Atomic Rockets have a new addition to the stable of atomic rockets

[DDE]

http://arc.aiaa.org/doi/10.2514/6.2016-4685

On the Use of a Pulsed Nuclear Thermal Rocket for Interplanetary Travel

Basically, they propose using brief bursts of extreme output - the kind that gets 100 Wt reactors to megawatt levels - to boost Isp's, potentially to ion thruster levels, approximately 13800 sec, heating the exhaust beyond core meltdown temperature through neutron flux heat transfer.

Edited February 3, 2017 by DDE

[p1t1o]

Dang, thats pretty interesting. The part about the 2nd law of thermodynamics (thats the entropy one) is particularly interesting, as it implies that the reactor itself will not have to actually reach ridiculous "nuclear light-bulb"-like temperatures, but the propellant can.

I wouldn't want to be the engineer and run the reactor that does this but you dont see a new method of propulsion every day!

 

[Scotius]

So, less extreme nephew of Orion Drive?

[DDE]

3 hours ago, p1t1o said:

The part about the 2nd law of thermodynamics (thats the entropy one) is particularly interesting, as it implies that the reactor itself will not have to actually reach ridiculous "nuclear light-bulb"-like temperatures, but the propellant can.

Not just that, the propellant can be hotter than the reactor itself, which is normally impossible.

1 hour ago, Scotius said:

So, less extreme nephew of Orion Drive?

So unextreme they let universities have them.

 

[Scotius]

Whoa that was bright! And that Cherenkov afterglow - so metal

[DBowman]

I cannot read the article, or find anything that describes 'how' this works. Is it something like 'the fuel gets heated faster than the core by the neutron flux' + 'shut the engine down before it gets too hot'+'cool for the next pulse'? Sounds like these things would need cooling apart from the fuel - just like the KSP nukes.

I did find that they think they can pulse a TRIGA (don't google without 'pulse reactor' qualifier) reactor 20 - 50 times a second. A non mechanical control scheme (i.e. not rods being spun 3000 rpm) involves lasers Helium 3 and electromagnets - so very scifi. If it's pulsed it will be lower thrust than regular nukes, I guess you could use it non pulsed for 'high' thrust and pulsed for high isp.

[DDE]

4 hours ago, DBowman said:

Sounds like these things would need cooling apart from the fuel - just like the KSP nukes.

Correct, they want a lithium cooling loop.

4 hours ago, DBowman said:

non mechanical control scheme (i.e. not rods being spun 3000 rpm) involves lasers Helium 3 and electromagnets

So, it's just like the final version of my old friend, the RD-600.

 

[radonek]

What about effects of neutron flux on on structure nad crew safety?  If I understand it correctly, this thing is gonna shine a lot more then regular NTR, and those already have quite a thing with shielding.

[Gkirmathal]

1 hour ago, radonek said:

What about effects of neutron flux on on structure nad crew safety?  If I understand it correctly, this thing is gonna shine a lot more then regular NTR, and those already have quite a thing with shielding.

I think a normal designed "shadow shield" would suffice, maybe it would need a slight adjustment in material composition/desntity, but far from an expert. I could be wrong.

[p1t1o]

2 hours ago, radonek said:

What about effects of neutron flux on on structure nad crew safety?  If I understand it correctly, this thing is gonna shine a lot more then regular NTR, and those already have quite a thing with shielding.

Thats the sort of thing they can deal with if they ever get it to produce thrust, if its performance is high enough, then any amount of shielding will be worth it. IF its performance is high enough.

Seems like a fairly new concept, things like "crew safety" can come once we know if it can push a crew.

Besides, its already known that the neutron flux is going to have an effect on the coolant, there are other problems to overcome before we start designing starships.

[magnemoe]

21 hours ago, Scotius said:

So, less extreme nephew of Orion Drive?

Not exactly, you have multiple pulsed fusion rocket ideas who works that way you charge up and then generate an burst of fusion. It uses more energy than it produces so you need to power it but the fusion give an far higher isp than using the energy running an vasmir or similar. 

This sounds more like using radioactivity to heat the reaction mass to higher temperatures than an nerva can operate under, this will also heat the reactor so they can not run it continuously. 
Benefit is higher isp, downside is lower trust and you need cooling for the reactor, later is nice if you want to use the reactor for power production too. 

[DBowman]

Boy this idea is wild.

For the pulse mode they arrange the reactor is 'super critical' which means it produces more neutrons than required to maintain the chain reactions - it's on the bomb side of 'critical' - but control 'rods' are in so all is well. Then they take out the rods, reactor goes super critical (nuclear excursion), on the way to a meltdown, and they jam the control rods back in to 'quench' the reaction. Maybe thousands of times a second ( not actually rods, but conceptually ) - produces 100 to 1000 times sustainable NTR power.

The heat source that is important for a normal NTR is considered 'waste' - there is too much (100x or 1000x) for hydrogen to remove in time for the next pulse. The normally inconsequential 6% neutron flux of a normal NTR has the convenient property of not being heat but will heat the propellant by colliding with it transferring 6 to 60 times NTR heat directly into the propellant - getting it (a lot) hotter than an NTR can run.

Controlling propellant mass flow controls how long it has to be heated, controlling pulses/sec sets how much heat there is to pick up. So they can make it do low isp and 'high' thrust or high isp and NTR level thrust.

Could one throw some other heavy high density more storable propellant into the hot H2 before the nozzle to reduce isp and increase thrust? 

Edited February 5, 2017 by DBowman
oops power of 10 error