What would happen if an energy storage device failed?

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

Post Reply
Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

To Dan and Stoney3K
That’s true Forces in ITER are very big
But this means nothing. We are talking about the system increasing its temperature on about 20 deg even in case of the worst scenario. Field in Alcator C-mod and projected Italian-Russian Ignitor Tokamak are here: correspondently on-axis toroidal field in ITER 5-6T vs. 8 and 13T for Alcator and Ignitor. The last two have resistive and not superconducting magnets. Also in 80s of last century there was a project of Ignitex with 20T of TF (toroidal field) and also with resistive single turn coil.
Why I have asked about teapot with whistling valve? One safety valve would solve the problem of overpressure in the cooling loop. Do not worry.
I note you about the real problem of TOKAMAK. The problem you are worrying is farfetched.

By the way what do you think, what type of cooling system is used for superconducting magnets? Opened loop of closed?

PS: lithium in breeding modules is Li6 or Li7 and if I recall correctly both isotopes are not radioactive.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

Stoney3K wrote:The LHe2
Would you like to say LHe? :)

JdeJ
Posts: 1
Joined: Mon Jul 28, 2008 8:06 am

Post by JdeJ »

Hi, Long time lurker, since NASA Spaceflight days.

This is the Bussard paper that analyses energy dump in ITER vs Polywell.

http://www.askmar.com/Fusion_files/EMC2 ... ystems.pdf

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

JdeJ wrote:Hi, Long time lurker, since NASA Spaceflight days.

This is the Bussard paper that analyses energy dump in ITER vs Polywell.

http://www.askmar.com/Fusion_files/EMC2 ... ystems.pdf
I saw, thank you. But as I have seen paper was written in 1992. So, no ITER. And estimation of million MJ is wrong, as largest TOKAMAK ITER stores (will store) only 44000 and not million.
"Explosive vaporization" is wrong too.
However, many magnetic fusion power systems, especially those using large fields over large volumes, have been designed to use superconducting magnets in order to reduce drive power requirements and thus improve system power balances.
Once I have calculated toroidal field coil made of copper with internal radius 2.8 m. And there the energy that should be spent on creation of mag field much exceeds ohmic losses per 5 sec. So, for that design superconducting magnets have not advantages.

Stoney3K
Posts: 154
Joined: Sun Jun 28, 2009 9:24 pm

Post by Stoney3K »

Joseph Chikva wrote:To Dan and Stoney3K
That’s true Forces in ITER are very big
But this means nothing. We are talking about the system increasing its temperature on about 20 deg even in case of the worst scenario.
Think again. The SC magnets need to be kept at a few Kelvin (-270C) to keep them superconducting. Moreover, they are not made of copper, but of some crystalline material having completely different thermal characteristics.

The reason liquid helium is used is because it can reach such low temperatures, not because it has a high thermal capacity (on the contrary). So the whole system, if it failed, would heat up by *hundreds* of degrees in an instant, going from 2-3K to 300K in a split second.

That is, if the machine isn't already running and the whole reaction heated up the machine considerably, which would mean a temperature climb of even more.

Remember, we're dealing with one of the coldest materials known to Man (liquid helium) being mere inches from the hottest substance known to Man (fusion plasma) in a single machine.
Why I have asked about teapot with whistling valve? One safety valve would solve the problem of overpressure in the cooling loop. Do not worry.
You're assuming the cooling loop would overpressurize because of a *global* increase in coolant temperature (like in a fission reactor with over-spec power output). In case of an SC quench, that is not the case.

The heat buildup would be very localized to the area of the quench which would cause the rupture in the first place. Bleeding off helium from a different part of the system would do nothing, except remove valuable coolant in a situation where you'd need it most.

It's not the overpressure you need to worry about here. The effect of a quench can be compared with someone taking a blowtorch and cutting a hole in a steam line somewhere. The fact that the helium would escape somewhere near the vacuum vessel is the cause for concern.

The only way to effectively prevent a quench from taking the entire facility down with it is to find some manner to shunt the current off the magnets once a quench is detected. For example, a set of crowbar relays which divert the SC current to an area where it can be disengaged in a controlled manner.
By the way what do you think, what type of cooling system is used for superconducting magnets? Opened loop of closed?
Closed loop, without a doubt! SC magnets need to be at cryogenic temperatures to work, and that's impossible to accomplish with open loop cooling systems. You'd constantly need to re-refrigerate fresh coolant which is very energy costly.

Furthermore, the losses in a SC magnet are fairly marginal, and the LHe cooling loop has a very limited cooling capacity to start with. LHe is also very valuable, so you don't want to lose it to the environment if you don't need to.

The system keeping the LHe at the required temperature (secondary magnet cooling loop) could be open-loop because it can use water.
Because we can.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

Stoney3K wrote:Moreover, they are not made of copper, but of some crystalline material having completely different thermal characteristics.
I think that copper content is high. See here http://www.iter.org/doc/www/edit/Lists/ ... ndbest.jpg
But it does not matter, only mass and specific heat does. Mass we have 365 tons. Not enough for high thermal capacity?
Stoney3K wrote:Remember, we're dealing with one of the coldest materials known to Man (liquid helium) being mere inches from the hottest substance known to Man (fusion plasma) in a single machine.
Divided with blanket.
Stoney3K wrote:You're assuming the cooling loop would overpressurize because of a *global* increase in coolant temperature (like in a fission reactor with over-spec power output). In case of an SC quench, that is not the case.
Wrong I am not assuming but know that helium coolant has not capability to take and then dissipate such a big energy. But 365 tons of not copper but matrix material can regardless to that material is crystalline or amorphous.
By the way what do you think, what type of cooling system is used for superconducting magnets? Opened loop of closed?
Stoney3K wrote:Closed loop, without a doubt!
I am not so sure.
See here http://www.iter.org/mach/cryostat Or "closed loop" for you and me are the different things.
And what are we arguing here. Wrong data which as I have understood today goes from Dr. Bussard’s words and repeating here as for majority Dr. Bussard is the man like messiah.
Can you please explain me via which mechanism energy stored in magnetic field transfers into explosion energy. Thanks.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

Stoney3K wrote:
By the way what do you think, what type of cooling system is used for superconducting magnets? Opened loop of closed?
Closed loop, without a doubt!
..................................
The system keeping the LHe at the required temperature (secondary magnet cooling loop) could be open-loop because it can use water.
I see the outer closed loop water cooling system: http://www.iter.org/doc/www/edit/Lists/ ... ling_1.jpg
Please show closed loop helium. Then let's talk about other doubts.

erblo
Posts: 29
Joined: Thu Mar 18, 2010 1:08 pm
Location: Sweden

Post by erblo »

Regarding the LHC 'explosion' (which was covered here in detail if I remember correctly) se for example http://iopscience.iop.org/0953-2048/23/ ... 034001.pdf page 12.

Summary: The SC is surrounded by a Cu bus bar that, in case of a local quench, takes the current while the SC is forcibly quenched with heaters and large external resistors are connected to absorb most of the energy. When the quench occurred (in a badly made SC-SC connection with ~100 times the normal resistance) the surrounding bus bar was also badly connected resulting in an arc. The arc dissipated 275 MJ (@2-6MW) but 'only' partially destroyed the joint (between two magnets) where it occurred. This would have meant repairing/replacing those two magnets and some cleaning.

The arc did however also breach the LHe cooling circuit and the cryogenic insulation vacuum envelope leading to a flow of 13 kg/s (average, peak 20 kg/s - 10 x 'maximum credible incident'). This flow was also heated by the arc. The pressure inside the envelope rose to 8 bar (design maximum 1.5 bar), things started to move and cause further arcs/leaks and ta-da! - the repairs and delays got a little bigger.

Relevance to ITER (and SC energy storage): it is probably designed to handle both a benign quench and some fairly unlikely series of failures without much damage, but even a little damage to one of the coils would likely be a major pain in the ***. For spectacular/catastrophic failures I think one would have to involve the other subsystems (like cooling, vacuum, Li blanket as already suggested) in very unlikely scenarios (like the LHC).

(personally, I'm more curious about how they will deal with any relativistic runaway electron beams...)
Last edited by erblo on Tue Nov 08, 2011 6:52 pm, edited 1 time in total.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Thanks JdeJ for the link.

I continue to disagree that comparing a helium coolent / quench failure as the same as a hole in a steam line. This ignores the liquid flashing to steam as the temperature increases (by even a small amount or not at all if there is any starting positive pressure in the system/ liquid container.
In a liquid helium system, if the internal pressure is small the only driver for vaporization is the temp. If due to ohmic heating the temperature increases a few degrees K, the helium is now boiling and gas overpressure is now building. Once the pressure builds enough that the vessel fails the pressure falls, and the superheated helium flashes into vapor. This provides a large volume of pressurized gas that progressively expands what ever is in it's way untill it disperses in air or meets a containment wall that resists the further expansion.

It is not that steam is leaking out via a small or large hole so much as what this does to the pressure in the system and if there is superheated liquid that can then flash into a much larger volume of steam/ gas.
This is what drives the explosion of Steam Locamotives, and steam boilers in general. A valve becomes stuck or for some other reason the pressure builds to a point where the containment fails, there is a very large hole which allows for rapid decompression and resultant vigerous boiling of the superheated liquid. Another good example is a video of the Italian (?) battleship around the time of WWI that rolled over. Presumably the boiler venting was blocked, the boiler walls failed and the the resultant superheated steam explosion blew the battleship in in half. It wasn't that the water (or helium) hat much of a change in temperature, it was that due to the drop in pressure the liquid became superheated, and the resultant rapid change in volume drove and explosive expansion. The temperature of the solid materials is trivial. What is important is the amount of liquid that can flash to vapor, how fast it does it, and volume considerations of surrounding structures that it can push against.

The best estimate of stored (superconducting) magnetic energy in ITER that I have seen comes from here. ~13 T central magnet with ~ 6.4 GJ of stored energy.

http://www.iter.org/newsline/122/182

PS: The steam explosion that destroyed the battleship was not due to a increase in temperature, in fact the temperature of the steam/ liquid water was probably significantly decreasing due to the heat of vaporization. It was purely a pressure containment issue, with subsequent physics effects.

Dan Tibbets
To error is human... and I'm very human.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

D Tibbets wrote:Thanks JdeJ for the link.

I continue to disagree that comparing a helium coolent / quench failure as the same as a hole in a steam line. ........................
Dan, again catastrophic scenario. But Italian battleship at WWI, many other explosions of pressure vessels teaches us how to design and build e.g. steam turbines with corresponding boilers. And we are still building them now. We (or at least someone of us) know how to do that well.

Please answer on question I've put you not one time: how you imagine the mechanism of transfer of energy initially stored in magnetic field. I understand that you think that energy will transferred to helium, helium will boil and then expand with catastrophic aftereffect. But before how energy goes to helium?

And finally: what do you think will happen with TOKAMAK with very strong field (e.g. 13 T on-axis http://www.frascati.enea.it/ignitor/ ) with not SC but resistive magnets?
Thanks.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

I couldn't find the Battleship explosion I mentioned, so it cannot be confirmed.

Alternately, a small quench in an MRI machine

http://www.youtube.com/watch?v=1R7KsfosV-o

And an apparent submarine steam explosion (sort of). The effects are surprising, especially when the resultant tidal wave reaches the ship!

http://www.youtube.com/watch?v=Jw--eCJi4P4

Dan Tibbets :)
To error is human... and I'm very human.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

D Tibbets wrote:I couldn't find the Battleship explosion I mentioned, so it cannot be confirmed.

Alternately, a small quench in an MRI machine

http://www.youtube.com/watch?v=1R7KsfosV-o

And an apparent submarine steam explosion (sort of). The effects are surprising, especially when the resultant tidal wave reaches the ship!

http://www.youtube.com/watch?v=Jw--eCJi4P4

Dan Tibbets :)
So, you are not answering on question. I know about threat. But it's solvable and solved. TOKAMAK as concept has less problems with magnetic field. TOKAMAK unlike to all other approaches has reached already Lawson criterion. But triple product has not been reached and very unlikely that ITER with his claimed 3keV from ohmic heating 20MW RF source and 1MeV NBI injector will reach desired 15keV temperature. That is a problem. Also problem is in joining of NBI injector with vacuum chamber. You are searching problems in wrong place.

ladajo
Posts: 6258
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

D Tibbets wrote:I couldn't find the Battleship explosion I mentioned, so it cannot be confirmed.

Alternately, a small quench in an MRI machine

http://www.youtube.com/watch?v=1R7KsfosV-o

And an apparent submarine steam explosion (sort of). The effects are surprising, especially when the resultant tidal wave reaches the ship!

http://www.youtube.com/watch?v=Jw--eCJi4P4

Dan Tibbets :)
Dan,
That was Base Surge from a nuclear weapon test at Bikini. That was not a "submarine steam explosion".

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

ladajo, well I did say 'sort of'. Of course it was a nuclear explosion. I added it purely because of the startling ending :) though it did involve liquid/ vapor phase changes and resultant gas expansion, along with shock wave effects in the liquid. The major difference is that the expanding shock wave was powered by a sudden large increase in temperature, phase conversion and possible hydrolic effects of the overlying liquid water, not a rapid (slow by comparison) boiling of a superheated fluid due to pressure release. Both need a material to carry the shock wave. In a steam explosion it is the liquid in the boiler, and surrounding air or other volitile elements and solids nearby. In in a chemical (small to moderate extent) or nuclear explosion most of the shockwave is due to heating of outside material.

And I used submarine loosely- implying an underwater ship... boat, mostly to justify watching the video till the end. But anything underwater could be called submarine, like a submarine volcano, or explosion of any origin.

Dan Tibbets
To error is human... and I'm very human.

Stoney3K
Posts: 154
Joined: Sun Jun 28, 2009 9:24 pm

Post by Stoney3K »

Joseph Chikva wrote:
D Tibbets wrote:I couldn't find the Battleship explosion I mentioned, so it cannot be confirmed.

Alternately, a small quench in an MRI machine

http://www.youtube.com/watch?v=1R7KsfosV-o

And an apparent submarine steam explosion (sort of). The effects are surprising, especially when the resultant tidal wave reaches the ship!

http://www.youtube.com/watch?v=Jw--eCJi4P4

Dan Tibbets :)
So, you are not answering on question. I know about threat. But it's solvable and solved. TOKAMAK as concept has less problems with magnetic field. TOKAMAK unlike to all other approaches has reached already Lawson criterion.
Isn't that a bit beyond the scope of the discussion?

What I'm interested in is whether or not a tokamak (or any other fusion facility, for that matter) could be operated safely.

Remember, if we manage to get fusion going in the coming decade, it means commercial deployment, and it also means that fusion plants stop being science facilities, but start becoming utilities which need to operate efficiently.

Operating efficiently means operating on a tighter budget than university research grants. So it may involve cutting corners during maintenance, operation and construction... which means serious safety risks.

I wonder how long it will take for the first fusion facility to go 'boom', either through an accident or due to deliberate (possibly politically motivated) sabotage, and what safety mechanisms would be implemented to make the facility go 'fizz' instead.
Because we can.

Post Reply