Bloom Energy

[Jammrock]

Excuse me, but I am still extremely sceptical about this kind of technology. In the first place, according to the description it still uses fossil fuels ...

Does not require fossil fuels. Fossil fuel is an option. One of the advantages of the Bloom Server, if it works as reported, is that is can use any number of "local sources" of gas as fuel, including biofuels. When using a biofuel they claim it is "carbon neutral". When using fossil fuels they claim it is more Eco-friendly because of the efficiency.

For example, the eBay installation uses biofuel. I believe in the 60 Minutes piece they say they use landfill gases to power their Bloom Servers.

http://bloomenergy.com/customers/customer-story-ebay/

http://bloomenergy.com/benefits/save-money-and-the-environment/

I admit they are all claims at this point, but seeing as they have working solutions in production it appears more feasible versus the other alternate energy options on the market or that have been announced.

While burning the gas may be slightly more efficient all of those solutions I know of involve moving parts which make it infeasible for anything but utility or government deployment. This re-introduces grid loss and ultimately makes the solution less efficient as local generation. If Bloom and their partners can provide ready sources of inexpensive biofuels and affordable home installations it could be the beginning of a real alternate energy revolution.

Could. As usual, time will be the ultimate judge.

[Brian Ellis]

Until it becomes mainstream for at least five years without major problems, it remains vapourware. Prototypes are vapourware (Greek protos means first). 5-10 years hence is hardly mainstream.

Biofuels are so dirty that the purificataion they would require would probably cost more than the energy produced. Pure methane is VERY difficullt to obtain chemically, even from so-called natural gas. All impurities would clog the electrolyte in time; remains to be seen the time scale. Ten years may be a commercial exaggeration (probably is).

Better for district, rather than home generation (say, 1 larger plant covering 1,000-10,000 homes) as the efficiency would be better and could be maintained by people that know what they are doing. Also, diurnal demand variations could be better coped with.

Agree that district cooling is feasible (there is an ongoing study here at this time, as it happens), but the efficiency is much lower than district heating.

The real clincher is that it works at elevated temperatures (like Ballard cells) and will take time to come on line.

Hydrogen is more easily made in pure form (electrolysis from ultra-pure water, NOT reforming NG) and would be better. Want hydrogen in your basement?

[Wulfman]

by the way, in the 60 minutes piece, the reporter asks about using solar (8:00), and the inventor says something along the lines of "yes, we can use solar". what did I miss? is he honestly suggesting using solar electricity to create hydrogen, to feed into the bloom box?

mfg
wulfman

[Jammrock]

Want hydrogen in your basement?

In steel lined storage container set inside a concrete bunk...sure

[Jammrock]

by the way, in the 60 minutes piece, the reporter asks about using solar (8:00), and the inventor says something along the lines of "yes, we can use solar". what did I miss? is he honestly suggesting using solar electricity to create hydrogen, to feed into the bloom box?

Yeah, lots of people have speculated about that. Nothing has been mentioned about what he meant since then. Consensus is that it was either a flub on his part, solar would be used to power some fuel generation process for the Bloom energy server, or used in tandem with a BES to lessen peak demand usage on a hot summer day.

[dshumake]

From Bob Park's What's New column

1. ENERGY: THE BLOOM BOX DEFLOWERED.
The Hubbert Peak, IPCC reports, melting ice caps, does all this foretell a
disaster? Maybe for polar bears, but for alternative energy pitchmen it's
the mother lode. The free-energy scams such as Joe Newman's Energy Machine
and Dennis Lee’s Hummingbird motor will still find mom-and-pop investors
hoping to stretch their meager retirements, but the big money awaits a more
sophisticated pitch. Consider the Bloom Box. The pitchman, K.R. Sridhar,
CEO of Bloom Energy in Sunnyvale, is not your typical scam artist, but an
artfully understated business executive. He can go on 60 Minutes or Good
Morning America and never make a slip. But there are warning signs. What is
it he’s selling? "It starts with beach sand” he says, opening a box of the
stuff. It's like saying a diamond is a lump of carbon. Silicon dioxide is
the most abundant mineral on Earth, and essential to modern electronics.
With the help of animation, he explains that plates made of the stuff
enable hydrocarbon gases to react with oxygen, producing an electric
potential. It's a fuel cell. He never says so, maintaining the fiction that
this is something really new. It’s not. Fuel cells date back to 1838, but
have found little application. The magic calls for painting the two sides
of the plate with secret green and black ink respectively. Well there's a
little more to it, but CEOs don't worry about details. Oh, and the Bloom
Box is not cheap. However, Google, FedEx and Wel-Mart can afford to test
the Bloom Box. Everybody loves the idea of distributed energy production,
where we have our own power plant in the backyard. But a Bloom Box is not
totally isolated; it needs to be tied to a gas pipeline. Is this the
future? Probably not.

[MultimediaMan]

Actually, if it works as advertised, it will be a huge step forward.

Consider: electric power is, on the average, only 10-15% thermally efficient from Heat energy source to the socket. That's pretty crappy.

We already have a pretty good natural gas infrastructure worldwide so infrastructure impact is less of an issue than with electrical grids. If Bloom can run with 30-40% thermal efficiency, you've got the grid beat by a long shot. Yes, you're still burning fossil fuels at this point, but the efficiency increase is tremendous.

If Bloom can scale, and will last as long as the advertised specs, then this could be a game changer. Fuel Cells have always been finicky, so we don't know how this play out: It could be that if they do require more maintenance or other special considerations, we might see Blooms at every substation rather than at powerplants or indivdual houses.

[Jammrock]

In the actual product press release and the 60 Minutes spot he actually says "fuel cell" and very specifically says they did not invent it, just made a recent breakthrough that allows Bloom to use cheap, highly available materials. The theory is that he can drive down costs and decrease the overall carbon footprint of the world power infrastructure. While nuclear has the best carbon footprint it's not exactly cheap to build or maintain a reactor.

As I've said before, time will tell.

[Brian Ellis]

Actually, if it works as advertised, it will be a huge step forward.

Consider: electric power is, on the average, only 10-15% thermally efficient from Heat energy source to the socket. That's pretty crappy.

We already have a pretty good natural gas infrastructure worldwide so infrastructure impact is less of an issue than with electrical grids. If Bloom can run with 30-40% thermal efficiency, you've got the grid beat by a long shot. Yes, you're still burning fossil fuels at this point, but the efficiency increase is tremendous.

If Bloom can scale, and will last as long as the advertised specs, then this could be a game changer. Fuel Cells have always been finicky, so we don't know how this play out: It could be that if they do require more maintenance or other special considerations, we might see Blooms at every substation rather than at powerplants or indivdual houses.

I'm sorry, but you have made several mistakes in your statement. A modern thermal power station does not work at 10 to 15% but more like 30 to 35%, overall. The very best CHP plants can work as high as 65% although 60 to 62% is more usual. That is not exactly crappy!

Bloom are claiming 60%, which is certainly in the same ballpark, but this must assume new, unpolluted, solid electrolyte which would have to be changed regularly. There is no significant efficiency increase. On the other hand, there may be less downtime as maintenance is probably much quicker.

The modern CHP plants are already working and online in mainstream applications, but are quite costly, although they do not require high purity gases which are very expensive. It is possible that the Bloom equipment requires less capital investment but will certainly require much higher running costs to purify the gases sufficiently so as not to clog the electrolytes. In addition, if using the natural gas network, there is absolutely no advantage over CHP in terms of greenhouse gas emissions. The same applies to hydrogen, if generated by reforming natural gas, as 98% is already done.

The few that are running are still prototypes and, as such, almost certainly have resident engineers working full-time on them to keep them running.

As far as I am concerned, these are still vapourware. In my opinion, it is very unlikely that they would not be suitable for use in individual houses, but would make district plants IF they lived up to their promise, and I'm still sceptical!

[Dr Mordrid]

I'm sorry, but you have made several mistakes in your statement. A modern thermal power station does not work at 10 to 15% but more like 30 to 35%, overall.

That may be the number for generation to the first power station outside the gate, but it isn't for from generation through the grid to a wall plug. Lots of losses in transmission and distribution.

[MultimediaMan]

Brian, the Doc has it... long line/transformer power losses are horrendous, especially here in the US, where powerplants are often sited in remote locations.

Also, I deliberately understated the efficiency of the Bloom Servers: they aren't exactly forthcoming with heat efficiency over time: I assumed a fairly rapid degradation in output efficiency. (Rarely do anyone's numbers jibe with reality when dealing with a sales brochure.)

If the Bloom Server can be made to work with a 5-10 year maintenance cycle at even a 30% thermal efficiency, you will have the grid beat by a long shot (at least 10-20%). A small Natural gas pipeline carries a staggering amount of energy around in it.

[Brian Ellis]

I believe you exaggerate greatly the I²R transmission losses in the USA. In 1995, Wiki claims "Transmission and distribution losses in the USA were estimated at 7.2% in 1995 [13]. In general, losses are estimated from the discrepancy between energy produced (as reported by power plants) and energy sold to end customers; the difference between what is produced and what is consumed constitute transmission and distribution losses. " which is not enormous. With Higher voltage transmission lines, it is now lower. According to the DTI, in 2002, it is as low as 8.4 quads nationwide, out of 96.2 quads. Your 10-20% is way out. I'll try and find more recent numbers.

Remember that most major power plants are not far from the great conurbations (NY, LA, SF, Chicago etc.), except for HE, which represent only 2.6 quads to domestic consumers, anyway.

Sorry, it is still vapourware for me.

[Dr Mordrid]

I believe you exaggerate greatly the I²R transmission losses in the USA. In 1995, Wiki claims "Transmission and distribution losses in the USA were estimated at 7.2% in 1995 [13].......Your 10-20% is way out. I'll try and find more recent numbers.

In line 2001 losses had risen to 9.5% (almost double 1970's number) and were still trending up sharply for several reasons, so if it's not well over 10% by now (fully 9 years later) I'd be very surprised. Of course with the turtle like speeds of US govt. statisticians we may not know current numbers for another 5 years. Also, these are an average, so local losses can be much greater - especially in locales like those described by MMM.

Numbers from the US Department of Energy....

Transmission and distribution losses are related to how heavily the system is loaded. U.S.-wide transmission and distribution losses were about 5% in 1970, and grew to 9.5% in 2001, due to heavier utilization and more frequent congestion. Congested transmission paths, or "bottlenecks," now affect many parts of the grid across the country.
>
America's electric transmission problems are also affected by the new structure of the increasingly competitive bulk power market. Based on a sample of the nation's transmission grid, the number of transactions have been increasing substantially recently. For example, annual transactions on the Tennessee Valley Authority's transmission system numbered less than 20,000 in 1996. They exceed 250,000 today, a volume the system was not originally designed to handle. Actions by transmission operators to curtail transactions for economic reasons and to maintain reliability (according to procedures developed by the North American Electric Reliability Council) grew from about 300 in 1998 to over 1,000 in 2000.

Additionally, significant impediments interfere with solving the country's electric transmission problems. These include: opposition and litigation against the construction of new facilities, uncertainty about cost recovery for investors, confusion over whose responsibility it is to build, and jurisdiction and government agency overlap for siting and permitting. Competing land uses, especially in urban areas, leads to opposition and litigation against new construction facilities.

[Marshmallowman]

The effeciency is certainly in the ball park for replacing power plants, but I think reliabilty/maintenance will be the issue.
Instead of overhauling turbines etc, you will be replacing the plates.
If they are as cheap as they say and last long enough you will have a viable replacement