PSA: Nuclear engine overheating

[THX1138]

Just a note for people using nuclear engines to keep an eye on the overheating because things have changed. I alt + tabbed out and came back to find I had no engines when previously they'd have been okay if left at 90% thrust (I have four grouped close together). Now they're overheating at 50% continuous thrust.

[timsfitz]

Will a single one overheat at max thrust?

[regex]

You need radiator parts. Some helpful poster pointed out in another thread that wing parts are probably the best way to go due to high heat tolerance and dissipation. Also, managing your throttle helps.

[sal_vager]

You can find it here.

It looks pretty cool having heatsinks like this actually

[moogoob]

I'm going to try mounting mine to the precooler - it apparently radiates heat well.

Also, the engines pass heat onto your craft. I'd never seen fuel tanks glow until 1.0. Oh, and my batteries exploded.

[RedDwarfIV]

I'm going to try mounting mine to the precooler - it apparently radiates heat well.

Going to try this too. There's no way I'm getting my SSTO to space if it doesn't work. Wing radiators certainly weren't enough.

[Brotoro]

Having to add 24 additional parts to every nuclear engine just so it can operate as intended is goofy. I use nuclear tugs to move around complex payloads, and the part counts already make things so laggy that it's no fun.

[r_rolo1]

Well, basing myself in Roverdude comments in one of the KSP-TV pre-release streams, solar panels supposedly are good radiators. Still have to try it though ...

[Wahgineer]

Also, the LV-N only burns LF now, not LF & LOX.

[Brotoro]

Torham234’s tests showed that solar panels are a very poor option for cooling.

[smjjames]

So.... what's the best option for cooling?

Would have been nice if SQUAD added actual radiator parts whose sole function is to dissipate heat.

[RedDwarfIV]

Tried the precooler. It is absolutely useless for radiating outside of an atmosphere.

Frankly, I don't think it's possible to get to orbit with a Mk-2 parts, all-LF spaceplane. Even with some small LFO engines for a boost, I could only get mine up to 1900m/s before falling back into atmo.

[CalculusWarrior]

Also, the engines pass heat onto your craft. I'd never seen fuel tanks glow until 1.0. Oh, and my batteries exploded.

Batteries are somewhat of a 'canary' part now, they are one of the first to explode when heated (along with solar panels and intakes). As such, placing them near heat-generating parts will let you know there's a problem quickly, if the batteries are dead, throttle down quickly!

[regex]

I have to say, this new heating system is awesome. Finally an engineering challenge that doesn't involve dumb physics/rubbery rockets! We could certainly use some dedicated radiator parts but wings should fit the bill quite nicely for the time being. This is the realism that KSP deserves.

[Frozen_Heart]

No idea how to use nukes now. Anyone found a cooling solution that doesn't involve multiplying the part count by 10?

[Jarin]

Torham234’s tests showed that solar panels are a very poor option for cooling.

That's testing in an atmosphere. I wonder if they're better in space? Precoolers seem to only radiate heat in atmo, so maybe solar panels are the opposite.

[Streetwind]

I have to say, this new heating system is awesome. Finally an engineering challenge that doesn't involve dumb physics/rubbery rockets! We could certainly use some dedicated radiator parts but wings should fit the bill quite nicely for the time being. This is the realism that KSP deserves.

It's a bit suboptimal that it comes at the cost of a huge increase in part count, though... dedicated radiators are very necessary.

[regex]

It's a bit suboptimal that it comes at the cost of a huge increase in part count, though... dedicated radiators are very necessary.

I agree. I'm a big proponent of procedural parts for simply reducing parts count. That being said, I'm having a blast figuring out the intricacies of the new system.

[Torham234]

Having to add 24 additional parts to every nuclear engine just so it can operate as intended is goofy. I use nuclear tugs to move around complex payloads, and the part counts already make things so laggy that it's no fun.

My radiators are aimed at a closed system with limited heat sink capacity. On large vessels, you can simply use the huge heat sink capacity of large fuel tanks. Big fuel tanks have big weight, and this means they can absorb a lot of heat. You can fairly safely fire a nuke on a big ship without dedicated heat radiators, because it will take a long time warm up the high mass of your ship. The bigger the tank right next to your nuke, the better. Avoid placing heat sensitive parts near the nuke ( batteries, sensors) and you are good to go. The radiators are mainly useful on smaller ships where the nuke quickly floods the whole ship with heat and you need a good way to radiate the heat into space.

PS: thanks to everyone who found my research useful, I think you guys just doubled my rep rating.

[Teutooni]

I think it's great you need to deal with some kind of complexity with the engine as a tradeoff for ISP. The heat management is kinda backwards though. As I understand NERVAs, the process of pumping fuel through the reactor and out the nozzle actually cools the system down. The more thrust, the more it cools down. The nuclear reactor would have some kind of control rods to control the reactor temperature. Higher temperature would produce higher specific impulse, and there would be a balance of reactor temperature and fuel flow rate and these would determine specific impulse and thrust. This might be too complicated for what squad wants for the base game, but I think it would be neat to have the engine heat up when you activate it (it could blow up if you leave it like that without thrust). The specific impulse would reach its peak 800 seconds at near max temperatures, and you could cool the engine down by applying more thrust or shutting it down completely. I imagine around 90% thrust should reach steady state at 800s ISP.

[Frozen_Heart]

I think it's great you need to deal with some kind of complexity with the engine as a tradeoff for ISP

Complexity is one thing, having to put 20 to 30 parts sticking out is completely another. There needs to be a proper heat sink / radiator part.

[Renegrade]

I have to say, this new heating system is awesome. Finally an engineering challenge that doesn't involve dumb physics/rubbery rockets! We could certainly use some dedicated radiator parts but wings should fit the bill quite nicely for the time being. This is the realism that KSP deserves.

Seconded.

Finally, those CPU cycles going towards heating actually mean something.

[regex]

If you feel like radiator parts are needed, up this thread: http://forum.kerbalspaceprogram.com/threads/117281-Dedicated-Radiator-Panel-parts

In the meantime, this is my solution. No idea if it'll work yet but the previous iteration did pretty well (four~five minute 100% burn before the solar panels blew) on only five control surfaces per engine. Large tanks help as well.

I'm pretty happy with the part count since this tug is pretty low on the count as it is. You could probably even go with five of the C-wings (I think that's what they are) per engine and large tanks. Further testing required, obviously.

Personally, I think this looks awesome. Much more like how a spacecraft should be.

[Streetwind]

Would look even better if the wing parts would start to glow though I'm not sure it'll push them that high.

[Gkirmathal]

Havent managed to play due to work. But I find it odd it needs radiators when under thrust.... :S

Like the guy on page 2 also said, sry typing on mobile, NTR systems dissipated heat via fuel flow when running under thrust and for startup/cooldown.

Bimodal system indeed needed radiators due to no fuel flow.

So I find it rather duff we need to use radiators for normal operation. Which is opposite to how, from my knowledge, most NTRs functioned.

Edited April 28, 2015 by Gkirmathal