | by Steve Hanley


Panasonic Quickly Beats SolarCity’s Solar Module Efficiency Record

October 9th, 2015 by  

Originally published on Solar Love.

Recently, SolarCity announced it will begin manufacturing the “world’s most efficient solar panels” at its factory in Buffalo, New York, starting in 2016. It claims it has designed a panel that converts 22.1% of sunlight into electricity.

Panasonic solar panel is world's most efficientNow, less than a week later, Panasonic says it has trumped that achievement. A Panasonic solar panel has established a new world record module conversion efficiency of 22.5% on a commercial sized prototype using solar cells based on mass production technology. The test results were confirmed by the renowned Japanese National Institute of Advanced Industrial Science and Technology. The 72-cell, 270-watt prototype incorporates newly developed enhanced technology that will eventually be scaled into volume production.

“The new panel efficiency record demonstrates once again Panasonic’s proven leadership in photovoltaics and our ongoing commitment to move the needle in advanced solar technology,” says Daniel Roca, senior business developer at Panasonic Eco Solutions Europe in a company press release by Panasonic Electric Works Europe AG.

Panasonic also says it is introducing the HIT® N330, the latest addition to the company’s high-efficiency hetero-junction photovoltaic module product line and its most powerful photovoltaic module to date. It will be available in the UK and other European markets starting in March, 2016. Manufactured at Panasonic’s state-of-the-art, vertically integrated solar fabrication facilities in Malaysia, HIT® N330 features 19.7% module-level efficiency and a nominal power output of 330 watts.

The new 96-cell module returns approximately 27% more peak-power compared to mainstream 260-watt multi-crystalline modules. Because of the increased power for the same number of modules and roof area, installers will have significantly lower system costs on residential and commercial installations by using panels with the N330 array.

The HIT® N330 offers higher energy density per square meter and reduces the total number of modules required, making it ideal for European ‘self-consumption’ markets,” says Roca. “For solar installers, the high-performance HIT® N330, represents the ideal complement to their existing Panasonic Solar portfolio, delivering the missing tool to achieve more power on any roof and to maximize balance of systems savings and lower installation costs.”

Between the SolarCity announcement and the one from Panasonic, it is clear that the future of solar power is only going to get brighter over time as manufacturers figure out how to get even more electricity from a given area of solar panels.

Complete our 2017 CleanTechnica Reader Survey — have your opinions, preferences, and deepest wishes heard.

Check out our 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

writes about the interface between technology and sustainability from his home in Rhode Island. You can follow him on Google + and on Twitter. "There may be times when we are powerless to prevent injustice, but there must never be a time when we fail to protest." Elie Wiesel

  • Bryan

    So where is SolarCity’s specsheet for their new solar module? Why the big secret? Does it really offer mono facial (single sided) high efficiency or does it rely on a bifacial (double sided design to attain its efficiency claim?

  • Also, cool article! I’ll have to look into that some more, hadn’t thought of the problem from that angle!

  • Well, that specific limit only applies to the common, single p-n junction silicon cells. Other materials definitely have the potential to exceed said limit. For example, off the top of my head, I believe that concentrated sun on multi-junction silicon cells can produce an efficiency approaching 86%. However, as the efficiency increases, the number of layers also approaches infinity (I’ll look up the specifics).

  • Philip W

    The rate of progress and amount of research is simply astounding.

  • tmac1

    Yingli not tingling darn autocorrect

    • JamesWimberley

      On this site you can edit your own comments. I do it all the time. Tingling would be a great name for a piezoelectric startup.

      • eveee

        Yes, but the spell checker won’t let you. No matter how hard you try. It butchers certain words. It doesn’t happen all the time and I don’t know why. Try typing a word with a period after it. Now try typing CAISO.COM in lower case. Do that as many times as you can stand without swearing. Now laugh. I hope I am not wrong and it behaves differently for you. In that case, I give up. I will now go use Google maps to be navigated in circles to the wrong address.

        • nakedChimp


          [firefox 37.0.1]

          • eveee

            Does that mean you tried the test? How did it come out?

  • tmac1

    Go Panasonic go solar city go tingling go LG go SOLAR !!!

  • Ray Boggs

    Now that’s funny. You go Panasonic !

  • Bob_Wallace

    Install either before the end of next year or expect costs to be higher for a year or so.

    The US average price for residential solar is about $3.50/watt. That is more than double the cost in Australia. And that suggests that there is a lot of fat in installation costs.

    Due to the expiring subsidies I expect high demand to support high prices. But once subsidies go away the market is almost certain to shrink which will send companies scrambling to cut costs and grab as much of the shrunken market as possible.

    Making up for the lost subsidies would require installation for just under $2.50/watt. Still very much more expensive than Australian prices.

    December, 2014 Germany was installing for 1.24 euros/watt. $1.41/watt.

    • Riely Rumfort

      You may just be right about the post incentive scramble, but if you were to do much of the installation yourself this year may be advantageous for panel/system cost alone.

  • WuestenBlitz

    Humans! We come up with a quantum energy converter that captures it’s power from from a fusion reactor in the sky and we find stuff to gripe about. 😉

  • mike_dyke

    I’m all for panels producing more output than other “standard” panels but what gets me in these announcements is that it is mainly the increase in number of cells that a specific panel has that causes the increase in headline power rather than better power production. i.e. If you took two 250wp panels and stuck then together, you’d have a 500wp panel but it would be twice the size of the standard panels.

    /Rant off

    • Bob_Wallace

      You sure? An increase in efficiency means more power out of the same panel area.

      • mike_dyke

        Yes, I agree about the increase in efficiency increasing power but that’s not my point. if you look at all the different designs/shapes of the panels (not just the panasonic ones) and count up the number of cells then you’d find that the only difference between the panels is the number of cells and so companies claim “Better than standard panels” when they are comparing, say, 80 cells vs 72 cells.

        • Bob_Wallace

          Some of the gain in panel efficiency has been achieved by decreasing the “wasted space”. Early panels had a lot of non-productive space. Wafers were very expensive and every bit was used. As wafers became cheaper we moved to trimmed square cells that packed closer together. And a lot of work was done to minimize the amount of area covered by ‘wiring’.

          But if the conversation is about cell efficiency all that stuff is not in the picture.

    • heinbloed

      You’re wrong.
      Read the technical data sheet:

      • mike_dyke

        Panasonic do N325/330, N285 and N240/245 according to the sheet. The only difference between them is the number of cells. The actual efficiency is the same (Although it has increased in this case)

        • heinbloed

          You have to look at the panel covered surface (W/m2).
          The larger panels allow for less gaps and frame material on the roof and facade.
          Therefore the output from the surface increases with the larger panels.

  • This is really of little interest as what counts is the price/performance ratio.

    As long as these prime modules cost almost twice the standard price for 17% efficiency modules I see no market for them.

    Unless you’re working with small roofs and small areas where you need to place large power solar fields, one does not really care for efficiency.

    Give us high efficiency with reasonable price/Wp and then we’ll talk,

    • Bob_Wallace

      Often improvements in efficiency do not add much cost to the process.

      The lead in efficiency passes back and forth from company to company. Overall we keep getting more efficient panels for less and less money.

      It’s good….

      • Nice theory, tell that to the likes of Panasonic/Sanyo.

        • Bob_Wallace

          Which part? There’s a lot of parts.

        • nakedChimp

          Hey, even the premium panels come down in price.. but as the name implies – still carry a premium.
          If you got enough roof space just ignore them and go for Wp/$ and/or Wp/install cost.

          Personally I already used some premium panels in a situation where space was on a real premium and I needed as much power as I could get for a given area, so we coughed up the premium and that’s it (2x Sunpower 328Wp panels to be exact).

    • Hazel

      That’s not the whole story. The whole story is *installed* cost/Wp. For example, if residential rooftop installation and “soft costs” of permitting, customer acquisition etc. are 70% of the installed cost and don’t change with module efficiency, then a 1.8% increase in module efficiency is worth 1.8% more installed cost, i.e. up to 6% higher module cost.

      Installation and soft costs are a smaller fraction of commercial rooftop solar installed cost, and even smaller fraction of utility. Hence module efficiency is less important for utility, and most important for residential rooftops.

      • Those costs you refer to are irrespective of module efficiency, so I really don’t see how the efficiency have to do with anything. Simply put the companies who make these modules ask for premium prices for them. They are not worth it. Panasonic modules are close to 1 Eur/Wp, while German quality modules are bellow 60 cents. There is simply no competition there.

  • JamesWimberley

    Mono or poly? It’s impossible to keep track if bloggers forget the basic distinction in silicon PV technology. From the context, the 19.7% module is enhanced poly.

    • Riely Rumfort

      I’m often annoyed the same way. Though of course a little digging into the announcement usually turns up a definitive answer.

    • GCO

      Once it’s on the roof, does it matter anyway?

      • Riely Rumfort

        Yes, slightly different life span and onset of efficiency loss between the two.

      • JamesWimberley

        No. But what we are concerned with on this thread is getting a handle on technical progress. Company PR types love to confuse us by comparing unlikes, so it’s the job of journalists to put back the essential context.

  • vensonata

    Until the price comes in there is no meaningful comparison possible between Panasonic and Solarcity panels. Kudos to both of them though for pushing up the efficiency.

  • Zorba

    If I’m going to have something on my roof for the next 20+ years then every 0.4% makes a difference. Great to see competition driving costs down and efficiency up.

    • Dag Johansen

      Meh. It doesn’t make much difference at all. And no matter what, the panels degrade over time.

      • jeffhre

        Everything degrades over time – even your objections. Unless they are presented in gold.

    • GCO

      I thought the same, so I was all over Sunpower panels. Then I saw the price…

  • Marion Meads

    I’m tired of overhyping eensy weensy teeny tiny bit of improvement, and so with sacrcasm: WOW an earth shattering super baduper outstanding improvement of 1.81% over that of SolarCity!!!

    • Bob_Wallace

      But were Volt to increase their range by 1.81% you’d be spamming the site with that info…. ;o)

      • Marion Meads

        Nope! You’re wrong!

        I am a statistician, so I would require an improvement that is statistically significant. In simple terms, the improvement should be greater than the variance. When you produce panels, the variance in their effiiencies are +/- 5% and that 0.4% increase is hardly significant as it is well below the manufacturing variance.

        • Larmion

          How can you say that an improvement is not statistically significant if you don’t know the sample size, population variance or even if the samples are identically and independently distributed?

          Reporting improvements smaller than the manufacturing variance is perfectly valid, if the sample size is sufficiently large (sufficiently being defined as meeting the Chebyshev inequality if the underlying distribution unkown or a considerably lower number if it is, which is the case).

          Perhaps you still practice statistics as it was before the days of Ronald Fisher and co. In that case, you must be over 100 years old by now…

        • Bob_Wallace

          If you’re moving the mean, you’re moving the mean.

          • Larmion

            Not necessarily. If the 95% confidence interval of the old mean overlaps with the confidence interval of the new mean, you can’t say that the mean has moved. That’s what’s meant by saying that an improvement is not statistically significant.

            (in practice, this happens when the sample size and/or the effect size being measured are tiny).

            Marion is suggesting that this is the case here, which suggests she has some deep insight in the data used by Panasonic and SolarCity or is psychic.

            I for one would be inclined to believe the Panasonic and SolarCity data. Confidence intervals on efficiency measurements by testing organisations are usually pretty tight, so overlap seems unlikely.

          • Bob_Wallace

            Yes, I understand that. I understood that as early 1964.

            I also assume that the testing procedure for solar panels is such that the average reported is meaningful.

          • Larmion

            Wow, you understood statistics long before I was born. It’s entirely off-topic, but you’re just a few years younger than my parents :O

          • Bob_Wallace

            Old, but feisty….

            (I probably wrote a textbook on research methodology which was about half basic statisticial techniques before you were born. But I do admit that I’ve forgotten much of what I knew…. ;o)

          • jeffhre

            Yes, um, if the variance of +/- 5% is over a very large sample (like panels installed world wide for example) then maybe the variations could average out through the large installed base and the new results could look sorta significant. I’m not a statistician though.

          • Babam

            When you get quotes from solar installers, the panels in their brochure has 19% rated efficiency but they don’t guarantee that you’ll get that due to variations in manufacturing of individual panels. The caveats that I have seen are plus or minus 2% absolute, meaning your panels could have between the 17% to 21%, so it gives you a good idea of the standard error of estimate. It’s all in the fine print. Manufacturers though tend to publish the highest range.

          • Babam

            You can read the spec sheets and the guarantee of the efficiency ranges and indeed as Marion has suggested, the manufacturing runs can result to variations in the actual efficiencies, some of them very high as much as 10% of each other. But it is here wherr QA/QC comes in. They test each panel and sort them their efficiency group that are within 2% of each other, again from the tech sheets that comes with the panels. The calculation done by Marion is 1.81% improvement which is within the variations of sorted panels with their guaranteed range of efficiency, so it is not significant based on data that were sent to me by various suppliers in their spec sheets.

          • Kevin Gregerson

            Be careful out there. I know solarcity has a bunch of stats guys on their engineering teams. I recall seeing a reports in great detail that highly tested the differences between each panel. I can’t discuss due to NDA. But I can say they have very very clear idea of how to test panels for efficiency in both the lab and the field and they wouldn’t have invested in this tech if it wasn’t up to the snuff they needed to compete in the coming market of decreasing government incentives and utility support.

        • NRG4All

          You are right, little tiny improvements are not Earth shattering in and of themselves. But when you step back you can see the tremendous improvement and why the cost is falling. Our array of 34 panels with each panel producing 235 watts (190 after inversion) was installed in June of 2010. Its stated efficiency is 14.4%. Thus, a panel that produces 19.7% is significant when put into a longer range perspective. As these costs drop it means that areas that don’t get as much sunlight can use solar, which is very important to world-wide adoption. However, the only way we have gotten to this point is by tiny baby steps over the course of 5+ years.

          • hank1946

            I bought my car new and it said 15 miles to a gallon around town 25 miles on the freeway average 19 Miles! We drive our car very carefully and seldom if ever get over 13 miles to a gallon it is a 2011 Chrysler 300 with a 5.7 Hemi! So even though they were not lying on the sticker it gave me a starting point as the 6 cylinder only averaged about 4 miles better. My point is 1.8 percent is not as important as cost per watt and size panel. So all things being equal the speck for anything only gives you a starting point. So if both companies offer the same wattage panels for the same price and they are the same size panel and they are the same quality! I am going to take the one that tested highest as it just might produce more power under under worse conditions.
            I come here to read and learn about solar and it seems like there is always a problem with people nitpicking over nothing. The articles are fun to read most of the time I wish I could say the comments were the same!

        • eveee

          Would you Marion? Or would you get the numbers and their significance wrong?

          The increase is approximately 10%. Not huge, but when you compare 22.5% to 20.5% and claim the increase is small, you ignore the relative improvement.

          I like the Volt range improvement, but its not the same kind of improvement.

          Adding more battery to get longer range is like adding more solar panels to your roof. There was no technological improvement, just more devices.

          On the other hand, we can compare efficiencies.

          The Volt improved gas mileage from 37 mpg to 42 mpg according to EPA, a 13.5% improvement.

          Shall we then say the same about the Volt, that it only increased 5 mpg?

          I think not.

          Exaggeration with sarcasm is OK for comedians, but for statisticians, its not so funny.

          • Bob_Wallace

            “The Volt improved gas mileage from 37 mpg to 42 mpg ” while adding more batteries? That’s impressive.

            2015 curb weight – 3,786 lbs.
            2016 curb weight – 3,543 lbs.

            Battery capacity jumped from 16.5 kWh to 17.1 kWh in 2015. An increase to 18.4 kWh for 2016.

            *How did they add more batteries but make it lighter? Any clues?*

          • eveee

            They improved aerodynamic efficiency which contributes more to range than weight.

            The weight was lowered 30 lbs. by increasing cell size. Pack weight was lowered, too. The rest of the car underwent the usual weight reductions.


            I like GM finally getting serious about efficiency. Its about time.

            The original Cd was a mediocre 0.30.

            The battery pack increased from 16.0 to 18.4 kwhr, about a 15% increase.

            ( wikipedia gives first gen as 16 kwhr)

            The range increased from 38 to 50 miles, a 31.5% increase. Thats not possible from the increase in pack size alone.


          • Bob_Wallace

            The Volt pack was enlarged to 16.5 in 2015.

            Some of the battery range comes from discharging batteries deeper, 9% more before the engine turns on.

            Sounds like with more kWh and a drop in battery pack weight Volt is enjoying higher capacity (more kWh/kg) batteries. More confirmation that battery capacity continues to improve.

            Any Cd rating for the 2016 Volt yet?

          • eveee

            Have not heard anything official yet. But a look at the car confirms my suspicions.

          • Bob_Wallace

            Just ran across this – talking about manufacturing batteries for the Leaf…

            “The new 30kWh lithium-ion batteries are made up of 192 cells in 24 modules packaged in four packs. The first generation battery was slightly different, but the changes increase the efficiency by 20 percent.”


            Another bit of info that battery capacity continues to rise.

          • eveee

            Here is a picture of the Leaf Pack. 48 modules, 4 cells per module. Each cell:

            avg voltage 3.8V

            33.1 A-hr

            They must be parallel series, two at a time. 192 cells would give way more than the pack voltage of about 360V.

            So 192/2 x 3.8V = 365V

            2 in parallel gives 66.2 A-hr x 365V = 24.16 khwr.

            They are side connected. The larger 4 pack is bulkier and packaging is more difficult.





            This is the best Leaf Pack dissection I have found.


            IMO, the cell is a pouch, not a prismatic. (Its has soft sides, no hard plastic packaging)


            You can see the advantages of smaller cylindrical cells. Cooling and packaging are both better.

            Also, pouch cells must be enclosed by an outside mechanical package to limit swelling and form them properly. Cells expand and contract with heat and charging, if not limited mechanically. Cells are kind of spongy that way.

          • eveee

            There is more detail, and from the GM aerodynamics engineer, there has been some real work going on.

            “With electric vehicles, such as the Volt, aerodynamics play a more important role in efficiency than conventional vehicles,” said Nina Tortosa, aerodynamics engineer. “The knowledge we took from the first-generation Volt and other recent Chevrolet cars allowed us greater leverage to work with a more flowing design.”

            The car also has a new shutter system behind the front grille that will reduce airflow at higher speed to optimize air drag vs. cooling requirements.



            The underside has seen some improvements. That would be panels to smooth the underside and keep air flow smooth.

            They might also employ deflectors in front of the wheels to move air around them.

            They added a shutter system behind the front grille to block the air flow ( cooling causes induced drag) when cooling is less needed.

            Now that I just analyzed the Leaf pack, we can compare. The new 2016 Volt pack is 192 cells, and this must also be series parallel, with 96 series.

            Since the pack is 18.4kwhr, we can tell approximate cell voltage, and capacity.

            Lithium cells are about 3.8V, so the pack voltage is about 365V. The pack A-hr rating is twice the cell rating, because the cells are parallel.

            The pack amp-hr rating is:

            18.4kw-hr = 50.4 A- hr.

            The cell Amp hour rating is half of that,

            25.2 A-hr

            Its a bit less than the Leaf rating of 33.1 A-hr per cell.

            The original Volt pack weighed about 400 pounds. The cells are also pouch type, not prismatic. Overall, they have quite a few similarities with the Leaf.


          • Joseph Dubeau

            There are fewer cells in the 2016 Volt battery pack.

          • eveee

            That’s right. 192 to be exact.

    • jeffhre

      You could (cynically?) look at it that way. Or you could see, “The new 96-cell module returns approximately 27% more peak-power compared to mainstream 260-watt multi-crystalline modules.” And that Panasonic’s announced high performing panels are a 13% improvement in energy collection over even the 27% higher than typical collection their newest high performing HIT 330 panels.

      As a statistician isn’t a 2.8% overall improvement added to the company’s current 19.7% module efficiency a 13% improvement in collection efficiency? I don’t think they will improve efficiency by 50 or 100% overnight. So this looks like pretty good work to me. If the costs continue to fall accordingly.

      Does including the words SolarCity in a post cause people to miss things like that?

      • Bob_Wallace

        “Does including the words SolarCity in a post cause people to miss things like that?”

        Ya think?

    • Joseph Dubeau

      Didn’t Pansonic hit 25% last year?
      This 22.5% is absolute hype. Just like solar city’s announcement is.
      I didn’t think there were Solar City fanboys. Wow!

      Panasonic HIT Solar Cell Achieves World’s Highest Energy Conversion Efficiency*1 of 25.6%*2 at Research Level

      • Bob_Wallace

        That’s cell performance, not an average output for a product (module) line.

        Companies report “champion” module/cell performance at times. That’s the very best cell/module picked out of a group.

        I’m not sure how you’ve decided that 22.5% is absolute hype. Are you accusing the company of dishonesty?

        • Joseph Dubeau

          Here is article I read yesterday. Look at the graph at the bottom of graph. Expand the graph and zoom in on the green and blue. They are all pretty close. In fact, the thin film is catching up.

          • Bob_Wallace

            Thin film has done a good job of efficiency growth. But it hasn’t been able to sell for considerably less than silicon which many people expected.

          • Joseph Dubeau

            I don’t understand why.
            I don’t have enough information to make any judgement.

          • Bob_Wallace

            If I’ve got things straight – processed silicon was sort of expensive. Thin film promised the ability to spray the photoreactive material in a very thin layer, saving money over thick silicon.

            Cutting the silicon thinner and thinner with less waste has allowed the price of silicon panels to drop rapidly and kept silicon heavily competitive with thin film.

      • Alex Maverick

        It says “research level”. Does this article relate to repeatable quality production samples?

    • Marion, considering that the maximum possible efficiency of a single-junction, silicon solar panel is 33.7%(Shockley–Queisser Limit), I’d say this is rather large news.

      • Bob_Wallace

        Put in that light, a 1.8% increase in output would mean getting 5.3% closer to the hypothetical maximum.

    • mike_dyke

      So if Usain Bolt broke the world record for the 100m sprint by 0.01 seconds you wouldn’t celebrate because it wasn’t statistically significant?

      • JamesWimberley

        To a statistician, all the finalists in an Olympic race are the same.

  • Dag Johansen

    It is interesting to see Panasonic compete with Tesla on battery packs and Solarcity on solar PV modules. It seems they are frenemies.

    • Larmion

      Compete with Tesla? I think you meant cooperate, as Panasonic produces the batteries Tesla slaps its fancy sticker on.

      • I believe they also have a major stake in the Gigafactory, unless I am mistaken.

        • Bob_Wallace

          Yes. Panasonic may well take up more floor space manufacturing cells in the GF than Tesla will use assembling them into battery packs.

          I seem to recall reading something a couple of days back about Tesla and Panasonic agreeing to split the storage market with Tesla taking the US and Panasonic Europe. Don’t recall details.

        • After I quit my previous job 13 months ago, I was blessed to get introduced to following job opportunity online which saved me … They offer online home base work . Last paycheck after being on this job for them for four months was 12,000 dollars …. Cool fact about that job is that the requirement for being able to start is basic typing and connection to the internet …..

          For the site address open my P R 0 ? Î L È


      • Dag Johansen

        I meant compete. Panasonic has its own home battery product available for sale. And that is why it is interesting . . . they both compete and cooperate in both cases.

Back to Top ↑