polywell reactor for booster rocket applications?

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

I can't imagine how a system like that could maintain a hard vacuum against tens or hundreds of atmospheres of pressure. Especially if you had to momentarily shut down the reactor for anything... Remember that these concentrated alpha streams cannot be allowed to touch any surface, because they're at a kinetic temperature of 20 billion degrees and contact WILL result in local damage regardless of cooling. This means that neutrals from the working fluid could easily slip around the core alpha flow and enter the chamber. Furthermore, in this scheme there are streams exiting at points all around the reactor; one per cusp. That's a lot of potentially heavy hardware, and a lot of engines that all have to work continuously if the reactor is to not choke...

It just seems ugly. Am I missing something here? It seems to me that it would be far easier and lighter to just use direct conversion (you've got to solve the neutral pumping problem anyway) and pipe the resulting high voltage current into the engines. That way you have no problems with vacuum seals, focused alpha beam impingement, multiplication of small engine modules (or worse, an attempt to pipe either fast alphas - that aren't all at the same speed! - or hot propellant from all around the reactor to a single engine/nozzle)...

On the other hand, if it does turn out that your way works better for some combination of reasons, so be it...

Regarding low-voltage electron guns, it's not feasible. You'd have to step down 6 GW from ~1.5 MV to whatever low voltage you're talking about, and you can't do that inside anything resembling an aerospace mass budget. For lower power requirements, like pumping the LH2 or running the reactor, you could use a heat engine powered by a mini-REB, but most of the power has to be used as is - HVDC - or else you're wasting a LOT of mass.

[An elaboration on my previous post: I forgot to mention that the required engine temperature is too high for any known substance, by a large margin. Since the nuclear lightbulb principle cannot be used here, this disallows wall-separated direct heating and leads to the requirement of some form of energy injection.]

I think there's a reason Dr. Bussard chose REB heating for high-pressure high-thrust modes, and direct ejection of alphas with small supplementary mass flows ("Diluted Fusion Product") for the long-range ultra-high-Isp deep space mode. As MSimon has noted, the things Dr. B said typically turn out to have had a fair amount of thought put into them. I've already tried and failed (twice) to invent a better atmospheric engine system than ARC-QED, and I am therefore of the opinion that this case is another example of Bussard knowing what he was talking about.

Also realize that the fusion output -> propellant heating efficiency of a REB engine is pretty much 100%, because it's "All Regeneratively Cooled". This means that any energy produced by the reactor that doesn't wind up in the electron beam is already in the propellant by the time the beam blasts it - including any refrigeration power necessary to get the propellant flow rate down to an optimum level...

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

Post by D Tibbets »

93143 wrote:I can't imagine how a system like that could maintain a hard vacuum against tens or hundreds of atmospheres of pressure. Especially if you had to momentarily shut down the reactor for anything......
Upon reflection and reading your input, I've concluded that my arguments are flemsy (that is being generous). I got hung up on how the fission nuclear rocket was supossed to work with a fluid flowing through the reactor core to cool it while at the same time heating the fluid. In this case there are no vacuum concerns. Also, the term 'relativistic electron beams' made me think of large and massive acellerators. I actually have no idea what machinery would be needed to produce relativistic electrons in sufficient quantity to heat enough fluid to produce the nessisary thrust.

As far as the back pressure for an alpha particle exaust port, I was figuring the alphas would be directed and focused into a highy collimated beam that would pass through a small orifice ( perhaps only a few mm-cm(?) wide . The tolorances would be so tight that nothing could enter the orifice of the port due to the alphas high density/ high pressure flow. The alpha flow would then be allowed to expand and mix with a working fluid to produce large hot exaust flows. Problems of pulsating operation, non perfect shielding of the walls and other trivial concerns are mearly engeenering problems. :wink:

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

Rick Kwan
Posts: 34
Joined: Sat Feb 21, 2009 5:26 pm
Location: Silicon Valley, California
Contact:

Post by Rick Kwan »

ohiovr wrote:But doesn't the space shuttle main engines have a thrust to weight ratio (shuttle and full tank included) less than 1?
I know is this from a while back, but I didn't see any clarification... so here goes...
Scanning some Wikipedia pages, the Space Shuttle thrust/weight ratio is
6.78 million lbf / 4.5 million lb == 1.51 (or so)
It has to be greater than 1 to get off the ground. (Commercial aircraft are a different situation. I see 0.373 for the Concorde SST. Fighter aircraft is typically above 1.0, which must help when dogfighting means getting above your opponent.)

And some other miscellaneous points....
* The Space Shuttle main engine (SSME) specific impulse is about 450 sec in a vacuum (about 373 sec at sea level).
* I recall something in one of the QED papers about specific impulse of 1500 seconds on the low end. (My notes say: "Inertial-Electrostatic-Fusion Propulsion Spectrum: Air-Breathing to Interstellar Flight", Bussard and Jameson, AIAA Journal of Propulsion and Power, March-April 1995 vol.11 no.2 (pp. 365-372))
* Regarding the use of LH2/LOX as propellant, I don't think there is any advantage over heavier propellants if you are not doing chemical combustion and using an expansion bell to shape the exhaust. You could use something heavier (methane? water?) with a lower exhaust velocity and get the same momentum.

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

Rick Kwan wrote:You could use something heavier (methane? water?) with a lower exhaust velocity and get the same momentum.
But then you need more of it, mass-wise. This is the whole reason they use LOX/LH2 in the first place.

You could use something heavy at the same exhaust velocity, and not have to deal with huge LH2 tanks. But then the chamber temperature would have to be very high... Temperature rise from cooling the reactor would be higher too...

Personally I would prefer to use onboard LH2 for coolant, and air as supplementary propellant. There's even a bit of performance enhancement from combustion of the hydrogen...

It's interesting to note that the TRITON trimodal nuclear engine uses LOX injection as an afterburner in high-thrust mode... it also has an NEP mode for long-duration high-Isp flight...

Roger
Posts: 788
Joined: Fri Jul 06, 2007 2:03 am
Location: Metro NY

Post by Roger »

Rick Kwan wrote: * I recall something in one of the QED papers about specific impulse of 1500 seconds on the low end. (My notes say: "Inertial-Electrostatic-Fusion Propulsion Spectrum: Air-Breathing to Interstellar Flight", Bussard and Jameson, AIAA Journal of Propulsion and Power, March-April 1995 vol.11 no.2 (pp. 365-372))
Like this chart?

http://www.dailykos.com/story/2007/5/8/ ... /57/332512
I like the p-B11 resonance peak at 50 KV acceleration. In2 years we'll know.

Rick Kwan
Posts: 34
Joined: Sat Feb 21, 2009 5:26 pm
Location: Silicon Valley, California
Contact:

Post by Rick Kwan »

93143 wrote:
Rick Kwan wrote:You could use something heavier (methane? water?) with a lower exhaust velocity and get the same momentum.
But then you need more of it, mass-wise. This is the whole reason they use LOX/LH2 in the first place.

You could use something heavy at the same exhaust velocity, and not have to deal with huge LH2 tanks. But then the chamber temperature would have to be very high... Temperature rise from cooling the reactor would be higher too...
Assuming that we're still in the realm of expanding gasses rather than accelerated plasmas, then I think you're right. And in the case of a heavy lift booster, perhaps that is indeed the case.

However, if you forget the gas expansion effect for a moment, if you maintain the same exhaust velocity but use heavier (denser) fluid, you will get proportionally greater thrust. ( F = d(mv)/dt ) This is the same effect as pumping more (actually equal mass flow) of the lighter fluid, e.g., LH2.

I was assuming we had a pure accelerated plasma, but I suspect I was dreaming.

Rick Kwan
Posts: 34
Joined: Sat Feb 21, 2009 5:26 pm
Location: Silicon Valley, California
Contact:

Post by Rick Kwan »

Roger wrote:
Rick Kwan wrote: * I recall something in one of the QED papers about specific impulse of 1500 seconds on the low end. (My notes say: "Inertial-Electrostatic-Fusion Propulsion Spectrum: Air-Breathing to Interstellar Flight", Bussard and Jameson, AIAA Journal of Propulsion and Power, March-April 1995 vol.11 no.2 (pp. 365-372))
Like this chart?

http://www.dailykos.com/story/2007/5/8/ ... /57/332512
If we're talking the same chart (exhaust velocity = 11,800,000 m/s), then that is about 1 million seconds in specific impulse. (I like the comment about " a one way trip to Saturn in 76 days".) I think that is interplanetary flight mode, where T/W > 1.0 is not required.

Rick Kwan
Posts: 34
Joined: Sat Feb 21, 2009 5:26 pm
Location: Silicon Valley, California
Contact:

Post by Rick Kwan »

By the way, to me, 11,800,000 m/s is a really awesome exhaust velocity. (1/25 * light speed :wink: )

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

Rick Kwan wrote:...if you maintain the same exhaust velocity but use heavier (denser) fluid, you will get proportionally greater thrust.
93143 wrote:You could use something heavy at the same exhaust velocity, and not have to deal with huge LH2 tanks.
I was assuming we had a pure accelerated plasma, but I suspect I was dreaming.
Well, a Polywell-powered atmospheric ion drive (straight-through E-field acceleration of ionized air) is probably not a great idea - I've already tried to design one, and I ran into a few problems (heavy voltage stepdown equipment, huge currents, too much ionization from hypersonic plasma choking the engine). It never really got beyond BoE... I haven't investigated the concept as a rocket, but I suspect that if comparable thrust is required from the same amount of power, the problems will be similar...

MHD crossed-field has been investigated by others. It requires a conducting fluid, so the air would have to be extensively pre-ionized, possibly with electron beams and maybe ECR... I'm not sure how well the idea works with the voltage you get from a Polywell, although I'm pretty sure it works better than the ion drive did...

Aaaand it looks like someone has already patented my idea for an MHD turbine on the back end of a scramjet. Darn...

Bussard's CSR-B engine was supposed to be a travelling-wave/MHD/thingy drive, but it was intended for deep space; it's one step below DFP... I really should read up more on these advanced electric propulsion concepts...

If we use magnetic shielding on the engine the thermal problems shouldn't be too bad. Denser propellants shouldn't be ruled out a priori, although I suspect that with Isp this high, the structural considerations will be less important...
By the way, to me, 11,800,000 m/s is a really awesome exhaust velocity. (1/25 * light speed)
Yeah, if you assume pure helium-4 exhaust that's almost 3 MeV. Sounds familiar somehow...?

And yes, that's the lowest-thrust interplanetary mode (DFP, apparently without the D). Actually it's not quite the lowest thrust/highest Isp you can theoretically get from this thing - you could use hydrogen as propellant for more exhaust velocity (about 16,600,000 m/s)... but of course if you separate fuel and propellant like that your mass fraction suffers... or you could try to step the voltage UP... but really, what's the point? The thrust is almost too low as it is. Come on, Dr. Woodward! One newton per watt! No propellant required!

...yeah, it's late and I'm tired...

Rick Kwan
Posts: 34
Joined: Sat Feb 21, 2009 5:26 pm
Location: Silicon Valley, California
Contact:

Post by Rick Kwan »

93143 wrote:Aaaand it looks like someone has already patented my idea for an MHD turbine on the back end of a scramjet. Darn...
You mean this thing?
"Magnetogasdynamic Power Extraction and Flow Conditioning for a Gas Turbine"
http://gltrs.grc.nasa.gov/cgi-bin/GLTRS ... 12612.html

It's late for me too, but I couldn't resist.....

djolds1
Posts: 1296
Joined: Fri Jul 13, 2007 8:03 am

Post by djolds1 »

Roger wrote:
Rick Kwan wrote:* I recall something in one of the QED papers about specific impulse of 1500 seconds on the low end. (My notes say: "Inertial-Electrostatic-Fusion Propulsion Spectrum: Air-Breathing to Interstellar Flight", Bussard and Jameson, AIAA Journal of Propulsion and Power, March-April 1995 vol.11 no.2 (pp. 365-372))
Like this chart?

http://www.dailykos.com/story/2007/5/8/ ... /57/332512
I think Rick means one of these:

http://nextbigfuture.com/2007/11/fusion ... usion.html
Vae Victis

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

Rick Kwan wrote:You mean this thing?
"Magnetogasdynamic Power Extraction and Flow Conditioning for a Gas Turbine"
http://gltrs.grc.nasa.gov/cgi-bin/GLTRS ... 12612.html
No, it was something else. Actually it looks like I spoke too soon; it looked like the same idea from the Google summary, but on further investigation there's a lot of weird stuff going on with steam turbines that I don't think is quite what I had in mind...

I saw the idea in your link too - it's more of a way to run a normal turbojet at hypersonic speeds by using MHD to soften up the airflow before it hits the moving parts, and then add back the energy at the other end. My idea was to use an MHD turbine in a scramjet the way a mechanical turbine is used in a turbojet - to bleed energy from the exhaust so as to do other things with it. One advantage might be that the airflow would probably already be ionized by that point...
United States Patent 4418294 wrote:Alternatively, the superheated steam could be used directly for heating homes and businesses in surrounding areas.
That's the first time I've ever seen someone propose using a ramjet as part of a power station.

Rick Kwan
Posts: 34
Joined: Sat Feb 21, 2009 5:26 pm
Location: Silicon Valley, California
Contact:

Post by Rick Kwan »

djolds1 wrote:I think Rick means one of these:

http://nextbigfuture.com/2007/11/fusion ... usion.html
I almost forgot about Tom's ISDC presentation. I'm sure he interprets the data better than I do. I think the first slide on that page sums up the booster scenario:
* specific impulse: 1538-3062 seconds

The curious part are the thrust and wet mass (uh, weight).
* Thrust: 208.6-83.2 T
* Wet: 250 T, Dry: 155 T
This yields T/W = 0.822, meaning that it can't do a vertical lift-off. (Units "T" stand for... English ton? Metric ton? Naw, the latter one is mass.)

However, for what's claimed on the next few lines, this is a great deal.
* Payload 35 T
* $27/kg to LEO

I knew I got the idea for water from somewhere. In the slides, it is the second one, on LEO to Luna transport/lander.

Roger
Posts: 788
Joined: Fri Jul 06, 2007 2:03 am
Location: Metro NY

Post by Roger »

Water injection ISP charts etc all in my prior link.
I like the p-B11 resonance peak at 50 KV acceleration. In2 years we'll know.

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

Rick Kwan wrote:This yields T/W = 0.822, meaning that it can't do a vertical lift-off.
Yeah, it's a spaceplane. HTHL, like Skylon but with much better payload fraction.
I knew I got the idea for water from somewhere. In the slides, it is the second one, on LEO to Luna transport/lander.
I forgot about that. Apparently there's a reason I couldn't come up with a convincing argument against it... looks like the Mars cycler uses water too...

That lunar cycler will definitely require shielding and cooling of the exposed walls - it's ARC/QED, ie: thermal...

It's also really tall and skinny for a lunar lander - based on current discussions surrounding Altair, I suspect that would have to change in the final design...

It occurs to me that VASIMR is more or less a thermal engine, and it uses very high temperatures. The key seems to be magnetic shielding. This is good because the required temperature for 7800 s with water as propellant is extremely high...

Post Reply