Pumping with no moving parts; will it work for the LFTR?

Threads relating to the design of liquid-fluoride reactors.
User avatar
jaro
Posts: 1938
Joined: Nov 30, 2006 9:18 pm
Location: Montreal
Contact:

Re: Pumping with no moving parts; will it work for the LFTR?

Post by jaro » Nov 06, 2008 4:49 pm

It also depends on other details of the PHT loop.

For instance, if it combines pumping with siphon fuel channels, a stoppage in pumping will quickly cause the siphons to suck air and unprime, draining the fuel channels & shutting down the reactor.

User avatar
Axil
Posts: 1055
Joined: Aug 21, 2008 12:57 pm

Re: Pumping with no moving parts; will it work for the LFTR?

Post by Axil » Nov 06, 2008 4:49 pm

STG wrote:And again: Low inertia! If you lose power to the pump the salts stop circulating more suddenly (so a Loss Of Flow accident becomes essentially alike to a blocked rotor accident). This results in a high and fast insertion of reactivity...which is an unwanted effect and you have to make sure that this won't result in any potential damage. Furthermore it's much easier to proof to and to convince a safety organization that the possibility for such a fast increase is very unlikely (for rotating pumps only in blocked rotor accident).

If you want to have a prototype MSR design, go conservative!
Multiple EM pumps operating in serial alignment on the double walled pipe will mitigate single EM pump failure. For total power failure, a capacitor bank can support a gradual current reduction to the EM pumps and associated magmatic field reduction to mimic the high inertia of a rotor pump causing fast reactivity to be minimized.
The old Zenith slogan: The quality goes in before the name goes on.

User avatar
honzik
Posts: 90
Joined: Apr 21, 2007 1:42 am
Location: Silicon Valley, CA

Re: Pumping with no moving parts; will it work for the LFTR?

Post by honzik » Nov 06, 2008 4:54 pm

I remember reading an patent by an Israeli who came up with a direct electromagnetic pump that require a two phase flow (with one phase being liquid metal) to pump effectively. If the metal were bismuth, you could pump and extract protactinium at the same time....

STG
Posts: 277
Joined: Oct 29, 2007 6:27 pm

Re: Pumping with no moving parts; will it work for the LFTR?

Post by STG » Nov 06, 2008 5:09 pm

OK all of those things are true, it depends on a lot of factors...

I would expect putting them in series at "strategic locations" in the primary loop is more likely to result in a less rapid decrease of flow and would result in less pressure fluctuation in the primary system and therefore in less catastrophic results from a leakage. So no complaints there. But the high magnetic field can result also in destabilization of the salt (read W. G. Homeyer, “Thermal and chemical aspects of the thermonuclear blanket problem,” Tech. Rep. 435, MIT, Cambridge, Massachusetts, June 1965.)

I know all reactivity effects can be controlled and so on...but the MSBR document mentions that the reactivity margins of the graphite control and boron carbide rods were all set to limit the effects of the loss of pumping power.

Furthermore the inertia of the system can be increased, not with a capacitor bank, but easily by using a motor-flywheel-generator set. Of course all of these tricks essentially make a normal pump with EM-aided inlet flow more interesting..
Liking All Nuclear Systems, But Looking At Them Through Dark And Critical Glasses.


STG
Posts: 277
Joined: Oct 29, 2007 6:27 pm

Re: Pumping with no moving parts; will it work for the LFTR?

Post by STG » Nov 08, 2008 8:25 am

Those works are all about liquid metal pumping and the article about salt is the one which I mentioned and describes the destabilization of the salt in high magnetic fields....
Liking All Nuclear Systems, But Looking At Them Through Dark And Critical Glasses.

User avatar
robert.hargraves
Posts: 302
Joined: Sep 10, 2008 7:40 pm
Contact:

Re: Pumping with no moving parts; will it work for the LFTR?

Post by robert.hargraves » Nov 11, 2008 11:07 am

The salts expand when heated, a criticality safety advantage. If the salt piping is vertical, can gravity pump the hot, expanded salt and return cooled, chemically processed salt?

User avatar
jaro
Posts: 1938
Joined: Nov 30, 2006 9:18 pm
Location: Montreal
Contact:

Re: Pumping with no moving parts; will it work for the LFTR?

Post by jaro » Nov 11, 2008 11:54 am

Well, yes, but there is an HX somewhere in that loop -- that's where the salt gets cooled.

User avatar
ondrejch
Posts: 391
Joined: Dec 19, 2006 11:01 am
Location: Knoxville, TN
Contact:

Re: Pumping with no moving parts; will it work for the LFTR?

Post by ondrejch » Nov 11, 2008 7:08 pm

jaro wrote:Well, yes, but there is an HX somewhere in that loop -- that's where the salt gets cooled.
This should be on top of the vertical core. Perhaps a cylindric vertical core (2-fluid design) not finished by conical end, but by a block (of tungsten :), serving as a reflector as well) with vertical holes for fuel salt (and horizontal for coolant salt), fuel salt is collected at the top of the HX and piped down (possibly with EM pumps and multiple loops, some with fuel salt processing) under the bottom of the active zone, where it joins freeze plug tube (going down) and a pipe back up to the core.

Fuel salt could move rather slowly, which creates a gradient T(r) (to be calculated), thus there will be a strong negative reactivity feedback if T(r=0 ) reaches salt boiling temperature (~1400C).

Due to "flat head", this geometry also allows to interlace core and blanket regions with heat pipe channels finished inside the tungsten block head. For the heat pipe fluid a simillar salt as in the core could be used, with slightly different chemistry (more Na?) to lower the boiling temperature to ~1100C.

Fast spectrum (~10% fissile salt concentration) allows compact design, together with faster actinide destruction, and is compatible with tungsten. IF there is a reason for thermal spectrum, lower salt concentrations and molybdenum may be better.

Cyril R
Posts: 5037
Joined: Jul 14, 2008 3:12 pm

Re: Pumping with no moving parts; will it work for the LFTR?

Post by Cyril R » Jun 10, 2010 7:44 am

Does anyone know the electrical conductivity of molten very hot UF4, UF3, PuF3 and ThF4?

Post Reply