Thermal Control Systems and Radiator Panels

[NewtSoup]

I've not yet needed to use a thermal control system except for on a mining rig.

What do we use the small fixed panels and radiators for?

My only guess is that they may be necessary when visiting the inner planets to stop the crew from cooking in their capsule.

Can they also be used to mitigate re-entry heat?

[Foxster]

Mining is the only use I've ever found for them. Even nuke engines don't overheat enough anymore to need them. And, no, they don't help with re-entry heating. 

[KerikBalm]

Mining is the primary use.

Looking cool for ISS replicas and such is another use.

There are mods with reactors and such that can make them useful (although they may add their own, better, radiators)

They can be useful for "sundiver" craft as well, ie craft that get fairly close to the sun.

[Brainlord Mesomorph]

On 2/7/2018 at 11:15 AM, Foxster said:

Mining is the only use I've ever found for them. Even nuke engines don't overheat enough anymore to need them. And, no, they don't help with re-entry heating. 

Your use of the words "enough anymore" give me pause, but  IME:

Yes nukes can overheat. It all depends on how many, how close they are to each other, the length of the burn, and the TWR of the ship.

I had a 950 ton ship w/ 12 nukes a few radiators and three medium Thermal Control units, and before it finished a Munar transfer burn, its overheating. The Thermals and Radiators were all at 100% and I was getting temp warning gauges on the engines and nearby parts. Nothing blew up. But I rebuilt the ship w/ 6 medium thermals. I didn't want to try an interplanetary burn like that.

OTOH: An 8 ton probe w/ one nuke doesn't even need a small radiator.

While were on the topic:

I need to do ISRU on Moho. Two large drills and one large Refinery. Think two large Thermal Control  units is enough?

Edited February 14, 2018 by Brainlord Mesomorph

[Snark]

I haven't found the fixed-radiator panels to be useful for, well, anything, for a very very long time.

I use LV-N engines quite a bit for interplanetary craft.  Once upon a time, they used to be explody and need radiators.  However, these days, they get a bit warm but I'm not sure it's even possible anymore for them to self-heat to the point of explosion, unless there's some external factor like heavy solar heating close to the sun or the like.  I can do a long, long LV-N burn and they never get even close to exploding.

Sundiver craft can get really hot from solar heating.  But I find the fixed radiator panels to be fairly useless there, too.  That's because they don't track to stay edge-on to the sun, which means that unless you manually orient the craft to keep them edge on (which might not even be geometrically possible, depending on how they're mounted), then they can get sunlight falling on their broad surfaces and can actually become net heat absorbers rather than radiators.  The deployable radiators get around this problem by always tracking to stay edge-on to the sun, so they can do their job.

As for mining... I suppose maybe one could use the fixed panels for that.  But I find them inconvenient for that purpose, because both the ISRU and the drills need to be cooled, and the static panels need to be mounted very close to them, and that's often inconvenient.  I've never had any occasion to use the fixed panels on a mining ship, since the deployable ones are so much more convenient and powerful.

So, the practical upshot:  as far as I can tell, anything the fixed panels can do the deployable ones can do better.  And there are things the deployable panels can do that the fixed ones can't.

2 hours ago, Brainlord Mesomorph said:

I need to do ISRU on Moho. Two large drills and one large Refinery. Think two large Thermal Control  units is enough?

You mean the largest-size deployable radiators?  My guess would be yes, those things are huge.  It's worth noting that solar heating even on Moho isn't really that big of a deal, in my experience-- the real problems come if you try to get significantly closer to the sun than Moho.

[Archgeek]

HEhEHE, it's like you guys are daring me to redo my old thermal experiments from back in 1.1 and post the results.  'Sounds like my Wednesday evening might get weird.

edit: for reference: 

 

Edited February 14, 2018 by Archgeek
added reference link to an old thread

[NewtSoup]

3 hours ago, Archgeek said:

HEhEHE, it's like you guys are daring me to redo my old thermal experiments from back in 1.1 and post the results.  'Sounds like my Wednesday evening might get weird.

edit: for reference: 

 

Ive been doing just that. Was writing up but had to go to work.  Spoiler - you now get shock heating at all altitudes.  TCS and Radiatiors do not mitigate shock heating. 

[Archgeek]

10 minutes ago, NewtSoup said:

Ive been doing just that. Was writing up but had to go to work.  Spoiler - you now get shock heating at all altitudes.  TCS and Radiatiors do not mitigate shock heating. 

Oooh, neat!  'Guess I'll just look forward to that, then.  Do some parts still transfer heat a lot less efficiently than others?  I remember docking ports and few other things acting as insulators, letting one get away with letting the bulk of the ship get very hot (so as to allow for a lot of surface radiation) while not making sensitive parts like probe cores and batteries explode, and holding the heat transfer up enough that the hot side cools swiftly enough to prevent leftover heat from blowing up sensitve parts post-burn.

[NewtSoup]

@Archgeek - here is the post I was constructing - you may still want to re-run your old experiments.  I have not exactly been thorough.  But it was fun hooning Jeb and Bill down the runway to almost certain death every time.

I am just conducting a test:

Mk 2 Cockpit
Mk 2 Cargo Short Liquid Fuselage
Mk 2 Cargo Bay ( long ) 
Mk 2 Bi Coupler
2 x AV8 Winglets 
2 x Vector
Medium Landing gear
Fuel For balance only
Infinite propellant

Hit space bar and the craft went down the runway @ 700m/s the speed stagnated.  Pulled in the landing gear and the speed went up to 1923 m/s at which point the cockpit exploded due to overheating  (3015 / 2500 K)
Note:  You now get shock heating effects at almost sea level - this craft didn't go above 200m - so presumably you get them at all altitudes considered within the atmosphere.

Reverted to  SPH

Added 2 fixed small thermal radiator panels to the cockpit

Ran the test again:

Craft went out of control and exploded due to g-forces

Added Canards
Wings
Elevons

Ran test again:

Craft exploded when the cockpit overheated at 1726 m/s and 2865K - actually slower and at a lower temperature than without the radiators.  However the aim was to make the craft run faster and cooler not explode sooner.  Suspect Drag from the radiators was slowing the craft while still allowing heat to build to the point where the cockpit failed.  

Remembered to turn the radiators on and re-ran the test:

Craft exploded when the cockpit overheated at 1836 m/s ( no significant difference to above )

Put 2 medium  TCS in the cargo bayt and extended them and relaunched with the cargo bay closed.

Craft exploded at 1926 m/s and 3021 K ( effectively the same result as the first test with no cooling )

Ran test again with the cargo bay open.

This time the TCS fell apart due to aero forces even though they were inside the bay.

Conclusions:

Radiators and TCS do not appear to mitigate shock heating.
Radiators are Robust however and will survive re-entry.

I did actually run each test several times.  I've just posted typical results
 

[Fearless Son]

On 2/7/2018 at 7:50 AM, NewtSoup said:

Can they also be used to mitigate re-entry heat?

Not by much.  The deployable panels will be destroyed by any atmospheric forces strong enough to cause enough aero-heating to make them necessary, and since the panels only work while deployed they just become dead weight while attempting reentry.  

However, I have included the small fixed panels in spaceplanes.  Mostly because they are light, and the planes tend to heat up while on ascent.  I like having a way for them to bleed off that extra heat once they exit the atmosphere, since they can no longer air-cool their engines.  I suppose I could just wait until they radiate out naturally, but I like option to use them for suborbital flights, and starting a reentry descent with a bunch of heat built up from the ascent makes me nervous.  This goes especially for Whiplash-driven spaceplanes, which tend to have a more aggressive ascent profile than R.A.I.P.I.E.R.-driven spaceplanes due to them needing to build up more speed in thicker parts of the atmospheric where more heating happens.

1 hour ago, NewtSoup said:

Craft exploded when the cockpit overheated at 1726 m/s and 2865K - actually slower and at a lower temperature than without the radiators.  However the aim was to make the craft run faster and cooler not explode sooner.  Suspect Drag from the radiators was slowing the craft while still allowing heat to build to the point where the cockpit failed.  

That would be it.

One does have to be careful about their placement.  I keep them on the dorsal side of the plane, toward the back and well away from any dragging edges.  The heat-flux works both ways, and placing them where they would be exposed to more atmospheric heating makes the plane heat up faster, which is the opposite of what I want.  There is obviously a balance because they only bleed heat from the parts they are directly attached to, so any heat elsewhere in the craft has to travel through each part between them via contact convection to be dispersed, so it cannot be too far from the center of heat build up.  

Edited February 15, 2018 by Fearless Son

[NewtSoup]

47 minutes ago, Fearless Son said:

One does have to be careful about their placement.  I keep them on the dorsal side of the plane, toward the back and well away from any dragging edges.  The heat-flux works both ways, and placing them where they would be exposed to more atmospheric heating makes the plane heat up faster, which is the opposite of what I want.  There is obviously a balance because they only bleed heat from the parts they are directly attached to, so any heat elsewhere in the craft has to travel through each part between them via contact convection to be dispersed, so it cannot be too far from the center of heat build up.  

The reason I'd placed the fixed radiators on the cockpit was because as far as I understood the fixed ones only cool the component they are attached to while TCS will draw heat from the entire structure.

My minmus base has a module of 4 large TCS for cooling the drill module ( 2 large drills)  and the ore processing module ( a single convertotron 250 )  
 

Incidentally 4 large TCS should be overkill for 2 drills and a large ore processor but if you time warp the base and then return to "normal" time the TCS glow red hot and spark and show temperature bars.  They do eventually cool down again.  This I think is a bug with the heating calculations on time warp. 

[AngrybobH]

Stock thermal management systems are mostly useless except for the case of ISRU operations. There does seem to be a bug when time warping or just loading the craft/base that has a lot of heat load. Sometimes I have to click the "ignore max temp" in the alt-f12 menu just to load a craft. If you are looking for better thermal management systems, look at near future and/or KSPI-E. They both have some nice options like a large reaction wheel that is also a radiator that performs well in atmosphere.  I haven't tested either mods thermal systems for mitigating shock heating. If they don't, I would suspect KSP has a buggy heat code.

[Archgeek]

6 hours ago, NewtSoup said:

@Archgeek - here is the post I was constructing - you may still want to re-run your old experiments.  I have not exactly been thorough.  But it was fun hooning Jeb and Bill down the runway to almost certain death every time.

Well shoot-shart, 'wish I'd caught this earlier this evening.  'Guess I'll be doing 'em Friday night, then.

Amusing write-up, though.

[Geschosskopf]

On 2/14/2018 at 1:55 PM, NewtSoup said:

Ive been doing just that. Was writing up but had to go to work.  Spoiler - you now get shock heating at all altitudes.  TCS and Radiatiors do not mitigate shock heating. 

Radiators should NEVER have been helpful while actually burning through an atmosphere.  In fact, they should be counterproductive.  It sounds like KSP has finally fixed this bug  

See, heat only flows from hotter things to colder things.  Thus, radiators only work if they're hotter than their surroundings.  When burning through an atmosphere, the flames outside are hotter than the ship.  And radiators, due to their surface area - to - mass, are just as good at absorbing heat from hotter surroundings as they are at radiating heat away into colder surroundings.  Thus, sticking a radiator into reentry plasma would pump heat into the ship, not radiate it away.  I suspect this is why your test ship exploded sooner when you added radiators to it.

As for how heat flows within a ship and various radiators work, that's kinda complicated.  Inside a ship, heat flows from hotter parts to colder parts via conduction, until every part has the same temperature.  But this heat flow depends on the the specific heats and masses of the parts involved, as well as their surface area-to-mass ratios, so the speed of heat propagation through a ship varies with the parts involved.  Also, part heat is divided into 2 types, surface and core, which function semi-independently, again based on the thermal properties of the parts.  Surface heat tends to move around a lot faster than core heat, and (IIRC) surface heat can exceed the explosion temperature of the part for a while---it's the core temperature (I think) that causes explosions.

Anyway,. fixed radiators are only "connected" to the part they're attached to, so indeed only really apply any cooling to that one part.  The heat from other parts only gets into the radiators and then out of the ship by the natural conductive processes, so fixed radiators have more impact on the more mobile surface component of part heat.  Surface heat eventually raises core temperatures so getting rid of it sooner keeps core temps lower overall.  But surface heat comes mostly from sunlight and atmospheric friction.  As noted above, radiators should be counterproductive during atmospheric friction so the main good of fixed radiators is on the shady side of things close to the sun, or on the underside (in the shade) of nuke-powered, low-speed (as in non-flaming) aircraft whose non-stock nuke jets get hotter than stock LV-Ns.  Such as the nuke jet in Mk2 Expansion.  But they can survive flight speeds and cost zero EC/sec.

Retractable radiators have a ship-wide plumbing system for coolant that costs EC/sec to pump around.  But it goes to every part of the ship and reduces core part heat directly.  Thus, heat from all parts  moves to the radiators quickly, so these radiators frequently become quite hot themselves, which is a good thing.  The greater the temperature difference between the radiator and its surroundings, the faster heat leaves the radiator into the surroundings.  This is another advantage of retractable radiators, as fixed radiators rarely get more than luke warm due to the comparatively slow rate of heat flow into the part they're attached to.

All in all, these days radiators seem only necessary anymore for ISRU, which is also a good thing.  The whole thing we had a while back about nuke engines running so hot was pretty bogus IMHO.  Those engines are designed so that the fuel is also the coolant.  Such engines ONLY work by transferring heat to the fuel, causing it to expand out the nozzle and carry the heat away from the ship.  Having the engine get dangerously hot is therefore a sign of a very faulty design   And hydrogen has one of the highest specific heats there is, so it can absorb a LOT of heat, which makes it excellent for this purpose.

[Archgeek]

5 minutes ago, Geschosskopf said:

Anyway,. fixed radiators are only "connected" to the part they're attached to, so indeed only really apply any cooling to that one part.  The heat from other parts only gets into the radiators and then out of the ship by the natural conductive processes, so fixed radiators have more impact on the more mobile surface component of part heat.

[...]

Retractable radiators have a ship-wide plumbing system for coolant that costs EC/sec to pump around.  But it goes to every part of the ship and reduces core part heat directly.  Thus, heat from all parts  moves to the radiators quickly, so these radiators frequently become quite hot themselves, which is a good thing.  The greater the temperature difference between the radiator and its surroundings, the faster heat leaves the radiator into the surroundings.  This is another advantage of retractable radiators, as fixed radiators rarely get more than luke warm due to the comparatively slow rate of heat flow into the part they're attached to.

[...]

The whole thing we had a while back about nuke engines running so hot was pretty bogus IMHO.  Those engines are designed so that the fuel is also the coolant.  Such engines ONLY work by transferring heat to the fuel, causing it to expand out the nozzle and carry the heat away from the ship.  Having the engine get dangerously hot is therefore a sign of a very faulty design   And hydrogen has one of the highest specific heats there is, so it can absorb a LOT of heat, which makes it excellent for this purpose.

I think that was at some point adjusted so that neighbor parts one connection out are also cooled.  I'm utterly curious about the skin/internal bit, though.  I knew about it, but I hadn't thought about which pool fixed radiators grab heat from.

[...]

Yup, radiative heat flux goes up with surface area and the fourth power of the temperature in Kelvins.  (I remember this made the Rhino nozzle surprisingly capable as a passive radiator -- the whole ship would develop a dull glow, but nothing would explode.)

[...]

Yeah, it would've been much cooler if we had NTRs where you could switch the core on and off, and had to have radiators to cool the core while it spins up (or hit the throttle and get lower ISp (and thus thrust) from the reduced core temp until it finishes) and when a burn ends (or you could turn the core off and let the fuel chill it down, again at the cost of reduced ISp while the burn tails off).  I just love the image of a nuke craft floating placidly in orbit, extending its radiators like sails that start to glow angrily, then doing its burn, after which the radiators flare again before cooling enough to retract.

[Geschosskopf]

15 minutes ago, Archgeek said:

I think that was at some point adjusted so that neighbor parts one connection out are also cooled.  I'm utterly curious about the skin/internal bit, though.  I knew about it, but I hadn't thought about which pool fixed radiators grab heat from.

Something like that.  Also, the fixed radiator makes the part it's attached to relatively cool compared to surrounding parts, which increases the rate of heat flow into that part from its neighbors.

 

15 minutes ago, Archgeek said:

Yup, radiative heat flux goes up with surface area and the fourth power of the temperature in Kelvins.  (I remember this made the Rhino nozzle surprisingly capable as a passive radiator -- the whole ship would develop a dull glow, but nothing would explode.)

This is also why engines on the bottoms of landers can explode during re-entry.  That nozzle's surface-to-mass ratio absorbs heat easily from the plasma.

 

15 minutes ago, Archgeek said:

Yeah, it would've been much cooler if we had NTRs where you could switch the core on and off, and had to have radiators to cool the core while it spins up (or hit the throttle and get lower ISp (and thus thrust) from the reduced core temp until it finishes) and when a burn ends (or you could turn the core off and let the fuel chill it down, again at the cost of reduced ISp while the burn tails off).  I just love the image of a nuke craft floating placidly in orbit, extending its radiators like sails that start to glow angrily, then doing its burn, after which the radiators flare again before cooling enough to retract.

Yeah, really and truly, LV-N burns should be rather complex.  Instead of a step function of zero-->full thrust-->,zero, it should be a bell curve with low thrust and Isp at both ends.  Or, as you say, use radiators at both ends as the core temp is rising to and falling from optimum.  Just definitely NOT radiators in the middle of the burn, as that's when the fuel would be cooling the engine to perfection.  But OTOH, that seems a bit overly complex from a coding standpoint to do all that for just 1 engine, when you can instead simply average out the thrust and Isp values to take the ends into account, and then let the engine use the same system as all the others.  After all, you're going to apply the same impulse to the rocket either way.

[Archgeek]

6 minutes ago, Geschosskopf said:

 that seems a bit overly complex from a coding standpoint to do all that for just 1 engine, when you can instead simply average out the thrust and Isp values to take the ends into account, and then let the engine use the same system as all the others.  After all, you're going to apply the same impulse to the rocket either way.

Curiously, it's rather not, relative to what they've already got.  Back when they fixed the aero and made nukes LF-only, they also fixed Isp -- Isp used to modify fuel flow, with engines producing constant thrust at any pressure.  That was more than a bit wrong.  Now, fuel flow is constant, and engine thrust varies with ISp, which varies with pressure, hitting full thrust in Vacuum.  So when a nuke's still in atmo and not getting it's full rated 900s, that's actualized by it not getting its full rated 60kN of thrust for the same amount of LF burned.  So it turns out if you're running your NTR suboptimally, you'd actually apply less impulse for the same burn time|fuel cost.

So since they're already got a system that varies Isp by a pressure curve, we can hook the same code to vary it with NTR core temp.  Just have an internal temp that ramps up when you turn the reactor on, have an internal flux that rises alongside it, and have Isp vary from zero to max based on that temperature.  The internal flux will blow things up if not dealt with, so there's your need for radiators, and you can set the internal flux to vary from its calculated amount based on core temp to zero as fuel flow varies from 0 to 100%  I suppose you would still need to respect the pressure curve, though...perhaps tell that bit that your max Isp is the current core-temp-based one?

[Geschosskopf]

6 hours ago, Archgeek said:

 So it turns out if you're running your NTR suboptimally, you'd actually apply less impulse for the same burn time|fuel cost.

Sure, I agree   But I was using "same impulse applied" in the sense of comparing 2 rockets with different engines doing the same burn.  IOW, same starting orbit, same final orbit.  The end result is the same either way, so why overly complicate the process of getting there?

 

6 hours ago, Archgeek said:

So since they're already got a system that varies Isp by a pressure curve, we can hook the same code to vary it with NTR core temp.  Just have an internal temp that ramps up when you turn the reactor on, have an internal flux that rises alongside it, and have Isp vary from zero to max based on that temperature.  The internal flux will blow things up if not dealt with, so there's your need for radiators, and you can set the internal flux to vary from its calculated amount based on core temp to zero as fuel flow varies from 0 to 100%  I suppose you would still need to respect the pressure curve, though...perhaps tell that bit that your max Isp is the current core-temp-based one?

I suppose it depends on how well you can throttle the fuel flow.  The cooler the core is, the less fuel (coolant) flow it needs to stay in the safe range.  Thus, while the core is below optimal temperature, you don't have to pour fuel out at the full-throttle rate, which would offset to some extent the lower efficiency at that time.

I understand that in real life throttling liquid fuel rockets is non-trivial and many engines can't do it at all.  Turbopumps spinning at ridiculous speeds and all that.  Still, some engines can throttle, and it's my understanding that NTRs don't need flow rates as high as combustion engines, so maybe throttling them would be easier.

But this brings up the fact that LFO engines aren't really step functions, either (although they're closer than NTRs).  They also have a bit of a bell curve as their turbopumps spin up and then coast down.  So if you're going to make a bell curve system for NTRs, then you should honestly also apply it to LFO engines.  However, this again brings up the question of "why bother?"  The result of any burn is changing the rocket's trajectory by the amount required to perform the desired maneuver.  That trajectory change is the end product of whatever system under the hood is accounting for fuel, so that the player can get his rocket to the intended destination.  As long as that happens and the fuel accounting is within acceptable tolerances, why load the CPU with a bunch of extra calculations that really have no effect on the bottom line?

[NSEP]

Too cool down from using ISRU and Nuclear Thrusters? I tried using them for re-entry, but i don't think the actually work well during re-entry.

[NewtSoup]

Regarding Nuclear engines, the simple reasoning to my brain is - 

Nuclear Engines would be prone to overheat when NOT being used if the reactor is turned on.

Engines which are spun down need their TCS deployed

Engines under full thrust are cooled in whole or in part by propellant being passed over the reactor.

To make it simple on the coding the reactor can be on / off. 

The exact amount of cooling required can then be inversely proportional to the amount of propellant being passed over the reactor.

[XrayLima]

I'm intrigued by LV-N thermal behavior.  I would love to see a basic mod that adjusted LV-N behavior simplistically to follow the following rules.

1. LV-N's have two heat parameters, internal and external.
2. Internal temperature builds over a time constant until it tends to a max temperature.over say 90 seconds  This is greater than the max temp of the engine.  Over this period Internal temp rate of rise is moderated by Internal Temp x (1/Throttle).  A bit of maths needed here!
3. External temperature equates to External temp=Internal Temp x (1/Throttle).  Again heat gain/loss occurs over a time constant  Therefore at 100% throttle, external engine temp is constant.  At 0% throttle, engine external heat will rise over the time constant until overheat.
4. Thrust = (Int temp / Int Temp Max) x Throttle.  Therefore thrust scales with reactor core temp. Fuel flow however scales with thrust, therefore low temp engines are very inefficient.

This quick thought would result in engines that on start need cooling unless you want to run cold and inefficient, and at a steady temp when at 100%.  You could even run them hotter than design for increased thrust for set fuel flow, but closer to the max temp.  I have no idea how to make this a mod, and the thought needs refinement  but this actually feels doable somehow.  And brings back a realistic need for cooling.

Thoughts??

[AngrybobH]

1 hour ago, XrayLima said:

I'm intrigued by LV-N thermal behavior.  I would love to see a basic mod that adjusted LV-N behavior simplistically to follow the following rules.

1. LV-N's have two heat parameters, internal and external.
2. Internal temperature builds over a time constant until it tends to a max temperature.over say 90 seconds  This is greater than the max temp of the engine.  Over this period Internal temp rate of rise is moderated by Internal Temp x (1/Throttle).  A bit of maths needed here!
3. External temperature equates to External temp=Internal Temp x (1/Throttle).  Again heat gain/loss occurs over a time constant  Therefore at 100% throttle, external engine temp is constant.  At 0% throttle, engine external heat will rise over the time constant until overheat.
4. Thrust = (Int temp / Int Temp Max) x Throttle.  Therefore thrust scales with reactor core temp. Fuel flow however scales with thrust, therefore low temp engines are very inefficient.

This quick thought would result in engines that on start need cooling unless you want to run cold and inefficient, and at a steady temp when at 100%.  You could even run them hotter than design for increased thrust for set fuel flow, but closer to the max temp.  I have no idea how to make this a mod, and the thought needs refinement  but this actually feels doable somehow.  And brings back a realistic need for cooling.

Thoughts??

KSPI-E has a couple of nuke engines. One is a replacement for the LV-N(you can still use the original unmodified, though). Heat management becomes the more important factor in thermonuclear power and ISP is tied to reactor temp. You don't get a whole lot of control, unfortunately, over the throttling of the thermal systems but it does most of those things automatically.

[qzgy]

I'm not gonna comment on actual thermal capabilities, I do use them kinda often - mainly for aesthetic purposes. By now I only really use the TCS for mining. Much more convienient with the place anywhere behavior.

I will say though - while burning on a craft with nukes I did see some mild warming up. Not near exploding levels but warm.

[Archgeek]

Thermal analysis forthcoming tonight -- I've already updated the methodology (using stock hyperedit this time to save some time) and whipped up the test designs (stationary opposing nukes).  I've thrown in some very stupid passive options this time around, like the kerbodyne adapter, and some girders.

As a curious bit of preliminary data, I've noted that the internal flux has gone up a lot from the old value, but I think thermal mass must've gone up along with it, as heating seems...slower.

 

[Geschosskopf]

6 hours ago, NewtSoup said:

Regarding Nuclear engines, the simple reasoning to my brain is - 

Nuclear Engines would be prone to overheat when NOT being used if the reactor is turned on.

Engines which are spun down need their TCS deployed

Engines under full thrust are cooled in whole or in part by propellant being passed over the reactor.

To make it simple on the coding the reactor can be on / off. 

The exact amount of cooling required can then be inversely proportional to the amount of propellant being passed over the reactor.

i dunno.  Seems to me, the only reason you'd ever need radiators with an NTR is if you can't throttle the fuel flow much if at all, and/or the reactor's rate of temperature change is very slow.

Let us assume the Boffins designed the NTR so that when its reactor is cranking out its max heat release rate, and the fuel is moving at its max flow rate, the engine is producing both its max thrust and max Isp, AND the fuel is carrying away enough reactor heat to keep the reactor below its explody temperature indefinitely.   It might get a bit toasty but reaches a certain temperature and stops there, well below its material limits.  Thus, no radiators needed when the engine is operating in this regime.  And it really makes no sense to design the engine any other way.  If the reactor can get hotter than the fuel pumps can cool, you have too much reactor or too little fuel flow.

So, if the fuel system can provide adequate cooling (as in not needing radiators) when the reactor is as hot as it can get, then keeping the reactor cool at lower temperatures is no problem for it.  Thus, the engine has no keep-it-from-exploding need for radiators at all, regardless of reactor output.  But that's only considering the internal workings of the engine.  We also have to worry about fuel mileage, which is the whole point of using NTRs to begin with.

If the fuel flow can be controlled in an essentially analog manner, then it can be scaled to match the heat output of the reactor at any point on its temperature curve and there's still no physical need for radiators.  And if the reactor's temperature increases and decreases fast enough (which depends on the design of the reactor), flowing fuel to produce less-than-max thrust and/or Isp might be of insignificant consequence.  OTOH, if you're stuck with always flowing fuel at or near max rate, and/or the reactor heats and/or cools slowly, then spewing fuel for reactor cooling in non-optimum temperature regimes will get expensive, not only because most of the reaction mass isn't getting as much energy as it can take, but also because excessive fuel flow will slow the heating of the reactor to max temp and speed its cooling at the back end, so the non-optimal regimes last longer.  In this case, providing whatever pre- and post-optimum cooling is necessary with radiators would be better.

HOWEVER, then the radiators would be scaled (within mass trade-offs) to handle all the non-optimal regimes.  In which case, the actual time when you're flowing fuel would be wholly within the optimal regime or close enough, so then the thrust/Isp output would really be a step function or close enough.  In which case the NTR is the functional equivalent (as far as the game is concerned) of an LFO engine, and there's no need to change the underlying code.