When I was a child, I spake as a child, I understood as a child, I thought as a child: but when I became a man, I put away childish things.
Judging by the excitement around the unveiling of the Bloom Box, you'd think it was the Second Coming.
The green blogs fell all over themselves repeating the breathless "Holy Grail" speculations in Lesley Stahl's 60 Minutes report, which was indistinguishable from an in-house marketing puff piece.
Bloom Energy's media blitz ended eight years in stealth mode, and the star-studded coming-out ceremony on Wednesday featured notables such as board member Gen. Colin Powell, California governor Arnold Schwarzenegger, legendary VC John Doerr of Kleiner, Perkins, Caulfield & Byers, Google co-founder Larry Page, eBay CEO John Donahoe, and Wal-Mart COO Bill Simon.
The market could dwarf the Internet, they said. It's like another Google. It's a "disruptive technology" akin to cell phones replacing land lines. It will slash CO2 emissions and replace the grid. It's twice as efficient as grid power generation. Its technology is based on common beach sand, not expensive and corrosive materials. It will "change the world." (Cue Tom Waits' "Step Right Up.")
They congratulated themselves with bear hugs and "I love you, man"s.
Bloom Energy CEO KR Sridhar's laid out his bold vision: Bloom Boxes would provide affordable, abundant, reliable, clean electricity to every home and business and light up the dark areas of the world — in a decade. It's "a market size that starts with a ‘T'," not a ‘B', he said at the company's 2002 launch.
After a decade of studying energy, such claims instantly arouse my skepticism. I had to take a closer look.
The Bloom Box is a mini power plant based on fuel cell technology. Fuel cells use an electrochemical (not thermal) process to separate the protons and electrons of the fuel and then pass the electrons through a circuit to generate electricity. At the end of the process, the protons and electrons recombine along with oxygen.
If the fuel is pure hydrogen, the emissions will be pure water, but if the fuel is a hydrocarbon, emissions will also contain carbon dioxide.
Fuel cells aren't new. The first was invented by a Swiss-German chemist named Christian Friedrich Schönbein in 1838, and at least 20 different designs now exist. A long history of companies has attempted to bring a cost-effective and practical device to market, and in the last decade hundreds of millions of dollars have been poured into their research and development.
None have achieved real commercial viability yet. According to data assembled by Fuel Cells 2000, fewer than 1,000 units are in operation or planned worldwide. One of the top public U.S. fuel cell manufacturers, Fuel Cell Energy, Inc. (NASDAQ: FCEL) has a mere 90 installations worldwide, and its stock has been moribund since 2002.
What's new about the Bloom Box is that it claims to be high efficiency (producing more power with less waste heat than other fuel cells), small, relatively cheap, and able to run on a variety of fuels including natural gas, landfill gas, and biogas. Its solid oxide design reportedly uses zirconium oxide for the proton-exchange membrane.
Let's have a look at the numbers, assembled from Bloom's statements and various blogs.
One cell produces 25 watts. A "residential sized" stack of cells produces 1 kilowatt (kW), which Sridhar claims could be on the market at a price point of $3000 in five years.
A 100 kilowatt system (before incentives) costs $700,000 to $800,000. For my purposes I'll take the high estimate and call it $8,000/kW.
Most units will run on natural gas. Assuming the gas costs $7 per million BTU, the cost of generation is in the $0.13 to $0.14 per kilowatt-hour (kWh) range, before subsidies. After factoring in the federal 30% investment tax credit, the $2,500/kW California rebate, and the claimed 10-year lifespan, the unit should produce power for roughly $0.08 to $0.10/kWh. The average retail grid power price in the U.S. is about $0.11/kWh.
Under those metrics, the company claims it will take three to five years to pay itself off, including the cost of swapping out the used-up fuel cell stack twice during the 10-year warranty period or not. This strikes me as a highly dubious claim.
The CO2 emissions when running on natural gas are reportedly 0.8 pounds/kWh, as compared with 2 pounds/kWh for coal-fired plants and 1.3 pounds/kWh for natural gas-fired plants. Hence the squishy claim that the unit produces power with "half the emissions" of grid power from natural gas.
First, a 1 kW unit isn't enough to power a house in the U.S. I know from my experience in the solar business that 2.5 kW on an averaged demand basis is more like it.
Peak demand loads can be much higher. For example, a hair dryer, a microwave, and a toaster together could draw more than 5 kW. Large houses can need 10 kW or more on average. So even at Sridhar's $3000 price point for a 1 kW unit, a residential application would cost more like $7500.
But long experience in watching "breakthrough" devices like this come to market tells me that one has to discount the initial claims by at least a factor of two. So the real price point will probably be closer to $6000/kW in five years, or $3000 in ten, in which case the unit might never pay itself off in a residential application over the 10-year warranty period.
A quick calculation by Editor Rembrandt Koppelaar at The Oil Drum also questions the payback period. Assuming $0.10/kWh for grid power and an $800,000 cost for a 100 kW unit, he calculates it would pay itself off in 15 years — five years longer than its expected lifespan.
The eBay installation featured in the rollout offers a final example. The scant available information about this installation suggests that it consists of five 100 kW units, at a cost of $800,000 each, which have saved the company $100,000 in grid power costs over nine months. If that surmise is correct, then the $4 million installation will pay for itself in 30 years.
Second, since nearly all customers will run the unit on natural gas, it doesn't fulfill the claims of clean, abundant, or cheap power.
If my expectation of a global natural gas peak in the 2020-2025 range is correct, it could in reality make the situation worse, by moving significant loads to natural gas just as supply starts to flatten out.
In effect, it would allow us to crawl farther out on the fossil-fuel limb just before it cracks.
It definitely won't make sense to use solar power to crack water into hydrogen and oxygen in order to use the hydrogen in a fuel cell. It would be far more efficient, and cheaper, to simply use the solar power.
Third, the suggestion that it will "replace the grid" is simply nonsense. Few of the customers in the commercial market will generate all of their power with Bloom Boxes (the high-profile campuses currently testing the units get 15% of their power or less from them), and most buildings already have grid connections. The units might eventually replace some of the load carried by utility power plants, but that's about it.
Even a residential application would not eliminate the need for grid power unless it was sized to meet peak loads, which would not make economic sense.
Finally, and most importantly, there is the scale problem. I don't know what universe you'd have to live in to think that a company currently producing one unit a day is going to put several billion of them into operation in one decade, or even five. Particularly if your outlook on capital markets for the next five decades is informed by an education in peak fossil fuels.
The Bloom Box doesn't belong in any discussion about renewable, clean power, or changing the world.
The main effect of the device would be to transfer some of the power generation load off centralized coal plants and onto distributed natural gas plants.
Few customers — and probably only commercial and industrial ones, at that — will have the option of running it on biogas or landfill gas. For the slightly more than half of the homes in the U.S. that even have a natural gas line, it won't make economic sense.
Therefore, I do not expect it to become a viable residential application. Nor do I expect it to light up the Third World without installing a network of natural gas lines — in itself, an unlikely proposition.
A fair comparison would be to a standard natural gas-fired backup generator. A quick Google search finds an 18 kW Briggs & Stratton natural gas backup generator for $4,200. If $3,000 will get me a 1 kW Bloom Box, then an 18 kW device would be $54,000. Is that a price premium any homeowner would pay for slightly reduced emissions?
For another cost comparison, at $8000/kW, rooftop solar (after incentives) is cheaper today. As long as you have a functioning grid, the 24x7 benefit of a fuel cell (assuming uninterrupted natural gas supply) wouldn't be worth the cost premium over solar. And once it's installed, a solar PV system consumes no fuel, and produces no emissions.
By time a Bloom Box goes for $3,000/kW, my bet is that solar will still be cheaper. Should natural gas prices go to $15 or $20 in the next decade (a not unreasonable proposition) then solar will be half the price, or less, and the payback period for the fuel cell would lengthen considerably.
What it can do is allow commercial customers to claim some green cred for reduced emissions while paying close to the going market rate for power.
However, I expect a solid handful of more mature companies (like FuelCell Energy, Kyocera, UTC, and Ballard Technologies) to give them a run for their money.
If the Bloom Box does, in fact, sport a 50% efficiency gain over utility plants — which I think still needs to be proved — then that should confer an advantage on it in the form of carbon reduction incentives. That may be the best advantage the Bloom Box has.
There are certainly important intangible benefits in distributed generation and baseload (24x7) capacity, as my readers well know.
However, the Bloom Box's reliance on natural gas cannot be overlooked, and it appears that nearly everyone has overlooked it here.
The short lifespan of the device and the need to swap out the cell stack every five years must be factored in as well. The cost of maintenance and the availability of service technicians are important questions that still loom over the Bloom.
By comparison, a rooftop solar installation is low tech, low maintenance, and far more durable.
In short, I view the Bloom Box as a modest gain over the status quo in natural gas fired power supply. A world-changer it is not.
The most interesting part of the Bloom Box story is the social aspect.
Lesley Stahl's gushing 60 Minutes take on the Bloom Box was, in so many ways, a paragon of everything that's wrong with energy coverage in the media.
She was in hot pursuit of "the next big thing," and found the unit to be "awfully dazzling" in a market "worth bazillions."
"I'm installing a power plant!" she exclaimed with childlike glee, as she peeled the shipping packaging off a new unit.
She was obviously very impressed that the technology was an inversion of an invention that could produce oxygen so people could live on Mars.
Her opening statement, "In the world of energy, the Holy Grail is a power source that's inexpensive and clean, with no emissions," is either a complete non-sequitur, or a concise demonstration of her energy illiteracy. One leans toward the latter explanation after watching her ask if the box could use solar as a power source, and Sridhar's humoring affirmation.
The pressure is clearly on the MSM to make some noise for Holy Grails.
There is also something telling about the appetite for hope in the way the blogosphere lapped up the excitement around the unveiling. The appeal to authority of the brass on stage clearly worked, producing uncritical comments like "Gee whiz, $400 million in capital, it clearly works, it's cheap, and fits in my backyard? I'm in!"
The fact is that the energy problem is too big to grail. Or, as the peakists say, "There are no silver bullets, only silver BBs."
BBs as in Bloom Box.
I want to be clear. I spent nearly two decades in the computer industry before I got seriously into energy. I used my first computer (a very early, educational prototype) at the age of five, in 1969. I saw the computer revolution firsthand, and I know the power of technological development.
But I also know that the ingrained optimism of Silicon Valley entrepreneurs — as much as I love them — simply does not translate to the challenge of generating or saving hard BTUs. No single technology will save us. Moore's Law does not apply here. The history of energy is littered with the bodies of enterprising souls just like them.
One thing I will say: Only a venture capital firm with the power of Kleiner, Perkins could coordinate such a media blitz and star-studded unveiling, and wow the socks off the media. My hat is off; they scored a major coup with this one.
I remain staunchly rooted in numbers and of the mind that it's better to have no hope than false hope, because it pushes us toward real solutions.
The Doomsday clock is ticking. It's time to put aside childish things, retire the phrase "Holy Grail" permanently, and get real about energy.
Until next time,
P.S. While the Bloom Box is no doubt a step in the right direction in the ongoing search for energy's Holy Grail, we still need to consider other sources of energy to deal with the scope of the oncoming energy crisis — and all signs point to nuclear as a viable option. Green Chip has uncovered a little-known metals company with a monopoly on the new "monter metal" that will change the nuclear game. This brand-new report reveals this $0.16 company about to launch to the top of the industry.
Editor's Note: Kudos to Lux Research, The Oil Drum, Wired and WattHead for their informed commentary on the Bloom Box.