Energy From Thorium Discussion Forum

It is currently Jan 20, 2018 8:12 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 86 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next
Author Message
PostPosted: Dec 29, 2013 3:42 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Luke wrote:
Cyril R wrote:
.......Cavitation occurs in liquid with very little subcooling. For MSRs, we have several hundred degree subcooling........

I'm not sure it's that simple. The salt is involatile, but in a running reactor it is saturated with volatile fission products and full of helium bubbles. Remember that in the MSRE a lot of the volatile products came out of the salt in the pump bowl.


But the volatiles plus helium are only a small fraction of the volume, so there's not much to cause gross voiding and cavitation (the designs I'm leaning towards are mostly without helium circulating though).

Quote:
The hydrodynamic forces on the trailing side of a pump vane create an effective pressure that is well below the inlet pressure. If the accelration of the fluid required for it to keep up with the pump rotor exceeds what can be supplied by the inlet pressure, the effective pressure behind the vane goes to zero and even a perfecly involatile liquid will cavitate. Big, slow rotors minimise the problem, but increase out-of-core volume.


I'm not sure if this happens. It seems what you are describing is a drying up of the pump bowl free liquid surface (sucked out). Then there is no cavitation but dryout of the impeller. Very dangerous for pump life, and something that must be avoided by design or incorporated (dry run in emergency pumps are available on the market).

I like big, slow rotors. A lower impeller speed relative to the fluid reduces erosion (lots of erosive noble metal particles in the salt).


Top
 Profile  
 
PostPosted: Dec 29, 2013 4:48 pm 
Offline

Joined: Sep 22, 2013 2:27 pm
Posts: 262
The KSB feed pumps for the KWU PWR reactors have their bearings as well only above the fuel level as well. The shaft is comparable thick and long. They does not have an issue with that. I attached a sketch of them.

Holger


Attachments:
ksb pumpe dwr.pdf [671.95 KiB]
Downloaded 90 times
Top
 Profile  
 
PostPosted: Dec 29, 2013 7:44 pm 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
Sirs:

I read a little about your European magnetic canned pumps. The magnets are supposed to go bad at 450 degrees C. So they probably wouldn't hold up.

:idea:
The idea of this is to lift this thick liquid back to the top of the reactor after it has given up some of its heat to boil water or some other purpose. It is not at high pressure. Is it out of the question to simply force it up with a compressed gas? This would be similar to automobile pistons except the gas would press down on the Flibe and not a piston. Solenoid valves could control the entry and exit of the gas and the FliBe. The gas would have to be a sealed system and only vented for maintenance. Only the gas, piping and expansion chamber would be exposed to the FliBe. This would minimize the exposure of mechanical parts to the radioactive FliBe. It would allow maintenance to be performed with less exposure (ALARA). Several of these would be needed to ensure a smooth flow of the Flibe throughout the system.

I don't envision a very high pressure since the idea is simply to lift the FliBe a minimum height. Let's say 10 meters. Per Wikipedia, the density of Flibe is 1.94 g / cc. This is almost twice water. One atmosphere would theoretically push water to 32 ft (9.8 m) Two atmospheres of pressure just might do it. Probably three but that is certainly manageable. My garage air compressor is higher than that.

Nitrogen is fairly inert. Will it work with FliBe or are there chemical or radiological consequences? GE BWRs use it to push rods so I know that there should be some radiological experience with it.

I was also thinking of a simple archimedes screw, but compressed gas may be simpler. There would be seal maintenance and compressor maintenance with this option, but it's something that plant mechanics are used to working on.

If the idea has no merit, I figure you guys will just ignore it and move on. 8)


Top
 Profile  
 
PostPosted: Dec 31, 2013 11:07 am 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
Well - I made a couple of sketches of the pump. It's actually like the old Newcomen engine. It's simple, but it should be computer controlled for the timing of the valves. Could dry Nitrogen be used for such a pump? It might be cheap. Now you guys can tell me why this is impractical,.......or not. I know there are some German guys out there. They are supposed to have the best engineers in the world. I'll bet they've already built a FliBe pump and have it ready to sell to the world.


Attachments:
FliBe Pump.jpg
FliBe Pump.jpg [ 830.85 KiB | Viewed 1623 times ]
FliBe Reactor.jpg
FliBe Reactor.jpg [ 669.43 KiB | Viewed 1623 times ]
Top
 Profile  
 
PostPosted: Dec 31, 2013 11:56 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
Valves that work with 700C fluoride salts are not easy to make. ORNL flattened a section of pipe and cooled it to freeze the salt within it to form a valve that could be opened by heating. This kind of valve worked but it takes minutes to open and close.


Top
 Profile  
 
PostPosted: Dec 31, 2013 5:56 pm 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
Well - Valves are going to be needed. This is something that I would think would be commercially available. What other industrial processes operate in this temperature range? I don't see valves needed for throttling, but primarily for isolation. Will throttling be needed for the chemical processes or are these batch processes? What kind is best? slide gate? Butterfly? Globe? This is as important as the pump. Valves will be needed for maintenance isolation as well as the process.

Even if you could freeze and thaw the material quickly enough, it would probably result in thermal stresses which could shorten material life.

How about tritium and other materials in the FliBe? What problems will these cause to materials? This is going to take some work to find materials that are properly environmentally qualified both for normal operation and realistic postulated accident scenarios.

Have you guys looked at valves already?


Top
 Profile  
 
PostPosted: Dec 31, 2013 7:10 pm 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
Well - I've been looking over the internet. I found this was a start of a research project for the government the description is attached. It looks like work started in 2010. It may be done by now and there may be a report out there on appropriate valves.

This high temperature is not a commonplace thing. I found the following for molten salt valves. These are not to the desired temperature rating. They are used for molten salt energy storage with solar power.

http://www.vanessavalves.it/applications/applications_detail.php?id=6

http://www.cesare-bonetti.it/valves/green-energy-valves-concentrating-solar.html

These valves are good up to 550 degrees C or so which isn't quite there.

Here's an article on how molten salt valves are made and some of the material problems.

http://www.solarpowerworldonline.com/2012/04/how-to-seal-molten-salts-in-concentrated-solar-plants/

Here's one where they tested the valves as part of a DOE study. Once again not over 550 degrees C

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&ved=0CGUQFjAG&url=http%3A%2F%2Fwww1.eere.energy.gov%2Fsolar%2Fsunshot%2Fpdfs%2Fcsp_review_meeting_042413_grogan.pdf&ei=dVHDUpn6NOqq2wXpoYGAAw&usg=AFQjCNF5OFr6LWsdNkPfLtIbY8NQLcfoVg&sig2=1F8clckzo3gEZs2cY4bILQ&bvm=bv.58187178,d.b2I

In any event it looks like valves for the first molten salt reactor will be custom and not off the shelf. I hope I am wrong because custom is more expensive than stock items.


Attachments:
10-857_Nagle_lmp_5-6-10(6).pdf [40.82 KiB]
Downloaded 83 times
Top
 Profile  
 
PostPosted: Dec 31, 2013 7:42 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
The original ORNL design used freeze valves but only very few and no applications where the speed of the valve was important. For flow control in the chemical plant I recall using gas pressure (for example in the distiller a reservoir with salt to be distilled was filled then the freeze valve closed. Then gas pressure would be applied above the salt level to push the salt out a pipe at the bottom of the reservoir. The pipe would run up to above the salt level so that it required gas pressure to get the salt to flow.

In general, there is a preference toward flow processing rather than batch processing - but truth be told the amount of processing to be done is quite small - roughly engineering prototype scale for a 1 GWe power plant.


Top
 Profile  
 
PostPosted: Jan 01, 2014 2:25 am 
Offline

Joined: Sep 22, 2013 2:27 pm
Posts: 262
The freeze valves face a major challenge in a commercial reactor with some fission product heat generation in the fuel. It rquire plenty of (cold) energy to close them. The heat generation in the frozen salt decreases quickly. The longer the time to reopening lasts the longer it lasts to open the valve.

Was there any reearch done on mechanical valves as used in chemistry as gate valves, slide valves, ball valves with grinded valve seats. The passing thru for the shaft could be made of graphite???

regards

Holger


Top
 Profile  
 
PostPosted: Jan 01, 2014 4:29 am 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
Happy New Year!

The Transatomic web page says they will be using Hastelloy N as a material for their molten salt reactors (if they ever get built). I also looked at a Chinese video on this and they referred to something they called superalloy. I think that is Hastelloy N.

Thomas register lists several companies that sell Hastelloy N valves.

Shipman valves comes up quickly on a search for Hastelloy N valves.

http://www.shipham-valves.com/en/materials/overview

This doesn't look to be a standard product but a custom order.

Here's a company that looks like they make what is needed.

http://guichon-valves.com/hastelloy-valves/

They state Hastelloy B and not N, but I'll bet they would accommodate a paying order. They are in the right temperature range. They list both globe and gate valves. (800 degrees C)
If a valve specification was written, bidders should be available.


Top
 Profile  
 
PostPosted: Jan 01, 2014 10:11 am 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
I was thinking about my silly pump idea while shoveling snow this morning. This pump is a Reactor Coolant Pump (RCP). I used to work at a relatively small PWR and the RCPs were 5,000 HP. A real pump needs to be designed. You'd quickly wear out valves if they had to be continually throttled. RCPs are custom animals. A great amount of fluid will need to be moved through the reactor and the heat exchanger. There will be more head loss than just the elevation difference. There will be pressure drop through the pipes, steam generator and of course the reactor. A real pump with a Variable Speed Drive (VSD) is needed. I have the gut feeling that a small pump will be needed to begin the process and recirc before the RCP is started.


Top
 Profile  
 
PostPosted: Jan 01, 2014 1:50 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
The fluoride salts are wonderfully at completely cleaning metal surfaces. So good that it is a challenge to avoid diffusion welding (parts in contact that get welded by simply virtue of their contact with each other). ORNL thought that building ordinary valves was more work than they were worth and settled for avoiding them. I haven't thought to challenge this and rather focused thinking on areas they identified as needing improvement.


Top
 Profile  
 
PostPosted: Jan 01, 2014 2:15 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
If you have a variable speed drive pump, what is the advantage of mechanical valves? Throttling with a pump is greatly preferable to throttling with valves as it avoids liquid hammer and similar pressure type transients. This is particularly true if we have a high inertia pump.

Freeze valves work well, there are other variations of them that do not involve mechanical valves.

Freeze valves are slow, but what is the advantage of fast? Fast means liquid hammer. That's bad. The salts have high density and we are moving thousands of kg per second through pipes. Don't want to shut that off abrubdly in a low pressure loop, trust me.

A (mechanically) valveless primary loop is a worthy design goal, IMHO.


Top
 Profile  
 
PostPosted: Jan 01, 2014 4:19 pm 
Offline
User avatar

Joined: Nov 14, 2013 2:34 pm
Posts: 178
Location: Here and There
"If you have a variable speed drive pump, what is the advantage of mechanical valves? Throttling with a pump is greatly preferable to throttling with valves as it avoids liquid hammer and similar pressure type transients. This is particularly true if we have a high inertia pump."

You will need to isolate the pump. I suppose you could put the freeze plug on each side, but I doubt whether a commercial facility will be built that way. Downtime is to be kept to a minimum. I'm sure a commercial reactor will not be able to take some of the tactics used in the 1960s research reactor. I'm not a mechanical engineer. I'm sure there are plenty out there who can give you a better answer. I'll bet there are other reasons.


Top
 Profile  
 
PostPosted: Jan 01, 2014 5:10 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Eino wrote:
"If you have a variable speed drive pump, what is the advantage of mechanical valves? Throttling with a pump is greatly preferable to throttling with valves as it avoids liquid hammer and similar pressure type transients. This is particularly true if we have a high inertia pump."

You will need to isolate the pump. I suppose you could put the freeze plug on each side, but I doubt whether a commercial facility will be built that way. Downtime is to be kept to a minimum. I'm sure a commercial reactor will not be able to take some of the tactics used in the 1960s research reactor. I'm not a mechanical engineer. I'm sure there are plenty out there who can give you a better answer. I'll bet there are other reasons.


Isolation valves or isolation flanges are preferred in non-salt non-nuclear applications for this. Typically my clients are oil industry folks that use isolation flanges. It works because the lines are accessible, low temperature, nonradioactive, and a small spill is easy to clean up. So what if a few liters of diesel drip into a holdup tray below the flange. Someone will just clean it up. And flanges and isolation valves can be closed manually.

In nuclear, salt environment, everything is different. A liter of salt spilling is a major issue, you now have beryllium out of containment and loads of radionuclides. Even the secondary loop is intensely radioactive from delayed neutrons in the HX. There is no possibility of manual operation. Avoiding mechanical valves and flanges is a major advantage in safety, high temperature design. It avoids silly scenarios like someone deciding to close an isolation valve just for fun to see what happens to the pump and lines (Chernobyl operators were like that, even at TMI the operators made a critical valve closure mistake). It avoids isolation valves suddenly not working, this is actually to be expected in this environment, a wetted mechanical component that is not used for years can't be relied upon to isolate upon demand.

I see no issues with using freeze valves for isolation. In fact, I'd use double freeze valves in parallel to assure a good seal. But I'd want a small improvement in the freeze valve design so that it can't damage itself from repeated freeze-thaw cycles.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 86 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  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