Page 1 of 1

Teller-Ulam fusion generator

Posted: Fri Jul 29, 2011 6:04 pm UTC
by idobox
Hi everyone,

Thinking about fusion again. While researchers have been trying to sustain plasma hot enough for a few tens of years, there have quite a number of devices built with net positive energy gain from fusion : Teller-Ullam Thermonuclear bombs.

From what I read, The device is ignited by a nuclear bomb. The radiation (X-ray) is focused on a pellet of fuel, that will implode, fuse, and explode with a net energy gain. Most devices are two stages, with the Tsar Bomba suspected to have three stages.

So I understand that the specifics of the design of thermonuclear bombs is not widely available, but it appears to me that the fusion of a pellet of D-T mix will release enough energy to make a larger pellet fuse.
My question is, would it be possible to use this for power production?
I imagine exploding a first pellet, with lasers for example, and focusing the energy of this explosion to fuse another bigger pellet, with as many stages as required to finnally get a net energy gain.
For additionnal awesomeness, we could imagine a form sustained fusion, by having two parabollic reflectors facing each other a few hundreds km away. The ignition of a pellet at the focus of the first reflector will cause the ignition of the other one at the focus of the second one. The round trip of the shock wave would take a few hundreds of µs, during which you would have removed the used fuel and put a new pellet.

Apparently, no one seriously studied this, so I suppose there is a very good reason it doesn't work.
Is it a probem of scale (like, nothing interresting happens under nuclear weapon-level energies, which would blow the power plant)?
Other engineering issue, like the durability of X-Ray reflectors or something?

Re: Teller-Ulam fusion generator

Posted: Fri Jul 29, 2011 7:00 pm UTC
by mfb
>> (like, nothing interresting happens under nuclear weapon-level energies, which would blow the power plant)
That should be the main problem. There is a good reason why all fusion bombs are quite large - you need a lot of energy to compress the material, and that is just impossible with some small home-made tools.

The National Ignition Facility uses lasers to start small fusion reactions. But the laser pulses have to be really short and really intense, and generating that needs a lot of power (much more than you get out of one shot). Another problem is that they can do ~2-3 shots per day - a power plant would need several per second.

X-ray / gamma ray deflectors? There is no material which reflects them with a reasonable efficiency under useful angles.

Re: Teller-Ulam fusion generator

Posted: Fri Jul 29, 2011 7:11 pm UTC
by PM 2Ring
mfb wrote:X-ray / gamma ray deflectors? There is no material which reflects them with a reasonable efficiency under useful angles.

Indeed, mfb.

Idobox, you can only get x-rays to reflect if you use very shallow angles, so a simple parabolic or elliptical reflector isn't going to do what you want.

Also, I assume you made some kind of typo / scale error when you wrote "by having two parabollic reflectors facing each other a few hundreds km away". That'd be a very large power plant! :)

Re: Teller-Ulam fusion generator

Posted: Sat Jul 30, 2011 7:56 am UTC
by sikyon
mfb wrote:>> (like, nothing interresting happens under nuclear weapon-level energies, which would blow the power plant)
That should be the main problem. There is a good reason why all fusion bombs are quite large - you need a lot of energy to compress the material, and that is just impossible with some small home-made tools.

Actually, you can create a fusion reactor at home. http://discovermagazine.com/2007/mar/radioactive-boy-scout, though not a bomb.


Idobox, you can only get x-rays to reflect if you use very shallow angles, so a simple parabolic or elliptical reflector isn't going to do what you want.


Actually, it is possible to create x-ray reflectors using dielectric mirrors that work on direct incidence for soft x-rays. A parabolic mirror of such design would be prohibitively difficult. Instead, you can use a zone plate as a focusing lens for these types of optics.

In any event, he is not talking about focusing x-rays.
The round trip of the shock wave would take a few hundreds of µs,

He is suggesting focusing the shock waves. This is, indeed, the critically difficult point in a regular nuclear bomb.

Consider a fission nuclear bomb. You grab a piece of U-235 and wrap C-4 around it. You then detonate the C-4 and compress the U-235 to critical mass. Then what happens? The thing explodes, leaving most of the uranium unused. In order to get your bomb to actually work well, you need to control how the explosions are shaped within it so that more than a tiny bit of the nuclear material can be used inside of it. You need to keep your nuclear pipe bomb from just leaking out one side, and to keep the explosion going stronger and stronger in it.

This took the greatest physicists in the world years to solve before they crafted the first atomic bombs. It is a virtually impossible scenario for a controlled system, with shockwave resonances that would need to last more than milliseconds and the added difficulties of introducing new fuel and actually harnessing the energy. In theory, in a perfect world, it would work. In reality, it is far beyond us to control such an energetic process so precisely.

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 1:46 pm UTC
by idobox
PM 2Ring wrote:Idobox, you can only get x-rays to reflect if you use very shallow angles, so a simple parabolic or elliptical reflector isn't going to do what you want.

I know that most of the info on bombs is classified, but people tried to understand how they work, and on the wiki page, they talk about focusing the radiation of the primary on the secondary. Obviously, the people who came up with this idea believe it is possible to reflect the radiation of a bomb with reasonable efficiency.

PM 2Ring wrote:Also, I assume you made some kind of typo / scale error when you wrote "by having two parabollic reflectors facing each other a few hundreds km away". That'd be a very large power plant! :)

No typo here. It would be two plants very far away. If you have a better idea to delay a X-Ray burst long enough to change fuel, I'm interested
sikyon wrote:This took the greatest physicists in the world years to solve before they crafted the first atomic bombs. It is a virtually impossible scenario for a controlled system, with shockwave resonances that would need to last more than milliseconds and the added difficulties of introducing new fuel and actually harnessing the energy. In theory, in a perfect world, it would work. In reality, it is far beyond us to control such an energetic process so precisely.
.
But the solution for the Teller-Ulam is knwon now. The NIF ignites D-T pellets with lasers. Using the released energy to fuse another pellet, with as many stages as needed to get more energy out than the lasers used, doesn't look very difficult to me. Unless, of course, if this kind of chain reaction can only happen at important energies.

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 2:12 pm UTC
by Soralin
idobox wrote:I know that most of the info on bombs is classified, but people tried to understand how they work, and on the wiki page, they talk about focusing the radiation of the primary on the secondary. Obviously, the people who came up with this idea believe it is possible to reflect the radiation of a bomb with reasonable efficiency.

Well, perhaps not:

http://en.wikipedia.org/wiki/Teller%E2%80%93Ulam_design
The calculated ablation pressure is one order of magnitude greater than the higher proposed plasma pressures and nearly two orders of magnitude greater than calculated radiation pressure. No mechanism to avoid the absorption of energy into the radiation case wall and the secondary tamper has been suggested, making ablation apparently unavoidable. The other mechanisms appear to be unneeded.

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 2:24 pm UTC
by thoughtfully
The Teller-Ulam design features prominently in one of Robert L. Forward's novels, Camelot 30K.

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 3:11 pm UTC
by idobox
Soralin, I just re-read the wiki page. And all seem to assume a significant part of the X-rays stay in the casing. Whether the pressure is direct radiation pressure or caused by ablation, you need X-Rays everywhere. Especially in the ablation process, they consider materials that absorb X-Rays for the pusher, so the only way for X-rays to heat all the pusher is to reflect on the outer casing.
Given the absurd energy of a nuclear bomb, it is possible that only a tiny amount of X-rays need to be reflected to ignite fusion.

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 4:01 pm UTC
by Soralin
idobox wrote:Soralin, I just re-read the wiki page. And all seem to assume a significant part of the X-rays stay in the casing. Whether the pressure is direct radiation pressure or caused by ablation, you need X-Rays everywhere. Especially in the ablation process, they consider materials that absorb X-Rays for the pusher, so the only way for X-rays to heat all the pusher is to reflect on the outer casing.
Given the absurd energy of a nuclear bomb, it is possible that only a tiny amount of X-rays need to be reflected to ignite fusion.

Well the point of the ablation thing is that most of it isn't being reflected, it's being absorbed in huge quantities by the surface, which causes the surface to heat up to very large temperatures, and causes atoms and such to fly off of the surface at very high speeds. And as a result (every action has an equal and opposite reaction), the rest of the secondary is pushed inwards. And that this ablation process provides about 50x as much energy as what the reflection amount is calculated to be. And in the same way, it's not really reflected by the outer casing so much that it's just heated up to such absurd temperatures that it starts emitting x-rays. You could try doing the same thing, but such a process is a destructive one, you'd have to replace your "mirrors" each time, and they'd have to be quite close to their target because it would get emitted in all directions, not just perpendicular to the surface. You wouldn't be able to aim it in other words, it would be closer to a white sheet of paper than a mirror, in terms of reflection.

And something like that is done, just to get single pellets to fuse, look for "hohlraum" in http://en.wikipedia.org/wiki/Inertial_c ... ent_fusion

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 7:26 pm UTC
by Technical Ben
With the cold war, and previous WWII industry would it not be possible to mass produce nuclear bombs? At a scale where you could run at least 1 fusions generator off of them?
Bombers constantly flying over head orr perhaps ICBMs constantly being launched. Set one site, with the continuous explosions, nations set up steam generators on the craters edge.

It would make for one great Sci-Fi. Not practical by any means.

Re: Teller-Ulam fusion generator

Posted: Sun Jul 31, 2011 9:13 pm UTC
by idobox
Soralin wrote:Well the point of the ablation thing is that most of it isn't being reflected

That I had understood

Soralin wrote:it's not really reflected by the outer casing so much that it's just heated up to such absurd temperatures that it starts emitting x-rays

That, I didn't understand.
If it is impossible to effectively reflect X-Rays, a sustained reactor could not work on this principle.

Technical Ben wrote:With the cold war, and previous WWII industry would it not be possible to mass produce nuclear bombs? At a scale where you could run at least 1 fusions generator off of them?
Bombers constantly flying over head orr perhaps ICBMs constantly being launched. Set one site, with the continuous explosions, nations set up steam generators on the craters edg

A steam powered catapult would be a more efficient way to drop them. The question is how to convert the energy of the blast into useful power? If you explode the bombs in free air, most of the power will be lost to radiation, and heating air. You couldn't heat water to more than 100°C, which would make it difficult to extract power from it. You could use another fluid, like molten iron or something, but still, the loss to the atmosphere would be huge.
I suppose a giant boiler would be a better solution, but I get headaches when I try to imagine a boiler capable to whistand tsar bomba explosions.
An alternative could be to drop the bomb to the bottom of a well, fill it with water, and put a turbine on top. Again, the scale of the device is mind boggling.

By the way, I've heard about people trying to use another concept from nuclear bombs: fusion-boosted fission. The idea is to use fusion as a source of neutrons, to irradiate nuclear material, and make it fission at a much higher rate.
This could be used to destroy waste, or to use low grade fuel, or even alternate fuels.
Sounds cool.

Re: Teller-Ulam fusion generator

Posted: Mon Aug 01, 2011 5:10 am UTC
by Technical Ben
Well, the idea was that it's a mind boggling over the top construction. But we still get sci-fi about dyson spheres right?

Re: Teller-Ulam fusion generator

Posted: Mon Aug 01, 2011 2:39 pm UTC
by idobox
And you can get 50% bonus awesome points by making it on the moon and beaming the power down to Earth.

Re: Teller-Ulam fusion generator

Posted: Mon Aug 01, 2011 5:28 pm UTC
by Technical Ben
True. But steam punk generators would look less awesome on the moon. Less trees for them to chop down.

Re: Teller-Ulam fusion generator

Posted: Mon Aug 01, 2011 8:30 pm UTC
by idobox
Steam punk generators on the moon powered by 3 or 4 stages nuclear bombs on the moon, not awesome? If this isn't awesome, what is?
Go go nuclear powered steam rocket made of brass and wood!

Re: Teller-Ulam fusion generator

Posted: Wed Aug 03, 2011 12:32 am UTC
by GreenTom
You might be interested in Project Gnome, a 1961 attempt to generate power by using atomic weapons to produce steam.

http://en.wikipedia.org/wiki/Project_Gnome
http://www.atomictourist.com/gnome.htm

A casual google on this doesn't reveal much about if it worked and why the concept was abandoned. I suppose common sense prevailed.

Re: Teller-Ulam fusion generator

Posted: Wed Aug 03, 2011 4:33 pm UTC
by Technical Ben
Or super conspiracies. ;)

Re: Teller-Ulam fusion generator

Posted: Thu Aug 04, 2011 12:59 am UTC
by Eebster the Great
idobox wrote:And you can get 50% bonus awesome points by making it on the moon and beaming the power down to Earth.

You'd need one hell of a rectenna :O.

Re: Teller-Ulam fusion generator

Posted: Thu Aug 04, 2011 2:14 am UTC
by thoughtfully
Practicality and cost effectiveness don't get any awesome points, silly. In fact, they seem to be anti-correlated.

It might be conceivable in a trans-galactic civilization in which funding such an absurdity counted as a rounding error, I suppose.

Re: Teller-Ulam fusion generator

Posted: Thu Aug 04, 2011 2:58 am UTC
by Eebster the Great
thoughtfully wrote:trans-galactic civilization

See, this is a bit difficult to imagine, given speed of light considerations.

Though I suppose it is possible given long enough life spans, etc.

Re: Teller-Ulam fusion generator

Posted: Thu Aug 04, 2011 3:00 am UTC
by thoughtfully
Sorry, the prefix I was looking for was pangalactic. That makes it all better, yeah? :)

Re: Teller-Ulam fusion generator

Posted: Thu Aug 04, 2011 4:25 pm UTC
by Technical Ben
Eebster the Great wrote:
thoughtfully wrote:trans-galactic civilization

See, this is a bit difficult to imagine, given speed of light considerations.

Though I suppose it is possible given long enough life spans, etc.

Guess so. We managed before the invention of the internet etc. A few hundred years ago communication could take years between civilisations. Even longer when it came to isolated communities. Only non time essential information would be useful though. No point asking what the weather will be like tomorrow.

Re: Teller-Ulam fusion generator

Posted: Thu Aug 04, 2011 8:02 pm UTC
by Eebster the Great
Technical Ben wrote:A few hundred years ago communication could take years between civilisations.

And this is one of the main reasons there was no single global civilization . . .

Re: Teller-Ulam fusion generator

Posted: Fri Aug 05, 2011 6:29 am UTC
by Technical Ben
Well, except the British Empire right?
I get your point though. If travel is 100+ years how do you know if your other planet is still in your control. Unless you use super skynet systems. :D

Re: Teller-Ulam fusion generator

Posted: Fri Aug 05, 2011 1:16 pm UTC
by Eebster the Great
Technical Ben wrote:Well, except the British Empire right?

The British Empire was not global "a few hundred years ago." India, for instance, wasn't absorbed until less than two hundred years ago. Still, this was before the invention of the telegram, which is impressive in itself.

Most examples of territorially huge empires before the nineteenth century did not last long, in part because it was simply almost impossible to govern such a large area.