Navigant Research Blog

Google’s Floating Data Center Is No Fantasy

— November 5, 2013

Back in 2009, in what already seems like the technology dark ages, the U.S. Patent and Trademark Office awarded Google a patent for a floating data center that would use the ocean to provide power and cooling.  The original patent reads as if someone at Google was channeling the late, and truly great, Nikola Tesla with plans for using wave and wind power, among other sources, combined in an islanded microgrid.  Since Google has a patent war chest of more than 18,000 U.S.-based patents, it was easy to see the floating islands as nothing more than a vision, similar to Daniel Zubrin’s floating cities as a first step to Mars.  Now, though, it seems that Google may in fact be building two of these floating data centers – so, we could actually see the creation of not only the first floating data center, but also the lowest carbon data center in the world.

If this is the case, how would this data center be powered and cooled?  First, Google will not be the first company to apply shipping container architecture, along with water cooling, to servers.  Rackable has designed a unit containing 28 server racks, with up to 1,400 servers, in a standard 40-foot shipping container.  Using water for cooling reduces energy demand, it is claimed, by 25 kilowatts (kW).  If, due to better cooling, the average server consumes 800 watts per hour at 28 servers per container, each container would require 268 kW of power per day.

A Tidy Fit

One photo shows 12 shipping containers going onto the potential Google barge.  If each is a server box, then this represents 3.2 megawatts (MW) of power demand, not including the power needed for the barge and any living compartments.  Let’s say the total power per day is 4 MW.  Assuming there is not going to be four 1 MW wind turbines and some heavy-duty energy storage on board, we are likely looking at some combination of solar and fuel cell power.

We already know Google understands this combination; it has operated a solid oxide fuel cell Bloom Box and has had solar panels at its headquarters since 2008.  Handily, fuel cells also can come in 40-foot shipping containers with solar panels attached to the top.  So, this would fit neatly into the configuration of the barge.  Using figures from a recent study looking at powering a barge in the United States with polymer electrolyte member fuel cells, 4 MW of fuel cells and solar would likely require between six to eight containers, with some built-in batteries for energy storage.  This would take the grand total up to 20 containers – 12 for the servers and eight for the power plant and fuel.  If the Pelamis offshore wave energy converter system, which was also mentioned in the grant, was working, this could be reduced somewhat. In other words, suddenly, these plans stop being visionary and start to be doable.  Using fuel cells, solar, and batteries, we not only could see the first floating data center, but also the first near-zero carbon data center.


United Kingdom Throws a FIT over Solar

— November 5, 2013

In two recent blogs, I discussed the fervent debate taking place in the United States over the efforts of some utilities to change their net metering policies in ways that critics call a “tax” on solar, and provided details on how Germany’s aggressive promotion of renewables has resulted in dire financial consequences for German utilities and high rates for consumers.  Now, on to the United Kingdom, where sharp changes to the Feed-In Tariff (FIT) program during just 3 years have placed considerable stress on the marketplace.

The FIT went into effect in April 2010, offering 43.3 pence ($0.69) per kilowatt-hour (kWh) to individuals generating solar energy with less than 5 megawatts (MW) of capacity.  (Notably, those who had installed solar panels prior to the FIT program were ineligible, and continue, to this day, to receive just 9 pence per kWh.)  The new generation tariff was paid whether the homeowner used the electricity or not, and an additional 3.2 pence per kWh was paid for energy exported back to the grid (the export tariff).  The costs of the program are paid by the utilities, which spread them across the entire customer base for recovery.

Falling FIT

By late 2011, it was clear that solar take-up rates were greatly exceeding the plan’s original expectations, and the Department of Energy and Climate Change (DECC) announced that it would cut generation tariffs by more than half, to 21 pence.  Lawsuits ensued, but by March of the following year, the cut was made.  Further cuts came in August of 2012, bringing the base rate down to 16 pence; and in the time since, the Office of Gas and Electricity Markets (Ofgem) has periodically lowered payments by a predetermined (and complicated) degression formula based on the rate of PV system deployment and the actual costs of solar panels.  As a consolation for the falling generation tariff, solar owners now receive 4.5 pence for exported kWh, rather than 3.2 pence – but the lower generation tariff is only good for 20 years, rather than for 25 years under the original scheme.

Rates Up, Installations Down

Partly as a result of the volatile policies, U.K. solar installations have slowed dramatically.  According to Ofgem data, nearly 470,000 small PV systems have been installed through the program.  But in the June, the monthly installment rate fell to just more than 6,000 systems, down from 14,500 per month 1 year ago.  About 1.7 gigawatts (GWs) of solar capacity have been installed through the program since its inception.

Solar advocates note that the program is still lucrative for homeowners, because the costs of the systems have also fallen sharply.  At the time of the last FIT cut, the Solar Trade Association in the United Kingdom said that PV system buyers still earn about a 9% return on their investment, and pointed out that electric rates were still rising.  Indeed, according to a recent report by DECC, electric rates across the United Kingdom have risen by nearly 50% since 2005 in real terms.

The FIT program in the United Kingdom was controversial and the abrupt policy changes have led directly to business failures, like the one described in this Dragon’s Den article.  But where the grumbling appears to have tapered off in the United Kingdom , ire over net metering policies in the U.S. is just hitting its stride.

Policies meant to drive usage of renewable energy must be sustainable (pun intended) for both customers and utilities.  In my next blog, I’ll discuss some of the proposals designed to align the longer-term climate goals of renewable integration with the nearer-term financial needs of utilities and consumers.


Fuel Cell Market Gets Real

— November 5, 2013

Is there such a thing as “the fuel cell industry”? The industry is really a collection of disparate applications and markets.  What exactly do companies focused on passenger cars, uninterruptible power, energy storage, residential power, or forklifts have in common? One thing that the hosts of this year’s Fuel Cell Seminar & Energy Exposition, in Columbus, Ohio, hoped they had in common was the supply chain, which is Ohio’s strength in this sector.  And, one thing I learned at the Seminar’s plenary sessions is that there is there is no fuel cell in the world that doesn’t have an Ohio component in it.

Beyond that, these markets have quite different stories to tell on where they are in the technology development timeline and where they are going.  But the one theme I heard repeated in Columbus was realism – realism about the need to reduce costs to compete in the commercial market.  Two companies, Honda and American Electric Power, stressed that fuel cell technology is ready, but the costs must come down to compete against the many other clean, efficient options available.  Honda’s Bill Konstantacos spent much of his talk  touting the advantages of its gas and hybrid vehicles, which seemed rather off topic, until the point was made that fuel cells have to compete with these technologies, and will not be adopted just because supporters think fuel cells are the best zero-emissions option.  Reducing costs brings us back to the supply chain, since that is where the costs are going to come out at this stage, more so than from any basic research and development.

New Markets for Natural Gas

Other speakers also veered off-topic, promoting their own fuel or technology in addition to fuel cells.  Thus, we had Kathryn Clay of the Drive Natural Gas Initiative touting natural gas vehicles – and, in spite of claims that natural gas infrastructure might be a pathway to hydrogen infrastructure, this does not seem likely.  That said, I credit H2USA, the group developing a road map for U.S. hydrogen infrastructure rollout, for getting the natural gas industry on board with its efforts.  Fuel cell vehicles that use hydrogen from reformed natural gas can offer another domestic market as U.S. gas supplies increase.  It will not be until the latter part of this decade at the earliest, but the U.S. natural gas industry has to be making long-term plans on how to utilize the supplies from the U.S. shale gas boom beyond the export option.

I was surprised to see fuel cell vehicles (FCVs) placed in a very prominent role in the seminar’s plenary sessions.  FCVs have long played an outsized role in the public face of fuel cells, thanks to the (mostly contrived) battle between FCVs and battery electric vehicles, and because the media finds it more exciting to talk about cars than power boxes.  Frankly, this is not helpful to the rest of the fuel cell world because it creates an impression that the technology is not yet ready for prime time when, in fact, more than 28,000 fuel cell systems were shipped in 2012.  Still, there was some news on the FCV front – Honda has finally committed to introducing a new FCV in 2015 (its FCX Clarity is from 2008) and Toyota has said it is on track to produce its first production FCV in 2015.  Add to that the commitment in California to fund hydrogen fueling during the next 10 years, and there is continued momentum in the FCV arena.  It just requires being realistic about the timeline: Navigant Research’s report, Fuel Cell Vehicles, marks 2020 as the tipping point for this market.  In the meantime, other fuel cell applications are quietly making inroads into their respective markets.


Winners Emerge in EV Battery Race

— November 4, 2013

In the fall of 2007, General Motors announced the launch of a new program to develop a plug-in car called the Chevy Volt, officially launching the advanced battery industry.  For such a car to work, a new kind of battery had to be created that was affordable but with more energy dense than existing batteries.  Soon, every carmaker, battery manufacturer, electric utility and consumer electronics company sidled up to the table and placed their bets on this emerging industry.  Now, 6 years later, the first stage of this horse race is over and the judgment can begin to determine winners and losers.  Here are the initial winners, in my view:

Lithium Ion: By far the biggest winner is the chemistry that has taken up more than 99% of the market: lithium ion.  While the drawbacks to Li-ion are well-known (fire potential from thermal runaway, cost, lithium supply constraints), each has been mitigated by a combination of engineering advances and economies of scale.  Li-ion batteries have completely taken over markets, a few years after entering them.  This happened in consumer electronics, power tools and electric vehicles.  And the price of Li-ion batteries has dropped dramatically—so much so that other, supposedly cheaper, chemistries have had no chance to compete.  For a closer look at the current state and the future of the Li-ion battery segment, please join us for our webinar, “The Lithium-Ion Inflection Point,” at 2 p.m. ET on Tuesday, November 5.

Tesla/Panasonic: Of all the players in this space, the ones who made the biggest bets on the future of advanced batteries either don’t exist anymore (Better Place, Coda Automotive) or have triumphed.  In the latter category, the best example is the Tesla-Panasonic partnership.  Both companies bet big that their battery solution would win out.  And both have reaped the rewards.

Tesla’s concept of using small-format batteries in combination with an expensive and sophisticated thermal management system has worked very well.  Panasonic put all of its chips on its new nickel cobalt aluminum cathode battery that powers the Tesla Model S.  The success of that car has saved Panasonic’s battery business and its factories are now operating at full capacity while its Japanese brethren such as Sony and NEC see the demand for their batteries declining.

The Observers: Ironically, the other big winners of the initial stage of the advanced battery race are the companies that haven’t placed any bets at all — yet.  Hyundai Motors is one example.  The company has not made any big investments in the electric vehicle space, but is now poised to enter that market in a big way, without being dragged down by stranded manufacturing investments.  Likewise, Johnson Controls, whose Li-ion subsidiary is overshadowed by its massive lead acid battery business, is now able to enter the market in manner of its own choosing, with a keg full of dry powder and a much more visible path to success.

In my next blog I’ll review the losers, so far, in the advanced batteries space.


Blog Articles

Most Recent

By Date


Clean Transportation, Electric Vehicles, Finance & Investing, Policy & Regulation, Renewable Energy, Smart Energy Practice, Smart Energy Program, Smart Transportation Practice, Smart Transportation Program, Utility Innovations

By Author