PSA: Nuclear engine overheating

[Radam]

This is a petty cosmetic thing, but I really wish the thermal display's color spectrum was blue for cold parts, transitioning to red, then yellow, then white. It's just odd for cold parts to be red.

Yeah maybe not what one would consider cool. But still, in the IR spectrum, 300k is preety much cold red, going ever brighter as temperature increases.

[regex]

is the heatConductivity suppose to be commented out. Has anyone removed the first set of // and tested the engine?

That was commented out on every part I looked at last night. emissiveConstant and HeatProduction seem to be much more important.

[RedDwarfIV]

FLIRs always output an image in blue as cold, red as hot. That's why it would be more intuitive to do it that way. After all, we're already looking at something that isn't visible.

[mariohm1311]

Maybe I'm wrong, but a NERVA transfers the heat to the fuel, producing thrust. So it would not need radiators.

[r_rolo1]

Maybe I'm wrong, but a NERVA transfers the heat to the fuel, producing thrust. So it would not need radiators.

In fact , in RL the nuke engines of the kind we presumably have in game would heat up when not thrusting, for that exact same reason ( and would probably need a minimum level of thrust to keep the temps from rising ). That would actually be a fun challenge in game: how to keep nuclear engines cool when not in use

Edited April 29, 2015 by r_rolo1

[regex]

There's also the fact that you're getting liquid hydrogen isp out of what is essentially kerosene with the thermal properties of Aerozine 50. Plus, you don't have to worry about liquid hydrogen boil-off. We gloss over a lot of things in stock KSP for interesting gameplay, so having the engines heat the craft when thrusting instead of dealing with radiation and shadow shields isn't so terrible.

[youkofoxy]

This is a petty cosmetic thing, but I really wish the thermal display's color spectrum was blue for cold parts, transitioning to red, then yellow, then white. It's just odd for cold parts to be red.

200 celsius,

that's is what I read, could be Kelvin...

but Kelvin???? I did not expect that.

[soulreaver1981]

the nuke is broken in my game. it use no oxidizer any more.... hey nice for a ssto lol

[Red Iron Crown]

the nuke is broken in my game. it use no oxidizer any more.... hey nice for a ssto lol

That's deliberate, real nuclear rocket engines don't use oxidizer either.

[sal_vager]

One reason for this is that oxygen is extremely damaging to the engine, seeing as it is an oxidizing agent

[Brotoro]

There's also the fact that you're getting liquid hydrogen isp out of what is essentially kerosene with the thermal properties of Aerozine 50. Plus, you don't have to worry about liquid hydrogen boil-off. We gloss over a lot of things in stock KSP for interesting gameplay, so having the engines heat the craft when thrusting instead of dealing with radiation and shadow shields isn't so terrible.

A Nerva's shadow shield is built into the engine, right on top of the reactor core where the smallest size shield can be the most effective. I assume the KSP nuclear engine has a particularly massive shadow shield because the engine has a rather high mass.

If Squad wants to screw over the nuclear engine, they have plenty of realistic ways to do it. First, they could lower its Isp, since it should be lower given the high molecular weight of liquid fuel (since liquid fuel obviously CAN'T be hydrogen...you'd need MUCH larger tanks to hold that much liquid hydrogen). Second, they could implement radiation in KSP (it seems very silly not to have radiation in a space game anyway) to make them more difficult to use by limiting how close they could be to kerbals while in operation). Third, they could give the nuclear engine a long tail-off thrust (like the spool-down on a jet engine), because once you scram the reactor, there is still going to be a lot of very short half-life nuclides producing heat, so you need to run fuel through the engine for a while after cutoff. Then using the engine becomes an interesting challenge.

They certainly didn't have to implement a faulty heat model that indicates and propagates misunderstanding of this particular rocket technology.

[r_rolo1]

Yup, oxygen + hot metal is not a recipe for long lasting metal structures , like this friendly PSA of ol'days shows

• www.youtube.com/watch?v=unki1tMWHt8
  

[architeuthis]

A Nerva's shadow shield is built into the engine, right on top of the reactor core where the smallest size shield can be the most effective. I assume the KSP nuclear engine has a particularly massive shadow shield because the engine has a rather high mass.

If Squad wants to screw over the nuclear engine, they have plenty of realistic ways to do it. First, they could lower its Isp, since it should be lower given the high molecular weight of liquid fuel (since liquid fuel obviously CAN'T be hydrogen...you'd need MUCH larger tanks to hold that much liquid hydrogen). Second, they could implement radiation in KSP (it seems very silly not to have radiation in a space game anyway) to make them more difficult to use by limiting how close they could be to kerbals while in operation). Third, they could give the nuclear engine a long tail-off thrust (like the spool-down on a jet engine), because once you scram the reactor, there is still going to be a lot of very short half-life nuclides producing heat, so you need to run fuel through the engine for a while after cutoff. Then using the engine becomes an interesting challenge.

They certainly didn't have to implement a faulty heat model that indicates and propagates misunderstanding of this particular rocket technology.

Agreed, still the thermo model is much better than before all things considered.

Unfortunately I haven't had time to play 1.0 yet. But I notice that people are mounting radiators radially very closely together. In reality this would be extremely poor engineering (except for convective cooling fins). Is there no view factor? Are the surfaces all radiating into absolute zero? How does 1.0 radiation heat transfer work exactly?

[regex]

They certainly didn't have to implement a faulty heat model that indicates and propagates misunderstanding of this particular rocket technology.

The heat system itself is hardly faulty, merely the particular implementation of a given part. And while I see your point about the misunderstanding, KSP glosses over so many points of realism in a way that propagates misunderstanding that I personally just chalk this one up to a game balancing mechanic.

[Brotoro]

I was referring to the heat model for the nuclear engine part.

If we are complacent with glossing over realism, Squad will have no incentive to fix any. So I will not be complacent. Especially if this is just a question of balance when there are better ways to balance the part.

[regex]

Unfortunately I haven't had time to play 1.0 yet. But I notice that people are mounting radiators radially very closely together. In reality this would be extremely poor engineering (except for convective cooling fins). Is there no view factor? Are the surfaces all radiating into absolute zero? How does 1.0 radiation heat transfer work exactly?

It seems to follow black-body radiation in general, but I don't think there's any ray-casting or whatever going on to ensure good spacing or angle to heat sources, etc... Heat transfers from one item to another, "pools", and radiates, in what seems to be a logical manner, although there are certain things that act weird E: like cargo bays seem not transfer heat on, although that could be because it's all pooling in my large tank...

- - - Updated - - -

If we are complacent with glossing over realism, Squad will have no incentive to fix any. So I will not be complacent. Especially if this is just a question of balance when there are better ways to balance the part.

As I said, there is probably a fix in the works to lower the LV-N's heat production value. I did a bunch of testing for NathanKell last night who assured me he will pass that on, and I'll bug him later.

And as far as realism, let's get on the solar system's un-realism, shall we?

Edited April 29, 2015 by regex

[Archgeek]

I should hope they plan to tweak the LV-N's thermal tomfoolery. Compressive heating and sound barrier have made for fun engineering and flight challenges, but the LV-N heating isn't fun -- its main job is long-ish transfer burns, and having it attempting to blow up the ship in exchange for doing its job when by all rights it should be slightly cooling the ship at high throttle settings, is really just annoying. I'd much rather have to mount secondary adjustment thrusters to get around the thing blowing up the ship at low throttle than crank my part count up trying to make ad-hoc radiative heat sinks.

As things stand I'll either have to mess with the part config or expand my XenonStorm line beyond the scope of just Project Helios.

And as far as realism, let's get on the solar system's un-realism, shall we?

HEHehEHhEhe, Tylo: Destroyer of Systems. I recall a simulation wherein it ripped the Jool system apart, ejecting either Pol or Bop into some messed up solar orbit, then managed to eject itself hard enough to go on to mess up Kerbin and Eve's orbits a bit.

[regex]

HEHehEHhEhe, Tylo: Destroyer of Systems. I recall a simulation wherein it ripped the Jool system apart, ejecting either Pol or Bop into some messed up solar orbit, then managed to eject itself hard enough to go on to mess up Kerbin and Eve's orbits a bit.

OT: Apparently it's actually stable if you model it realistically but I'm talking about the solar system in general. Going purely off of body radii (and correcting density to reasonable levels), Kerbol is a low-end M class star orbited by a bunch of Ceres-class objects and a single Earth-size object. Kerbin couldn't hold an atmosphere.

[Temeter]

like cargo bays seem not transfer heat on, although that could be because it's all pooling in my large tank...

Cargo buys are bugged and don't transfer heat.

[Brotoro]

The things that I care about in Kerbal Space Program are, in order of importance:

1. Proper handling of rocketry (the rocket equation must work, etc. KSP does acceptably well here).

2. Proper handling of orbital mechanics (we don't have n-body, but the patched conics works quite well).

3. Everything else.

The only reason I play KSP is because it meets points 1 and 2. If it didn't, I'd drop it like a hot Nerva.

The fact that KSP has purple planets and ice moons too close to the sun, and such things as these, does not annoy me a whole lot. I don't even mind the overly dense planets, since the devs need only posit the existence of a super-dense form of matter (perhaps baryons made from the heavier quarks are stable in the KSP universe)...and since that dense material only needs to live in the cores of the celestial objects where is isn't going to affect points 1 and 2 above, it won't hinder me enjoying the game. It can also explain why a low mass star can produce the luminosity we see from Kerbol, since the pressures and temperatures of the normal matter just outside the tiny super-dense core will allow conditions for fusion of the normal matter.

The water physics is moderately annoying...but this is a "fly rockets around" game and not a boat game...but I surely hope they fix this at some point. But it doesn't mess with points 1 or 2 much, except for spacecraft recovery. KSP's aerodynamics were terrible before, but they are greatly improved in 1.0, so I mark those as acceptable now, but still with room for improvemment.

The lack of proper chemistry of the propellants annoys me...what exactly ARE fuel and oxidizer? I wished squad would pick some real propellants and use realistic numbers to determine their tank volumes and masses (and, no, the players who aren't interested in such things don't need to be concerned with the details)...since this affects point 1 for those who want to dig deep into the rocket science.

I don't mind little green aliens. I don't mind the kerbal sensibility and humor. Those things can be in the game without affecting points 1 and 2...and they make the game fun.

Edited April 29, 2015 by Brotoro

[regex]

fun.

This is why I argue for the LV-N heat, honestly, I find it immensely fun. And the solar system annoys the living crap out of me. It's too tiny which means launches still don't look or go proper, being in orbit lacks all the "majesty" it should have, sense of scale is completely screwed up, spaceflight is way too easy, people can make completely silly SSTO planes and rockets, etc... The point being, there are a lot of things in KSP that are unrealistic and get glossed over in the name of "fun". We just differ on what is "fun" in this case, and I totally understand where you're coming from.

Anyway, I confirmed the suggestion being passed on, hopefully the heating will be reduced in the next update.

[Brotoro]

This is why I argue for the LV-N heat, honestly, I find it immensely fun.

And I'm happy for you that you could enjoy it.

But the kinds of missions I like to do involve sending lots of large payloads to Laythe or Duna or such. I have literally spent hours and hours taking care of tedious interplanetary transfers with long nuke burns to get this stuff done. 18-minute burns for Jool injection were common, with armadas of half a dozen ships. And physical time warp often didn't help because the ships had lots of parts...so even at 2x or 3x time warp, the game was still running at slower than real time for these burns.

Therefore, the prospect of having to run my nukes at less than full throttle does not appeal to me at all. What used to take a few hours could take days of free time now. And having to add lots and lots of extra parts to combat the problem would only exacerbate the laaaaaaaag issue. So, at the moment, I don't think I'd want to do any more missions with the new nuke limitations. So this makes me very sad...because I used to like playing KSP.

Anyway, I confirmed the suggestion being passed on, hopefully the heating will be reduced in the next update.

And thank you for that. But I missed where Squad has officially expressed a concern about the overheating nukes problem, and saying that it may be a problem that needs to be addressed.

[regex]

But I missed where Squad has officially expressed a concern about the overheating nukes problem, and saying that it may be a problem that needs to be addressed.

Squad hasn't, I talked with NathanKell last night, who helped develop the heat system, and he asked me to test out some different configurations. Just to be safe, though, you might want to put in a proper bug report, although I'd assume there are other people who have an issue with this. Squad may have simply dropped the values from the previous configs (the LV-N always ran hot) by a set amount and released, we'll probably never know.

[Torham234]

Squad may have simply dropped the values from the previous configs (the LV-N always ran hot) by a set amount and released, we'll probably never know.

I bet its probably this. Like you said, the nukes were always hot, it just became a concern when the heat system suddenly started taking into account the whole craft.

[Redshift OTF]

As I thought. Heat shields protect the rest of the craft from LV-N heat remarkably well. I had to run some fuel lines though but that's cool.