Nuclear Rockets [WIP]

[jimmyjack]

I read something were kommit said he was going to redo the parts, this makes me sad I actually love the nuclear engines as is.

[kiwiak]

I think you have a number of definitions a little wrong, which may make it difficult for you to understand how things work.

- Specific Impulse (Isp) is not entirely a design target, but also a design result of your engineering approach. The Isp an engine can develop in an ideal condition (i.e. vacuum with a proper nozzle) is directly related to the energy put into the propellant, as well as the propellant itself. A kerosene/oxygen reaction might do 310s to 320s, whereas a hydrogen/oxygen reaction might do 450s to 460s, and a nuclear thermal rocket might achieve 800s to 2000s or more... completely dependant on how hot it runs, because there is no chemical reaction, only hydrogen gas being superheated by cooling the reactor core. You can use other, heavier gases aside from hydrogen in a NTR, which results in less Isp (but more thrust, see below).

- The LV-N atomic rocket motor is a nod to the NERVA XE, the only nuclear thermal rocket engine in human history that has ever been assembled in full flight configuration. It happened to have an Isp of around 800 and a TWR around 2.75, which is what the LV-N replicates fairly exactly. Now, the NERVA XE was a first generation NTR used for proof of concept; theoretically there exist ways to make NTRs with much higher Isp, simply by using a reactor core that runs much, much hotter. The so-called gas-core nuclear lightbulb could theoretically achieve over 2000s Isp with hydrogen. Of course, nobody has actually tried building one because nobody wants to be caught dead tinkering with nuclear fission tech in today's political climate, so we don't know if the concept would actually work.

- Isp is measured in seconds, which sounds odd at first for a figure that's effectively a fuel efficiency indicator. It works out like this: an engine burns for X seconds at Y thrust using Z propellant. If you have an engine developing 1 N of thrust, and this engine then burns for 800 seconds on 1 kg of propellant, then that engine has 800 seconds of Isp. It is quite literally "my fuel lasts this many seconds long".

- From this relationship, you can immediately see that in order to have a higher Isp, the engine must take longer to expel 1 kg of propellant. However, a rocket is defined by the act of expelling propellant. Doing so is the very thing that makes something a rocket. By expelling mass, it gains forward momentum, much like jumping off a boat causes the boat to float away from you. And if it expels mass more slowly, it moves more slowly: it has lower thrust. Higher Isp directly begets lower thrust, and there is no way around it. If you want to have the same thrust at higher Isp, you must not only choose the right fuel (see above), but also put an enormous amount of energy into the fuel, to offset that natural limitation by sheer brute force. And there are physical limits to how much energy you can impart - for example, limits to how hot you can run a reactor before your engine simply melts.

- Ion thrusters have gigantic Isp because their thrust is so low. Sounds silly at first, but it's in the way they are designed: they don't expel much mass at all. No, they work by taking a really really small mass (single gas atoms) and pumping that tiny mass full of gigantic amounts of energy. That way, expelling this tiny mass gives a lot more forward momentum than the same tiny mass used in a different engine that imparts less energy. You're shooting it out at screaming velocity instead of just lobbing it casually out the back, so to speak. Of course, as said above, you are still limited by the amount of energy your engine can handle at once. So if your finite amount of energy goes into a very small amount of mass, you only have a very small amount of thrust (again, rockets are all about the mass they expel). The beautiful part is, energy weighs nothing*, but propellant does. The ion thruster is simply the result of a human saying "what if I can trade using less heavy fuel for using more weightless energy". So merely by throwing gobs of energy at the problem, the ion thruster gets away with taking much longer to go through its fuel. 4200 seconds - the time it takes an ion thruster to go through 1 kg of propellant while producing 1 N of thrust. It burns more than five times as long as a LV-N would, at the same thrust, with the same amount of fuel. Absolute thrust is lower, unavoidably - but so long as you have time to make that slow, efficient burn, you're going to be accelerating at the same rate for more than five times as long, making your spaceship that much faster in the end and being able to go that much more distance.

- Ion engines do have nozzles: their exhaust stream is shaped by a magnetic field. Just because you cannot see it doesn't mean it's not there It works because the ionized gas atoms are charged particles and respond perfectly and predictably to magnetic forces.

* Energy production equipment, however, can get quite heavy...

This is pretty good summary of topic, thanks for that.

[LameLefty]

I read something were kommit said he was going to redo the parts, this makes me sad I actually love the nuclear engines as is.

Ditto. I really hope he simply creates alternative .cfg's for the FTmN series engines. I use them as-is for almost everything - the modeling, the textures, the thrust/mass/ISP balance ... all perfect for KSP. *sigh*

[Streetwind]

Why not wait and see how they're going to turn out? I mean, even if he redoes them, nobody said that the changes have to be huge, right? And if he made cool stuff in the past, chances are good that he'll make more cool stuff, and not sucky stuff

[Yilmas]

Question: What program(s) do you use for modeling these engines/parts ?

[Kommitz]

Time for some necromancy.

This is just playing around with an idea I had (and not really important enough for the actual thread), but I thought I'd let you bask in the warm, radioactive glow:

Failure states! Mostly meltdown themed, I think nuclear engines could really do with soem focus on their risks and unique properties past just having great Isp.

Maybe I'll look to contracting a plugin dev one day.

Edited February 13, 2015 by Kommitz

[Tommygun]

I wish all parts had damage states like this instead of just turning to dust.

Think of all the lovely debris fields when landings go bad.

[sojourner]

Any chance we'll see these beautiful engines updated for 1.0.2?