Energy From Thorium Discussion Forum

It is currently Aug 15, 2018 4:23 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 86 posts ]  Go to page 1, 2, 3, 4, 5, 6  Next
Author Message
 Post subject: Pumps for MSR Reactors
PostPosted: Dec 17, 2013 4:18 am 
Offline

Joined: Sep 22, 2013 2:27 pm
Posts: 262
A very important component of MSR reactors are the pumps required to pump the salt thru the reactor and as well intermediate coolants thru intermediate cycles. I did not find a lot of informations about the pumps used.

According to my studies the design of a pump is a challenge. It needs to resist up to 800°C, corrosive product, high mass/volume flow, controllability...

From my point of view magnetic drive or canned pumps seems an option if it is suitable to make use of materials that does not absorb too much magnetic power by eddy currents. What kind of pump types seems suitable (I attached some ideas)?
What kind of gaskets are foreseen?
What kind of bearings are foreseen?

Are there any studies about the pumps used for a MSR and the intermediate coolants as lead, salt?

Best regards

Holger


Attachments:
Pump Technology.docx [227.8 KiB]
Downloaded 122 times
Top
 Profile  
 
PostPosted: Dec 17, 2013 9:16 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Holger,

The holy grail for pump in MSRs is likely a canned rotor pump of the switched reluctance type. It is very difficult to develop because of the challenging conditions; you mentioned a nonmagnetic containment material for the can. That is made particularly difficult in this application due to a combination of gamma, neutron radiation, plus high temperature, plus corrosive fluoride salt (not corrosive to Hastelloy N but it may not be an ideal nonmagnetic alloy for this application). Control of the rotor position is also hard for the same environmental reasons. You'd have sensors right inside a high gamma and neutron flux.

The AP1000 has such canned rotor pumps, so we know we can deal with gamma rays and pressurized water. Probably the low pressure will balance the higher temperature for the MSR application, but the higher gamma and much higher neutron flux (ap1000 pumps have virtually no neutron flux), is a big unknown.

The AHTR/FHR groups are putting some effort in developing a fluoride canned rotor pump for a future application. If that is successful I suspect the canned rotor pump will also be used in MSRs (still need some work on neutron flux).

A more near term target would be a fairly conventional centrifugal pump. With dry gas seals as primary seal and additional metallic backup seals (and likely also a metallic standstill seal as added backup). Likely salt lubricated bearings consisting of carbides (SiC, WC). Such advanced bearings are available commercially these days.

In my opinion we should go for a relatively low pump speed to avoid abrasion of the noble metal fission products. The impeller speed relative to salt should be as low as we can make it while keeping a reasonable pump size. However, it seems that all groups working on molten salt pumps have ignored my opinion. :lol:


Top
 Profile  
 
PostPosted: Dec 17, 2013 10:11 am 
Offline

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

it was a pleasure to read your reply!

If I see my own design ideas the primary pumps hide behind the HX (1.5m thick Mo, salt, LBE). The pump control unit can be placed even further away. .hence I do not see the neutron flux or gamma as a main challenge for the pumps.

My idea is as well to use a conventional zentrifugal pump. An idea is to use labyrinth sealings and a He gas or non radioactive salt as flushing liquid.

SSIC bearings are on the market since decades. I remember a catalogue from 1989. SSIC shows somes corrosion in contact with liquid chloride salts.

I do not see the point of the rotor speed. "to avoid abrasion of the noble metal fission products". The blades of the pumps are cm thick and hence are by far more robust than the tiny structure of a HX. Could you please explain this point.

Thanks

Holger

PS: The AP1000 pumps work in < 300°C. It seems to me very different.


Top
 Profile  
 
PostPosted: Dec 17, 2013 1:20 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
If you are talking about the primary pump then you are pumping fuel salt. It has delayed neutron releases. Roughly 0.1% of the neutrons will be released outside the core with perhaps 20% of that in the pump. So 3-4 orders of magnitude less than the reactor core but still significant. This is just for the impeller.

ORNL had the pump motor many feet above the impeller due to temperature concerns. They also had the bearings only above the impeller - again due to temperature and sealing it away from the fission products. As a result had challenges with vibration. Their bearings were hydrocarbon lubricated which limited the temperature at the bearing (and also managed to leak into the offgas system which caused some clogging). They started looking into salt lubricated bearings.


Top
 Profile  
 
PostPosted: Dec 18, 2013 4:09 am 
Offline

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

Thank you very much for your comment.

I didn`t see the point of the delayed neutrons. Due to the heat it seems favorable to have a long shaft and place the motor a bit away from the pump.

Concerning the bearings I got in 89 working in the chemical industry in Germany a catalogue from Cerobear promoting ceramic roller bearings. Ceramic bearing allows a lubrication by the salt or even dry running. They are suitable for high temperatures and corrosive media. The corrosion topic would require further attention. It is as well to investigate if there are situations of high impact on the pump that might be critical example during draining or filling the cycle.

Another idea could be to investigate plain bearings with graphite.

Holger

http://www.cerobear.com/fileadmin/image ... ochure.pdf
http://www.cerobear.com/fileadmin/image ... ations.pdf


Top
 Profile  
 
PostPosted: Dec 18, 2013 8:31 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
HolgerNarrog wrote:
I didn`t see the point of the delayed neutrons. Due to the heat it seems favorable to have a long shaft and place the motor a bit away from the pump.


There are two points.

1. This is, by definition, impossible with canned rotor pumps; they are wet rotor pumps so the stator and positional electronics are directly around fuel salt. A small quantity of fuel salt, but it gives off one heck a lot of gamma and considerable neutron flux. The upside to this is that IF it works, it likely also works with magnetic bearings so you have no issue with molten fuel salt wetted bearings, making it potentially the ultimate pump for this application.

2. Even with conventional centrifugal pumps, the distance you need to get reasonable neutron fluxes (in terms of what electronics can take for years), is very large. This then results in new issues such as vibration. It can be solved with intermediate dry and wet bearings, but this means you need a new technology: a molten fuel salt wetted bearing. It is likely feasible as the state of the art in high temperature, aggressive media bearings is silicon carbide bearing with tungsten carbide contacting parts - shafts/sleeves/sliders (I don't know what to call them in English).

An interesting deduction is in order... IF the canned pump doesn't work, we need to use salt lubricated bearings and have higher risk of seal leakage (rotating seals... me no like).


Quote:
Another idea could be to investigate plain bearings with graphite.


Graphite is very soft and weak, I doubt it will do.


Top
 Profile  
 
PostPosted: Dec 19, 2013 2:30 am 
Offline

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

in a hydrodynamic plain bearing there is usually a hard shaft rotating in a soft, lubricating surrounding. Graphite is a good lubricant. A main question is if the chloride or fluoride salts have at least some lubricating properties.

If the motor is placed some several - 100 mm away from the pump it can be protected against neutrons and gamma by a thick layer of cast iron. It might be heavy but is not expensive.

Holger


Top
 Profile  
 
PostPosted: Dec 20, 2013 9:02 am 
Offline

Joined: Jun 05, 2011 6:59 pm
Posts: 1335
Location: NoOPWA
I'm still not sure what the issue is. Why not just use an external pump to pressurize a clean FLiBe and use that to drive an internal hydraulic motor driven pump? If any of the FLiBe leaks... well, who cares?

_________________
DRJ : Engineer - NAVSEA : (Retired)


Top
 Profile  
 
PostPosted: Dec 20, 2013 9:58 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
KitemanSA wrote:
I'm still not sure what the issue is. Why not just use an external pump to pressurize a clean FLiBe and use that to drive an internal hydraulic motor driven pump? If any of the FLiBe leaks... well, who cares?


Good thinking again, Kiteman. Solve the solid shaft problem with a liquid in a pipe. After all, its a liquid reactor, why not a liquid driven pump?

Have you looked at specifics? What is the peak pressure in the hydraulic circuit? Normally hydraulic circuits operate at low temperature where creep is not an issue, not so with FLiBe (what would be the minimum acceptable temp for the FLiBe hydraulic circuit?).


Top
 Profile  
 
PostPosted: Dec 20, 2013 10:09 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
HolgerNarrog wrote:

If the motor is placed some several - 100 mm away from the pump it can be protected against neutrons and gamma by a thick layer of cast iron. It might be heavy but is not expensive.

Holger


Good enough for primary gammas. Iron isn't very good at getting rid of neutrons though. It would have to be heavily borated iron or something. I think borated steel is better and not much more expensive. Still you get secondary radiation from the activation of the borated steel. Conventional electronics will be hard. I think we need an expert on radiation resistant electric motors and control electronics here. Maybe someone from NASA?


Top
 Profile  
 
PostPosted: Dec 21, 2013 1:50 am 
Offline

Joined: Jun 05, 2011 6:59 pm
Posts: 1335
Location: NoOPWA
Cyril R wrote:
KitemanSA wrote:
I'm still not sure what the issue is. Why not just use an external pump to pressurize a clean FLiBe and use that to drive an internal hydraulic motor driven pump? If any of the FLiBe leaks... well, who cares?


Good thinking again, Kiteman. Solve the solid shaft problem with a liquid in a pipe. After all, its a liquid reactor, why not a liquid driven pump?

Have you looked at specifics?
Nope. I'm retired. I will leave that to an active engineer. But I have to guess that since the pump head is not THAT high, the hydraulic circuit won't be too significant either.

_________________
DRJ : Engineer - NAVSEA : (Retired)


Top
 Profile  
 
PostPosted: Dec 21, 2013 5:52 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
I asked around with a few pump suppliers I've worked with before. They all mentioned the same thing: high temperature hydraulic pumps don't exist. Certainly not for fluoride salts.

The highest temperature they could offer was 370 Celsius, and only a tiny one. There was apparently mostly an issue with the synthetic heat transfer oils available. None of them were stable, except for a few that had unattractive physical properties (viscosity way too low or too high). FLiBe of course is stable, but I don't know if its lubricating properties are sufficient. Possibly it is with ceramic/composite materials, but then you'd have similar development issues ahead as a salt wetted bearing.


Top
 Profile  
 
PostPosted: Dec 21, 2013 1:37 pm 
Offline

Joined: Jun 12, 2011 2:24 pm
Posts: 92
Location: Taunusstein, Germany
Cyril R wrote:
Conventional electronics will be hard. I think we need an expert on radiation resistant electric motors and control electronics here. Maybe someone from NASA?


Why has the drive electronics to be located close to the motor? Conventional inverters can easily handle tens of meters of cable length.


Top
 Profile  
 
PostPosted: Dec 21, 2013 7:06 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3065
It helps me to summarize the requirements:
Pump Impeller
1) needs to handle the fluoride salts with fission products
2) needs to tolerate gamma and neutron flux of the fuel salt.
3) needs to handle 700C
Pump motor
1) wants to be in a serviceable area
2) should not become radioactive (to keep costs reasonable for servicing)
3) wants to be at a lower temperature (magnetics and wiring are more manageable at lower temperature
Pump controls
1) wants to be closer to human temperature, no radiation
Power Transfer
1) can be solid rotor
2) could be hydraulically coupled
Bearings
1) want to be close to the impeller to minimize vibrations
2) cantellevering is possible (was used in MSRE) but is (significantly?) more prone to vibrations than putting bearings on both sides of the impeller
Bearing lubrication:
1) hydrocarbons is the traditional solution (used in MSRE) but requires sealing from fuel salt which is challenging (and in fact was the cause of one of the shutdowns of MSRE). Cannot take full temp or radiation field. Hydrocarbons leaking into the fuel salt will create clogs in the offgas system.
2) magnetic - exceptional performance but need to isolate from temperature so a real challenge.
3) salt lubricated - develop bearings that use the fuel salt (or blanket salt or clean salt) as the lubricant. ORNL was working on this. But a fluoride salt will tend to clean metalic surfaces so well that they can be pressure welded together. If this can be done then we can deploy the bearings close to the impeller.

Canned motors have been mentioned several times. It seems like we can do this regardless of how we solve other challenges. We can make a canned motor that transfers power from the external motor to the inside the can rotor using magnetic coupling. This can be done far from the neutron and gamma fields and cooled to reasonable temperatures. Normally a canned motor is small using a single mass as both the impeller and rotor but if we can solve the bearings challenge then I see no reason we can't make the region inside the canned motor extend several meters so that the magnetically coupled rotor is far from the impeller. This extension could be either s solid rotor or hydraulically coupled.

The key issue I think is the bearings. Seems like the first choice would be salt lubricated bearings. Fortunately, I think this development could be done primarily outside a radiation field. I'm thinking 100mm of carbon boron should attenuated neutrons dramatically (especially so since they will be delayed neutrons that start out more thermal). Add 100mm of steel to attenuate the gamma and I think your bearings can be mostly out of the radiation field.


Top
 Profile  
 
PostPosted: Dec 21, 2013 11:48 pm 
Offline

Joined: Jun 05, 2011 6:59 pm
Posts: 1335
Location: NoOPWA
Cyril R wrote:
I asked around with a few pump suppliers I've worked with before. They all mentioned the same thing: high temperature hydraulic pumps don't exist. Certainly not for fluoride salts.
If you can make a high temperature pump for FLiBe fuel salt, why not a pump for CLEAN FLiBe?

_________________
DRJ : Engineer - NAVSEA : (Retired)


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 86 posts ]  Go to page 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:  
Powered by phpBB® Forum Software © phpBB Group