Energy From Thorium Discussion Forum

It is currently Aug 18, 2018 1:28 pm

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 49 posts ]  Go to page Previous  1, 2, 3, 4  Next
Author Message
PostPosted: Nov 28, 2013 4:40 am 
Offline

Joined: Sep 22, 2013 2:27 pm
Posts: 262
Hi Lars, Hi Cyrill, Hi Everyone,

please find a sketch of my idea of making a control rod.

The advantage of the dry rod is that it can be driven according to the reactor requirement. It does not require a feed-thru to the fluid fluid with a very challenging gasket topic. It would work like a baffle plate in the flow.

The disadvantage is the more complex shape of the reactor vessel.

Question: You mentioned control rods in the wet part of the reactor. Do you have any document, sketch... about it?

Best regards

Holger


Attachments:
1311 controlrod sketch.jpg
1311 controlrod sketch.jpg [ 88.75 KiB | Viewed 3297 times ]
Top
 Profile  
 
PostPosted: Nov 28, 2013 6:45 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Any electromagnet in this environment is very challenging. High temperature, high gamma and neutron flux. Also if the gears are stuck they are not fail safe.

More importantly I think is the fact that any active component can be abused (like the operators in the Chernobyl plant deciding they don't need the control rods, pesky things just get in the way of fun experiments with the reactor).

If there's a Stuxnet type virus of sort, repeated insertion and retraction of control rods could be used to damage the reactor.

Also control rods in the blanket aren't very efficient. A large number would be needed there to achieve a shutdown of <0.95 criticality. Though the blanket is a nice neutron and gamma shield, I admit.

I'd rather have fast salt drain than electric control rods.

Passive wet operation control rods are very easy. No control circuits, no computers, no logic voting, no electromagnets, no gears, no couplings. Just a welded extension (orifice) where the control rod gets pushed into. The pump then becomes your control rod. The control rod would be heavy like a high gadolinium alloy, so it would need a minimum flow to be pushed out, say 10%. The operators then get a hydraulic operating range of up to 10% to control post shutdown temperatures as needed. If a Stuxnet virus has taken over the pump, it can't do much because the control rod is slow and doesn't respond to the pump at critical low flow levels (ie below 10% flow). So Stuxnet throttling of the system would not be damaging by design.


Top
 Profile  
 
PostPosted: Nov 28, 2013 10:39 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
A dry control rod has the challenge that you need to be in a high neutron flux region so that there are neutrons to absorb AND you need to be in a low neutron flux area or the cylinder walls will suffer neutron damage and have a shorter lifetime.

Putting it in the blanket would mean that your drywell cylinder should have a similar lifetime as the first wall so if you make the drywell cylinder replaceable like a laser toner cartridge this would be feasible. But now you are in a low neutron flux region so the challenge is absorbing enough neutrons to serve as an effective stop.

As Cyril has explained I don't see where a wet control rod would require a gasket but agree that we should not have one.


Top
 Profile  
 
PostPosted: Nov 28, 2013 5:42 pm 
Offline

Joined: Sep 22, 2013 2:27 pm
Posts: 262
Dear Lars, Dear Cyrill,

manufacturing a reactor vessel made of two egg shape vessels in each other for the fissile and the blanket zone with pipes passing thru out of a fissile zone is a bit tricky. It is to make the inner vessel first out of 2 halfs with the inflow and outflow pipes welded in...weld vertical section to the pipes and the already welded in vessel segments. If it is the concept to make it by a molybdenum alloy that requires electron beam welding in a vacuum chamber of a size that does not exist yet it is a challenge.

If there are a couple of thin outer control rod pipes to cut thru both vessels it might become a manufacturing challenge and requires more and difficult welding. This was the reason to put them in the less efficient (lower neutron flux) blanket only. It is to make them large enough to absorb enough neutrons when required. Please don`t hesitate to comment.

Mechanic drives, springs and electromagnets are used in LWR control rod system since a long time.

As far as I remember the draining was calculated in the ORNL 4541 for the MSR of the 70ies to some minutes. A lot of the time was calculated for the freeze valves. It requires some time to drain some XX tons of liquid salt ot it requires very thick pipes.

Does liquid salt mixtures tend to create foam if they flow in a drain tank?

Do you have any document or illustration on the wet control rods?

In the wet system you mention the power is somehow linear to the flow of the pumps in the short run. You can increase the criticality in the long run by adding fissile material and reduce it by not replacing fissioned actinides. There is no option to run the fluid system without creating fission and heat. If you shut-down the reactor you have a lot of decay heat and risk overheating in some parts of the primary circuit.

Holger


Attachments:
1311 assembling the reactor vessel.jpg
1311 assembling the reactor vessel.jpg [ 58.56 KiB | Viewed 3260 times ]
Top
 Profile  
 
PostPosted: Nov 29, 2013 8:00 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Quote:
In the wet system you mention the power is somehow linear to the flow of the pumps in the short run. You can increase the criticality in the long run by adding fissile material and reduce it by not replacing fissioned actinides. There is no option to run the fluid system without creating fission and heat. If you shut-down the reactor you have a lot of decay heat and risk overheating in some parts of the primary circuit.


Sure there's an option, a minimum flow is needed to push the rod out, and this will be, by design, a flow rate much higher than the maximum decay heat (>7%). So you could use 1% flow or 7% flow (possibly with pony motors) and the control rod stays in. Enough operational freedom without any new risks.

Quote:
Do you have any document or illustration on the wet control rods?


Can't find any I like so I quickly sketched one up:

Attachment:
passive control rod MSBR.jpg
passive control rod MSBR.jpg [ 74.63 KiB | Viewed 3257 times ]


Top
 Profile  
 
PostPosted: Nov 29, 2013 11:30 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
Power in an MSR is not at all proportional to the flow in operation. In fact, I suspect that during operation we will allow different power levels but the flow will be held constant. If you run the reactor at full flow rate, with the full_stop control rods completely out of the fuel but you withdraw no power the fluid temperature will become uniform at the average temperature. So a reactor designed to have 700C hot outlet and 600C cold inlet temperatures during full power will have 650C inlet and outlet during full flow at zero power if you don't pull power out by extracting heat from the heat exchangers.

I understand that it is possible to design good reactors with active controls like motorized control rods BUT part of the design goals here is to make the reactor as stupid proof as we can. If there are going to be 10,000 reactors deployed in the world we won't have the same quality control of reactor operators at every one of them. But we still need to decrease the risk of scaring the public by 10x or more. So the goal is to increase the number of reactors 20x, and simultaneously increase the combined safety by 10x or more - while suffering a loss in quality of the operators. Part of that is to eliminate where-ever possible required operator actions to provide safety and to eliminate the possibility of operator actions being able to cause a problem.


Top
 Profile  
 
PostPosted: Nov 29, 2013 11:48 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Yes, one of the interesting things that happens without control rods in throttling the reactor but keeping the flow constant is the temperature swing changes. This does have implications for any power reactor. For one thing, we can't throttle too quickly to avoid rapid thermal cycling. So going from 600 C in, 700 C out, @ 100% power, to 50% power, would mean something in the ballpark of 625C in, 675C out (simplifying with the assumption on constant alpha throughout the temp range). So if you want to keep conservative thermal limits (1C/min), then the throttle down to 50% power would take 25 minutes. Doing this every day would consume thermal cycle life even with the rate of change limits being obeyed. Another issue has to do with the steam or brayton cycle. Running the primary loop at a different temperature would result in different steam/gas temperature inlet/outlet.

A flow controlled reactor would have fewer issues, right?


Top
 Profile  
 
PostPosted: Nov 29, 2013 1:21 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
I would expect the steam/brayton inlet/outlet temperatures will be much more a function of the flow rate changes in the steam than in the primary fuel salt. So I don't think this factors into the selection of whether to reduce the flow with lower power draw or not.

Second, the general topic here was going into or coming out of a full stop situation where the control rods are active. So, I presume this isn't a daily occurrence but rather something that happens once a year or less for main servicing/inspection or emergencies.

The separate topic of what is the wisest thing to do with the flow during when operating at less than full power is something to be explored. But the very fact that we do have the option to change the flow or not highlight the fact that flow and power production are not tightly coupled. I have the impression but no good engineer that the desire was to maintain the flow rate constant. But ORNL did vary the flow rates to test the pumps for vibration (and after San Onofre likely the HX too). Higher delta T brings its own disadvantages and changing the flow rate would change the fluid dynamics inside the core which might bring in issues of stagnant pools of salt in corners at lower flows.


Top
 Profile  
 
PostPosted: Nov 29, 2013 1:52 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
I'd rather see a change in flow rate than a change in temperatures. Variable speed or variable frequency drives are commonly used today, that is different from the 50's and 60s.

Vibration, if it occurs, always occurs at the highest flow rates not at lower flow rates.

Proper mixing of the salt is an important design requirement. Thermal hydraulics is easy to predict using modern codes and software. The reactor needs to be started up and shut down at some point, that is much more natural with variable pump flow. Thermal control means the pumps wham up to full speed instantly as the startup procedure, which seems crazy. But if you use flow control as startup procedure then you might as well use that to throttle the reactor.

It is a bit weird to throttle the steam/feedwater pumps to keep steam temperature constant but not throttle the reactor pumps. I don't see the advantage of this approach. If you can manage flow/dead spots with the changing feedwater then you can manage it with salt. Just gang them all up. Much easier. Make the feedwater pumps the master, reactor pumps the slave.

Pretty much anything I work with in industry in flow controlled.


Top
 Profile  
 
PostPosted: Nov 29, 2013 3:01 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
Sounds attractive. I'm not sure where the idea of constant flow comes from - I think I read it in an ORNL report but memory is fuzzy.


Top
 Profile  
 
PostPosted: Nov 29, 2013 3:30 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Lars wrote:
Sounds attractive. I'm not sure where the idea of constant flow comes from - I think I read it in an ORNL report but memory is fuzzy.


Some of the ORNL aqeous homogeneous reactors were constant flow. But electric motor technology has come a long way since those times.

Was the MSRE temperature controlled? Might make more sense with that design since it had such a small delta T across the core anyway and it didn't have a power cycle to worry about steam temperature/pressure, steam quality etc.


Top
 Profile  
 
PostPosted: Dec 01, 2013 8:44 pm 
Offline

Joined: Feb 05, 2013 5:24 am
Posts: 108
First I think we should try to maintain a separation in terminology of control rods and shutdown rods. It seemed to be very mixed, with "contol rods" being used most of the time interchangeably. Control rods are used to control power during operation, and can/are used as shutdown rods. But, shutdown rods are only used to shut down or maintain the reactor shutdown. For the MSR, I think we have come through that control rods (i.e. rods used during operation to control power and excess reactivity (we have little or no excess reactivity). Therefore, most of the time we should say "shut down rods", but be careful.

Second, why can't we use boxed rods held out of the core by fuel flow rate? Boxed rods are rods that have an open center filled with a good moderator, with a high thermal absorber for the box/structure. You could use any good high temperature moderator, (LiH, 6LiH, Be, BeO, ZrH2, YH1.6) inside a B4C composite box, or Hf box, or maybe even just plain Hastelloy-N box (it's a pretty good absorber in a thermal spectrum). The moderator would not need to be a pure moderator, as in the case of LiH or 6LiH, and it could even be liquid. A boxes of 10B4C might even work. You get larger fuel displacement, plus slightly better moderation than graphite, and lots of thermal and epi-thermal neutron absorption, of course they would have to be appropriately weighted.

If you are getting rid of the graphite to go to a fast fluid reactor, then you have more shape/space options.


Top
 Profile  
 
PostPosted: Dec 02, 2013 5:20 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Yes, shutdown rods are the more appropriate term here. We don't need control rods. Just shutdown rods.

Regarding boxed rods, what is the advantage of this approach? The neutrons would already be well thermalized by the graphite so you can use a nonmoderating absorber like gadolinium. We'll want something that doesn't generate loads of helium or tritium and something that is heavy enough to sink into the core against a small flow (maybe 20%), so gadolinium is attractive.

The material could be tailored to the right density by adding B-10 (pure Gd is likely heavy enough to sink in too early). So you'll get a B-10/Gd alloy/cermet. This would result in ultra high shutdown rod worth, minimizing the number of shutdown rods needed.


Top
 Profile  
 
PostPosted: Sep 09, 2014 1:53 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
Sorry, I've come very late to this discussion. I like Cyril's flow activated shutdown rod. Regarding the poor absorption in fast spectrum, can we get enough delta keff by simply displacing the the salt in the center of the reactor core, bad from a flux perspective, but if it is a shutdown rod it won't have a high average flux to deal with.

Regarding the original question, no, I do not believe that fast MSR's need or should have control rods, we should have no difficulty in getting what we want through negative and positive shimming of the fuel salt. The shutdown rod may be unavoidable, but in many ways I think that it is easier to deal with. What concerns me with the shutdown rod is the reactivity insertion when it is removed AND the fact that we really don't want the core to get so cold that the fuel salt starts crystallizing or worse freezing in bulk.


Top
 Profile  
 
PostPosted: Sep 09, 2014 2:13 pm 
Offline
User avatar

Joined: May 24, 2009 4:42 am
Posts: 823
Location: Calgary, Alberta
Lars wrote:
Sounds attractive. I'm not sure where the idea of constant flow comes from - I think I read it in an ORNL report but memory is fuzzy.
I believe that the following technically feasible and desirable.

A core idling temperature of 600C, fixed speed fuel salt pumps and fixed speed secondary salt pumps. At the full load condition the core inlet temp is 500C and the outlet 700C . Sliding pressure steam turbine, where inlet pressure and flow are proportional to mass flow through the turbine in a linear relationship. Steam generator provides almost constant steam outlet temperature with some active temperature control where appropriate. The master control element, controlling this orchestrated symphony of machinery, the feedwater control valve.

As feedwater mass flow increases so does STG power, as more heat is demanded by the steam generator, the returning salt temperatures drop, increasing power, raising the core outlet temperature to maintain the same approximate mass of fuel salt in the core.

No control rods, no significant swings in steam temperature, fast ramping. The temperature swings on the core during ramping are significant, and perhaps that should be tuned down to 550C/650C, but I'm not sure that is required, as the wall thickness could be quite small, meaning that thermally induced stresses will be similarly small. If the reactor vessel has thick walls this concept would not be acceptable.

It's a challenging thought, just one control lever in real-time. Controlling the flow of high pressure feedwater is many times easier than changing salt flow or the flow of high pressure superheated steam.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 49 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:  
Powered by phpBB® Forum Software © phpBB Group