OK, sure. For 'tiny' I was thinking centimetres rather than nanometres! There again, it's just down to the magnetic field you use to cause the ions to orbit.
Let's say you could generate 20T in a tiny volume. There may be some experimental setup for which the maximum w (omega) can be achieved. Recall Bq=mw so for a deuteron in a 20T field, w would be up to 1G.rad/s. Recall the lower range of fusion velocities is 1Mm/s, so that would be a radius of 1mm.
How tiny do you want to get?...How much magnetic field have you got?...
Magneto-Inertial Fusion
What's to prevent us from building small synchrotrons?chrismb wrote:[I'll say it just once more!]] Any tiny cyclotron should be able to generate energies high enough for fusion. A synchrotron would put too much energy onto an ion.
Remember - 10Mm/s is the fastest velocities you will need for fusion, and cycltrons are often tuned to run around the industrial 13MHz frequencies - viz the total orbit needs be only <1m (<15cm radius). Any faster or bigger and you'll be knocking off nuclear chunks and no longer fusing.
I am also wondering what kind of (dis)advantages a cyclotron accelerator would have instead of a synchrotron one. Or just a two-some of linear accelerators for that matter, but in that case, you're almost back in Farnsworth territory....
Because we can.
A synchrotron is needed because when particles go relativistic they get out of synch with the electric fields. Ions of fusible energies aren't relativistic.Stoney3K wrote:What's to prevent us from building small synchrotrons?
I am also wondering what kind of (dis)advantages a cyclotron accelerator would have instead of a synchrotron one.
The purpose of a synchrotron is to get particles relativistic and then to smash atoms up with those particles. Too much energy - we want to fuse, not to smash.
Sure, if you want to run Rubbia's energy amplifier idea of a proton gun initiating fission in thorium, then that's a different story, but it ain't fusion.
SYNCHROTRON = MeV TO TeV = WILL SMASH ATOMS UP = CAN'T DO FUSION
CYCLOTRON = keV TO 1MeV = WILL TRANSMUTE ATOMS = CAN DO FUSION
CYCLOTRON OVER MeV = TOO MUCH ENERGY = WILL CAUSE OPPENHEIMER-PHILLIPS STRIPPING = NOT FUSION
gottit??
-
alexjrgreen
- Posts: 815
- Joined: Thu Nov 13, 2008 4:03 pm
- Location: UK
Hmm, that is an interesting idea Chris.
I like it.
I wonder what the always critical Art Carlson has to say about it. He is very good at crunching the numbers. I would like to see what his opinion is on this. I only get the essences of everything, but dont have the time and understanding to get the details. I still like reading about these ideas though.
I like it.
I wonder what the always critical Art Carlson has to say about it. He is very good at crunching the numbers. I would like to see what his opinion is on this. I only get the essences of everything, but dont have the time and understanding to get the details. I still like reading about these ideas though.
-
Art Carlson
- Posts: 794
- Joined: Tue Jun 24, 2008 7:56 am
- Location: Munich, Germany
(Huh? Did somebody mention my name? Sorry. I was napping.)Skipjack wrote:Hmm, that is an interesting idea Chris.
I like it.
I wonder what the always critical Art Carlson has to say about it. He is very good at crunching the numbers. I would like to see what his opinion is on this. I only get the essences of everything, but dont have the time and understanding to get the details. I still like reading about these ideas though.
What 's the point? Just to say you got some atoms to fuse? Sure, you can do that with a cyclotron. I think it would be easier with a linear, DC machine, say a Van de Graaff generator. Anything that will get you a few tens of kV. But that's a question for MSimon, not me. Will you discover anything new and relevant to fusion as an energy source? No.
Absolutely! You've nailed the issue on the head there. A cyclotron is a very lossy machine, and there are many ways that the input energy is lost by ions nailing the chamber wall.kcdodd wrote:What about all the thermalized stuff that didn't fuse? It will just hit the chamber and be lost.
Even if you had a machine without the particular losses of the cycltron [it also ionises background at a non-central point, then those new 'non-beam' ions execute their orbits off-centre] then you still have to understand the losses you get at any given point of 'impact': Ions don't fuse - I say that as a statistical comment, the number of actual fusion events following collisions is almost statistically insignificant. Out of a trillion collisions, maybe you'll get a single fusion event.
As Art says, there is nothing to work out with a cyclotron. You'd just be calculation cycltron inefficiencies, not much to do with fusion. However, at the right energy levels and only for only certain fuels, there are small windows of opportunity for pure beam systems (viz not cyclotrons, per se) where the energy loss from a trillion non-fusing collision is still less than the fusion energy out of that single fusion event, so if a mechanism can be constructed where the energy loss in a non-fusing collision is fed back into the ion, with 100% efficiency, then you could theoretically defeat that issue. (This is as per Todd Rider's thesis appendix E2.2, so even he has thought over this conundrum!)
The Q is poor, the best you can hope for is around 10. But I think this is actually the same theory for all IEC/beam-target type systems anyway. Only thermal systems can 'ignite' with effectively unrestricted Q.
They key to this approach, then, is to find a viable solution to satisfy Rider's E2.2 configuration.
Trying to following this but too much bush is being beaten around I think (at least for me to understand).chrismb wrote:They key to this approach, then, is to find a viable solution to satisfy Rider's E2.2 configuration.
Are you saying that you may have come up with a configuration of a cyclotron/Penning trap/magnetron thingy that accomplishes this?
Any less 'bush' and I don't think it would make coherent sense. I'm afraid that if 'advances' aren't un-obvious and are easy to comprehend, then it probably isn't that much of an advance. We've done the easy stuff, we've done the complicated stuff, we've done the non-intuitive stuff in science. Now we need to go beyond-non-intuitive and if it's easy to understand, it probably isn't much progress. I don't think the argument that you should be able to explain anything in a sentence applies any more, if we want higher levels of sophistication that what we've got then its gonna be more tough to explain and to understand.seedload wrote:Trying to following this but too much bush is being beaten around I think (at least for me to understand).chrismb wrote:They key to this approach, then, is to find a viable solution to satisfy Rider's E2.2 configuration.
Are you saying that you may have come up with a configuration of a cyclotron/Penning trap/magnetron thingy that accomplishes this?
Anyhow, that's not to say I've done a good job at describing it, though I'm not sure where I could have improved it much.
In terms of what I am saying; I wasn't trying to say I had come up with such a design, though the inference, that I have, is correct. What I am saying is the underlying basis for why you might want to generate a beam of particles at fusible-energies and that you need to have some way to put the energy back in to the ions after scattering, else you've got no hope. That's why I pay keen attention to the fabled annealing of Polywell; if that doesn't happen, then the rest can work a treat but will be ultimately pointless.
... heres a patent application i found, seems sort of plausibly similar idea perhaps... http://www.osti.gov/bridge/servlets/pur ... 757040.pdf ?
uses a two phase target coating on inside of cyclotron drum + tuned RF driver. no sign of anyone actually building one (beer can size, allegedly).
i also wonder wheter the semicondustor industry might yield promising techniques (ion focussing)?
uses a two phase target coating on inside of cyclotron drum + tuned RF driver. no sign of anyone actually building one (beer can size, allegedly).
i also wonder wheter the semicondustor industry might yield promising techniques (ion focussing)?