Why do NTRs need radiators?

[MaverickSawyer]

Seriously, though. I though the entire point of an NTR was to get really freaking hot and then transfer that heat to the propellant, which keeps the engine from overheating. If you need to use radiators, your design is horribly off-balance and you're wasting thermal power.

So why do we need radiators for NTRs in KSP?

[W. Kerman]

Most times, you don't need radiators, but remember, heat transfer is not 100% efficient, so it generates waste heat.

[Guest]

The entire point of an internal combustion engine is to get gases really hot and convert that thermal energy into kinetic energy, then expel the hot gases. If you need to use radiators, it means your engine is wasting thermal power.

Yet all internal combustion engines need radiators. That's because it's not possible to make a 100% efficient thermal insulator. Heat will make its way where it's not wanted.

With an NTR, the temperatures involved are many orders of magnitude higher. If you want to heat your propellant to 1,000,000 degrees K, it's going to take work to keep the waste heat from melting your engine.

[Sandworm]

37 minutes ago, MaverickSawyer said:

Seriously, though. I though the entire point of an NTR was to get really freaking hot and then transfer that heat to the propellant, which keeps the engine from overheating. If you need to use radiators, your design is horribly off-balance and you're wasting thermal power.

So why do we need radiators for NTRs in KSP?

They don't.  While running the constant flow of cryogenic fuel makes them irrelevant.  They are needed when the engine is not providing thrust.  Either the reactor's lowest-level power setting still produces heat which must be dealt with, or you want to get the engine up to something hotter than 'off' before you start throwing fuel into it.

 

Now that thrust varies by ISP, rather than fuel flow, it would be perfectly reasonable for NTRs to have 'warm up' and 'cool down' periods where the less-than-optimal temperature of the engine resulted in much lower ISP.  Or you could install radiators so as to pre-heat the engine.

Edited October 1, 2016 by Sandworm

[KerikBalm]

1 hour ago, Sandworm said:

Now that thrust varies by ISP, rather than fuel flow, it would be perfectly reasonable for NTRs to have 'warm up' and 'cool down' periods where the less-than-optimal temperature of the engine resulted in much lower ISP.  Or you could install radiators so as to pre-heat the engine.

I've suggested such things before. I'd like to see them implemented. You could have the engine constantly produce heat like an RTG, and additionally have it warm up like a drill - lower temps = lower insulation.

You could allow the core temp to reach 3200 K, but have the main part temp have a maximum of 2500, and significant transfer between the core and main part while the engine isn't running (LF functions like ablator and carriers away heat in the LV-N?). You'd need radiators to keep the engine from melting if you want to pre heat the engine, otherwise you start your burn with lower Isp, and need to reduce Isp towards the end of the burn

[Temeter]

KSP is lacking the functionality to monitor heat on high timewarp. And it's also a bit questionable if there is a good way to do that in the first place. Following, you can't really go for the 'heating outside burns' thing.

I guess the general idea of heavin to boot up a nuclear engine for a few minutes doesn't really fit the idea of KSP as a game either, compared to how straightforward engines are otherwise. Not to mention you'd just want to timewarp during the startup, which, as said, doesn't really work with the current heat system.

 

Because of this, creating heat when the engine is running is a lot better for gameplay, simulation-wise, and gameplay-wise since it effectively limits the maximal length of your burn, or at least forces you to use the necessariy cooling equipment. At least as long as Squad isn't adding some complex and reliable system around the Nerva heat.

Sadly, people then complained because they didn't want to adapt to the changes and instead heat got nerfed so hard that it doesn't really matter at all. Radioators are pretty much completely useless at this point. :/

Edited October 1, 2016 by Temeter

[ZooNamedGames]

3 hours ago, KerikBalm said:

I've suggested such things before. I'd like to see them implemented. You could have the engine constantly produce heat like an RTG, and additionally have it warm up like a drill - lower temps = lower insulation.

You could allow the core temp to reach 3200 K, but have the main part temp have a maximum of 2500, and significant transfer between the core and main part while the engine isn't running (LF functions like ablator and carriers away heat in the LV-N?). You'd need radiators to keep the engine from melting if you want to pre heat the engine, otherwise you start your burn with lower Isp, and need to reduce Isp towards the end of the burn

I doubt Squad would go for degenerative forms of propellant, BUT what they might go for is something called "Liquid Ablator"  or "Liquid Coolant" which would function as you mentioned and slowly dissipate heat at the cost of this resource but the big difference being that it can help more parts than just the NTR. 

[cantab]

5 hours ago, Brikoleur said:

With an NTR, the temperatures involved are many orders of magnitude higher.

Point of order, they aren't. In fact the NERVA actually ran cooler than chemical rockets, the operating temperature being limited by what a nuclear reactor can take without having a meltdown. The efficiency is because lighter molecules move faster at the same temperature and the NERVA has pure hydrogen as its exhaust.

[Van Disaster]

4 hours ago, Temeter said:

I guess the general idea of heavin to boot up a nuclear engine for a few minutes doesn't really fit the idea of KSP as a game either, compared to how straightforward engines are otherwise.

This is what's put me off implementing proper jet engine startups - even military engines start slowly enough that it'd be an appreciable wait. Nertea's nuclear reactors do have a startup/shutdown time though.

[ZooNamedGames]

Simply add a jet engine-esque spool up/down for the NTR then?

[Temeter]

31 minutes ago, Van Disaster said:

This is what's put me off implementing proper jet engine startups - even military engines start slowly enough that it'd be an appreciable wait. Nertea's nuclear reactors do have a startup/shutdown time though.

Yup, those work fine, but they also aren't really a gamechanger or so. You activate them, timewarp to online. Or just set them to 1% to get costant low energy support.

Still interesting element considering how powerfull the equipment with NFT is, but I don't think there is that much point to put it into the main game.

Edited October 1, 2016 by Temeter

[Laie]

5 hours ago, MaverickSawyer said:

So why do we need radiators for NTRs in KSP?

Because someone thought this would make a good gameplay mechanic.

 

7 hours ago, Brikoleur said:

it's not possible to make a 100% efficient thermal insulator. Heat will make its way where it's not wanted.

You're not trying to insulate one part of an otherwise static or closed system. You have coolant/propellant flowing in from one side and out to the other.

Dumping so much heat into the vessel that the whole thing will take on a red-hot glow is a Kerbalism that has nothing to do with reality.

[Corona688]

You have coolant flowing into non-nuclear liquid fueled rockets, too, but they still ride the edge of thermal breakdown.  The hotter you make it run, the more efficient it is, so they will always ride the edge.  Properly cooling them was -- and is -- one of the biggest challenges in actual space rocketry.

These days, you don't get NERVA's overheating unless you use them in clusters, though, which some big enough liquid engines do too, like mainsails.

Edited October 1, 2016 by Corona688

[Archgeek]

Yup, by all rights, nukes should have an-off switch for their reactors, a spool-up period where their Isp wanders up from crap to 800, during which you either need radiators to prevent the tank above it from exploding, a trickle of low-efficiency thrust to cool the engine with its fuel.  They should be self-cooling while burning, but have a spin-down time after turning the reactors off where you either keep burning at decreasing efficiency to keep them cool, or cut throttle and give some radiators a nice workout.  Doesn't one of the interstellar mods do just that?

As things stand though, 2 small static radiator panels will cool a nuke on a nearly empty nose adapter tank with infinite fuel to allow an unrealistically long burn to equilibrium without anything exploding.