|
Post by darthroach on Jan 16, 2017 20:49:46 GMT
I decided to make this thread in the science discussion section rather than in suggestions because I am not exactly an expert, and am more interested in seeing some discussion on this subject. And considering that lots of you guys here are a hell of a lot smarter than me, this is as good a place as any for it. I don't think I have to make any introductions to the idea since pretty much everyone playing this game has been to AtomicRockets at some point in their lives, and mr Chung has gone into a fair bit of detail about the specific engine design here: Open Cycle Gas Core NTRThe gist of it is that instead of the engine temperature (and therefore exhaust velocity) being limited by the melting point of the solid fuel nuclear reactor parts, the fission reaction takes place in a cloud of Uranium plasma. Around this hydrogen or other propellant (doped with radiation-scattering dust) is channeled, getting superheated by the ongoing nuclear reaction in the center. That way, the temperature is limited by our ability to cool the combustion chamber's walls, which is considerably higher. The reaction is confined by cleverly arranging the surrounding propellant flow to keep the central vortex in for as long as possible, and is started by any number of AND THEN A MIRACLE HAPPENS proposals. Some of the fission fuel inevitably goes out the back end, of course, and pointing the thing at a surface launchpad will slag the county. But thankfully CoaDE's setting does not require us to pay any attention to the environment. Now it seems some people from NASA or related have done a fair bit of theoretical consideration on this already. In short: Pros: - Isp in the 3000-7000 range (limited by the need to keep the combustion chamber walls solid)
- Thrust comparable to solid core NTRs, which is to say, a whole damn lot
- Cool
Cons:
- No one has actually tried to confine a fission reaction inside a vortex of gas moving at supersonic speeds
- Starting the engine could prove a rather difficult task
- Will ruin the day of anyone sitting in the way of the exhaust. Maybe not the best idea for a fleet of battleships in close formation.
The pros are such that they make normal NTRs look rather weak and obsolete in comparison. The cons are mostly the lack of real engineering work done to make this work. Going by the general philosophy of the game, there is more than enough information to implement this as a workable engine family - we know how it will perform if it works, the things we don't understand have more to do with actually making the engine work in the first place. So the question is whether you guys think something like this could ever actually be built and perform to spec.
|
|
|
Post by bigbombr on Jan 16, 2017 21:53:30 GMT
I decided to make this thread in the science discussion section rather than in suggestions because I am not exactly an expert, and am more interested in seeing some discussion on this subject. And considering that lots of you guys here are a hell of a lot smarter than me, this is as good a place as any for it. I don't think I have to make any introductions to the idea since pretty much everyone playing this game has been to AtomicRockets at some point in their lives, and mr Chung has gone into a fair bit of detail about the specific engine design here: Open Cycle Gas Core NTRThe gist of it is that instead of the engine temperature (and therefore exhaust velocity) being limited by the melting point of the solid fuel nuclear reactor parts, the fission reaction takes place in a cloud of Uranium plasma. Around this hydrogen or other propellant (doped with radiation-scattering dust) is channeled, getting superheated by the ongoing nuclear reaction in the center. That way, the temperature is limited by our ability to cool the combustion chamber's walls, which is considerably higher. The reaction is confined by cleverly arranging the surrounding propellant flow to keep the central vortex in for as long as possible, and is started by any number of AND THEN A MIRACLE HAPPENS proposals. Some of the fission fuel inevitably goes out the back end, of course, and pointing the thing at a surface launchpad will slag the county. But thankfully CoaDE's setting does not require us to pay any attention to the environment. Now it seems some people from NASA or related have done a fair bit of theoretical consideration on this already. In short: Pros: - Isp in the 3000-7000 range (limited by the need to keep the combustion chamber walls solid)
- Thrust comparable to solid core NTRs, which is to say, a whole damn lot
- Cool
Cons:
- No one has actually tried to confine a fission reaction inside a vortex of gas moving at supersonic speeds
- Starting the engine could prove a rather difficult task
- Will ruin the day of anyone sitting in the way of the exhaust. Maybe not the best idea for a fleet of battleships in close formation.
The pros are such that they make normal NTRs look rather weak and obsolete in comparison. The cons are mostly the lack of real engineering work done to make this work. Going by the general philosophy of the game, there is more than enough information to implement this as a workable engine family - we know how it will perform if it works, the things we don't understand have more to do with actually making the engine work in the first place. So the question is whether you guys think something like this could ever actually be built and perform to spec.
Starting the engine is hard, but doesn't seem to be an insurmountable engineering challenge. Stopping the engine without ejecting all your fissionable material seem a lot harder. Orion drives, nuclear-saltwater rockets and metalic hydrogen engines seem more feasible.
|
|
|
Post by theholyinquisition on Jan 16, 2017 22:20:13 GMT
metalic hydrogen engines seem more feasible. I'm sorry, what? I thought you could only get metallic hydrogen with gas-giant core pressures and temperatures.
|
|
|
Post by bigbombr on Jan 17, 2017 7:33:35 GMT
metalic hydrogen engines seem more feasible. I'm sorry, what? I thought you could only get metallic hydrogen with gas-giant core pressures and temperatures. Metallic hydrogen is theoretically meta-stable. We'll find out soon if that's true, as a research group claimed a few months back the successfully made metallic hydrogen.
|
|
|
Post by darthroach on Jan 17, 2017 7:41:47 GMT
Orion drives seem more feasible, true, though they require massive shock absorbers and are not exactly efficient. A gas core NTR offers roughly similar performance while also keeping a far more traditional ship architecture. NSWR would give us torchship-tier performance if Zubrin is right, but it also has some very, very shiny failure modes. So far no one has gotten around the problem of fuel storage, and there is no way I would take that thing into a battle. As for metallic hydrogen, there is the problem of producing metallic hydrogen and even then we're looking at only 1/2 to 1/3 of the Isp. Considering the fact that at any given time only a fraction of your fuel will be in the combustion chamber, dumping it is not exactly a be-all end-all problem. It does, however, make firing the engine in short bursts rather wasteful. The question is, does that really outweigh the advantage in sheer power? A ship with 5000s of Isp can afford to have a very reasonable mass ratio of 3 and still have around 50km/s of delta v. Hell, you could run the thing with a mass ratio of 2 and still have about 35 km/s! Carrying a bit more fuel for times when you might need to dump the combustion chamber seems like a bargain.
|
|
|
Post by Enderminion on Jan 17, 2017 12:06:38 GMT
I ean you could run your gaseous nuclear fuel through a cooling loop... its only 4000K. Edit: stick a thermocouple or two on that and you make power too
|
|
|
Post by darthroach on Jan 17, 2017 12:23:01 GMT
I ean you could run your gaseous nuclear fuel through a cooling loop... its only 4000K. Edit: stick a thermocouple or two on that and you make power too It's actually far higher than 4000K, so you might be thinking some of the weird liquid-or-vapor core designs. No coolant loops here I'm afraid. The entire point is that the fuel itself does not need to be cooled, since it's not in direct contact with any bits and pieces that need to remain solid to work. To get a perspective on the amount of heat that this thing puts out you just have to consider the fact that even though only about 7-10% of the radiation hits the combustion chamber walls (in theory, anyway), it requires a radiator cooling system. Well, when running at lower temperatures you could make do with regenerative cooling, but might as well go for broke. The upper limit on the Isp is defined by the amount of heat the shell can take before being compromised instantly. This paper offers a fairly good estimate of the heat properties of this engine: High Specific Impulse Gas-Core ReactorsAccording to this, even for a modest 2500s of Isp you'd need to exhaust THE PROPELLANT (hydrogen) at 8300K. For 5000 - at 22'000K. Which means the core has to be at least that hot, and in reality hotter.
|
|
|
Post by newageofpower on Jan 17, 2017 14:23:21 GMT
Orion drives seem more feasible, true, though they require massive shock absorbers and are not exactly efficient. A unique feature of Orion propulsion is that Orion efficiency scales with size of the propelled craft; Orion starships are viable even with just Teller Ullam, at a sufficiently large scale.
|
|
|
Post by darthroach on Jan 17, 2017 14:26:11 GMT
True enough, but not every ship can or needs to be gigantic. Put it whichever way you like, Orion would probably not fill the same niche as the average NTR-propelled ship. The only thing that seems more brute-force than Orion is skyrocketing mass ratios. Which is yet more reason to include both these propulsion systems into the simulation - so we can see what they are like up against each other Honestly, if either of these systems is workable (which they probably are), that's what the admirals of Sol system would pay an arm and a leg to get. The performance gain is just too much not to look into.
|
|
|
Post by Pttg on Jan 17, 2017 19:46:27 GMT
Might be interesting if GCNTRs can be scaled down to missile sizes. Then turning the engines off is a question for the enemy to deal with after the 30km/s missile says hi.
|
|
|
Post by darthroach on Jan 17, 2017 20:32:35 GMT
Might be interesting if GCNTRs can be scaled down to missile sizes. Then turning the engines off is a question for the enemy to deal with after the 30km/s missile says hi. As far as I can tell, no. In order to reach criticality, the core needs to be of either sufficient mass and sufficient density. If you wanna make it smaller, you will have to increase the working pressure - and there is only so much that real materials can take. Not to mention a smaller radius means far, far more heat getting dumped per unit of combustion chamber surface area. Then again I only have the vaguest idea of how nuclear reactions work so I might be wrong - but I believe it was mentioned in one of the papers cited on AtomicRockets. ETA: even if you could, 30km/s missiles don't seem like the best idea. At that kind of relative velocity, they're basically dumb projectiles as far as evasion is concerned. And evasion will be concerned because the exhaust will light up the sky like Vegas on New Years.
|
|
|
Post by zuthal on Jan 18, 2017 10:11:43 GMT
The main advantage of Orion is that it is the only remotely feasible engine with torchlike performance - high exhaust velocity AND high thrust, so you can use it to get a TWR>1 even for very large payloads with high double or triple digit km/s delta-V, allowing you to boost straight from Terra to Mars on a brachistochrone trajectory.
Of course, even when using pure fusion detonators (and thus having more or less absolutely no fallout) you would still want to keep the launch site away from population.
Though, Orion wouldn't be nearly as environmentally bad as usually thought, because it only uses very small, sub-kiloton devices.
|
|
|
Post by darthroach on Jan 18, 2017 13:55:14 GMT
The main advantage of Orion is that it is the only remotely feasible engine with torchlike performance - high exhaust velocity AND high thrust, so you can use it to get a TWR>1 even for very large payloads with high double or triple digit km/s delta-V, allowing you to boost straight from Terra to Mars on a brachistochrone trajectory. Of course, even when using pure fusion detonators (and thus having more or less absolutely no fallout) you would still want to keep the launch site away from population. Though, Orion wouldn't be nearly as environmentally bad as usually thought, because it only uses very small, sub-kiloton devices. Orion is a separate topic, though. Orion will also take a lot more work to implement in the game than the Gas Core rocket, since the latter shares many properties with the NTRs already present. The biggest difference is the need for a cooling loop and pressure shell. The specifics of the fluid flow needed to contain the core are not actually needed, what matters is the mass flow, core temperature, fuel leak rate (biggest blackbox function involved, seems) and propellant heat absorption. From the combustion chamber into the nozzle and out it's like any other rocket.
|
|
|
Post by zuthal on Jan 18, 2017 14:35:01 GMT
Definitely, yes - I think a closed cycled GCNTR, i.e. nuclear lightbulb, might actually be somewhat easier to implement in game - as you do not have to worry about fuel loss. And might be preferred for military use anyways, if possible, since I am not sure if you could maintain fuel containment in an open-cycle GCNTR under high acceleration
|
|
|
Post by darthroach on Jan 18, 2017 14:58:15 GMT
Definitely, yes - I think a closed cycled GCNTR, i.e. nuclear lightbulb, might actually be somewhat easier to implement in game - as you do not have to worry about fuel loss. And might be preferred for military use anyways, if possible, since I am not sure if you could maintain fuel containment in an open-cycle GCNTR under high acceleration 1. While the open cycle engine requires some tinkering with fluid flow to work, the closed circle one requires quartz "lightbulbs" that can actually withstand the intense heat of fissioning gas. Overall a much more complicated system. While it does not eject any fuel out the back, it runs into several far bigger problems. How do you cycle the 20 thousand K or hotter gas and replace the spent fuel with unspent? How exactly do you cool the lightbulbs? How do you even build them? How strong can you make the thing before the lightbulb walls absorb too much heat? The only real argument I see for this system is that it could be used inside the atmosphere. With the open cycle version at least the heat management is rather straightforward. 2. Define high acceleration. Something makes me think if we're using high-pressure expanding gas inside the combustion chamber of a rocket - at tens of thousands of degrees kelvin no less - a couple hundred miligees should really be fine. I mean, if you can calculate a way to confine the core under no acceleration, you can probably figure out a way to do it with acceleration too.
|
|