If we allowed the first wall on the magrid to operate at 1-2000Kelven, how much radiative cooling might we expect?
Presuming power output limited to R^2 by cooling requirements, how would reactor mass scale? For many space apps volume isn't a killer.
Heat Transfer Limitations Re: Power Plants and Rockets
This confuses me.The actual pipe area will be approximately 2 pi times the intercept area because of the toroidal nature of the magnet. Assuming skinny toroids.
The intercept area is a projection of the pipe area.
I picture an imaginary spherical shell with the same radius as the device, upon which our little artificial sun shines a (roughly) uniform intensity of radiation at all points. I assume the 20% intercept means that rough mapping of the 3D pipes onto that shell works out to 20% of the interior surface area of that shell. Shouldn't 20% of the radiated power thus give us the total energy the pipes have to absorb and cool away, as in scareduck's post? Or am I missing something?
You are correct about total energy flow. What I am concerned about is total energy flow per unit area of heat transfer surface.TallDave wrote:This confuses me.The actual pipe area will be approximately 2 pi times the intercept area because of the toroidal nature of the magnet. Assuming skinny toroids.
The intercept area is a projection of the pipe area.
I picture an imaginary spherical shell with the same radius as the device, upon which our little artificial sun shines a (roughly) uniform intensity of radiation at all points. I assume the 20% intercept means that rough mapping of the 3D pipes onto that shell works out to 20% of the interior surface area of that shell. Shouldn't 20% of the radiated power thus give us the total energy the pipes have to absorb and cool away, as in scareduck's post? Or am I missing something?
Engineering is the art of making what you want from what you can get at a profit.
It's an approximation to the integral over the source to surface angle. Since it's a pipe, the surface is curved, so the sine (or cosine depending on your reference) of the angle reduces the amount of energy from the source. But if you don't assume the source is a point, you have to integrate all possible source points in the volume relative to the line of sight to the point on the pipe, and take _that_ angle into account for the actual cross section equivelent.
That's too big a pain. A little guessing and some experiments should give an order of magnitude estimate which will prevent catastrophy for a bit experiment. Then use real data and empirical plots to figure out how to build a reactor that works.
It's fun to do the integrals though, I'm sure a few PhD students will earn degrees working it out!
That's too big a pain. A little guessing and some experiments should give an order of magnitude estimate which will prevent catastrophy for a bit experiment. Then use real data and empirical plots to figure out how to build a reactor that works.
It's fun to do the integrals though, I'm sure a few PhD students will earn degrees working it out!
Intercept area is not a solution for heat flux per unit area of pipe. If you use intercept area you will get a heat flux that is higher (in most places) than the actual. Thus you will be over designed. Thus less power available than with a better design.TallDave wrote:Right, that part I get. But why can't you ignore the back side of pipe, since the radiation comes from the center, and just use the intercept area?
As Dr. Mike points out - you can solve the equations for the problem and get an answer. Or you can do measurement - for one thing you will get some deflection of alphas. Magnetic and electrostatic. This will help reduce the heat loads. Hard to predict in advance.
Engineering is the art of making what you want from what you can get at a profit.
How come nobody caught my mistake? Hydrogen is only three and a half times better per unit mass than water at holding heat. Plus the density is so low that pumping enough mass through is going to be tough, particularly since the compressibility is non-negligible.
Maybe we're better off just using high-pressure water after all, especially since it probably has to be a closed cycle anyway. I'll try to spec a system some time, if I can squeeze it in; it looks like fun...
I wonder if any of the Nusselt number correlations in my textbook go high enough...
Maybe we're better off just using high-pressure water after all, especially since it probably has to be a closed cycle anyway. I'll try to spec a system some time, if I can squeeze it in; it looks like fun...
I wonder if any of the Nusselt number correlations in my textbook go high enough...
For a rough estimate: 1 MW/ sq m pipe outside dia = 8" inside dia 6"93143 wrote:How come nobody caught my mistake? Hydrogen is only three and a half times better per unit mass than water at holding heat. Plus the density is so low that pumping enough mass through is going to be tough, particularly since the compressibility is non-negligible.
Maybe we're better off just using high-pressure water after all, especially since it probably has to be a closed cycle anyway. I'll try to spec a system some time, if I can squeeze it in; it looks like fun...
I wonder if any of the Nusselt number correlations in my textbook go high enough...
cyclonic turbulent flow - ie. the whole flow helps cool the "front" of the pipe.
20 MW energy dissipation
any delta Ts and operating temperatures you like.
Cu tubing preferred. Or a low induced radiation Stainless Steel.
Low pressure operation preferred - 300 psi minimum pressure - pumping pressure 100 psi.
That is about the range I would expect. I'm sure you would have a better idea.
Engineering is the art of making what you want from what you can get at a profit.
Just thought I'd mention, increasing the B-field will both lead to higher power densities and shield the alphas from the coils. If you built a smaller device at ultra high B-fields you could then put something at a highly positive potential outside the earthed cage and use it to slow down alphas coming out of the cusps. (assuming you could block all electrons from that region, if you couldn't it would be a disaster)
That is in fact what will be done to capture alpha energy without a thermal cycle.jmc wrote:Just thought I'd mention, increasing the B-field will both lead to higher power densities and shield the alphas from the coils. If you built a smaller device at ultra high B-fields you could then put something at a highly positive potential outside the earthed cage and use it to slow down alphas coming out of the cusps. (assuming you could block all electrons from that region, if you couldn't it would be a disaster)
Engineering is the art of making what you want from what you can get at a profit.
Rosenbluth and Hinton in Plasma Physics & Controlled Fusion v. 36 p. 1225 ("Generic issues for direct conversion of fusion energy for alternative fuels") claimed you couldn't get there from here with an electrostatic confinement scheme. Their principle argument was that power density would be too low because of Coulomb scattering, i.e. Rider's principle argument. I've been trying to find it and have been so far unsuccessful, but IIRC the p-11B reaction has one problem I haven't seen anyone address WRT direct conversion, and that is there are three alphas created but each with its own energy. That would have implications for direct conversion as well as exhaust scavenging.
I think it was covered at NASA Spaceflight a while back (maybe here - or on the IEC Fusion group - I'm not sure).scareduck wrote:Rosenbluth and Hinton in Plasma Physics & Controlled Fusion v. 36 p. 1225 ("Generic issues for direct conversion of fusion energy for alternative fuels") claimed you couldn't get there from here with an electrostatic confinement scheme. Their principle argument was that power density would be too low because of Coulomb scattering, i.e. Rider's principle argument. I've been trying to find it and have been so far unsuccessful, but IIRC the p-11B reaction has one problem I haven't seen anyone address WRT direct conversion, and that is there are three alphas created but each with its own energy. That would have implications for direct conversion as well as exhaust scavenging.
The deal is that there will be losses. Tom Ligon in his analog article suggested collection at the energy of the two low energy (2.46 MeV - IIRC) alphas and just thermalize the energy difference from the high energy alpha.
There is so much we have to learn.
Engineering is the art of making what you want from what you can get at a profit.
Let me add that losses would average 50KeV/alpha if there was a 100 KeV spread in exiting alpha energies - not unlikely.
The trouble is of course that we know nothing about the particle energy distributions from an operating reactor.
The trouble is of course that we know nothing about the particle energy distributions from an operating reactor.
Engineering is the art of making what you want from what you can get at a profit.