[1.0.x] Heat Management Development - 1.0.4 Beta version available

[drdeath]

I did activate heat dissipation, but I couldn't entirely figure it out. It showed an additional -28 something to -28000-something heat flow when activated but the temperature kept rising at about the same rate. Could that be because I didn't have radiator panels? Was it supposed to work like an active heatsink with NH3 circulation and stuff? It did't show any NH3 content, is that the expected behaviour?

About game balance, as I wrote, it felt completely comfortable, the engines couldn't overwhelm it easily but it didn't feel cheatlike either. I ran it up to about 1650 degrees, the thermal washer held and worked as expected.

About the visuals, I liked them. The isolator could be a little thinner, but IIRC you just used the drone core model there for now. Here's some shots for you (don't mind the flight attitude, mechjeb was making an ass of himself again): http://imgur.com/a/h1qNo

[Randazzo]

I just stuck the module on the heatsink for now essentially for testing purposes, with radiators it would theoretically work in the same fashion. It wasn't supposed to have NH3 in it or anything like that yet.

Thanks for the feedback!

[drdeath]

As I said, I couldn't figure out exactly how that module is supposed to work. I'm not an expert for the KSP heat system like you neither. I can figure out most of the stuff, but what "Int Flux" is supposed to mean is something I simply don't know. I'll gladly test it again if you tell me what it is supposed to do and what I'm supposed to see when it's active as opposed to inactive.

[Randazzo]

As I said, I couldn't figure out exactly how that module is supposed to work. I'm not an expert for the KSP heat system like you neither. I can figure out most of the stuff, but what "Int Flux" is supposed to mean is something I simply don't know. I'll gladly test it again if you tell me what it is supposed to do and what I'm supposed to see when it's active as opposed to inactive.

Oh I'm no expert, but thanks for the compliment

The internal flux is just what magically removes the heat and sends it to the fifth dimension underground on Phobos. It drops off over 1000 degrees because of some code Roverdude implemented for the ISRU system throttling itself at high temps, and this re-uses that module. If I were to make active radiators, they would work this way as well, so I thought it would be a reasonable test just to apply the function to the heatsink itself to see how it worked out.

Most engines have internal flux as well, but they generate heat with it.

I wasn't implying you did anything wrong, and I just wanted your impressions to refer to as I move forward with the parts.

Edited June 20, 2015 by Randazzo

[drdeath]

Ah, ok. So I guess "Int Flux" is supposed to be some measure of "heat generated by this module" and by making it negative, it turns into "Extra heat defeated by this module"

[Randazzo]

BETA Version for 1.0.3 is available on DROPBOX

This includes updated models and active radiators.

[drdeath]

BETA Version for 1.0.3 is available on DROPBOX

I'll give it a spin this weekend. Busy week, haven't launched a single rocket yet

[Randazzo]

I'll give it a spin this weekend. Busy week, haven't launched a single rocket yet

Thank you sir, be aware the Mk2 parts aren't in there yet. I do have a radiator put together for the MK2 frame though, but 1.0.3 happened.

[Terwin]

I suspect you are aware of this already, but in case you are not, the AHMS and large radiators attached to my Atomic Age lightbulb engine did not heat up or use any ammonia before the engine burned up at 3000 kelvin when I updated to 1.0.4.

I suspect that it is more related to the engine not transferring heat properly then the radiator parts, but though it might be useful for you to be aware of.

(note: in a newer design the parts connecting the stock active radiators are heating up more than the heat sink attached directly to the engine, so that also points to a heat-transfer failure in the engine itself)

[Randazzo]

I suspect you are aware of this already, but in case you are not, the AHMS and large radiators attached to my Atomic Age lightbulb engine did not heat up or use any ammonia before the engine burned up at 3000 kelvin when I updated to 1.0.4.

I suspect that it is more related to the engine not transferring heat properly then the radiator parts, but though it might be useful for you to be aware of.

(note: in a newer design the parts connecting the stock active radiators are heating up more than the heat sink attached directly to the engine, so that also points to a heat-transfer failure in the engine itself)

Is this with the beta version available in this thread or the 1.0.2 release version?

[Terwin]

Originally I was using the 1.0.2 version.

I moved the ksp\Game Data\Heat Management folder to ksp\hide\Heat Managment and copied the folder from drop-box in it's place.

after restarting (and saving with alt-F5) I re-deployed the radiators (may have become undeployed due to changing versions), topped off amonia, and fired up the engine (already around 1500k with 3500 fuel left to burn)

@ 2550K engine temp: (skin temp 503.0!! this looks like the problem, Atomic Age engine needs conduction to skin, only 1.0.2 version is available at the moment)

Radiators: 419.1 temp, 419.4 skin temp

AHMS: 388.1 temp 385.7 skin temp, 550 ammonia, pump still off

At Engine temp 2904 (skin temp 548.5, skin to int flux: still 1.39!)

Radiator temp: 460.5 skin: 460.8

AHMS: 415.0 Skin: 412.1

Apparently the engine actually goes over 3300K before exploding(if not smashed into the Mun), but I was not watching my altitude and it was smashed to bits around that time

(still 550 unused ammonia with the pump off and the engine skin temp finally topped 600, but only by a degree or two before the crash)

Design notes:

4x Large Radiator Pannels attached near the top of the engine and engine attached to 2.5M AHMS with 550 ammonia

Also, landing struts consisting of Long I-beams are attached to engines, everything else is above the AHMS

Summary: Looks like the (version 1.0.2) engine would have melted internally and exploded with the skin temp barely getting up to a point where the heat management systems would even notice it.

- - - Updated - - -

Just noticed the 'Activate' button on the radiators.

Re-running the test with the radiators both extended and activated(Burning prograde this time):

@2500 engine temp(less than 350 skin temp) radiators are at 208.8 and 209.0 skin temp Others about the same

@2900 engien temp(366.3 skin temp) Radiators @ 192 and 193.2 skin temp AHMS: 402.6 with 396.9 skin temp

@3300 engien temp(381.6 skin temp, Skin to Int Flux still 1.39)

Radiators: 185.1 and 186.5

AHMS: 436.1 and 429.4

@3500 and engine failure(skin temp still under 400):

Radiator temp: 175.3 and 175.7 skin

AHMS: 453.0 and 444.2 skin

Summary: looks like activating the radiators cause them to cool off but not to suck any additional heat form anything but the skin they are attached to.

note: apparently you can still get heat info off something not still touching your craft, and when detached form anything, active radiator panels cool off quickly

Final note:

The stock radiators(on a different ship) would quickly heat up to the same temp as the engine(same engine) when activated(and the engine gets over a certain temp), presumably pulling heat from the interior of the engine as the engine slows it's heat gain and the skin to Int flux stays at 1.39.

(hmm, stock panels seem to stop keeping up around 1600 and looks like they would stop gaining heat around 1700 due to intake and radiation balancing out, even though the engine continues to heat up)

Edited June 24, 2015 by Terwin

[Terwin]

Hmm, once the radiators get down to 4 degrees K they do seem to suck quite a bit of heat from the non-runnning engine...

[Randazzo]

Hmm, once the radiators get down to 4 degrees K they do seem to suck quite a bit of heat from the non-runnning engine...

Thanks for the detailed testing, I sincerely appreciate the help.

The dropbox link has just been updated with some new balance variables introduced. I have balanced the AHMS units to the Atomic Age engines (with the exception of the multi-node units, those are balanced for stock LV-Ns), and while it doesn't seem really possible to maintain indefinite burns any longer, they do support around 10 minutes full burn on the Nuclear Lightbulb. Adding HM radiator panels should extend this further. All active parts now use the stock radiator module to draw heat regardless of the conductive resistance of the engine alongside the original heat dissipation properties.

The heatsinks+radiator panel combos are now intended for stock use. Using them with Atomic Age engines will likely result in rapid failure. The AHMS units can still be used if so desired, and will essentially provide for indefinite burns.

DOWNLOAD: DROPBOX

---

Some detailed testing information with a 2.5m AHMS unit and 4 large HM Radiator panels with the Nuclear Lightbulb:

Prior to initiating the burn, I accelerated to 1000x timewarp to allow the ship to reach heat equilibrium. All parts maintained a temperature of 251.5 +/- .4

@~6 minutes (5 minutes 57 seconds)

Lightbulb

• 
  [*=1]Temp: 2004.7
  [*=1]Skin Temp: 702.9
  [*=1]Cond Flux: -30422.76
  [*=1]Int Flux: 79590.52 (Radiator panels and AHMS unit actively drawing heat at this point)
  

AHMS (pump active)

• 
  [*=1]Temp: 609.2
  [*=1]Skin Temp: 600.3
  [*=1]Cond Flux: 13868.62
  [*=1]Int Flux: 150000.00 (exceedingly high, but designed specifically for the Lightbulb)
  

LRG Rad Panels

• 
  [*=1]Temp: 525.8
  [*=1]Skin Temp: 525.5
  [*=1]Cond Flux: 3877.51
  [*=1]Int Flux: -1662.50 (Sum of active draw + active heat dissipation = net loss )
  

@~9 minutes (9 minutes 10 seconds)

Lightbulb

• 
  [*=1]Temp: 2505.8
  [*=1]Skin Temp: 744.5
  [*=1]Cond Flux: -41356.56
  [*=1]Int Flux: 79590.52
  

AHMS

• 
  [*=1]Temp: 608.3
  [*=1]Skin Temp: 589.8
  [*=1]Cond Flux: 31949.32
  [*=1]Int Flux: 150000.00
  

LRG Rad Panels

• 
  [*=1]Temp: 561.6
  [*=1]Skin Temp: 561.5
  [*=1]Cond Flux: 2127.09
  [*=1]Int Flux: -1662.50
  

@~13 minutes (12 minutes 41 seconds)

Lightbulb

• 
  [*=1]Temp: 3002.3
  [*=1]Skin Temp: 785.1
  [*=1]Cond Flux: -52141.74
  [*=1]Int Flux: 79590.52
  

AHMS (pump active)

• 
  [*=1]Temp: 609.6
  [*=1]Skin Temp: 586.4
  [*=1]Cond Flux: 44839.74
  [*=1]Int Flux: 150000.00
  

LRG Rad Panels

• 
  [*=1]Temp: 573.5
  [*=1]Skin Temp: 573.6
  [*=1]Cond Flux: 1615.01 (I believe this is dropping off due to higher rad panel temp and lower AHMS skin temp)
  [*=1]Int Flux: -1662.50
  

@~18.5 minutes (18 minutes 31 seconds)

Lightbulb

• 
  [*=1]Temp: 3401.0
  [*=1]Skin Temp: 816.7
  [*=1]Cond Flux: -60797.73
  [*=1]Int Flux: 79590.52
  

AHMS

• 
  [*=1]Temp: 610.8
  [*=1]Skin Temp: 581.4
  [*=1]Cond Flux: 53803.52
  [*=1]Int Flux: 150000.00
  

LRG Rad Panels

• 
  [*=1]Temp: 576.1
  [*=1]Skin Temp: 576.1
  [*=1]Cond Flux: 1548.06
  [*=1]Int Flux: -1662.50
  

Engine Failure @ 20 minutes 24 seconds

---

The AHMS and Radiator Panels still have plenty of breathing room, so it would technically be possible to make the lightbulb indefinitely sustainable by further raising the magic heat sucking number, but I'm pretty satisfied with a 20+ minute burn.

Edited June 25, 2015 by Randazzo

[Terwin]

Just as an FIY:

Using the second beta you provided, the radiators are keeping the lighbulb from heating as quickly as it did before, and are helping it cool off more quickly afterwards. (I have seen it over to 1200 degrees but I don't think it has gotten close to 2000 as of yet, even for burns over 2 minutes)

I am not 100% sure, but it looks like if a radiator is active but not extended, it continues to work as well as it did when extended.

It also seems that for most burns inside the Kerbin SOI 4 large radiators are enough to keep the AHMS from using coolant except near the tail end of the burn.

(not sure where you want that balance point, but in the past the AHMS used a lot more coolant in the same configuration)

Not sure if this is related but I tried to insta-cool at 1000x warp a couple times last night and it did not work(I do not remember trying this since I upgraded to 1.0.4, and I later noticed that I had a couple small nuclear generators running and producing heat in a different part of the ship, so there are several possible sources of this behaving differently, but I thought I would mention it just for the sake of completeness)

Thanks for your work, I am now much more confident that the upgrade to 1.0.4 will not cost me my Duna mission(I just hope I finish it before 1.1 is released...)

[Randazzo]

I am not 100% sure, but it looks like if a radiator is active but not extended, it continues to work as well as it did when extended.

Yes, since I'm just using stock modules, a radiator with active heat dissipation turned on will continue to function whether deployed or not. There is no way I know of to tie the activation to deployment without a plugin. It is possible to animate a converter module, but it loops continuously, so I'd have panels flapping around.

Not sure if this is related but I tried to insta-cool at 1000x warp a couple times last night and it did not work(I do not remember trying this since I upgraded to 1.0.4, and I later noticed that I had a couple small nuclear generators running and producing heat in a different part of the ship, so there are several possible sources of this behaving differently, but I thought I would mention it just for the sake of completeness)

I have noticed this too, it seems to be related to having any sort of active heat moving/negation going on. Shutting off all the panels seems to fix it.

Thanks for your work, I am now much more confident that the upgrade to 1.0.4 will not cost me my Duna mission(I just hope I finish it before 1.1 is released...)

You're welcome!

[RoverDude]

Yes, since I'm just using stock modules, a radiator with active heat dissipation turned on will continue to function whether deployed or not. There is no way I know of to tie the activation to deployment without a plugin. It is possible to animate a converter module, but it loops continuously, so I'd have panels flapping around.

Incorrect.

The stock radiators are designed to automatically shut down active transfer if they are not deployed. This implies you're using the stock deployable radiator module and the stock active radiator modules.

[Randazzo]

Incorrect.

The stock radiators are designed to automatically shut down active transfer if they are not deployed. This implies you're using the stock deployable radiator module and the stock active radiator modules.

He's referring to the active heat dissipation from Heat Management, not the stock radiator module. Since it's done via a negative temp modifier on a resource converter module, it remains in whatever state you set it to regardless of the panel deployment or the state of the stock radiator module.

I'm using some overlapping terms for different things, I realize. I was referring to ModuleResourceConverter when I said stock modules in the previous post.

Edited June 26, 2015 by Randazzo