What are the design principles of Nuclear Rockets?

[AdmiralTigerclaw]

Exactly what it says on the tin.

To elaborate: In general, what principle elements of design should be used when building a nuclear rocket, and what situations should you/I be building them in lieu of a conventional rocket?

I ask this because the few times I've tried nuclear rockets. Projected Delta V behavior, fuel usage, and thrust never seemed to add up. So I'd like to know how you guys build your nuclear rockets so they actually make sense to use. (And don't run out of fuel before a maneuver is complete and completely ignores what the Delta V was projected to be.)

[kfunk]

i only usually use nuclear engines for transfer burns. when you plot your maneuver node it'll tell you the burn time, you need to start it so that you do half the burn before and half the burn after.

if you're getting burn times that are too long you have a few options, you can use the oberth effect, performing multiple burns at Pe, or you can go Kerbal on it and add MOAR BOOSTERS (or ... more nuclear engines as the case may be)

hope this helps.

[Kasuha]

Assuming you're asking about the game and not real-life nuclear engines, the LV-N is (by Isp) the most effective rocket engine with Isp 800 in vacuum. That means it can pull more than two times the "work" from given amount of fuel compared to any other rocket engine in the game. With its mass, though, it only translates to higher dv if you are ok with lower TWR (TMR) and subsequently longer burn times. For these reasons it is ideal engine for interplanetary transfers and decent engine for landing on airless bodies, especially those with low gravity.

They are surpassed in efficiency only by jets (in breathable atmosphere) and ion engines (even less thrust, suitable for light payloads).

Some people say that to achieve ideal dv, you need to use just one LV-N. In my opinion, though, at certain total mass of the spaceship the added 2.5 t of another engine (or more of them) does not play big role in reducing the dv and you can actually save more fuel by making better use of Oberth effect, i.e. by burning closer to the periapsis than a very long burn would allow you.

Not mentioning my beloved LV-N based lander module which I used to land on nearly every body in the system, including Duna (shown docked with some fuel tanks it was carrying around and leaving in orbit):

Edited August 20, 2014 by Kasuha

[capi3101]

You want to be using nukes pretty much for transfer stages only - they work as part of an orbital insertion stage, but they really shine in situations where you need high efficiency without a great deal of thrust. You don't want them on your payload-to-orbit booster, that's for sure (to be fair, I did this a couple of times, but only because I was having a heavy transfer stage launch itself into orbit with the intent of refueling it once it was up there)

If you do find yourself in need of additional thrust, you might try designing a transfer stage kinda like this one:

Javascript is disabled. View full album

Though it's perfectly okay to just use a single bigger can of fuel if that's what you want to do

Edited August 20, 2014 by capi3101

[Red Iron Crown]

AdmiralTigerclaw, can you expand on what you mean by nukes not meeting delta-V projections? I've had exceedingly good results with them for the reasons stated by the other posters so I'm curious as to the nature of the difficulty you're having.

[juanml82]

Or you can make an "L" with girders in order to attach them to the side of an orange tank

[AdmiralTigerclaw]

AdmiralTigerclaw, can you expand on what you mean by nukes not meeting delta-V projections? I've had exceedingly good results with them for the reasons stated by the other posters so I'm curious as to the nature of the difficulty you're having.

I had a rocket that projected something like 1000 dV in orbit while I was building it. I got it into orbit, did my burn, and got something akin to a dV of 300 instead.

[MockKnizzle]

I'm not exactly what numbers you've been looking at... the tooltips in-game clearly state the thrust, mass, and ISP of the engine. Are you using some kind of crazy fuel-flow setup that is breaking Mechjeb or Engineer's dV readouts? Please elaborate on what exactly your problem is, and post pictures of the rockets.

[Red Iron Crown]

I had a rocket that projected something like 1000 dV in orbit while I was building it. I got it into orbit, did my burn, and got something akin to a dV of 300 instead.

How are you measuring dV? Are you using a mod, calculating by hand, or something else? Are you able to show a pic of the rocket or share the .craft file?

One part may be the nuke's abysmal Isp at very low atmosphere (though it catches the chemical rockets by about 1500m or so).

[Moonfrog]

if hes using kerbal engineer redux, it has a button that switches between atmosphere delta V and Vacuum delta V. Its possible hes seeing that.

[Sirrobert]

Nuke engines are only usefull for big crafts that stay in space for a long time.

They are very efficient, but they are also very heavy. It's not worth it if the rocket engine takes up half the weight of the entire rocket.

So really big transfer stages, or, if you use Kethan/Karbonite, they are really usefull for ferrying mined fuel back and forth between Ground station and Orbital station (on worlds without atmosphere), or between Orbital stations. (also usefull as reusable aircrane with a docking port on the bottom)

[AdmiralTigerclaw]

I have no specs/pictures on any of the rockets I was building because my attempt was back at the beginning of the year. I simply dismissed it until now as 'not worth it' until I randomly clicked on a page for a nuclear rocket mod. Then I was like. 'Okay, if these rockets aren't worth it, why do people keep making more and more fancy mods with them? I better ask how people properly build these things.'

As for Delta V calcs. I cross reference with both KER and MJ2's delta V systems because they'll disagree depending on certain configurations. Thus comparing the two allows me to see how it's getting predicted so I can smooth it. But in the case of the nuclear rocket, it was KER and it was a pretty straight forward linear stack. I'm likely to just dismiss the reading as 'something dun be messed up jeb' and leave it at that. Likely in eight months whatever issue it had been has been fixed.

(And KER was set to Vacuum Delta)

[Red Iron Crown]

I think we were all thinking about the stock nuclear rocket engine, added in 0.17 back in fall 2012. I can't really speak to how the mod parts you tried performed.

[AdmiralTigerclaw]

I think we were all thinking about the stock nuclear rocket engine, added in 0.17 back in fall 2012. I can't really speak to how the mod parts you tried performed.

I AM talking about stock nuclear engines. I've not even toyed with nuclear rockets from mods. (Aside from ol boom boom. Heheh... BOOM!) I just randomly clicked on a mod page and literally thought 'okay, if nuclear rockets are popular enough for entire mods, what am I doing wrong?'

[Kasuha]

I had a rocket that projected something like 1000 dV in orbit while I was building it. I got it into orbit, did my burn, and got something akin to a dV of 300 instead.

1000 dv is ridiculously low dv for a rocket with this engine. Maybe it could help if you post a screenshot or .craft file with your design so we can point out mistakes. My guess is you just used too little fuel.

What's below is what I used to distribute my probes around the system. Each probe has about 0.5 t and the nuke and the tank give it over 5000 m/s of transfer dv, not counting what ion engines on these probes can do.

(edit: I tested it and in fact it has almost 7500 m/s dv)

Edited August 20, 2014 by Kasuha

[Red Iron Crown]

I AM talking about stock nuclear engines. I've not even toyed with nuclear rockets from mods. (Aside from ol boom boom. Heheh... BOOM!) I just randomly clicked on a mod page and literally thought 'okay, if nuclear rockets are popular enough for entire mods, what am I doing wrong?'

Oh, sorry for the misunderstanding, I took your post to mean you used some of the mod ones. On a second reading you were pretty clear, the fault is mine.

Anyway, there's not much difference between designing a nuclear rocket and designing a chemical one in KSP. It's more efficient at the cost of lower thrust, but the fundamentals are all the same. You would generally not want to use them for lower stages of an a Kerbin ascent, but they are great once in orbit. I would suggest you try a few simple designs and see if you are still seeing the same lack of expected performance.

[r_rolo1]

Yeah, the best way of treating a nuclear engine is as it is: a relatively heavy engine ( for a 1,25m engine, that is ) that has a very good Isp in vacuum ( but far from that good in anything with atmosphere ) but a dismal thrust. In other words, it excels as a transfer engine and if you see yourself in a situation you need more than 2 or 3, you maybe should consider another options ...and also if your payload is ligher than it, as well It does not mean that you can't use it in other situations: most of the airless moons in KSP can be visited economically with nuclear powered landers and I've seen enough of spaceplanes that use them for the space part of their trips ...

[GoSlash27]

Same thing everyone else here has said; they're heavy, weak, and have really crappy efficiency in atmosphere. Naturally, you want to use them in situations where you're in vacuum and don't need a lot of thrust, but are looking for good efficiency.

As with every other rocket, the key to good DV isn't so much how big the rocket is, but rather how much of your rocket is fuel.

Best,

-Slashy

[Red Iron Crown]

The low atmospheric Isp is a bit overblown. The LV-N's efficiency is higher than any chemical rocket's by 1500m in Kerbin's atmosphere, starts about the same as the best chemicals on Laythe, and is much more efficient than chemicals on Duna. Eve is right out though.

[Jett_Quasar]

You can actually use nukes to get into orbit, the efficiency is about the same as a mainsail at about 3500m. The thrust to weight is crappy but if you kick them in after that height the isp is 350-400 good enough to take advantage of the extra boost.

JR

[AdmiralTigerclaw]

So the general advice on Nuclear Engines:

1: Go Big Fuel or Go Home.

2: Low ATM Pressure or Bust.

3: Best for Transfer Stages.

[Zablonski]

One other little hint I can offer - don't run your FTX-2 fuel lines too close to or directly into the glowing pink part of the LV-N. One fuel line always melts and stops fuel flow from that tank leading to asymmetrical fuel usage, course corrections during burns and general annoyance. While I can't replicate it in a static test @ KSC, I've had it multiple times in a vacuum.

[Zablonski]

One other little hint I can offer - don't run your FTX-2 fuel lines too close to or directly into the glowing pink part of the LV-N. One fuel line always melts and stops fuel flow from that tank leading to asymmetrical fuel usage, course corrections during burns and general annoyance. While I can't replicate it in a static test @ KSC, I've had it multiple times in a vacuum.

Correction: After studying the excellent fuel flow rules by Kasuha, I can confirm it is less the fuel lines melting and more I'm an average rocket designer.

[Jarin]

Again, it must be emphasized that nuke engines are utter ....e in atmosphere. They are far and away the worst engines in all categories when not used in a vaccum. This is the most likely problem I can see.

And as someone else said... if you had a LV-N stage that only had 1000dV, you need to redesign. That engine is heavy.

[Zylark]

I know, not very relevant in game-play terms, but very interesting nonetheless:

Gives a good walk-through of the advantages of nuclear engines and how they work.