Energy From Thorium Discussion Forum

It is currently Jul 19, 2018 2:38 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 63 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
PostPosted: Sep 13, 2012 8:12 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
jon wrote:
The Sandia guys are proposing to run S-CO2 directly through the reactor and into the turbine. Surely if you can make a simpler design, even at the cost of some plant contamination, it would have to be cheaper to build


That is for a gas cooled reactor. Makes sense there to simplify by going for a direct cycle, because gas cooled reactors have higher pressure in the primary loop anyway, and have highly retentive ceramic fuel (TRISO/BISO). Such reactors are best built as direct cycle for simplicity and cost reduction (and, surprisingly, safety is improved as well).

Our situation is exactly the opposite: the whole primary loop is intensely radioactive with nothing to keep the fission products from being blown everywhere if there's a pressure wave, and the vessel is low pressure, so it's not designed to deal with such pressure waves. This saves cost and eliminates a number of failure modes, but ONLY if there's absolutely no possibility of pressurizing the primary loop.

Even our primary heat exchanger has lots of neutrons and gamma's.

This is one of the disadvantages of a molten salt reactor that advocates are reluctant to talk about. Because of the neutron activation and radiation problems, you need at least 3 loops, preferably 4. According to the ORNL estimates the added cost of an extra loop isn't too bad, and the salts have low pumping power requirements. With a steam or CO2 cycle I suspect 4 loops will actually be cheaper on the whole because then we can put the steam turbine plant outside containment, in just a turbine hall. Because the 3rd loop has no gamma/neutron activity. ORNL used a lot of containment and steam suppression in the steam cell, because of the high activity from activated sodium. Also the lower melting point of an extra nitrate loop for instance means we can eliminate the booster feedwater pump (that is required to prevent fluoride salt freezing).


Top
 Profile  
 
PostPosted: Sep 13, 2012 12:28 pm 
Offline

Joined: Apr 28, 2011 10:44 am
Posts: 249
Finally, somebody talking sense about what we call the triple rupture casualty
I would add that a viable design must contain the salt/steam mess
within the plant if high pressure steam does somehow get
to the primary loop.


Top
 Profile  
 
PostPosted: Sep 13, 2012 1:38 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Actually I think a viable design would have to prove that this is physically impossible. This is done through conservative engineering and passive devices, such as redundant rupture discs all throughout the secondary/third loop.

Though, playing devil's advocate for a second, there will probably not be a release of radioactivity even if there's a steam generator tube rupture. Such a rupture is fairly small so would be a gradual pressurization of the salt loops. Pretending that the rupture discs don't work (which is impossible) at some point the primary HX tubes would also fail. But again there's limited break area so there's a small steam inflow. This would either be pushed back through the other loops or pass superheated out through the reactor vessel seals (which will likely be ruined in the process). Then the excess hot cell pressure, steam, and entrained particles would be passed through the filter because it is a filtered confinement space. There shouldn't be a release beyond noble gasses, even in this unplausible event.


Top
 Profile  
 
PostPosted: Sep 13, 2012 3:32 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
>But again there's limited break area so there's a small steam inflow.
This is an important part of design process, looking at failure modes, the available cross sectional area for flow in different parts of the system. Very carefully and in a step wise manner the design will step through the system calculating flows and pressure drops and applying redundant pressure relief devices to ensure that the maximum pressures developed anywhere are within the capability of the piping fittings and vessels.

As for a viable failure mode that allows steam to pressurize the primary loop to the point of structural failure or uncontrolled release of fuel salt, I'd have to think long and hard about that one.

A small sidebar: We tend to think of these high pressure systems as containing a lot of energy at very high pressure and imagining the consequences of a crack or defect becoming burst pipe or a ruptured whatever. In actuality most high pressure piping failures resolve through a crack that grows over time eventually releasing the working fluid. The leak is detected and a repair made. The higher the pressure, the thicker the pipe and more likely for the slow crack and leak scenario. Thin wall pipes are more likely to burst than thick wall pipes. One of the worst incidents ever in a NPP was the rupture of a low pressure pipe carrying hot condensate at less than 200C in a Japanese plant which suffocated and cooked anyone anywhere near the rupture, just from the flash steam at atmospheric pressure.

The other comment is generally speaking, while I would love to get away from the secondary loop, the potential consequence of water/steam/CO2/helium pressurizing a MSR core in during a major event could be very severe and a major liability from a public safety perception perspective. MSR tech would only ever have one chance to get that wrong and any kind of failure could be as chilling and overblown as Three Mile Island or Fukushima. A place best to avoid IMO.


Top
 Profile  
 
PostPosted: Sep 13, 2012 4:18 pm 
Offline

Joined: Apr 28, 2011 10:44 am
Posts: 249
If rupture disk failure is impossible, then why do you prefer
a 4 loop system to a 3 loop system?

Yes with proper design we can argue that pressurizing
the primary loop with steam is virtually impossible.
But the public distinguishes between 0.9999..9 and 1.0
even if engineers and computers dont.

I beoleive that we have to be able to say that,
if this impossible event happened,
we can contain it.

Our calcs indicate that this is economically feasible.


Top
 Profile  
 
PostPosted: Sep 14, 2012 1:19 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
djw1 wrote:
If rupture disk failure is impossible, then why do you prefer
a 4 loop system to a 3 loop system?


Gamma active salt expulsion & aerosol creation, through the rupture discs. Versus non-radioactive salt expulsion for the third loop. The second loop is quite radioactive from delayed neutron activation, unless it is FLiBe, in which case we get chemical contamination of water soluble BeF2 aerosols, through the steam jet from rupture discs.

ORNL's choice was a sodium fluoroborate intermediate loop, in which case you have toxic BF3, toxic HF, and highly radioactive activated sodium, going through your rupture disc. Nasty.

Apart from avoiding chemical and gamma rad contamination, a nitrate salt is also chemically compatible with steam, so leaks are not such a big deal. The nitrate salt tends to suppress the steam pressure by forming a eutectic with water. Nitrate is cheap so one way to go is to make the nitrate inventory large. We may not even need the rupture discs then.

Also, nitrate traps tritium very well.


Top
 Profile  
 
PostPosted: Sep 14, 2012 8:42 am 
Offline

Joined: Jun 05, 2011 6:59 pm
Posts: 1332
Location: NoOPWA
Seems a bit of complication. For a 100 MWe plant, how bout just lining up 4 (or maybe 5) LM2500s?

_________________
DRJ : Engineer - NAVSEA : (Retired)


Top
 Profile  
 
PostPosted: Sep 15, 2012 1:12 pm 
Offline

Joined: Feb 28, 2011 10:10 am
Posts: 355
KitemanSA wrote:
Seems a bit of complication. For a 100 MWe plant, how bout just lining up 4 (or maybe 5) LM2500s?


The GE LM2500 is an open-cycle gas turbine. How would you implement it in a closed-cycle system like a nuclear power plant ? I guess you would have to use He or S-CO2 as fluid, which would require redesigning the turbine. S-CO2 turbines are still in their infancy. The S-CO2 turbine developed by researchers at Sandia is just a couple of KWe.


Top
 Profile  
 
PostPosted: Sep 15, 2012 5:05 pm 
Offline

Joined: Jun 05, 2011 6:59 pm
Posts: 1332
Location: NoOPWA
Instead of the combustor, I'd install a salt/air HX. So, it may only be 25% efficient to begin with, but they are cheap!!

Perhaps I should have said a modified LM2500?

_________________
DRJ : Engineer - NAVSEA : (Retired)


Top
 Profile  
 
PostPosted: Sep 18, 2012 4:02 am 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
MW for MW steam turbines are cheaper than gas turbines, and in this application more efficient that open cycle gas turbines. In a head to head comparison with modified LM2500's I would expect the STG to win out on efficiency and $/kWe


Top
 Profile  
 
PostPosted: Sep 19, 2012 6:44 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
KitemanSA wrote:
Instead of the combustor, I'd install a salt/air HX. So, it may only be 25% efficient to begin with, but they are cheap!!

Perhaps I should have said a modified LM2500?


Apart from the poor efficiency, a far more serious concern is fission product contamination into the air. If you have a salt air/HX, which I remind that we don't have right now (and it may not be cheap once developed), then any leak of the HX would pressurize the primary loop, and possibly provide a direct path for fission products, especially the volatile ones, to go out into the air. Either through an overpressurized primary loop/containment, or through backflow of salt through the salt/air HX. In terms of releases, we are talking about a potential similar magnitude event as Fukushima here.

This is very very risky. I would like to design the reactor such that there are no failure modes that put fission products into the air. Design it out of the equasion. So I like to have more loops and a simple superheat steam generator or a compact supercritical water generator.

In terms of managing rupture of steam generators/supercritical water heater tubes, how about this? Attach a pipe onto the rupture disc of the third nitrate loop. The pipe terminates into a cross shaped sparger submerged in a tank of water. There you have it, a simple cheap pressure suppression system. Good for protecting the rooms against overpressure (and protect personell). Good for condensing the steam so that any activity (say traces of tritium oxide that are in the nitrate salt) will be condensed in the tank.


Top
 Profile  
 
PostPosted: Sep 19, 2012 7:34 am 
Offline

Joined: Mar 07, 2007 11:02 am
Posts: 911
Location: Ottawa
Just to chime in. I keep trying to upload a neat power point I found (but can't find online again). Only 200k but doesn't seem to want to load. Anyhow it compares superheat to supercritical for coal plants (a bit dated at 2003) and even when fuel costs are your main driver the superheat cycles were almost identical economically because of the high cost of exotic materials needed for supercritical (and ultrasupercritical). So especially for smaller units but maybe even large ones for some time it is likely simple subcritical Rankine is a likely winner.

I'd also add the warning for Brayton fans which I've brought up before. It is amazing how many reputable researchers are using the same bogus graph to compare Brayton to Rankine

http://i186.photobucket.com/albums/x110 ... 95b242.png

(hmmm, first I couldn't load a simple paper now it doesn't want to take a simple image, oh well, here is a link to look at the graph)

The huge mistake here is the Rankine data are directly taken from coal plants but don't take into account that those numbers are HHV% which compare joules of energy in the coal to joules of electricity. When you burn coal though, 15 to 20% of the energy goes right up the chimney in the heat of the gasses and most importantly non condensed water vapour of moisture in the coal. So the Sub and Supercritical values should all be close to 50% efficiency when you have a nuclear heat source and make any Brayton very hard to compete on efficiency until you are way into very hot gas temperatures.

This original graph goes back to Dostal's Ph.d on Supercritical CO2 but as far as I can tell the references he gives for the data show nothing like these values and I think the Rankine all came from the one reference which was basically a phone call to someone at PG&E who just quoted him some of their coal plant numbers. Its amazing how many places you see this graph pop up though (the screen shot was from the SmAHTR report from ORNL).

David LeBlanc


Top
 Profile  
 
PostPosted: Sep 19, 2012 9:53 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Good point about stack losses there David. Another benefit of going closed-loop nuclear systems: we get to use all the heat.

So the consensus looks to be like this. For a small reactor the path of least resistance (to get started) is to go with modern superheated steam cycles. For a big reactor there is a bit more choice. Either subcritical or supercritical. With supercritical being a bit more efficient and compact, at the cost of more expensive steam cycle materials (but the compactness of a supercritical water generator may outweigh this?).

For superheated steam smaller reactors, we can consider the CSP systems operating. These already have nitrate salt steam generators and superheaters, so it is proven.

For a larger reactor, I noticed a really attractive supercritical steam cycle, configured by Burns and Roe. It was optimised for a supercritical water reactor. So they kept things simple and low temperature. No reheat, no superheat. Just 500 degrees Celsius steam. Yet the efficiency is impressive, a net 44.8%.

http://nuclear.inl.gov/gen4/docs/scwr_a ... _fy-03.pdf

Of course with MSRs it is easier to add superheat and reheat. But simplicity is attractive - the molten salt to steam generators, and especially supercritical water heaters will need some R&D and engineering yet. Such a simple cycle would therefore be attractive for a large MSR. With such a low "steam" temperature we could reduce the primary temperature a bit too, helps make the Hastelloy last longer. Or with the same primary temperature it would make for more compact supercritical water generators and also primary HXs. Something to think about: all things being equal, the lower your power cycle temperature, the more compact your HXs become. Another reason why steam couples well to MSRs (we need compact HXs).


Top
 Profile  
 
PostPosted: Sep 19, 2012 2:32 pm 
Offline

Joined: Jun 05, 2011 6:59 pm
Posts: 1332
Location: NoOPWA
All good, but you will have to be near water while with a Brayton you don't. Oh, and if near sea water, no desal.

Just a thought.

_________________
DRJ : Engineer - NAVSEA : (Retired)


Top
 Profile  
 
PostPosted: Sep 19, 2012 4:50 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
David wrote:
I'd also add the warning for Brayton fans which I've brought up before. It is amazing how many reputable researchers are using the same bogus graph to compare Brayton to Rankine

http://i186.photobucket.com/albums/x110 ... 95b242.png

(hmmm, first I couldn't load a simple paper now it doesn't want to take a simple image, oh well, here is a link to look at the graph)

The huge mistake here is the Rankine data are directly taken from coal plants but don't take into account that those numbers are HHV% which compare joules of energy in the coal to joules of electricity. When you burn coal though, 15 to 20% of the energy goes right up the chimney in the heat of the gasses and most importantly non condensed water vapour of moisture in the coal. So the Sub and Supercritical values should all be close to 50% efficiency when you have a nuclear heat source and make any Brayton very hard to compete on efficiency until you are way into very hot gas temperatures.

This original graph goes back to Dostal's Ph.d on Supercritical CO2 but as far as I can tell the references he gives for the data show nothing like these values and I think the Rankine all came from the one reference which was basically a phone call to someone at PG&E who just quoted him some of their coal plant numbers. Its amazing how many places you see this graph pop up though (the screen shot was from the SmAHTR report from ORNL).

David LeBlanc

Yes, very good point David, I've be guilty of using that graph myself, but now that I have a STG model that I'm happy with I can say with confidence that there is something horribly wrong with the STG values on that chart, it could be power station HHV efficiency as you suggest or it could be describing a cycle without any regenerative feedwater heaters, however it has come to pass it is not a fair comparison, it's comparing apples to bananas. Also the rate of rise on that curve WRT temperature for the STG is quite wrong. All these systems when locally optimized show a strong efficiency response to temperature.

>Kiteman
Braytons definitely have there uses and use in locations where water is unavailable or normally very solid and hard to work with, I believe that Braytons have quite a bit to contribute.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 63 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group