New Nuclear Thermal Rocket testing planned - decided not to use bomb grade U

[DBowman]

http://aviationweek.com/space/nasa-keeping-nuclear-thermal-option-open-mars

It's an interesting read but if you are not inclined to register to get through the wall the summary is:

1. test with depleted uranium 'dummy' core and heat it with radio frequencies to test that the core will hold together ok,
2. subsequently test with live uranium in some closed system where they combust the H2 exhaust and check that the water isn't radio active (subtle core breakdown / erosion),
3. not use bomb grade uranium for the core ...

Bomb grade uranium? was that the original plan? I learned here that some nuke plants can use natural uranium (0.7% U235), most use 5%, bomb grade is 90%, and some modern smaller designs want higher grade. Can anyone shed any light on questions like:

1. Is the mass of fissile U is the key?
2. If so then using bomb grade is about 1/18 the mass of 5% U. Does this mean the U is only a minor part of the 3,000 engine mass? or that some safety considerations make the extra mass worthwhile?
3. I guess the idea with a bomb grade nuke is to build the fuel rods so they can never get critical mass? Cladding, alloying, and geometry?
4. if nuke weapon proliferation from space based reactors is not an issue then is there any low mass materials that the 90% enriched U can by combined with to make safe light fuel elements?

[magnemoe]

As discussed in the smallest reactor possible, how smaller reactor you use who more U235 do you need, power reactors are huge so they don't need strongly enchanted uranium. 
Reactors on ships and small research and isotope generating ones need far stronger enchantments 40-60% U235.
As weight is an issue on an thermal nuclear rocket you will go even higher up to weapon grade. 

Testing with depleted uranium makes sense as you will not generate radiation, makes it simple to pull the rocket apart and rebuilding it. 
You can also check for contamination this is important for an life test as you don't want to end up with lots of hot hydrogen with radioactive dust in it. 
Note that unlike RTG who contains highly radioactive material an reactor has low radiation until run. 

Before an manned mars mission you will need an in space test, best would be to use it on an deep space probe.

Personally I think the train has gone for the nuclear thermal at least for an manned mission, an reactor+ and high ISP engine like vasmir looks better and is easier to test. 

[Bill Phil]

Submarine reactors use high purity uranium, however I don't recall if it was weapons-grade (probably not). I think higher purity lets you get away with higher power to weight ratios... Not quite sure, though.

[hms_warrior]

It is quite simple. The Uranium is what creates the heat. Uranium is HEAVY. Heavier than lead (170% mass of lead).

The Enrichment % tells you how much of that Uranium is actualy "fuel" and how much is just useless junk. so a 5% Enriched Uranium is 95% dead weight you carry around....not good for rockets.

Summary: using anything other than weapon-grade material is stupid. REALY stupid. From a science-point-of-view of course.

Mix in politics and you may have no other choice but to go with stupid...

(Hey let's use carbon-composite instead if steal for the fueltanks so we can save 1ton but triple our costs...but let's carry an extra 5 tons weight in our engine cause...just cause.)

Edited December 7, 2016 by hms_warrior

[Jonfliesgoats]

Politics in aerospace engineering: 

Soviet transport planes were built with the ability to mount tail gun turrets well into the 80s due to an order by Stalin.  This significantly reduced the effective rudder surface area on airplane like the AN-12.  Just because something is dumb doesn't mean we won't do it.

[wumpus]

On 12/7/2016 at 3:13 AM, hms_warrior said:

It is quite simple. The Uranium is what creates the heat. Uranium is HEAVY. Heavier than lead (170% mass of lead).

The Enrichment % tells you how much of that Uranium is actualy "fuel" and how much is just useless junk. so a 5% Enriched Uranium is 95% dead weight you carry around....not good for rockets.

Summary: using anything other than weapon-grade material is stupid. REALY stupid. From a science-point-of-view of course.

Mix in politics and you may have no other choice but to go with stupid...

(Hey let's use carbon-composite instead if steal for the fueltanks so we can save 1ton but triple our costs...but let's carry an extra 5 tons weight in our engine cause...just cause.)

Presumably you would want "just under" weapons grade, simply to make sure that a launch failure didn't crush the bits together "just right" in a way that went critical.  The politics also means [in the US] following all the "top secret" procedures laid out by the DoD.  Expect the price of everything this touches (and the secret part has a tendency to grab everybody else's budget on the grounds of "I need it but can't tell you why, that's classified") to go up by an order of magnitude (higher salaries plus mind-boggling inefficient procedures to keep things secret).

Another US thing about such things is that it almost certainly has to fly out of Vandenburg (for security reasons), thus giving it a polar inclination.  Not my first choice for a Mars Transfer Injection.

Remember there is a long, long way between "weapons grade" and "[commercial] reactor grade".   You should be able to get decent efficiency with uranium that can't be directly used for a bomb [but is rather close].  This would at least mean that anybody who got there hands on it would find it useless without massive centrifuges (easy enough to track down) or a weapons program far enough along not to bother with this.  Politically, there would still be an enormous trust gap, but rolling it out on KSC would at least demonstrate that the DoD believed that relatively normal security would be sufficient.

[TheSaint]

Hi, ex-nuclear mechanical operator here. Couple of things:

Reactor design is a staggeringly complicated field. I spent six years running nuclear reactors in the Navy, and I couldn't even begin to scratch the surface of it. But it is not at all like putting gasoline in the tank of a car, or wood on a fire. The amount and/or concentration of uranium-235 in a reactor is almost secondary. It has much more to do with the configuration of that fuel in the core, especially in relationship to the neutron moderator (whether that is the cooling water, or graphite, or whatever). The concentration of U235 has much more to do with the length of useful core life before it needs to be replaced. But even that can be increased or decreased by design factors. Remember that when your core reaches the end of its useful life, the vast majority of the uranium-235 that it started with is still in the core. Only a very small percentage of it is fissioned before the concentration is reduced to the point that a chain reaction is impossible. (That's why reprocessing is such a big issue.)

So, if your 20% enriched reactor core produces enough heat and has enough core life to accomplish your mission, you probably aren't going to see a massive weight reduction by moving to a 95% enriched core. (Although you might see a much longer core life, which for a spacecraft probably isn't a big deal.) You will however see a much higher cost associated with a highly-enriched core, because highly-enriched uranium is much more expensive to produce.

5 hours ago, wumpus said:

Presumably you would want "just under" weapons grade, simply to make sure that a launch failure didn't crush the bits together "just right" in a way that went critical. 

Um, no. Just, no. Repeat after me: You cannot take a reactor fuel element and make a nuclear bomb out of it. It is not improbable. It is not a small but unlikely possibility. It is impossible. Nuclear weapons require large masses of solid, elemental uranium in very specific physical configurations. Nuclear fuel elements contain discrete elements of uranium dioxide that have been deliberately designed to prevent their being arranged into a critical mass. Now, if you had a highly-enriched fuel assembly you could conceivably take it and reprocess it to recover the U235 and use that to create a bomb. (This is why reactors that use highly-enriched uranium are a non-proliferation issue.) But you couldn't take a highly-enriched fuel assembly, pack a bunch of explosives around it, and expect a nuclear explosion to result. That would be like blowing up an auto parts store and expecting to find a Cadillac in the rubble.

[DBowman]

4 hours ago, TheSaint said:

So, if your 20% enriched reactor core produces enough heat and has enough core life to accomplish your mission, you probably aren't going to see a massive weight reduction by moving to a 95% enriched core. (Although you might see a much longer core life, which for a spacecraft probably isn't a big deal.) You will however see a much higher cost associated with a highly-enriched core, because highly-enriched uranium is much more expensive to produce.

I found this that says for NERVA-Alpha (1972 study) the core 'and hardware' ( whatever that means - not reflectors or shielding, so maybe just 'racks and cladding'? ) is 870 kg of 2250 kg engine total, about 38%. I couldn't find any info on the actual mass of fissile material it used or what level of enrichment it was.

Maybe for these new tests they are just going with the easiest to get and handle fuel as a proof of concept.

Do you think it's possible to choose the reactor geometry so the vast majority of neutron flux is planar? or even one dimensional (beam like)?

[RCgothic]

There are several reasons for going for a highly enriched fuel. The first is that U238 is heavy and you don't want to be bringing any if it's nut going to be doing you any good.

The second is that a 'fast' reactor can function without a moderator if the U235 concentration is high enough. Moderators are heavy, so you don't want to bring one if you don't have to.

I would also expect the fuel to be Uranium Dioxide, because Uranium metal undergoes a crystal phase change as it heats up that massively increases its volume. Ceramic UO2 is much more stable. UO2 has a density similar to lead.

[p1t1o]

12 hours ago, TheSaint said:

That would be like blowing up an auto parts store and expecting to find a Cadillac in the rubble.

XD XD XD

[kerbiloid]

On 07.12.2016 at 5:05 PM, Jonfliesgoats said:

80s due to an order by Stalin

80s = Stalin? Really?
 

On 07.12.2016 at 5:05 PM, Jonfliesgoats said:

This significantly reduced the effective rudder surface area on airplane like the AN-12

Significantly decreased? Or just shifted up? Also, this was a world common practice for ages.
 

Spoiler

 

[todofwar]

I'm willing to bet it might be easier, politically, to get a centrifuge into orbit to enrich the uranium in orbit and avoid launching weapons grade stuff. Not that this would be anywhere close to feasible, and it would actually be less safe most likely, but look at what politics did to the shuttle.

[magnemoe]

1 hour ago, todofwar said:

I'm willing to bet it might be easier, politically, to get a centrifuge into orbit to enrich the uranium in orbit and avoid launching weapons grade stuff. Not that this would be anywhere close to feasible, and it would actually be less safe most likely, but look at what politics did to the shuttle.

That would require an structure way larger than ISS. 

You would not need weapon grade, nor would you want, as other say you want uranium oxide. No issues with 60% U235, 
Main issue with nuclear thermal is that its an engine without an mission, only relevant if you send something else than small probes. The stuff in RTG is far more radioactive than U235
Now if you start the reactor you get high radioactive stuff, this was why they wanted to test fuel flow with external heating and U238, that they they can pull the engine apart and rebuild it easy. 


 

[wumpus]

15 hours ago, TheSaint said:

Hi, ex-nuclear mechanical operator here. Couple of things:

That would be like blowing up an auto parts store and expecting to find a Cadillac in the rubble.

That's good to know.  But you still have to keep the U235-U238 ratio low enough to be useless to non-nuclear hungry nation states.  Although this is likely meaningless: non-bomb grade stuff is likely useless to anyone but Iran and North Korea, and anything significantly better than commercial grade would be of interest to those two.  Since neither has much of a deep-water Navy in the Atlantic (nor any other possible candidate), I suspect it is safe to launch from KSC (although I'm sure there will plenty helping the Air Force push for Vandenburg, if only claiming security for pork/turf).

You might want to be a bit more careful to have a set of abort modes that lets you get to the wreckage first.

[p1t1o]

1 hour ago, wumpus said:

That's good to know.  But you still have to keep the U235-U238 ratio low enough to be useless to non-nuclear hungry nation states.  Although this is likely meaningless: non-bomb grade stuff is likely useless to anyone but Iran and North Korea, and anything significantly better than commercial grade would be of interest to those two.  Since neither has much of a deep-water Navy in the Atlantic (nor any other possible candidate), I suspect it is safe to launch from KSC (although I'm sure there will plenty helping the Air Force push for Vandenburg, if only claiming security for pork/turf).

You might want to be a bit more careful to have a set of abort modes that lets you get to the wreckage first.

Even if it is bomb-grade, it is likely to be incorporated into ceramic fuel elements, so you couldn't build an atomic bomb out of it, but you could build a "conventional" dirty bomb. 

Fun Fact: Modern RTGs are designed to survive reentry intact in the case of an accident.

[TheSaint]

10 hours ago, DBowman said:

Do you think it's possible to choose the reactor geometry so the vast majority of neutron flux is planar? or even one dimensional (beam like)?

Not really, no. Neutrons go wherever neutrons go. You can't really herd them like cattle. You could shield the reactor to prevent them from going where you don't want them to, but that's about it.

7 hours ago, RCgothic said:

There are several reasons for going for a highly enriched fuel. The first is that U238 is heavy and you don't want to be bringing any if it's nut going to be doing you any good.

True. But with everything else considered the fraction of your reactor weight that winds up being U238 is probably going to end up being very small. I think a lot of other factors will end up outweighing it, both figuratively and literally.

7 hours ago, RCgothic said:

The second is that a 'fast' reactor can function without a moderator if the U235 concentration is high enough. Moderators are heavy, so you don't want to bring one if you don't have to.

Interesting option. But even then, IIRC, fast reactors don't need highly-enriched fuel, can't they get by with like 20-30%?

2 hours ago, todofwar said:

I'm willing to bet it might be easier, politically, to get a centrifuge into orbit to enrich the uranium in orbit and avoid launching weapons grade stuff. Not that this would be anywhere close to feasible, and it would actually be less safe most likely, but look at what politics did to the shuttle.

Enriching the uranium with one centrifuge would take a bazillion years. Enrichment facilities generally have hundreds of centrifuges going at once. And that's just the centrifuges, there are lots of other processes involved as well, many of which require gravity. Your orbital enrichment facility would probably be the single most expensive thing ever built by mankind.

3 minutes ago, p1t1o said:

Fun Fact: Modern RTGs are designed to survive reentry intact in the case of an accident.

Fun Fact: One of them already has.

[wumpus]

1 minute ago, p1t1o said:

Even if it is bomb-grade, it is likely to be incorporated into ceramic fuel elements, so you couldn't build an atomic bomb out of it, but you could build a "conventional" dirty bomb. 

Fun Fact: Modern RTGs are designed to survive reentry intact in the case of an accident.

Actually, a better means would be to surround the fuel core with U238, in such a way to mix (as a solution) during impact.  It might make a better "dirty bomb", but there wouldn't be the issue of a chemist being able to separate the uranium from everything else and getting the purified deal.  Once you mix the U238 in, the chemist is at step 1.

I suspect this would get axed pretty quick, even a small bit of U238 would be extra mass you don't want, and the means to create the solution would be a danger after a successful flight (chuck it out the airlock with the rest of the trash and detonate?  I guess let it explode on its own during reentry, less space junk).

If a spacecraft fell near me, I'd be a lot more worried about the hydrazine and similar stuff than any plutonium (although I might be worried about the types that might be attracted by missing plutonium even though its the wrong stuff to build a bomb).

[Jonfliesgoats]

Very good question and observation!

It was actually a holdover from a Stalin-era edict that wasn't removed from design bureau requirements until sometime in the 80s.  Beurocracy being what it is, once a design requirement mad it out to the manufacturers, Soviet design bureaus were very, very slow to petition for changes to these requirements due to political considerations. 

As for rudder authority, it significantly reduced it from what could have been provided.  Obviously, the remaining surface has been moved up and the hinge line moved forward to increase its area.  Other devices can be used to increase rudder authority as well, like increased travel, servo tabs, etc.

However, the aerodynamic efficiency of the rudder would have been significantly better without having a turret, especially a manned one back there.  More rudder means more controllability with the loss of an engine, the ability to fly slower without compromising controllability with asymmetric thrust, etc.  For a tactical airlift plane, trading low speed rudder authority for a turret that still does not protect you from air to air or surface to air threats is silly.

Even if we say that available Soviet turboprop engine/propeller combinations didn't compromise the control authority of the plane, with existing empennage, the extra rudder authority allows for more engine options in future design iterations, greater single engine maneuverability and crosswind authority, etc.  There are also limitations in load distribution and other practical considerations when control authority is limited.

For their part some of the Russian decisions were really good.  Their airplanes, the AN-12, 26 and 32 in particular, are all very sustainable in remote, commercial operations.  This is a vestige of decisions to ensure the Soviet Air Force could operate from relatively austere airfields.

The human stories in aerospace engineering decisions are pretty cool.  Many WW2 British Bombers have fuselages that look like locomotives because they actually employed locomotive engineering, for example.

3 hours ago, kerbiloid said:

80s = Stalin? Really?
 

Significantly decreased? Or just shifted up? Also, this was a world common practice for ages.
 

  Reveal hidden contents

 

 

[p1t1o]

1 minute ago, Jonfliesgoats said:

Very good question and observation!

It was actually a holdover from a Stalin-era edict that wasn't removed from design bureau requirements until sometime in the 80s.  Beurocracy being what it is, once a design requirement mad it out to the manufacturers, Soviet design bureaus were very, very slow to petition for changes to these requirements due to political considerations. 

As for rudder authority, it significantly reduced it from what could have been provided.  Obviously, the remaining surface has been moved up and the hinge line moved forward to increase its area.  Other devices can be used to increase rudder authority as well, like increased travel, servo tabs, etc.

However, the aerodynamic efficiency of the rudder would have been significantly better without having a turret, especially a manned one back there.  More rudder means more controllability with the loss of an engine, the ability to fly slower without compromising controllability with asymmetric thrust, etc.  For a tactical airlift plane, trading low speed rudder authority for a turret that still does not protect you from air to air or surface to air threats is silly.

Even if we say that available Soviet turboprop engine/propeller combinations didn't compromise the control authority of the plane, with existing empennage, the extra rudder authority allows for more engine options in future design iterations, greater single engine maneuverability and crosswind authority, etc.  There are also limitations in load distribution and other practical considerations when control authority is limited.

For their part some of the Russian decisions were really good.  Their airplanes, the AN-12, 26 and 32 in particular, are all very sustainable in remote, commercial operations.  This is a vestige of decisions to ensure the Soviet Air Force could operate from relatively austere airfields.

The human stories in aerospace engineering decisions are pretty cool.  Many WW2 British Bombers have fuselages that look like locomotives because they actually employed locomotive engineering, for example.

To be fair, B-52s had tailguns until the early 90s (and the soviet cargo craft with them were designed for combat drops and things so potentially would be exposed to enemy aircraft).

Last combat usage of a tailgun was apparently in 1972, B-52 vs. MiG-21:

https://en.wikipedia.org/wiki/Tail_gunner

 

[Jonfliesgoats]

That's true.  American planes aren't removed from the effects we describe.  Americans seem obsessed with finding the most expensive and engineering intensive solution to simple problems, for example.  Perhaps it is more accurate to think that the West tries to simplify with multi-role aircraft and develops engineering nightmares with huge max hour to flight hour ratios.  Again, this is an artifact of political factors in the acquisition process.  The West then cancels successful programs to support these things.

Look at the CV-22 and F-35 for examples of doing less with more.  

It's easy to be a naysayer, however.  From experience, getting something to the field with constantly changing requirements and oversight is challenging.

[TheSaint]

13 minutes ago, wumpus said:

Actually, a better means would be to surround the fuel core with U238, in such a way to mix (as a solution) during impact.  It might make a better "dirty bomb", but there wouldn't be the issue of a chemist being able to separate the uranium from everything else and getting the purified deal.  Once you mix the U238 in, the chemist is at step 1.

Interesting idea, but probably not feasible. You would have to be assured that the explosion and impact would so pulverize the fuel assemblies and U238 poison as to irretrievably intermix them. I can't imagine that would be the case in any conventional rocket accident.

[wumpus]

23 minutes ago, TheSaint said:

Interesting idea, but probably not feasible. You would have to be assured that the explosion and impact would so pulverize the fuel assemblies and U238 poison as to irretrievably intermix them. I can't imagine that would be the case in any conventional rocket accident.

Yes.  I think the real answer is to send it up on an unmanned flight (but presumably a man-rated craft with a bunch of abort modes) and then arrange all the abort modes to land where you have a collection ship.  It might take a lot of ships, but arranging that shouldn't be such an issue.  And make it so it takes a bit more refining (and resources unavailable to anything less than a nation state) to make a bomb (I'm assuming this wouldn't kill the efficiency, especially if you can go as pure as you want as long as the total mass isn't large enough to build a bomb.  Politics being what they are, if you lose one reactors worth of fuel I doubt you'll have a chance to lose another).

A belt and suspenders solution to be sure, but nuclear proliferation is one of those things were you have multiple highly reliable and redundant protections.

[TheSaint]

10 minutes ago, wumpus said:

Yes.  I think the real answer is to send it up on an unmanned flight (but presumably a man-rated craft with a bunch of abort modes) and then arrange all the abort modes to land where you have a collection ship.  It might take a lot of ships, but arranging that shouldn't be such an issue.  And make it so it takes a bit more refining (and resources unavailable to anything less than a nation state) to make a bomb (I'm assuming this wouldn't kill the efficiency, especially if you can go as pure as you want as long as the total mass isn't large enough to build a bomb.  Politics being what they are, if you lose one reactors worth of fuel I doubt you'll have a chance to lose another).

A belt and suspenders solution to be sure, but nuclear proliferation is one of those things were you have multiple highly reliable and redundant protections.

I think you're overthinking things. We have (unfortunately) in the past lost actual nuclear warheads to accidents. There are, at this moment, at least half a dozen still sitting on the bottom of the ocean that I can think of off the top of my head. The reason they're still there is because the entire combined capability of the United States Navy couldn't find them, or because they sit at a depth that makes them essentially unrecoverable. I'm going to guess that any of the rogues gallery of possible nuclear actors (none of which have any significant blue water navy, never mind any deep submergence capability) wouldn't stand a snowball's chance in Hell of recovering them. So why would we imagine that they would be able to (or even stage an attempt to) recover a reactor core from a failed spacecraft launch? If I were Kim Jong Ahmadinejad III (aka Tinhorn) and I had even remotely that sort of capability, I would be combing the waters where warheads are already known to lie, not lurking around waiting for a rocket to fall out of the sky. 

A much better option would be to do everything you can to harden the reactor to withstand a failed launch, and then mobilize a recovery effort if an accident happens. I don't think that anyone will be racing us to the crash site.

[todofwar]

3 hours ago, TheSaint said:

Enriching the uranium with one centrifuge would take a bazillion years. Enrichment facilities generally have hundreds of centrifuges going at once. And that's just the centrifuges, there are lots of other processes involved as well, many of which require gravity. Your orbital enrichment facility would probably be the single most expensive thing ever built by mankind.

I wasn't making the case that is in any way a good idea, I was just saying that is more likely to happen than to actually load weapons grade uranium onto a rocket regardless of the science or engineering concerns. 

Edit: And to be clear, it is not likely to happen at all, just trying to illustrate how politically impossible it would be to launch something like that into space. 

Edited December 9, 2016 by todofwar

[Wallygator]

On 08/12/2016 at 11:10 PM, TheSaint said:

That would be like blowing up an auto parts store and expecting to find a Cadillac in the rubble.

That is the best metaphor I've seen in a while

Also, much appreciate your factual input in this thread.