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


Clean Power Hyperion Power Module Feature Image

Published on August 14th, 2008 | by Rod Adams

4

Hyperion Announces First Customer For Small Nuclear Reactor

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

August 14th, 2008 by  

Hyperion Power Module Feature Image Hyperion Power Generation issued a press release on August 12, 2008 announcing that their first customer had signed a letter of intent (LOI) to purchase 6 Hyperion Power ModulesTM (HPM), which the company describes as “a small, compact, transportable, nuclear power reactor”.

Each HPM will be priced at approximately $25 million. The company did not disclose an expected shipping date for the first HPM, but the US Nuclear Regulatory Commission currently lists their scheduled manufacturing license review as starting in 2012 with an projected completion sometime in 2015. (Ref – Periodic Briefing on New Reactor Issues dated February 20, 2008 – PDF.)

Hyperion Power Generation’s web site is a bit shy on the details of the design, which is to be expected. The site’s evident goal is to attract interested customers using rather conventional marketing techniques; no one will get involved in the purchase of even a “small” nuclear reactor without serious reviews of significantly more detailed technical materials.

After all, each module costs roughly $25 million (when ordered in significant quantities) and will supply enough heat to produce about 25 MW of electricity when connected to an efficient steam turbine cycle. Those are small numbers in the nuclear power world, but still significant to most businesses and small towns.

I am sure that the company will provide those details to serious inquirers who are willing to sign agreements to protect the information. As an early stage technology innovator, it would not be in Hyperion’s best interest to share too many technical details with the online world – which includes their competitors.

Hyperion’s marketing materials describe the Hyperion Power Module as a “battery”, but I think the company is reaching a bit with that analogy. Unlike a battery, an HPM is not a source of electricity that can simply be plugged into a circuit, it is a source of heat. Just like the heat released by burning fossil fuels, the heat produced by an HPM must be captured, focused and converted into motion and then into electricity typically through the use of the same kinds of heat engines available for converting fossil fuels into electricity.

Still, the concept is intriguing and the developments worth watching.

Here is a comparison to help put the system’s potential into perspective. A single truck can deliver the HPM heat source to a site. The device is supposed to be able to produce 70 MW of thermal energy for 5 years. That means that the truck will be delivering about 10.5 trillion BTU’s to the site. Buying enough natural gas at today’s New York City Gate price of $8.55 per million BTU (according to my favorite energy commodity price web site for August 14, 2008) would cost $90 million.

That is more than 3 times as much as the announced selling price for an HPM, but the advantage does not stop there – the HPM is targeted for places where there are no gas pipelines to deliver gas, so natural gas is not available at any price.

Instead, it would be better to compare the HPM to diesel fuel, which currently costs about 2 times as much per unit of useful heat as natural gas and still requires some form of delivery for remote locations. In some places, fuel transportation costs are two or three times as much as the cost of the fuel from the central supply points.

In certain very difficult terrains, or in places where there are people who like to shoot at tankers, delivery costs can be 100 times as much as the basic cost of the fuel.

As one of my heroes would say, one more thing. The HPM, like other fission heat sources, does not release any pollutants to the environment. After five years of operation, all of the waste products would still be sealed up ready to be picked up by a truck and taken to a company or government owned storage location.

Image by permission of Hyperion Power Generation all rights reserved

Related posts

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.



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

loves and respects our common environment, but he has a fatal flaw in the eyes of many environmentalists -- he's a huge fan of atomic energy. Reduce, reuse, and recycle have been watchwords for Rod since his father taught him that raising rabbits is a great way to turn kitchen scraps into fertilizer for backyard fruit trees and vegetable gardens. They built a compost heap together in about 1967, when he was 8 and when Earth Day was a mere gleam in some people's eye. During his professional career, he has served in several assignments on nuclear submarines, including a 40-month tour as the Engineer Officer of the USS Von Steuben. In 1994, he was awarded US patent number 5309592 for the control system for a closed-cycle gas turbine. He founded Adams Atomic Engines, Inc. in 1993, started Atomic Insights in 1995, and began producing the Atomic Show Podcast in 2006. He is currently an active duty officer (O-5) in the US Navy. He looks forward to many interesting discussions.



  • Intranick

    Roland had it right. the problem is they are talking about just the reactor. Perhaps thats a good excuse for an extra million dollars or so, would create a pool of water that even in times where there was a water shortage, would hold enough for a day or two of running at or near peak to use the heat from the reactor to spin the turbine that I imagine it would have to

    Iduno maybe thats thinking too far ahead but I could honestly see small cities, maybe even the really small ones partnering up, and using these units.

    very cool stuff.

  • Intranick

    Roland had it right. the problem is they are talking about just the reactor. Perhaps thats a good excuse for an extra million dollars or so, would create a pool of water that even in times where there was a water shortage, would hold enough for a day or two of running at or near peak to use the heat from the reactor to spin the turbine that I imagine it would have to

    Iduno maybe thats thinking too far ahead but I could honestly see small cities, maybe even the really small ones partnering up, and using these units.

    very cool stuff.

  • Roland Riemers

    I see I made a few typos is the first submission. Here is a somewhat better corrected one.

    The Hyperion concept is very interesting, but I think they are missing the boat slightly. Just supplying the heat source for power generation means each user has to re-invent the power generation wheel again. It would make more sense to have a total generation package where everything goes in every 5 years for a complete update and refurbishing. In fact, I recall reading of a concept some years ago where nuclear power packages could be built on barges, and the barges transported to any city that needed power. Every 5 years the barge would be replaced with a new one and the old one is recycled and the nuclear fuel enriched for another 5 years. What makes the barge system work is that most major cities are located on large rivers or other bodies of water. The barge would also have an available source of water for cooling as well.

    A smaller package on one or two trucks would be especially useful in cold remote areas where the generator would not only supply relatively cheap and reliable electrictiy, but could heat all the buildings (and maybe some of the streets) for a double savings.

    Getting back to the Hyperion unit, I suspect that after 5 years the Hyperion heat source would not be scraped, but instead would be refurbished and the fuel enriched. That means the costs of recycled units, and resulting electricity, would continue to come down with each replacement, as the refurbished units would obviously cost less then the new units.

    But, if the Hyperion unit had to be disposed of after 5 years, that brings on another big headache and maybe even a cost far exceeding the initial purchase as we are all well aware how inefficient the current nuclear waste disposal program is.

    I am also wondering, if the Hyperion unit is not run at full capacity for 5 years, if its life could be extended for much longer? For instance, you might have 2 Hyperion units to meet peak load, but only need one most of the time when the load is not peak. Or maybe the 5 year old unit would still produce a lot of heat for the next couple decades and thus could be used for a cheap building heat source for nothern towns?

    Roland of North Dakota

  • Roland Riemers

    I see I made a few typos is the first submission. Here is a somewhat better corrected one.

    The Hyperion concept is very interesting, but I think they are missing the boat slightly. Just supplying the heat source for power generation means each user has to re-invent the power generation wheel again. It would make more sense to have a total generation package where everything goes in every 5 years for a complete update and refurbishing. In fact, I recall reading of a concept some years ago where nuclear power packages could be built on barges, and the barges transported to any city that needed power. Every 5 years the barge would be replaced with a new one and the old one is recycled and the nuclear fuel enriched for another 5 years. What makes the barge system work is that most major cities are located on large rivers or other bodies of water. The barge would also have an available source of water for cooling as well.

    A smaller package on one or two trucks would be especially useful in cold remote areas where the generator would not only supply relatively cheap and reliable electrictiy, but could heat all the buildings (and maybe some of the streets) for a double savings.

    Getting back to the Hyperion unit, I suspect that after 5 years the Hyperion heat source would not be scraped, but instead would be refurbished and the fuel enriched. That means the costs of recycled units, and resulting electricity, would continue to come down with each replacement, as the refurbished units would obviously cost less then the new units.

    But, if the Hyperion unit had to be disposed of after 5 years, that brings on another big headache and maybe even a cost far exceeding the initial purchase as we are all well aware how inefficient the current nuclear waste disposal program is.

    I am also wondering, if the Hyperion unit is not run at full capacity for 5 years, if its life could be extended for much longer? For instance, you might have 2 Hyperion units to meet peak load, but only need one most of the time when the load is not peak. Or maybe the 5 year old unit would still produce a lot of heat for the next couple decades and thus could be used for a cheap building heat source for nothern towns?

    Roland of North Dakota

Back to Top ↑