Jump to content

I've just got 14 minutes out of the nuclear engines. Stock parts


Recommended Posts

I don't really know anything about real life nuclear thermal propulsion, so help me out here.

I gather that the idea is this: "We cool off the engine itself by dumping heat into the reaction mass. The hot reaction mass is obviously being discarded from the ship. So we can get rid of the engine heat via this mechanism."

My question, though, is: how much of the heat can you get rid of via this mechanism? Let's say the exhaust is at a temperature of 2000K. It is at that temperature because it has passed through the engine. then the engine itself is also at a minimum temperature of 2000K, yeah? Probably somewhat higher, unless the fuel cycles through the engine for quite a while.

So my question, then, is: how are you going to get rid of this 2000K worth of engine heat, ie, the heat left over after the equilibrium is reached between fuel and engine?

if we use some kind of heat pump to move even more heat from the engine to the fuel/exhaust (ie, beyond equilibrium), well, that doesn't seem to help, because the heat pump must also be on the ship, and in order to work it will generate even more heat than it pumps.

The solid core of a Nerva motor has a temperature that goes from 25 Kelvin or so where the liquid hydrogen is injected, up to around 3000 Kelvin at the end where the hydrogen leaves for the nozzle. You have the control drums (like control rods, only they rotate to face neutron absorbers toward or away from the core) set so that chain reactions in the core can maintain these temperatures as the fuel passes through. You don't run any part of the core hotter than this because the fuel elements need to say solid and not melt. (The gasses in the combustion chamber of the Space Shuttle Main Engine can be hotter than this because they are gaseous, and the walls of the chamber are cooled by passing propellant through them...same as in the NERVA).

When the NERVA nears the end of the burn, the control drums are rotated to stop the chain reactions in the core. The core will stay hot for a while after this because there are very-short-half-life fission products in the fuel elements. The heat production from these nuclides drops exponentially, but you have to keep running fuel through the reactor core until it cools down enough that the hottest part of the core will not exceed the melting point of the materials once the fuel flow is stopped. So, with a NERVA you have to plan your burn carefully and begin the shutdown soon enough. (The engines in KSP, both the NERV and the chemical engines, are ridiculously throttleable and instantly restartable...but that's because it's a game.)

Once the fuel flow in the NERVA stops, the heat equalizes between the hot end of the core and the cold end (where the fuel was coming in). Then the heat works its way outward, and will continue doing so at a decaying rate as the short-lived isotopes decay away. So, the outside of an NTR should get to its highest temperature AFTER shutdown of the reactor and tail-off of the thrust. The Isp during the tail-off is lower than during the main part of the burn, of course, because you are letting the reactor cool down, so the exhaust velocity gets lower as well.

Edited by Brotoro
Link to post
Share on other sites
Is it? The only place you'll need a 2+ km/s burn to return to Kerbin from is Moho, and that's problematic already because a typical 3,5-4,5 km/s circularization burn is already out of the question (fancy maneuvering can reduce that by a lot, or so people on these forums claim). Jool could require that much - but only if you aren't using Tylo to give you a gravity assist. I don't think Eeloo requires so much delta-v for the trans Kerbin injection but even if it does, you can easily split the burn.

It could be problematic if you want to return with a large amount of equipment but why should you? After all the effort in delivering whatever cargo you delivered to Jool, what's the point in return it back? Head back with kerbals and the upper stages of the landers (which are way lighter than the vehicles you took there to begin with) and leave everything else in the Jool system, to be used by future expeditions.

I guess it all depends on how much you are sending. This was my usual Jool ship. Once in the upper atmossphere it made it the rest of the way to orbit with 12 Nerv's in three quads and carried three two man landers each with two nukes. Once in system the landers could go anywhere and land and return to the mothership. I don't see any mission like this being possible in the new model no matter how you build your ship.

Screen%20Shot%202015-04-30%20at%2010.26.29%20PM_zpsxy8xf510.png

Edited by Aerindel
Link to post
Share on other sites

I bring back my crew in a set of habitat and control modules (because I don't want them to feel like they are in prison). I also include fuel tanks around part of the habitat to act as radiation shielding against cosmic rays. So my return ships can be more massive that you might expect.

- - - Updated - - -

...Now, back to the real issue, which is that the LV-N is broken. It does NOT need radiators, it needs its ridiculous and unrealistic overheating problem totally taken out, knocked in the head, and buried in an unmarked grave.

Geschosskopf, like me, does grand multi-ship expeditions to do interesting things in the far reaches of the Kerbol system. We have to do a LOT of interplanetary transfers. Having to deal with ridiculous amounts of heating or protracted burns at lower thrust just to get all of our ships to where we can have fun is going to discourage us mightily.

Link to post
Share on other sites
The solid core of a Nerva motor has a temperature that goes from 25 Kelvin or so where the liquid hydrogen is injected, up to around 3000 Kelvin at the end where the hydrogen leaves for the nozzle. [...]

Once the fuel flow in the NERVA stops, the heat equalizes between the hot end of the core and the cold end (where the fuel was coming in). Then the heat works its way outward, and will continue doing so at a decaying rate as the short-lived isotopes decay away. So, the outside of an NTR should get to its highest temperature AFTER shutdown of the reactor and tail-off of the thrust.

OK, so how are we we getting rid of this heat? IE, the heat that is not leaving with the reaction mass, both during the actual burn and the residual heat after the burn. This heat will build up through the ship and has to be radiated away, yeah?

Edited by allmhuran
Link to post
Share on other sites
OK, so how are we we getting rid of this heat?, IE, the heat that is not leaving with the reaction mass, both during the actual burn and the residual heat after the burn. This heat will build up through the ship and has to be radiated away, yeah?

The hot outer structure of NERVA radiates the heat away to space, yes. But the engine is designed and operated to stay at a temperature that is low enough to keep from being damaged as it does this.

Link to post
Share on other sites

I see. So you're saying that the surface area of the engine itself is sufficient to radiate away all of the heat generated by the operation of the engine (which seems like quite a lot of heat!). Or, in other words, the engine shouldn't overheat in KSP no matter what you have it attached to.

Link to post
Share on other sites

Btw, what's the big difference between using wings as radiators and using radiators as radiators? Heaving true parts might be nice, but it shouldn't change to much.

I guess it all depends on how much you are sending. This was my usual Jool ship. Once in the upper atmossphere it made it the rest of the way to orbit with 12 Nerv's in three quads and carried three two man landers each with two nukes. Once in system the landers could go anywhere and land and return to the mothership. I don't see any mission like this being possible in the new model no matter how you build your ship.

Sounds somewhat like you just want an easy engine. Then why not modify the config or just use cheats? The LV-N is still magnitudes better than every other engine in the game, it was probably the engine which was hit the least by a nerfbat, since it's usage increased in complexity rather than loosing efficiency (aside from 0.5t).

Even more so, the engines is, in comparision, actually an even stronger engine than it ever was. The (nvm the ion) 2nd most efficient engine the game is the poodle at 350 ISP, other than the KR2 with it's extreme thrust/weight ratio. Our nuke still runs at 800 ISP. If you want to talk about old ship-designs not working anymore, then the LV-N is the last engine you should worry about.

If you want fast and secure burns, use conventional engines.

Edited by Temeter
Link to post
Share on other sites
I see. So you're saying that the surface area of the engine itself is sufficient to radiate away all of the heat generated by the operation of the engine (which seems like quite a lot of heat!). Or, in other words, the engine shouldn't overheat in KSP no matter what you have it attached to.

If you design and operate your NTR correctly, yes. But the devs may decide they want to use heating as some sort of game balancing device...that's up to them. I'd prefer that they balance the NERV is a more realistic manner if they think it's OP...there are easy ways to do that. I would prefer it if KSP got its rocketry as close to realistic as possible because that makes me enjoy it more. If KSP had goofy rocketry, I wouldn't be playing it.

Link to post
Share on other sites
Sounds somewhat like you just want an easy engine.
I want an engine that is hard to design for, but easy to use.

I actually thought the Korburundom engines fit this bill. It took a lot of work to use once, and you needed a nuclear reactor (with cooling fins!) a capacitor bank, and mining trips to eloo to use it, but one you did all that you had a real spaceship on your hands, but one that you had to go to the ends of the system to get so it didn't feel like a cheat, it felt like a reward for dozens of hours of work.

If you think of Kerbal like an RPG the point of challenge is not so that you never become powerful, its so that you have to earn it.

Yes, the LVN is still a the best engine in the game on paper, but with the heat issue it is zero fun to use.

Edited by Aerindel
Link to post
Share on other sites

The question I am asking myself is:

Is there really a root cause of concern?

How long are you planning to run the nuclear engine?

A 10 min burn does not overheat the engine enough to cause it to explode, which far exceeds any burn I have ever done with it so far.

+ Given I use it for interplanetary transfer, there is a VERY long time for it to cool down until my next burn.

Edited by TimePeriod
Link to post
Share on other sites
Umm... yes, that's why I said there must be some radiator in that design somewhere. If the tank isolation is good, they won't have to dump a lot of waste heat to keep the hydrogen cold. But what the design does NOT have giant radiator wings to cool down its nukes. The reactors of the NTRs are cooled by their fuel. And, yes, these are design concepts by people showing how nuclear propulsion systems that we could build today can be used for Mars missions. These were not made by people playing games. Their primary concern is the nuclear propulsion system.

Once you turn off the reactor, it has fission products inside. Their quantity depends on the enrichment of the fuel, duration of chain reaction and neutron flux (which is the consequence of how much the control rods are doing their job). Relation is positive in all three cases.

Granted, the engine does not work for a month. It would work for a few hours at most, but it has the problem of being in vacuum. How is that heat removed? We can't do any calculations because we'd need in depth specifications, but for me it's pretty much obvious there would be pretty massive problems with avoiding meltdown.

I'm just asking because I'm curious.

Hmmm, what makes you say there needs to be a "heatsink", as in a solid object that absorbs heat? There's no necessity for such a thing. Anything cooler than the heat source will absorb heat, and that something can be nothing (as in vacuum).

That's what you said. Heatsink is anything that removes heat. Radiator-vacuum system is one. It's pretty obvious what a heatsink is.

So at the bottom line, hydrogen is far from "pathetic" as an absorber of heat. Sure, you do need "olympic swimming pools" of it, but that's why those real rocket designs shown by Brotoro have so many friggin' huge tanks.

I mean it is pathetic in the sense you won't be able to cool down a shut down reactor. That LH2 in the tanks is your propellant. If you spend it on cooling the systems, not only you lose propellant, but you also change your orbital parameters by venting it out.

Those tanks are nothing compared to an olympic sized swimming pool.

Anyway, I found this handy document. It could help, but I'm not into doing calculations now.

http://www.bnl.gov/magnets/staff/gupta/cryogenic-data-handbook/Section3.pdf

The hot outer structure of NERVA radiates the heat away to space, yes. But the engine is designed and operated to stay at a temperature that is low enough to keep from being damaged as it does this.

This thing

NERVA%20ENGINE.jpg

will be capable of getting rid of heat via radiation in vacuum only by its casing? I have strong doubts about that, but then again, I'd have to know all the details and do lots of calculations.

NERVA was tested as engine only, in atmosphere. It was not a system ready for implementing into flight. There wasn't a finished ship design for it.

I'd still go with KSP getting radiator parts for this engine. This is an engine for vacuum, not for flying in atmosphere and having radiators for it would just make the game more interesting.

After all, addition of radiators would benefit the station building.

110713-coslog-station-1p.photoblog900.jpg

Edited by lajoswinkler
Link to post
Share on other sites

I'd prefer for the engine not to require separate radiators. Why? Part count.

If Squad feels that radiators on the LV-N are the way to go, great! There are a couple of possible solutions I see that don't involve adding radiator parts specifically for it:

1) Remodel the engine. Give it a ModuleRadiatorFX module or whatever, and some cool animation where vanes fold out to radiate into space. Awesome! No new parts.

2) Damp the heat generation. Put a blurb in the description about how it uses fuel as coolant, or how it contains its own radiators, or whatever.

Note that I'm not talking about whether stock KSP needs radiators generally, just the LV-N. Specifically, I'd go with this approach even if we get radiators in stock KSP (and I think we should, for various purposes like orbital bases). However, no one sane is going to fly the LV-N without attaching enough radiators to make it usable, so save us all the time in the VAB and the extra part count and the illusion of choice.

Link to post
Share on other sites
I guess it all depends on how much you are sending. This was my usual Jool ship. Once in the upper atmossphere it made it the rest of the way to orbit with 12 Nerv's in three quads and carried three two man landers each with two nukes. Once in system the landers could go anywhere and land and return to the mothership. I don't see any mission like this being possible in the new model no matter how you build your ship.

http://i1232.photobucket.com/albums/ff366/Aerindel_Prime/Screen%20Shot%202015-04-30%20at%2010.26.29%20PM_zpsxy8xf510.png

Are you using mammoth engines as mid-stage? That makes no sense, those are ascend engines, too inefficient to use in vacuum where twr isn't critical. The picture isn't big enough, but that ship seems over engineered. In 0.90 I used a kerbodyne tank with four nukes to deliver three two passenger rovers to Jool. Now, those rovers fell short on delta-v and required refuel within the Jool system, but that could have been solved by giving an earlier stage with a single nuke, which could be offset in the transfer stage by using six nukes instead of four.

If you still want to send inefficient rockets, get the MRS mod, and put its quad nuclear engine beneath a short mk3 fuel tank. You save in part count and likely heat generation.

I bring back my crew in a set of habitat and control modules (because I don't want them to feel like they are in prison). I also include fuel tanks around part of the habitat to act as radiation shielding against cosmic rays. So my return ships can be more massive that you might expect.

Good! Then you have a lot more stuff to distribute heat around. That ship with the half Mk3 and the four Mk2 tanks I've put together has a defect: the central tank only conducts heat upwards. Had I split the payload and put stuff below it, it would have been spreading its heat even more. It needs testing, but it's like that would extend the burning time even longer.

You don't want protracted burns, how long were your usual burns in 0.90 while also keeping a large dV? We're discussing ten minutes burns here. And when they are ejection burns, those burns are inaccurate anyway and work best by splitting them.

I don't think nukes are broken any more than anyone would think the aero model is broken because rockets can get out of control if control surfaces aren't used. They changed, and the old way of using them doesn't work anymore. They are still perfectly usable.

Link to post
Share on other sites
I see. So you're saying that the surface area of the engine itself is sufficient to radiate away all of the heat generated by the operation of the engine (which seems like quite a lot of heat!). Or, in other words, the engine shouldn't overheat in KSP no matter what you have it attached to.

Your fuel is your cooling medium, you use the fuel to cool the outside before you inject it into the core.

This is also done on chemical engines, one bonus with LH2 is that its good at cooling

Note that the nerva has some other downsides, after the burn it would be radioactive, you would not want to attach it radially to an crew compartment like we do in KSP for one.

I have no issues with LV-N and the ISRU module require radiators even if real world nerva don't need it.

An better solution might be to make the LV-N create lots of heat but its also good at radiating it, this would work well it the LV-N was alone but hurt if you clustered or used them radialy.

Yes I know players would then use them as radiators :)

Another idea would be to use fuel as cooling, if the engine run to hot it would have to use extra fuel for cooling reducing ISP.

Fuel cooling could also be used especially on plane parts to protect against heat. Think SR-71 used fuel for cooling, yes same rule as for LV-N isp might go down and you would use fuel during reentry.

Edited by magnemoe
Link to post
Share on other sites
OK, so how are we we getting rid of this heat? IE, the heat that is not leaving with the reaction mass, both during the actual burn and the residual heat after the burn. This heat will build up through the ship and has to be radiated away, yeah?

Think of the nozzle of a rocket engine (any type, not just nukes) as a gun barrel. It's designed to accelerate a "projectile" (the reaction mass) along its length to achieve the desired "muzzle velocity" at the end to produce the desired amount of thrust. Thus, the highest velocity of the "projectile" in the system is at the end of the nozzle. The highest velocity is associated with the highest heat due to the way gas math works, so the hottest point of a rocket engine is the end of the nozzle, farthest away from the rest of the ship, and not at the core itself closer in.

The nozzle is a thin sheet of material (especially at the end) so has a very large surface area compared to its mass. This means that the rate at which it absorbs and radiate heat is very high compared to a cubical block of the same material. IOW, the nozzle itself is 1 big cooling fin. So heat from running the engine easily gets into the nozzle on the inside, and easily escapes to the cooler surrounding environment on the outisde. Even if that environment is vacuum. The most important way heat transfers is via radiation in the form of photons, and photons don't need a medium to propogate through.

So, you design a nuke engine to handle 3 heat regimes. First, you make the nozzle out of a material that can take the heat of full-power operation, either alone or by running fuel through it en route to the core. LH2 is really good at absorbing heat so this works pretty well, and preheating the LH2 like this increases the efficiency of the engine. Second, you need a way to dispose of the waste heat of the "idling" core. Ideally, just circulating some LH2 between the core and the pipes surrounding the nozzle (which is 1 big cooling fin) would handle this. If not, the ship will already have a refrigeration plant, necessary to keep the LH2 cold to minimize leakage during long-duration trips, so this could be added to its workload. And finally, you just build the core in such a way, by selecting the fuel, its amount, and such things, that the rest of the system can handle the heat it produces in all phases of its operation.

The bottom line is that the LV-N is broken and the whole overheating mechanic for it is misguided and inaccurate. The whole point of nuke engines is to run at high Isp, to enable carrying usefully large payloads to far places. This high Isp comes at the cost of low thrust so the engines have to run at full power for long periods of time to produce the dV needed by the heavy payload for a given maneuver. These are the unalterable design parameters of a nuke engine. Therefore, any operational nuke engine SHALL be designed to work under such conditions without imposing undue thermal loads on the rest of the ship. Period, end of story. It is just ridiculous to have them overheating at all, because they MUST be designed from the get-go NOT to overheat, or they can't do their jobs. Because KSP now has them overheating, they can't do their jobs and are effectively useless.

BTW, somewhat OT, has anybody else noticed that the Mainsail no longer overheats at full power? Because energy must be conserved, that heat has to be going somewhere. And now the LV-Ns are overheating. Coincidence? Has the evil Mainsail worked some foul magic to shovel its heat into the innocent LV-N? :D

Link to post
Share on other sites
The whole point of nuke engines is to run at high Isp, to enable carrying usefully large payloads to far places. This high Isp comes at the cost of low thrust so the engines have to run at full power for long periods of time to produce the dV needed by the heavy payload for a given maneuver.

Later designs didn't even have that low thrust. Their TWR was pretty good.

Link to post
Share on other sites

Can we please -NOT- have game mechanics which force us to add parts onto ships which would not have previously been there? In a part-count limited game any function of a part which is mandatory needs to be included in that part. If an engine needs radiators to prevent overheating, those radiators need to be included; not force use to use even more parts and result in an even worse game play experience.

Link to post
Share on other sites
Bit pointless having an overheat mechanic if no engines can overheat...

Too flippant, sal. "Bit pointless having an engine that cannot be prevented from overheating" is the obvious (and equally flippant) response. There's more depth here than either statement acknowledges, and I think that's what this thread is about.

Link to post
Share on other sites
Nope, overheating engines is fun ;)

This.

Course I'd like to see more meaningful differences between engines anyway. Also, have the tech tree do something actually useful, and have upgrades to existing engines (advanced materials, etc). Then we can have tattling improve with tech, etc.

Link to post
Share on other sites

Having read the flight manual for a 1960's jet, I am not convinced "... any operational ... engine SHALL be designed to work under such conditions without imposing undue thermal loads on the rest of the ship" applies to non-civilian aerospace engineering, especially of certain timeframes or the engineering standards of little green men. It seems entirely possible "maximum rated power" for the NERVA is 90% and "emergency power" is 100%. Jeb pushes the throttle past the gate when he has a narrow burn window, and sometimes he blows up. Big deal.

Additionally that big picture of the NERVA shows two pipes which run around the nozzle throat. Propellant pre-heating and nozzle cooling?

Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...