Energy From Thorium Discussion Forum

It is currently Aug 15, 2018 11:46 pm

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 53 posts ]  Go to page Previous  1, 2, 3, 4  Next
Author Message
PostPosted: Sep 29, 2011 11:13 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
The He sparge is done by taking a fraction of the salt flow and pushing it through a pipe that induces swirling. In the acceleration the gases are pressed to the center (kindof like increasing the buoyancy by increasing the effective gravity). The pumps provide the power, the salt is returned to the inlet to the pump. I don't believe this is unrealistic and everything I've read in the ORNL reports say that the MSRE system worked well and they were confident that they could substantially increase its performance for MSBR.

Second, the sparge also removes tritium and is our first line of defense for tritium leaks. We generate around 2400 Curies of tritium per day initially. In the long run this will reduce to about 1/2 (I think) as the 6Li burns off. The legal limit for release is 10 curies per day. But the political limit appears to be several orders of magnitude lower - witness Vermont Yankee. If you slow down the sparge then you will capture much less of the tritium in the offgas system and leave a much bigger job for the secondary salt etc.

Recapturing the heat for productive use shouldn't be too hard. Put a heat exchanger with cool secondary salt inside the gas holding tank. Though I might add that we should not worry too much about passive heat leaks from the core + decay holding areas. We need to design in a 1% leak rate in emergencies and want it to be extremely reliable. We could try to be cleaver and arrange things so that during normal operations we avoid most of that heat loss - but it comes at the cost of increasing complexity of the passive cooling system and the risk that when you need it most something in that increased complexity breaks on you. Losing heat from the core that doesn't go through the pumps/HX/turbine/generators is much less painful than losing efficiency in the turbines since the core itself is only a portion of the total system cost.


Top
 Profile  
 
PostPosted: Sep 29, 2011 12:50 pm 
Offline

Joined: Mar 07, 2007 11:02 am
Posts: 911
Location: Ottawa
Quote:
If you're not sparging fast enough to get the Xe-135 out before it steals a neutron, is there any point having a sparging system at all? Just let the Kr and Xe diffuse out of the salt, and pipe it away.


Yes Luke, that is actually what I've been doing thought experiments on for the past few weeks (but wary of tomatoes thrown my way for even suggesting it!). I discussed this with a few, including Dick Engel with mixed results. Dick was pretty negative and was optimistic that the off gas system could be made quite rugged if that was my worry.


Quote:
Where do I find capture half lives for a given power density?

More specifically, does anyone know what it is for Xe-135 at DMSR and MSBR power densities?

It occurs to me that a DMSR would have less losses to Xe and any fission product for that matter because of the lower power density (similar to lower losses to Pa). As such it might do with a slower sparging system. That's good since DMSR is likely to be a first MSR development.


Cyril,
I think the effective half live (i.e. before half absorb a neutron) will still be relatively long. For PWRs with typically a higher power density I've heard much less than half of the Xe135 absorbs a neutron before decaying. As well, in App A of ORNL 4541 they show a Xenon poison fraction curve for bubbling rates. Their goal was 0.5% (under 1% was hard) and the curve seems to go to only 4% if they they don't strip the bubbles. So if I'm reading that right, we'll only have 4 to 8 times as many neutrons lost if we don't sparge but just let it collect in a plenum space and slowly pipe it off. As others have said, there are other tricks besides sparging to get out the gas.

Yes, the DMSR will lose a lower percentage to Xe since it has a higher fissile load than MSBR. However, for harder spectrum like the TMSR (MSFR) it is tough call. They have much higher fissile loads and a harder spectrum but also very high power densities.


However, a big issue with just trying to let the Xenon come out itself or even sparging slower might be a safety implication of sudden bubble collapse (which I've never understood how that could happen in practice). I guess the physics is if you have a certain percentage by volume as gas bubbles, if they all collapse then it is like suddenly having that much more fuel salt in the core. I'm not sure if I've really seen this as an issue of concern from ORNL documents though, but from Uri Gat's paper "MOLTEN SALT REACTORS - SAFETY OPTIONS GALORE"

There are two safety concerns for the MSR that can
lead to a power excursion. The first of these concerns is the
accumulation of gases and volatile materials in the fluid fuel
that would coalesce into bubbles that could then collapse at
once in the core, resulting in a reactivity excursion. A
careful design will ensure that such an event is avoided. The
dispersed gases must accumulate over an extended period of
time, which allows for removal by sparging, and
recognizing and noticing the failure of the gas and volatile
removal system. By removing the gases early in the cycle of
the fuel from the core to the heat exchanger, the likelihood
of the collapse of a bubble in the core can be minimized.
The geometrical design of the core can also assure that the
added volume has a small reactivity contribution

(the other by the way was the "cold slug" event of colder fuel salt suddenly returning from the heat exchangers which is easy to guard against).

Have others ever noticed this "bubble collapse" concern covered well in any ORNL document?

David LeBlanc


Top
 Profile  
 
PostPosted: Sep 29, 2011 5:16 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Bubble collapse by heating it up? That is kind of like worrying about water freezing if you put in a fire. The opposite will happen! Ideal gas law is pretty simple. Bubbles will only collapse when pressurized as in the heat exchanger. The pump will be before the heat exchanger not behind it. Pressure drop in the core is tiny. What physics can cause bubble collapse in the core?

Low losses to the gasses was in one of the TMSR documents, where they said the difference between 30 seconds sparging and a few days sparging was very little in breeding ratio.

I agree that a short sparge is reasonable if your sparging system is actually the primary loop and you don't need to add helium (ie let gasses come out of the lowest pressure part of the system and collect them). But certainly not 30 seconds at 100% efficiency.


Top
 Profile  
 
PostPosted: Sep 29, 2011 6:09 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
I'm not entirely confident that I have kept up with the preceding discussion, but here are some thoughts that I hope will be relevant.

If the off gas is conducted through to the dump tank, then presumably the dump tank has to have a passive cooling system capable of dealing with the full decay heat load which I understand is about twice as much as that associated with the off gas. So if by having longer sparge time the heat load is reduced from 1/2 design heat load to 1/4 design heat load I fail to see the benefit in doing so, apart from the heat loss angle.

If the dump tank decay heat can be reliably and simply inserted into the heat engine, then there is no efficiency benefit associated with longer sparge times.

[edit] If one could eliminate then helium sparge all together then I suppose that would be an advantage to be able to drop an entire system, but if we only get to make it smaller than ORNL envisaged, then the benefits are likely to be modest. I did once look at the idea of making it much bigger, effectively turning it into a gas lift system with view to eliminating the primary salt pumps. It looks pretty horrible, well worth avoiding.

Re bubble collapse, I wonder if the actual meaning is different to what the words would suggest. I imagine a process that starts with the nucleation of a small bubble which then rises through the core getting bigger as more gas enters the bubble and as the hydrostatic pressure reduces allowing the bubble to significantly expand as it rises, but then as that bubble leaves the active core zone it is replaced by fresh fuel momentarily increasing the reactivity of the core.

I also wonder what the difference is between a genuine void with nothing inside and a bubble of xenon, I presume that a bubble of xenon will have a measurably different impact on reactivity than a simple void containing nothing.


Top
 Profile  
 
PostPosted: Sep 30, 2011 6:58 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Lindsay wrote:
If the off gas is conducted through to the dump tank, then presumably the dump tank has to have a passive cooling system capable of dealing with the full decay heat load which I understand is about twice as much as that associated with the off gas. So if by having longer sparge time the heat load is reduced from 1/2 design heat load to 1/4 design heat load I fail to see the benefit in doing so, apart from the heat loss angle.


The offgas heat load drops off rapidly so decay heat will replace it. The dump tank needs to be preheated to avoid thermal shock.

Quote:
If the dump tank decay heat can be reliably and simply inserted into the heat engine, then there is no efficiency benefit associated with longer sparge times.


Yes, but this requires safety class cooling pipes. You then need to deal with a scenario of earthquake breaking the pipes or endogenous pipe failing. I'd much rather not play games with the decay heat. It demands respect. I'll give it no chance and have fully passive always on heat removal to the outside air with rediculously reinforced steel plate concrete chimneys. So we lose 1% of our heat, big deal. Cost of safety grade HX, pumps, and extra components could well be more than this.

Quote:
Re bubble collapse, I wonder if the actual meaning is different to what the words would suggest. I imagine a process that starts with the nucleation of a small bubble which then rises through the core getting bigger as more gas enters the bubble and as the hydrostatic pressure reduces allowing the bubble to significantly expand as it rises, but then as that bubble leaves the active core zone it is replaced by fresh fuel momentarily increasing the reactivity of the core.

I also wonder what the difference is between a genuine void with nothing inside and a bubble of xenon, I presume that a bubble of xenon will have a measurably different impact on reactivity than a simple void containing nothing.


That's an interesting suggestion. ORNL did mention the bubble collapse a few times but then when they looked at how quickly this would have to happen to cause a reactivity excursion, they concluded that such fast collapse would not be possible in this system (the fuel dilatation coeffient in any MSR is very negative and very fast).

Bubbles of xenon should be worse than voids because replacing a neutron poison with fuel inserts more reactivity than replacing empty space with fuel.


Top
 Profile  
 
PostPosted: Oct 01, 2011 7:00 am 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
Cyril R wrote:
Lindsay wrote:
If the dump tank decay heat can be reliably and simply inserted into the heat engine, then there is no efficiency benefit associated with longer sparge times.

Yes, but this requires safety class cooling pipes. You then need to deal with a scenario of earthquake breaking the pipes or endogenous pipe failing. I'd much rather not play games with the decay heat. It demands respect. I'll give it no chance and have fully passive always on heat removal to the outside air with ridiculously reinforced steel plate concrete chimneys. So we lose 1% of our heat, big deal. Cost of safety grade HX, pumps, and extra components could well be more than this.


I don't think that we'd have too many differences on the performance requirements for any backup cooling system. I like the term walkaway safe, my definition of that term would be that one could sever every single wire connection to every single device in the entire facility, step out the door and walk away while coping with any specific safety cases at the same time as the 100% failure of all things electrical; and have no loss of containment or radioactive emissions. Any backup cooling system that can achieve that is doing pretty well IMHO.

I believe that it is possible to meet that walkaway safe criteria and still recycle most of the normal heat load. I have devised a passive cooling concept that I believe will work and fulfil all of those criteria as well as all of the following ones. I think that it's important to start with performance criteria, then evaluate the options, rather than taking a position that this or that is unacceptable. If we can satisfy the performance criteria AND recover some heat economically, then that's fine. But the moment that 'nice to have' starts affecting the reliability of any backup cooling system, then it is definitely time to step away from that nice to have and stick to basics.

Cooling Concept Characteristics
Fully passive: Tick
Always on: Tick
No valves required: Tick
Temperature regulated: Tick
No pumps required: Tick
Passive failover to alternate heat sink/s: Tick
Powered by gravity and heat: Tick
Will not freeze, FS, pipes, HX or other components: Tick (as long as some decay heat remains)
Heat transfer fluid will not freeze in heat sink: Tick (depends on heat sink details)
Choice of common or independent cooling systems: Tick
Setpoint temp for heat source fixed and independent of sink temp: Tick (within wide limits)
Standing losses less than SmAHTR: Tick
Can cater for multiple heat sources: Tick
Acceptable seismic performance: Tick (with appropriate care during design)
Acceptable aircraft resistance: Tick (with appropriate care during design)
Can integrate off gas and and Dump tank decay heat back into thermal power cycle: Tick
Operates at low pressures and elevated temperatures: Tick
Disadvantages: Relies on Na or NaK as heat transfer media, any significant leak of heat transfer fluid will disable or seriously affect operation.


Top
 Profile  
 
PostPosted: Oct 01, 2011 7:22 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Fair enough Lindsay, lets first look at what can be done while retaining walk-away-safe.

NaK is interesting, but quite reactive to air, concrete, the fuel salt, and water. If using heat pipes or thermosyphons of it in contact with water or air for the ultimate heat sink, then this becomes a scenario that we have to deal with.

How about the buffer salt idea, where we put the gasses in a calandria that is on the floor of the buffer salt pool? This allows full passive cooling with a fluoride salt, buffer salt flows through the inside of the calandria tubes to remove heat by natural circulation. Normal operation would be to suck some of that buffer salt that floats up to the surface of the pool, into a heat exchanger to preheat the working fluid. The low melting point of the buffer salt (KF-ZrF4, 390 Celcius) will allow decent preheating. If the power fails or the HX/pump breaks the high thermal buffer capacity of the pool buffer salt will soak up the short term heat from the offgas, with the passive hot cell cooling system taking up the remainder on a longer term.

If you do a heat balance over the reactor pool you'll find that the time that the decay heat cooling system kicks in is very important to prevent excessive temperature rise. The logical extension to that is to have it always operating, losing maybe 0.5% of the heat.

I think it will be attractive to keep any temperature rise gentle. This will prevent stresses in the components and prevents us needing special materials just because an unlikely event such as station blackout without scram, could cause a high temp rise. I think we can do a lot better than the AHTR due to lack of excess fuel stored energy (fuel temp=coolant temp) and the rapid reactivity shutdown of fuel dilatation (insignificant in the AHTR). On the negative side we have a lot of delayed neutrons insertion that will add some heat (but pump and fuel salt inertia seem to take care of this inherently).


Top
 Profile  
 
PostPosted: Oct 01, 2011 4:29 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
The chemical reactivity of Na and NaK is major issue. Some of the posters here have come up with some good suggestions that support a Na based system, such as using radiative heat transfer at the heat sink end with a gap in between, requiring a double failure to permit a reaction to take place. I really don't like Na due to it's reactivity, but the thermo-physical characteristics as they might support a passive backup cooling system, are simply stellar.

Salt based systems using natural circulation hold great possibilities, but I don't think that we can achieve passive temperature regulation, and I have concerns about reliably managing the salt freeze risk in the heatsink, which is why I first took a look at Na. But if we look at the backup cooling system on SmAHTR, that does look pretty good.

Cyril R wrote:
Fair enough Lindsay, lets first look at what can be done while retaining walk-away-safe.

NaK is interesting, but quite reactive to air, concrete, the fuel salt, and water. If using heat pipes or thermosyphons of it in contact with water or air for the ultimate heat sink, then this becomes a scenario that we have to deal with.

How about the buffer salt idea, where we put the gasses in a calandria that is on the floor of the buffer salt pool? This allows full passive cooling with a fluoride salt, buffer salt flows through the inside of the calandria tubes to remove heat by natural circulation. Normal operation would be to suck some of that buffer salt that floats up to the surface of the pool, into a heat exchanger to preheat the working fluid. The low melting point of the buffer salt (KF-ZrF4, 390 Celcius) will allow decent preheating. If the power fails or the HX/pump breaks the high thermal buffer capacity of the pool buffer salt will soak up the short term heat from the offgas, with the passive hot cell cooling system taking up the remainder on a longer term.

If you do a heat balance over the reactor pool you'll find that the time that the decay heat cooling system kicks in is very important to prevent excessive temperature rise. The logical extension to that is to have it always operating, losing maybe 0.5% of the heat.

I think it will be attractive to keep any temperature rise gentle. This will prevent stresses in the components and prevents us needing special materials just because an unlikely event such as station blackout without scram, could cause a high temp rise. I think we can do a lot better than the AHTR due to lack of excess fuel stored energy (fuel temp=coolant temp) and the rapid reactivity shutdown of fuel dilatation (insignificant in the AHTR). On the negative side we have a lot of delayed neutrons insertion that will add some heat (but pump and fuel salt inertia seem to take care of this inherently).

I think that a big tank of salt concept is a sound one, it provides a lot of grace when dealing with contingent events. The thermal buffer of the fuel salt and any salt in a backup cooling system is a very valuable thing for limiting temperature rise.

What I'm less confident of is how much cooling one can reasonably expect from the hot cell, but I've not done any calculations there. As you will have probably noticed the thermal inertia of the system is critical if one wants to have modestly sized heat transfer arrangements, this where the big tank of salt can be extremely useful and extremely reliable. If the thermal inertia is great enough, natural heat losses from the hot cell may be good enough.


Top
 Profile  
 
PostPosted: Oct 01, 2011 5:46 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Well, the idea of the AHTR is to transfer heat from the vessel to the buffer salt and then a seperate natural circulation system from the buffer salt to a HX with another heat transfer fluid in a chimney. They do suggest to use FLiNaK which I agree is asking for freezing in the chimney. The AHTR group admits this but don't suggest any low melting stuff. NaK solves the freezing issue but brings the chemical reaction issue back.

There may also be non-benign failure of the pipes of the last system, in the event of earthquakes and such, as heat transfer depends on the pipes being filled.

If we don't have thin pipes we don't have the freezing issue. A million liters of buffer salt in a pool would be much easier to manage in terms of freezing.

The heat transfer in my idea would be to use the same AHTR first system from primary to buffer salt, but then let the buffer salt rise to the top of the pool and radiate on a passively cooled hot cell wall (this would be our 'pressure vessel'). That wall then transfers heat to a second container which is the containment, which is passively cooled by air from outside (the AP1000 containment cooling concept). The buffer salt wants to rise to the surface so radiating heat there is going with the flow.

I'm not as good with thermo software but I've done some calculations by hand (Stefan-Boltzman's, Newton's law of cooling). Then I found out it was difficult to get accurate because I had to make assumptions on things that were really iterative. So I also used some raytrace software. The results are close to what we'd need so could probably work with further optimizing. Basically the bottleneck is the pool surface area; only about 200 square meters. That's a fairly small radiator for just radiative heat transfer and natural convection. It seems that the hot cell, which is the absorber, can be kept below the 370 Celcius limit imposed by US ASME nuclear pressure boundary code qualified materials. But it can't be too cold since it still has to dump heat to the second shell, the containment, which must also be hot to heat up air in the chimney.

Convection of the argon is mostly useless, only one or two hundred kilowatts. Argon is a good insulator gas. But even helium doesn't get much above a megawatt and I don't want to use helium as the hot cell gas (nor a very tall hot cell).

But that's good, it makes the things simpler for me - we're down to mostly radiative heat transfer. It isn't hard to get 2-3 MW of radiative heat transfer for the fluoride salt surface (assuming emissivity of 0.5-0.7 for the fluorides). Letting the buffer salt heat up to 600-650 Celcius allows up to 8 MW to be removed (for a bigger hot cell that has a cooler surface). That's almost as much as I'd like. It shouldn't cost much more to increase the pool liquid surface area a bit.

Losing a few MW in normal operation seems quite reasonable. We can keep the buffer salt from heating up too much from the offgas decay heat by using it productively in a HX.

From the AHTR modelling which is far more advanced, the buffer salt temperature rise at the bottom where the reactor is, is very small. Heated buffer salt simply rises to the top of the pool. That's great since there is good heat transfer from the primary to the buffer salt.


Top
 Profile  
 
PostPosted: Oct 02, 2011 8:26 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
Sounds like you've done more here than me, one of the challenges I bumped into early on was that the heat rejection is quite high unless you have a big thermal mass. In my case it was 1.3% of full load power (similar to SmAHTR) for a peak temperature of 50C above normal, that's 29 MWt on a 2,250 MWt core, with compact cores that can be quite hard to do. I once looked a an external cooling jacket for a simple two fluid design and was stunned to find that the removal of blanket heat could not be safety achieved with simple boundary cooling.

When you do your radiative heat transfer calcs I presume that you factor in the heat sink temperature, so as that as the receiving wall heats up it's ability to receive more heat is reduced, which is presumably one of the factors driving the need to iterate to a solution. Thank goodness for computers.

How do you propose to limit the heat losses during normal operation, is that primarily by the radiative heat transfer being proportional to T^4 (where T is absolute temperature)?

How do the numbers look for normal operation vs loss of cooling event?

Is 200 m2 for 8 MW?


Top
 Profile  
 
PostPosted: Oct 02, 2011 11:49 pm 
Offline

Joined: Dec 26, 2007 11:45 am
Posts: 191
Cyril R wrote:
But that's good, it makes the things simpler for me - we're down to mostly radiative heat transfer. It isn't hard to get 2-3 MW of radiative heat transfer for the fluoride salt surface (assuming emissivity of 0.5-0.7 for the fluorides). Letting the buffer salt heat up to 600-650 Celcius allows up to 8 MW to be removed (for a bigger hot cell that has a cooler surface). That's almost as much as I'd like. It shouldn't cost much more to increase the pool liquid surface area a bit.


Covering the surface of the pool with graphite floats can dramatically increase effective emissivity. Is the thermal conductivity of graphite good enough for this to actually improve total heat transfer rather than block it? The floats can also have fins for more effective convective heat transfer.


Top
 Profile  
 
PostPosted: Oct 03, 2011 3:04 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Hi Lindsay. The 8 MW heat loss is at peak during a loss of forced cooling, for a pool with 300 square meters radiative surface area, the containment reaching equilibrium at 252 Celcius, salt emissivity of 0.7 and the peak buffer salt temperature in the pool at oscillating around 670 Celcius.

During normal operation at 500 Celcius peak buffer salt temp the heat loss is 2-3 MW. We use the buffer salt in heat exchangers to use the heat productively in the power cycle, this prevents more the temperature from rising.

Per Peterson says his AHTR is 2400 MW thermal and has a decay heat removal system that can remove 10.2 MW of heat which is equivalent to the decay heat about 30-40 hours after the loss of forced cooling event.

If I try to achieve the same result the same amount of buffer salt then it has to run hotter, but with a bigger salt pool (to get to the 300 square meters that is bigger than the AHTR salt pool) the buffer salt temp does not go over 670 Celcius as the extra salt soaks up the rest. Alternatively the same amount of buffer salt in a conical pool (smaller at the bottom but with more surface area) achieves the 10.2 MW of heat loss result with a peak buffer salt temp around 720 Celcius.

To achieve this result I had to couple the containment and hot cell thermally using aluminium shunts.

I don't know how to model floating graphite. It will certainly help because radiative heat transfer is dominant and graphite has about 200 W/m/k at operating temps so little conductive resistance. If the graphite is thin then we might assume that the heat impediment is negligible and the emissivity simply increases to 0.9. This increases the heat emission by 25-30% so you can then get to the 10.2 MW of heat dumping with under 680 Celcius.

One of the problems is that I'd like the entire pool to be visually inspected from the hot cell. With graphite floating on it we can't do inspection in visible light.

I will also try to model a copper plating extending from the pool sides to the floor next to the pool, as an extra hot component to see if this will increase the heat emission surface area.

Does anyone actually know the emissivity of a molten fluoride salt?


Top
 Profile  
 
PostPosted: Oct 03, 2011 10:06 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
Cyril R wrote:
One of the problems is that I'd like the entire pool to be visually inspected from the hot cell. With graphite floating on it we can't do inspection in visible light.

This is why I don't like the whole idea of a high-temperature hot cell ("hot cell" in the first instance was always intended to refer to high radiation levels, not high temperatures).

Inspection, maintenance, repair, and replacement activities require, at the very least, robotic devices and associated electronic equipment that can tolerate the operating environment.
Similarly, the ports and other interfaces in the hot cell walls are less technically demanding for low temperature, rather than high.

Maybe the top of the huge buffer pool could be thermally shielded during inspection, maintenance, repair, and replacement activities ? .....though that would likely require a lengthy shutdown, for even a minor inspection look-see.
Have you thought about this ?


Top
 Profile  
 
PostPosted: Oct 03, 2011 11:41 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
High temp is challenging, but the pool does help in this design. If the pool is the radiator and the hot cell the heat sink the hot cell operating temperature will be much lower - I'm guessing in the ballpark of 300 degrees Celcius based on what the model is saying, for normal operations. Possibly somewhat less with a bigger hot cell.

But any major repairs or replacement of entire modules will be infrequent and we'll just drain the buffer salt to a lower level by pumping it up to an insulated tank outside containment (with the fuel salt drained to a tank inside the pool at a low level). This allows replacement of anything in the pool. I'm guessing we can let the lower layer buffer salt freeze if we want to, to operate at really low temperatures, but doubt it would even be necessary. I'd much prefer to keep the buffer and fuel salt liquid at all times. Doing some welding and cutting with robot arms at 300 Celcius, seems reasonable. The actual sensitive electronics can be cooled with fluorocarbon and will be outside the radiative buffer salt surface area. If lower temps are required there's another option: cover the lowered buffer salt level with insulating floats (insulating carbon of some sort) and cool the buffer salt below actively with a heat exchanger. But I don't like this option since it is not passively safe; we need to cool the fission products that are contained below the buffer salt.


Top
 Profile  
 
PostPosted: Oct 03, 2011 11:42 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Regarding inspection, Luke suggested to use glass fibers looking at the pool routing the light to a camera outside containment. Transmit light, but not heat and radiation.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 53 posts ]  Go to page Previous  1, 2, 3, 4  Next

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


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