QED drives vs. Orion drives?

If polywell fusion is developed, in what ways will the world change for better or worse? Discuss.

Moderators: tonybarry, MSimon

cgray45
Posts: 93
Joined: Thu Apr 22, 2010 10:15 pm
Contact:

QED drives vs. Orion drives?

Postby cgray45 » Tue Jun 01, 2010 9:51 pm

Here's a question-- in terms of thrust and range, how would Polywell powered QED drives do vs. the Orion style drives that are such a staple of science fiction?

WizWom
Posts: 371
Joined: Fri May 07, 2010 1:00 pm
Location: St Joseph, MO
Contact:

Postby WizWom » Wed Jun 02, 2010 2:49 am

http://www.askmar.com/Fusion_files/The% ... Engine.pdf

Is the paper properly about the suggested engines.

It suggests a thermal drive for raw thrust/high mass ratio. High ISP is accomplished by using ion drives such as in VASIMR.

He's honestly suggesting Isp of 200,000 to 1,000,000 for his vacuum drive, which uses a tokomak-style toroidal magnetic field to guide escaping plasmas out the rear directly from the core of the IEF.
Engine System mass of about 52 tons is calculated.

http://en.wikipedia.org/wiki/Project_Or ... propulsion)

This includes the calculation for Isp, and plugging numbers in gives an Isp of from 2500 to 500,000, making the ship larger to get the higher Isp.

Isp is, to some degree, range - it's actually what you need to calculate an energy budget. For intrastellar missions you can use it as range (interstellar will start seeing efficiency loss from Einsteinian effects).

IEF will be a much smaller project, which means much cheaper. I think the thrust to mass ratio is better for ARC drive, too.

Honestly, I trust Dyson's numbers for Orion more than Bussard's for the IEF core drives. I suspect the actual energy produced in Bussard's system would be much higher, given a modest scale up. That is, I think Bussard was purposefully conservative in his Isp numbers.
Wandering Kernel of Happiness

93143
Posts: 1130
Joined: Fri Oct 19, 2007 7:51 pm

Postby 93143 » Fri Jun 04, 2010 8:15 pm

T/W is much, much, much better for Orion. You have to dial the Isp way down to get past 1:1 on a non-airbreathing QED-ARC engine, to the point where you're much better off just using chemical.

Bussard's numbers for Polywell masses were almost certainly optimistic. Radiation shielding is still required because of X-rays and (probably) gammas, as well as very low but nonzero levels of neutron radiation. The machine in the first link (which is a paper published in 1993 and thus does not represent the latest understanding) used a 1.35 T magnetic field, which based on current predictions would make a pretty anemic reactor, especially with p-¹¹B; 10-20 T is more likely. Bigger, more powerful magnets, direct conversion and power handling equipment, vacuum vessel, support structure for more powerful magnets... a real multi-GW Polywell could easily weigh a couple hundred tonnes even without shielding, and shielding could weigh between 50 and 1300 tonnes depending on gamma emission rate (or bremsstrahlung spectrum if the gamma rate is low enough), reactor size, and whether you want full shielding or just shadow shielding.

Let's say the reactor+engine weighs about 300 tonnes. Getting 3 MN out of 6 GW, even at 100% efficiency, requires an Isp of 4000 N­·s/kg, or about the same as an RS-68. Doubling the thrust (so you can actually lift something besides the reactor) requires halving the Isp, bringing it down to 2000 N­·s/kg, or 204 s - worse than the V-2. At a best-case mass ratio of 2:1, this gives you a vehicle delta-V of 1386 m/s over a 100-second burn time - accelerating straight up (980 m/s of gravity losses, plus drag), I seriously doubt this vehicle would even go supersonic before it ran out of hydrogen...

Airbreathing and horizontal takeoff improve the picture dramatically, of course...

You can only crank a Polywell so far before magnet technology craps out on you. Several gigawatts is probably the limit for the near future. (Actually, for a given field-strength limit on the superconductors, smaller reactors end up being more powerful, because the field seen at the wiffleball is higher...)

Orion can produce high thrust and high Isp at the same time. The momentum-limited starship design had an average Isp of more than 700,000 s (less than half the theoretical limit) at a vehicle T/W ratio of 1:1.

Of course, Polywell doesn't produce fallout or EMP, and it doesn't require mass production of nuclear warheads... also, with full-coverage (ie: ****ing heavy) shielding, it can operate safely in close proximity to space stations and such...

cgray45
Posts: 93
Joined: Thu Apr 22, 2010 10:15 pm
Contact:

Postby cgray45 » Fri Jun 04, 2010 10:20 pm

The shielding quesiton has been one that fills me with the most confusion-- I've read in some place that the clver use of magnetic fields might essentially eliminate neutron/X-ray radiation, at least from the P+B11 reaction. In other places, I've read that this would be impossible.

93143
Posts: 1130
Joined: Fri Oct 19, 2007 7:51 pm

Postby 93143 » Fri Jun 04, 2010 11:02 pm

Neutron production in a Polywell (aneutronic, p-¹¹B cycle) is suppressed by the non-Maxwellian velocity distribution and by the poor ash confinement. It's supposed to be ~1e-8 of what you'd get in a neutronic reactor. This requires perhaps 6" of water, a thin layer of boron-10, and 1.5" of lead.

Bremsstrahlung (X-ray) production in a Polywell is suppressed by careful balance of the non-Maxwellian, spatially inhomogeneous velocity distributions of the electrons and ions, which are at different energies from one another in most regions of the plasma (roughly speaking, you get cold ions/hot electrons at the edge, and hot ions/cold electrons at the centre, with different densities of course). Bremsstrahlung is supposed to be about 5% of net power in a reactor, which is still a hell of a lot of X-radiation and could require 2-5" of lead (possibly less) depending on spectrum.

Bremsstrahlung production in a DPF machine is suppressed by the enormous magnetic fields encountered in the pinch. It's supposed to be higher than in a Polywell (enough so that photovoltaic conversion is an important part of the power balance) but lower than predicted for a thermal machine without the pinch effect.

A Polywell doesn't produce magnetic fields anywhere near strong enough to attenuate bremsstrahlung (in fact, over most of the plasma the magnetic field is essentially null, due to the wiffleball effect). It depends on the non-Maxwellian two-temperature trick, which requires controlling the virtual anode height and probably other stuff and, well, it's complicated and no one outside EMC2 really knows much about it...

Gamma rays (4, 12, and 16 MeV) are supposed to result from 1 in 10,000 p-¹¹B fusions. This is a substantial amount of gamma radiation and requires something like a foot of lead to attenuate (at which point you can probably dispense with the water and boron-10; the neutrons from a p-¹¹B core shouldn't be able to penetrate a foot-thick wall of lead...). On the other hand, it is possible (though dubious) that the non-Maxwellian distribution could substantially suppress the gamma branch, resulting in lower shielding requirements...

cgray45
Posts: 93
Joined: Thu Apr 22, 2010 10:15 pm
Contact:

Postby cgray45 » Sun Jun 06, 2010 9:54 pm

A Polywell doesn't produce magnetic fields anywhere near strong enough to attenuate bremsstrahlung (in fact, over most of the plasma the magnetic field is essentially null, due to the wiffleball effect). It depends on the non-Maxwellian two-temperature trick, which requires controlling the virtual anode height and probably other stuff and, well, it's complicated and no one outside EMC2 really knows much about it...

Gamma rays (4, 12, and 16 MeV) are supposed to result from 1 in 10,000 p-¹¹B fusions. This is a substantial amount of gamma radiation and requires something like a foot of lead to attenuate (at which point you can probably dispense with the water and boron-10; the neutrons from a p-¹¹B core shouldn't be able to penetrate a foot-thick wall of lead...). On the other hand, it is possible (though dubious) that the non-Maxwellian distribution could substantially suppress the gamma branch, resulting in lower shielding requirements...


Now bear with me, because I'm a historian specializing in the British slave trade, and my general response to numbers like this is to raise my hand, make a series of monkey shrieks and retreat into a tree . :D

But it sounds like this is a case of "it *may* be possible to reduce gamma and neutrons, BUT nobody is likely to know for quite some time, because these are the sorts of questions that can only be answered once we have a working polywell (energy producing), and enough time to really tinker with said working polywell, so the answer to this is likely to be decades after we get the general "power producing" polywell.

MSimon
Posts: 14330
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Postby MSimon » Sun Jun 06, 2010 10:26 pm

he answer to this is likely to be decades after we get the general "power producing" polywell.


More likely weeks. Because you can take measurements. The theory might take decades to refine. But I doubt it.
Engineering is the art of making what you want from what you can get at a profit.


Return to “Implications”

Who is online

Users browsing this forum: No registered users and 4 guests