CleanTechnica is the #1 cleantech-focused
website
 in the world. Subscribe today!


Clean Power Google Little Box Challenge for inverters

Published on July 24th, 2014 | by Tina Casey

28

Google’s $1 Million Little Box Challenge: Invent A Smaller Thingy

Share on Google+Share on RedditShare on StumbleUponTweet about this on TwitterShare on LinkedInShare on FacebookPin on PinterestDigg thisShare on TumblrBuffer this pageEmail this to someone

July 24th, 2014 by
 
Google has just officially launched the Little Box Challenge, a $1 million competition that could help push the cost of rooftop solar panels down through the floor and into the basement. The prize goes to anybody who can come up with a tablet-sized version of an inverter. If you don’t know what an inverter is, that’s okay. It’s this thing which everybody ignores because it’s not particularly sexy and it doesn’t have a reality show (yet), but it is an essential part of the rooftop solar power package and other solar harvesting systems.

Google Little Box Challenge for inverters

The Little Box Challenge (cropped) courtesy of Google and IEEE.

The Little Box Challenge

So, what’s up with the Little Box Challenge, and what’s the big deal over inverters?

An inverter is a device that converts DC (direct current) to AC (alternating current). If you have rooftop solar panels you gotta have one of these things because when solar cells take in sunlight, they spit out electricity in the form of DC, and that would fry the brains out of everything in your house including your home charging station (that’s probably why Google included that generic EV in the infographic above).

You need something to convert DC to AC, and that’s where the inverter comes in.

The problem, at least as far as Google sees it, is that a typical inverter is the size of a cooler, which Google feels is too big. Google feels (don’t laugh, companies have feelings, too) that a much smaller inverter would “enable more solar-powered homes, more efficient distributed electrical grids, and could help bring electricity to the most remote parts of the planet.”

While size matters to us, too, we’re not quite as obsessed as Google is.  Smaller is not necessarily more efficient. However, we can think of lots of ways in which smaller is better, and smaller can also lead to improved efficiencies. The more efficient the inverter, the better, since more of the collected solar energy gets whipped into the form of a usable current.

Specifically, Google says it is aiming for the size of a tablet. Go ahead, Google, you can say iPad — kidding!

Hey, What About Microinverters!

Actually, there already are these things called microinverters, which can be built into individual solar panels. We took note of them back in 2011, when a company called BenQ Solar introduced its “plug-and-play” rooftop solar panel with a built-in microinverter, which looks suspiciously like, you guessed it, a tablet.

The problem is that one conventional microinverter can’t handle a whole array of rooftop solar panels, so then the question is which is more cost-effective and/or has broader potential for applications: installing an array of panels each with its own built-in microinverter, or coming up with one new, superpowerful, hyper-efficient, cutting edge microinverter that can handle the whole thing at once.

Just a wild guess, but we’re thinking that relatively conventional, relatively low cost microinverters will continue to play a role in the rooftop solar panel market, but if the Little Box Challenge pulls in a winner then relatively expensive new microinverter technology could make gains in the building-integrated solar market as well as in rooftop arrays.

What About The Gallium Nitride?

Since the Little Box Challenge suggests that you take a look at gallium nitride if you want to grab the prize money, let’s take a closer look at that gallium nitride.

Gallium nitride is a compound material with a crystalline structure, commonly used as a semiconductor in LEDs as well as inverters.

 

CleanTechnica first took note of the Little Box Challenge when Google leaked word out last May, further noting that Google Ventures is an investor in Transform. That company is focusing on making gallium nitride more efficient, so Google stands to gain big if you can make more inverters while using less gallium nitride.

For the record, the Little Box Challenge also suggests that you explore the possibility of using silicon carbide.

Also for the record, Google’s partner in the Little Box Challenge is IEEE, which stands for the Institute of Electrical and Electronics Engineers. They will love you if you pronounce their acronym “Eye-triple-E” instead of sounding like you just broke your toe.

Follow me on Twitter and Google+.

Keep up to date with all the hottest cleantech news by subscribing to our (free) cleantech newsletter, or keep an eye on sector-specific news by getting our (also free) solar energy newsletter, electric vehicle newsletter, or wind energy newsletter.

Print Friendly

Share on Google+Share on RedditShare on StumbleUponTweet about this on TwitterShare on LinkedInShare on FacebookPin on PinterestDigg thisShare on TumblrBuffer this pageEmail this to someone

Tags: , , , ,


About the Author

Tina Casey specializes in military and corporate sustainability, advanced technology, emerging materials, biofuels, and water and wastewater issues. Tina’s articles are reposted frequently on Reuters, Scientific American, and many other sites. Views expressed are her own. Follow her on Twitter @TinaMCasey and Google+.



  • Orson Cart

    Hi all, I’ve been reading about the 2kVA, 40 cu In google inverter challenge, as I design inverters for a living, then I found http://www.magnetpowernz.co.nz seems like they already have the technology in cut and dried form but want someone to build it to their design… is this possible?
    Cheers, Orson

  • vensonata

    One wonders why the huge inverter companies from Europe have not already maximized the efficiency and size of these things, they are not basement workshop companies. Anyway inverter cost is a big part of the solution. They add 30-50cents per watt to a pv system. That is about 40% addition to the cost of the solar panels.

    • nakedChimp

      All gear I’ve seen so far from the inside looks pretty well designed from what the electronics engineers can do with what they get.. if there are any gains in size possible it will have to come from the semiconductor manufacturers.. these solid state switches are needed in some numbers (10-20) in there and produce most of the heat which implies some sort of heatsink.. passive = big+silent, acitve = small+noisy. Then there are the coils and capacitors which take most of the remaining volume.. banks of 600V capacitors where the energy is buffered and then the coils for the same, just different, one needs both for power conversion.

      The funny thing is, if everybody would just go for 380/400 VDC on the outputs instead of the 230/120VAC you could spare some of the stuff inside… probably 1/3 to 1/2, but yeah.. there is just no 380/400VDC distribution system yet for normal households. The only ones doing this NOW are 2 sectors afaik:
      1) EVs (the DC power bus usually is around that number)
      2) IT (DC distribution in datacenters, Google/FB very active here..)

      • GCO

        Indeed.
        My observations so far match yours, but see actual or potential improvements in most areas:
        - Semiconductors get better, e.g. silicon-carbide, although that’s probably still a couple years down for PV inverters.
        - Not tying the current-carrying conductors of the PV array to ground enables transformer-less inverter topologies — big space and weight savings (plus, no buzzing).
        - Even without switching to DC, the large capacitors requirement shrinks dramatically with 3-phase AC, common across Europe.

  • MarTams

    INCORRECT: “….would fry the brains out of everything in your house including your home charging station”

    You can use direct DC charging to recharge your battery!!!

    • Bob_Wallace

      Yep. That is common practice.

      Array and battery bank voltage are matched. Only differs where arrays are wired for very high voltage in order to reduce conductor size.

      • nakedChimp

        “Array and battery bank voltage are matched.”
        These days with MPPT chargers the charger happily does this for you.. you can use grid-tie panels (Voc per from 30 to 70 V) to charge 24/48 V battery banks, as long as the power being converted is within the chargers specs..

    • GCO

      Obviously batteries are DC, but whether chargers accept DC (and at what voltage) is dependent on their design.

      More generally, while transformers and AC motors won’t tolerate DC, a lot of electronics would probably just work fine actually: the first stage of most power supplies is either a simple rectifier, or an active PFC (power factor correction) circuitry, both of which take DC as happily as AC.

      Some manufacturers (e.g. Meanwell) even make it explicit, e.g: “Input range: 88~264VAC, 124~370VDC.”

      • nakedChimp

        “a lot of electronics would probably just work fine actually: the first stage of most power supplies is either a simple rectifier, or an active PFC (power factor correction) circuitry, both of which take DC as happily as AC”
        absolutely true, now the interesting point would be to take 380/400 V DC solar and feed it directly to appliances like those..

        Hell, even the Inverter Aircon I got here would run like this, the first thing it does is to rectify the AC that comes in just to make DC and than to create it’s own 3-phase directly in the outdoor unit to run the compressor from that. It’s really funny.

        • GCO

          Yup. 240VAC peaks at “only” 340V though, so care would probably need to be taken to keep the DC input at or below that value (most of the AC power supplies I’ve seen also spec’d for DC tolerate 360 or 370V max).

          A PV array sized to deliver maximum power around 380V could reach 500V when cold and under no/light load, too high to feed directly to appliances not specifically designed for this.

    • jeffhre

      Yes, but that is a solution that skips the home charging station and is connected directly to the cars on-board DC charging unit.

  • http://www.michaeljberndtson.com/ Michael Berndtson

    I’ve learned over the years as a chemical engineer to contract with electrical engineers, really good electricians and hands on I&C techs. Regardless of the fact that power feed, electrical work, instrumentation and controls (I&C) is an essential ingredient to process engineering, any process. So I’m out. I’ve worked in hazardous environments (explosive) most of my working life so I like well trained technicians and bright folks helping out. Nonetheless, I did like how Tina kind of suggested we first google “inverter” before going to our basements and garages.

    Serious question to anybody. Why not transition our energy economy to include both DC and AC transmission and distribution for awhile? Isn’t this AC to DC conversion/inversion and visa versa simply a waste of energy? I’m assuming an inverter, though necessary, exhausts heat. Add up the little devices, computers, servers and what not, I’m guessing a bunch of energy is wasted. For instance, solar and wind generates DC. Many end use equipment and miscellaneous stuff works better or requires DC to operate properly. Why not just transmit and feed DC instead of converting/inverting at the power source, transmit as AC, convert/invert at the end use to DC. The really expensive infrastructure and electricity distribution appurtenances are already in place. That would be power line right-of-ways and races (buried or not). Though costly, it may not be too expensive. AC lines could still feed whatever is necessary.

    Here’s where I’m getting at. One of the biggest users of electricity in the world are pumps. Pumps for moving any type of liquid, but mostly water. We’ll be doing a lot more pumping and piping of water as climate change impacts accumulate. An AC motor is pretty efficient, but is limited on speed control without a variable frequency drive. A DC motor pump is more controllable and therefore uses much less power compared to a AC pump that is pumping way off the curve. So why not feed solar and wind power as DC directly to water supply and agriculture areas, i.e. California. And maybe internet server farms as well. End of rant.

    • vensonata

      off grid people know this. Variable speed dc pumps use a fraction of the energy of AC. Although, for instance, I replaced a standard Ac deep well pump which brought up 2.5 gallons per minute from 300 feet. It used 1250 watts/hour and spiked at start up to 2500watts. The replacement was the German made Sq flex. Now get this: it uses 220 watts! to do the same thing, 2.5gal minute 300feet. Also no spike at start up. 80% reduction in energy! You have to wonder if it is all a conspiracy to squander energy. So now the other amazing thing: the pump use either DC direct variable right off of DC solar panels or it uses Ac 120v. It is automatic, no inverters or switches. We really are hillbillies here in North America…what is wrong with us?

      • http://www.michaeljberndtson.com/ Michael Berndtson

        Awesome stuff. It’s amazing how long we’ve been valving open or closed for flow control rather than controlling the motor speed for flow control. It’s becoming common for industrial facilities to relay directly to the pump motor rather than a control valve for flow. There’s a lot of old installations throughout the world. Once a pump falls off it’s design curve, which happens much of the time from valving one way or another, it wastes a lot of power.

    • nakedChimp

      depending on the electric motor you need a ‘wandering’ electrical field to move something.. in households those usually are compressor pumps for fridges/freezers and the aircon, more remotely some water pumps also work that way. Most if not all other loads can be dealt with by DC.
      As for the AC motors.. no problem giving them some semiconductors + controllers and they can be run from DC by creating the phases right there, even with speed control so they soft start etc.. latest developments of the controllers include all kinds of neat and nifty things to make them more economical too.

      So yeah, get the 380/400VDC distribution going.

      Don’t forget connectors for this.. so far the only I know is from Anderson, called Saf-D-Grid®. Looks similar to the connectors for PSUs on your computers, but they’re made for DC.. must be patented or something as they’re still the only manufacturer of those and they’re pretty expensive.

      400V 20A without rendering the connector totally useless while breaking the load with it is kinda impressive.. afaik Google/FB are using those in their DC distribution systems.

      • http://www.michaeljberndtson.com/ Michael Berndtson

        Great stuff. I learned something today.

    • GCO

      For AC generation, going with a DC grid doesn’t save any conversion.
      For DC generation (e.g. solar PV), you’d replace DC-AC then AC-DC conversions with DC-DC (to match your new grid voltage) then DC-DC (to whatever the application needs).

      Yes, DC-DC is simpler and should be more efficient, but DC-AC has become pretty damn good too: for example, PV inverters get 95~98.5%.

      Problem is, the grid also requires conversions, from high-power, high-voltage lines, to lower-voltage distribution, etc down to something adequate for household use.
      AC-AC is simple and scales well: transformers can be made of almost any size. DC-DC remains challenging at very high power levels.

      While DC can bring advantages, I don’t think they’re significant enough to ever justify the kind of dual grid you suggest. What I could imagine however is a slow, progressive “DC-ification” of the grid as power semiconductors improve, starting from high-voltage transmission, but I wouldn’t expect to see this trickle down to us end-users for several decades.

      • http://www.michaeljberndtson.com/ Michael Berndtson

        I’m even more convinced now entering Google’s competition would be hopeless. I can’t even bend wire conduit right for my home improvement projects. Chicago has a code. No Romex. Outstanding stuff, sir/madam.

        So to recap. For the time being we’re not in bad shape with transmission and delivery, but all things considered, there could be efficiency gains and performance improvements. Power generation and grid stability has more room for improvement. End use from gizmos to industrial equipment, in terms of efficiency and reduction of use, has enormous room for improvement. Correct?

        I have another question. The blog Seeking Alpha is an investor blog and pretty popular. One of the writers who focusses on energy made some interesting statements in this post last June:

        “How Energy Storage Will Help Wind And Solar Power Clean Up Their Act,” by John Petersen, 6/16/14

        http://seekingalpha.com/article/2270163-how-energy-storage-will-help-wind-and-solar-power-clean-up-their-act

        Copied from the post:

        “Let’s forget about ideology for a few minutes and focus on facts. Photovoltaic panels can only produce power for six hours a day and wind turbines can only produce power when the wind is blowing. Even with good conditions the electric current from wind and solar power facilities fluctuates wildly. When the sun doesn’t shine or the wind doesn’t blow, those facilities are little more than landscape art. The only way renewable power technologies can be truly useful to a society that can’t function without clean electricity is if backup facilities are waiting on standby to fill the gaps.”

        I’m not sure how I feel about this guy – I believe he’s in the energy storage business – so there’s that. I’m sure he’s pretty knowledgeable. However, Seeking Alpha does get read by a lot of folks with money to invest. They read this type of stuff over technical blogs.

        He later goes into how PV solar PV and wind are “dirty electricity”

        “Is Residential PV Solar More Trouble Than It’s Worth?”

        http://seekingalpha.com/article/2287833-is-residential-pv-solar-more-trouble-than-its-worth

        “The costs utilities pay to smooth and stabilize the filthy electric current from residential PV solar and provide standby facilities for the inevitable cloudy, rainy, gloomy days are huge. Sadly, the residential PV system owners who created the instability in the first place do not pay the costs of intermittency abatement. Instead, those costs are simply folded into the rate base and paid by the utility’s other customers. It’s a great deal for the PV solar system owner and abusive for all other users of electric power.”

        Maybe this is where I got my musing on separating DC from AC. Is any of this true? That PV and wind produce so called dirty electricity?

        • GCO

          Persons post on Seeking Alpha for financial gain, to support their own investments and nothing else. When not flat-out lies, the points made there are very, very biased to say the least.

          Here we have the CFO of a company into batteries and diesel generators. Well, what would you expect?

          His claims about residential PV being filthy are beyond idiotic. First, grid-tie inverters produce super-clean power.
          Not just pollution-free, electrically too: beautiful sinewave, perfectly in sync with the rest of the grid, much better than, say, generators (oops).

          Next, solar, and residential PV in particular, destabilizing the grid, is also false: it’s actually the opposite.

          First just some common-sense evaluation here:
          I have a PV system: it starts smoothly in the morning, reaches its max around 2pm (about 5kW these days), and slowly fades back down, probably correlating nicely with air conditioning usage in the neighborhood.
          I also have an electric dryer: it instantly goes from 0 to 7kW, then keeps switching between ~200W and 7kW several times per minute to maintain temperature.
          Which one does disturb the grid more? Neither actually, they’re both utterly insignificant.

          If any nation ought to know the “dangers” of going all-out on wind and solar, it’s Germany. Yet

          Germany ranks #1 in European grid reliability, Denmark just behind, both about tenfold better than the U.S.
          Across Europe, renewable expansion correlates with more reliable power.

          More on the topic => http://energytransition.de/2013/08/energiewende-separating-fact-from-fiction/

        • Bob_Wallace

          Petersen is a long time anti-EV, anti-renewable energy guy who is far from honest.

          I went a few rounds with him a couple/three years back and I put him in the same basket with Robert Bryce.

          Sometimes there are decent pieces on Seeking Alpha, even some authors who are willing to take on new facts and work with them. But there are also people who seem to be either pushing stocks or trying to make money short-selling.

          I’d recommend staying away from the site as much as possible. If you can’t generally trust the people running the site to keep things somewhat honest then it’s just too risky.

    • jeffhre

      Long range transmission with DC would get really expensive with a lot more copper required for high power transmission, in most cases. Until superconducting systems cost less, long range AC is generally quite a bit less expensive to build.

      • Bob_Wallace

        Just the opposite. DC is more efficient for long distance transmission. More power per diameter of copper/whatever.

        There is some power loss at each end (IIRC ~0.75%) converting from grid AC to high voltage DC and back down again. That makes short runs of HVDC inefficient but over modest distances that loss is recovered.

        http://en.wikipedia.org/wiki/High-voltage_direct_current

        • jeffhre

          I don’t understand this one. Wouldn’t the copper depend on the amperage? And isn’t it expensive to convert DC? If converting from rotating generators to DC is expensive and raising the voltage and lowering the amperage is expensive with DC, how is it ultimately less expensive to use DC for long distance transmission? Not to be argumentative, I just do not know the answer :)

  • upnorthenergy

    I understood micro-inverters were the recipe for rooftop installations. For larger arrays, one shadow across any portion of the array will affect the entire array due to the layout of panels. With a micro inverter per panel, the outputs are all summed together and a shadow only affects the efficiency of the panel in the shadow.

    • johnBas5
    • GCO

      Inverters use MPPT (maximum power point tracking) to continuously adjust how much the current they get from the PV modules, so as to get as much power as possible.
      When multiple modules are stringed together, this optimization is done on the while array, and therefore indeed, a single shaded or dirty module (or even cell) end up having more effect overall than ideal.
      [Bypass diodes, which are integrated in all modules I know of, are important to prevent shaded cells/modules from being damaged by the high current which would otherwise be driven across them by the rest of the string. They don't allow the bypassed cells/modules to contribute any power however.]

      To minimize the impact of partial shading, soiling and in general any discrepancy between modules, independent MPPT can be done for each one.
      Micro-inverters inherently do that. They also do the DC to AC conversion at each module though, and use proprietary wiring, so are only cost-effective for relatively small systems (e.g. a dozen modules max).

      Another approach is power optimizers (from e.g. Tigo or SolarEdge). Those do per-module MPPT and monitoring, but leave the DC-AC conversion to central inverter(s), effectively combining the advantages of micro-inverters with the lower cost per watt of larger inverters.

  • Boris

    Generic EV it is not, it’s Tesla Model X

Back to Top ↑