EM2 Reactor Design: Facts, Pros & Cons

[Keln]

I read a little bit about General Atomics EM2 reactor in the December issue of Nuclear News that someone had brought to work recently. I haven't had a chance to read all of the article and I am curious about this reactor design. Are any of you involved in it?

It claims to be able to run for 30 years without re-fueling on even DU as an initial fuel. Then it goes on to claim that it has a low weapon proliferation, even though the thing is obviously a breeder reactor that runs most of its lifetime on plutonium (from U238 neutron capture/-beta decay to PU-239).

It seems to present a reactor design that would put uranium enrichment out of business, at least for a long time. Even though I work in the nuclear fuel cycle side of things, I still find this design exciting if it is true. But it seems too good to be true.

What are the cons to this? Is it even a proven design? There is actually so much nuclear waste available, if even half the claims by General Atomics are accurate, this could provide power for a long long time if this thing works.

 

[etudiant]

The concept is surely attractive, as always, the devil is in the details.
This is a high temperature gas cooled reactor, using helium as the coolant and as the working fluid for the associated turbines.
Making that work in real life is a bear. Google Fort St Vrain reactors, an earlier GA design.
The system worked, but had enough troubles that eventually the utility shut the reactors and switched to natural gas.

 

[jim hardy]

The system worked, but had enough troubles that eventually the utility shut the reactors and switched to natural gas.

I talked with some engineers at that plant. They were enthusiastic about the equipment.
According to them, management was a bit of a mess. Corporate headquarters in Holland and engineering in California, so getting a decision was quite a gauntlet.
The project was behind schedule and over cost. The utility had contracted for kilowatts so there were gas turbines burning expensive jet fuel to provide it... imagine the money pit for GA.

I think the failure was more one of management science than of nuclear science.

 

[etudiant]

Management is always the biggest risk in any complex endeavor, as there are many more ways to screw up than to succeed.
That said, behind schedule and over cost does suggest engineering issues of some size, even if the engineers involved were completely on board.

 

[jim hardy]

That said, behind schedule and over cost does suggest engineering issues of some size, even if the engineers involved were completely on board.

yes there was significant trouble with "Pelton Wheel" coolant pumps.
It was 1973 i think... i just don't remember details...

 

[jim hardy]

probably this:

from
https://inlportal.inl.gov/portal/server.pt/directory/nrc_published_material/2656?DirMode=1# https://inlimages.inl.gov/imageserver/plumtree/portal/public/img/sp.gif
at

https://inlportal.inl.gov/portal/server.pt/directory/nrc_published_material/2656?DirMode=1#

2.8 Metallurgy of Primary Coolant System Boundaries
The event that finally brought FSV operations to an end was the severe cracking of the incoloy-800 steam generator super-heater headers. Replacement of the headers was deemed too expensive to justify a plant restart given the long history of operational problems at the plant. The header cracks were caused by vibration and thermal cycling of the header material which turned out to have large course grain sizes that made the metal structure prone to cracking. Had the microstructure of the inconel been held to a fine grain size during fabrication and acceptance inspection, the problem may never have occurred leading to a much longer plant life. Although the header cracks were not really a safety issue, attention to the acceptability of metallurgy is a key consideration. During the DOE-funded work on the New Production Reactor Modular HTGR, which was being pursued in parallel to the civilian Modular HTGR, the reactor vendor argued that the once-though steam generator superheater section should be fabricated from Alloy 800H with a bimetallic weld to the 2.25Cr-1.0Mo ferritic stainless steel evaporator section. However, Alloy 800H has a higher carbon content and was thus very susceptible to sensitization and stress corrosion cracking when exposed to water.

they had water ingress from coolant pump bearings .

 

[etudiant]

Thanks for the extra clarification.
The failure of the FSV installation was a tragedy, imho. Success might have shifted the US towards a much more attractive nuclear reactor design and clearly the people who worked there understood that they were pioneering a potential game changer.

 

[mheslep]

...

I think the failure was more one of management science than of nuclear science.

I've found that the two, management and design, are typically tightly coupled. A robust, well understood, hopefully elegant design tends to be tolerant of minimally competent management. But take a fragile design which forces compounding complexity, in this case very high temperature gas against an incomplete understanding of the required metallurgy, and the most competent management is apt to fumble.

 

[Astronuc]

Management is always the biggest risk in any complex endeavor, as there are many more ways to screw up than to succeed.
That said, behind schedule and over cost does suggest engineering issues of some size, even if the engineers involved were completely on board.

I still see a lot of that even today.

they had water ingress from coolant pump bearings.

There was an effective solution to this based on gas cooled bearings and magnetically supported bearings. Interesting comment about Incoloy 800H, since Incoloy 800 has been successfully used in steam generator tubing in Siemens PWRs. However, the industry was young in the 1960s, and there were assumption made about steels and nickel-bearing alloys that proved incorrect. Inconel 600 was used in most of the initial steam generators (even thought it was supposed to last 40 years), and most of that has been replaced at great expense to utilities. Had utilities realized that they would have to replace the steam generator tubing after 15 to 20 years, most might not have ordered nuclear plants, or rather they would have pushed for a better material.

 

[jim hardy]

Had utilities realized that they would have to replace the steam generator tubing after 15 to 20 years, most might not have ordered nuclear plants, or rather they would have pushed for a better material.

Indeed early stainless tubed boilers were disappointments. We had to replace ours, also get all copper(brass) out of the feedwater system and change water chemistry. Condensers we retubed with titanium and feedwater heaters with suitable stainless.

I believe FSV's superheater could have been replaced.
Perhaps because it was only one plant there was not much of a perceived return on investment for developing the skillset to do it... or perhaps it was an excuse to get the financial albatross off their necks...
I'm no business guy.That plant made 1000 degree steam at 2400 psi, same as fossil plants of its day.

A quick Carnot with Tcold at 75F and Thot at 1000F compared to 525F(typical PWR?).
Absolute °F being °F + 460,
1- 535/1460 = 63%
1- 535/985 = 46%
real plants approach 70% of those numbers,

That nuclear heat is "too cheap to meter" means it's too valuable to be thrown away in low temperature cycles. It can displace those expensive BTU's from burning carbon.
And we owe Mother Nature that consideration

I was really disappointed to see Fort St Vrain fail.

old jim

 

[Astronuc]

Back in the 70s, there were plans for at least 8 gas-cooled reactors after FSV. IIRC, the thermal efficiency was expected to be about 42%, which was much better then the ~32-35% for LWRs. I'll have to dig up the information later.

It was a host of technical problems that led to the shutdown of FSV. The Incoloy degradation and steam intrusion into the primary circuit were perhaps the two critical problems at the top of the list. At the time, I was hoping to help solve some of those problems, but the utility decided to give up.

In my first job out of grad school, I worked at a company that consulted with PS of Colorado and FSV. In my current job/location, I just learned today that there are some folks who worked on FSV. I'll have to track them down.

With regard to Carnot efficiency, one has to look at T_hot and T_cold across the turbine, or T_in/h_in to the HP turbine and T_exit/h_exit from the LP turbine. The work output comes from the change in enthalpy across the turbines.

List of operated and canceled US graphite-moderated, gas-cooled reactors.

    Unit        MWt   MWe   Dates
                     Gross
Peach Bottom 1  115   42    January, 1967
                           November, 1974
                        
Fort St. Vrain  842  342    January, 1974
                            August,  1989

Planned, but canceled ~1974-1975
Fulton 1       3000 1200    **
Fulton 2       3000 1200    **
Summit 1       2000  781    Planned, 1974
Summit 2       2000  781    Planned, 1978
St. Rosalie 1  1300 3333    Planned, 1978
St. Rosalie 2  1300 3333    Planned, 1978
Vidal 1        2000  779    **
Vidal 2        2000  779    **