[1.12.x] SystemHeat - a replacement for the CoreHeat system (October 9)

[Spaceman.Spiff]

7 minutes ago, si2504 said:

that's what I thought which is why I wondered why the coolant tanks show up the overheat icons...

I’m also very interested, where is that craft going, it looks cool....

[si2504]

1 minute ago, Spaceman.Spiff said:

I’m also very interested, where is that craft going, it looks cool....

eveeerywhere! haha. 5 Ion engines for propulsion, 4 Large Xenon gas tanks providing 118,775 dV

[panarchist]

4 minutes ago, si2504 said:

eveeerywhere! haha. 5 Ion engines for propulsion, 4 Large Xenon gas tanks providing 118,775 dV

How are you putting it in orbit?

[si2504]

1 minute ago, panarchist said:

How are you putting it in orbit?

In 4 split sections. Crew module, control module, Xenon tanks & solar panels and finally the rear main propulsion engine & triple radiators. RSS also so its quite fun building the rockets to lift such heavy payloads (mainly the tanks & engines).

[Inqie]

On 3/9/2021 at 6:43 AM, JadeOfMaar said:

This new issue is due to my expecting that users would have System Heat's generic extras patch that changes all supported ModuleResourceConverter into ModuleSystemHeatConverter. The errors point at my patch for a B9PS module that changes the options of an EC-consuming converter in the Dark Drive engine.

Am I misunderstanding this? Are the SystemHeat extras the cause or fix of this problem? I have all the extras installed (via CKAN) and can confirm they are in GameData, yet I am still getting a similar problem: https://pastebin.com/xMRVjPea 

[JadeOfMaar]

@Inqie I posted a quick fix to OPT Reconfig that cures the problem by no longer assuming that the extras patch is there, so now, having the patch causes the  problem. That patch will need to be updated to declare a mod let so that other mods can detect it. 

[Zacho]

Not sure if its a KA bug or systemheat. But on the Liberator and the Emancipator when you go to shut down the reactor the engine turns off, thats fine but you lose the ability to turn it back on. 

[aristurtle]

Hey, I added SystemHeat to an existing game and I'm trying to get the hang of it (love the UI, by the way). I had a couple questions:

So I've got this ship that I was using before:
https://imgur.com/r7yq4Kf

It has the gas core nuclear engine at the back and a power reactor in the middle. I separated the power reactor and the engine into two separate cooling loops:
https://imgur.com/tmtZHin

But I have some problems:
1) On the engine loop - I used to be able to provision way fewer radiators than the nuclear engine's thermal output would imply, by using the propellant to handle most of the thermal flux. By carefully starting the engine's reactor after the burn started and shutting it down before the burn ended, I'd only need those tiny radiators on the back there (at the cost of some wasted propellant at the start and end of the burn, and of course playing russian roulette with the main engine). I can't seem to do this anymore -- is this no longer a valid conops?

2) On the power reactor loop - that's the MX-1 reactor with four stock "Thermal Control System Medium" radiators. Used to be, those radiators would be sufficient for that reactor but now they're only rejecting 40kW each. The XR-500 from Heat Control weighs about the same and rejects 500kW. Am I missing something important? What is the stock radiator good for, now?

[Nertea]

On 3/16/2021 at 1:36 AM, Zacho said:

Not sure if its a KA bug or systemheat. But on the Liberator and the Emancipator when you go to shut down the reactor the engine turns off, thats fine but you lose the ability to turn it back on. 

Should be sorted in the next update.

On 3/19/2021 at 4:37 PM, aristurtle said:

1) On the engine loop - I used to be able to provision way fewer radiators than the nuclear engine's thermal output would imply, by using the propellant to handle most of the thermal flux. By carefully starting the engine's reactor after the burn started and shutting it down before the burn ended, I'd only need those tiny radiators on the back there (at the cost of some wasted propellant at the start and end of the burn, and of course playing russian roulette with the main engine). I can't seem to do this anymore -- is this no longer a valid conops?

I'm still trying to get this to where I want it to be. The considerations are:

1. You can't have instant response with a nuclear engine.
2. Operating an engine's reactor at lower power should still need radiators.
3. Higher power engines with low mass flow shouldn't be able to reject all their waste heat through the propellant flow. Going above ~1400s seems about the limit, so that applies to the Liberator, Emancipator and probably the Deliverance

I haven't quite got it sorted out yet. Next version should have what I think is a somewhat better model but *shrug* will need testing. Regardless you probably won't be able to do that with the Liberator anymore, you'll need at least some radiative capacity. 

On 3/19/2021 at 4:37 PM, aristurtle said:

2) On the power reactor loop - that's the MX-1 reactor with four stock "Thermal Control System Medium" radiators. Used to be, those radiators would be sufficient for that reactor but now they're only rejecting 40kW each. The XR-500 from Heat Control weighs about the same and rejects 500kW. Am I missing something important? What is the stock radiator good for, now?

The stock radiators seemed like they looked similar to the ISS TCS, so they were given similar operating parameters to the ISS TCS in terms of max useable temperature. As a result they are not useful for reactor radiators  and are really only appropriate for ISRU.

Edited March 21, 2021 by Nertea

[Nertea]

I've been pretty busy but working hard on 0.4.0. This will bring a number of key changes:

Nuclear transfer improvements

The EVA-based system has been shown to be tedious and will be removed and replaced with a much simpler system. If you have an engineer of sufficient level on board, you will be able to transfer waste/fuel through the standard stock methods. Nothing else required. This is effectively a streamlined version of the old NFE mechanic.

Nuclear engine improvements

The handling of nuclear engines has been tuned to provide what I hope is a more intuitive experience. The split between needed cooling and exhaust cooling will be more evident and a better progression through the low power to high power engines will be provided. 

Radiator convective transfer

 Radiator convective transfer has been added, which allows radiators to reject heat via convection in atmosphere as well as radiation. This is highly dependent on where you are and how fast you're going, but could be useful in increasing cooling on planets that have atmospheres. 

Naturally this comes with simulation capabilities to plan it. Here's a small UI preview. 

Probably still a week away from release but slowly squashing bugs.

 

[toric5]

12 hours ago, Nertea said:

I've been pretty busy but working hard on 0.4.0. This will bring a number of key changes:

Nuclear transfer improvements

The EVA-based system has been shown to be tedious and will be removed and replaced with a much simpler system. If you have an engineer of sufficient level on board, you will be able to transfer waste/fuel through the standard stock methods. Nothing else required. This is effectively a streamlined version of the old NFE mechanic.

Nuclear engine improvements

The handling of nuclear engines has been tuned to provide what I hope is a more intuitive experience. The split between needed cooling and exhaust cooling will be more evident and a better progression through the low power to high power engines will be provided. 

Radiator convective transfer

 Radiator convective transfer has been added, which allows radiators to reject heat via convection in atmosphere as well as radiation. This is highly dependent on where you are and how fast you're going, but could be useful in increasing cooling on planets that have atmospheres. 

Naturally this comes with simulation capabilities to plan it. Here's a small UI preview. 

Probably still a week away from release but slowly squashing bugs.

 

will the nuclear transfer improvements be available for parts not covered by systemHeat? I currently use the old NFE system to let my engineers do things like transfer solid fuel/refurbish ablator, wondering if I will be able to switch this over to the new system.

[Nertea]

1 hour ago, toric5 said:

will the nuclear transfer improvements be available for parts not covered by systemHeat? I currently use the old NFE system to let my engineers do things like transfer solid fuel/refurbish ablator, wondering if I will be able to switch this over to the new system.

It's a module like any other implementation, you can try using it in your context and see what happens. 

[Ozzy.R]

Hello there.  About the Cherenkov nuclear engine from Restock+. It doesn't seem to have System Heat support and I think it should. Is there a way I can add System Heat support myself? Is this forum the best place to be asking about that?

[Nertea]

4 hours ago, Ozzy.R said:

Hello there.  About the Cherenkov nuclear engine from Restock+. It doesn't seem to have System Heat support and I think it should. Is there a way I can add System Heat support myself? Is this forum the best place to be asking about that?

Support is being added in 0.4.0

[Ozzy.R]

2 hours ago, Nertea said:

Support is being added in 0.4.0

Aight, that's good to know. Thank you.

[panarchist]

On 3/24/2021 at 1:22 PM, Nertea said:

It's a module like any other implementation, you can try using it in your context and see what happens. 

I actually really liked that mechanic. I mean, it's not like my engineer has to frequently swap that out anyway, but I liked the idea of a refuelling actually requiring some time and effort.

[Nertea]

43 minutes ago, panarchist said:

I actually really liked that mechanic. I mean, it's not like my engineer has to frequently swap that out anyway, but I liked the idea of a refuelling actually requiring some time and effort.

It just wasn't fun, honestly. I tried playing with it and refuelling a tiny reactor - fine. Refuelling anything larger needed so many trips and clicks that it made me feel like I was babystitting MPLs or something equally mind-numbing. 

Anyways probably a lot of small bugs to deal with because of the new features, but here you go:

SystemHeat 0.4.0

• Reorganized some UI elements
• Added convection simulation to radiators
  • Convection can increase radiator efficiency in atmosphere
  • Altitude, body and velocity-based in a similar fashion to stock
  • New UI components in VAB to simulate situations to check convection on bodies at locations
• Redid nuclear transfer mechanics
  • Presence of an engineer of Level 3 on the vessel will allow nuclear transfer using the stock transfer system
• Changed how NTRs behave slightly
  • Internal 'exhaust' cooling is now exposed, previously it was implied
  • Engine thrust will reduce the amount of flux contributed to the system
  • Higher energy engines need a lot of extra radiation cap (up to 15% of engine power), lower energy don't need a lot (1-2%)
• Fixed engine module reactivation issue on fission engines
• Fixed several incorrect temperature curves on HeatControl radiators
• Changes pass specifiers in Extras patches to conform to the Extra's name
• Added support for Restock+ Cherenkov to NTR patches

[aristurtle]

I like the exhaust cooling! You can underspec the radiators on the nuclear engines quite a bit, but then one wrong move and the engine overheats, and shuts down, and this causes the propellant flow to stop so it heats up more and immediately melts down completely. It's a neat mechanic for those gas-core engines, I'm quite enjoying it.

One problem I'm having is that the heat simulation in the VAB doesn't seem to work for vacuum anymore; it displays NaN everywhere. Do I have something set up wrong?

Edited March 27, 2021 by aristurtle

[Nertea]

11 hours ago, aristurtle said:

One problem I'm having is that the heat simulation in the VAB doesn't seem to work for vacuum anymore; it displays NaN everywhere. Do I have something set up wrong?

I'm going to need more details, ran a very quick test and everything seemed fine (1.25m reactor, 4 small radiators in VAB, Kerbin situation, Vacuum button clicked).

[Gallert3]

Hello, I'm struggling with making a interplanetary spaceship using FFT and SH. I built the ship in the VAB, then  separated it to launch it. Now, I cannot get the engine and the radiators on the same cooling loop. (if i put the engine on loop 1, the radiators cannot use loop 1) I looked through here and the wiki, but couldn't find out what my problem is. Does anyone know how to solve this?

[yjchun]

Hello @Nertea and everyone.

I am playing with System Heat along with other compatible mods (Near Future, Far Future, Kerbal Atomics, ...) for some time and I am having a few issues I wish to be addressed.

1. There are not enough aero dynamic radiators which makes atomic motors and reactors practically not usable in atmosphere. There is only one small YF series for each fuselage size but you need 100ish such radiators to cool 1500kW Stubber Atomic Motor for example.

2. There is no thermal consumers other than radiators. I wish there is thermal electricity generator to make use of the heat.

3. Atomic reactor's efficiency is too poor. Here, atomic reactor includes thermal electricity generator which is its main role. But it seems most of reactor's heat is wasted instead of generating electricity. Or generating too much heat for its generated electricity. Convert-O-Tron, a small chemical factory, generate mere 40kW heat but a compact fission reactor is wasting a thousand kW heat after generating electricity. In RL, Nuclear submarines are (almost) closed system as well and I've read, discharged cooling seawater is not even boiling. Most heat is used by power generator.

4. It seems Heat energy is just added and converted to temperature increase linearly. I suggest heat increment should be inverse exponential because there is no perfect heat loop. I mean 0.4 version's new convection should be more evident, even in space and even more in atmosphere, and exponentially as temperature go up, and temperature and convection reach Equilibrium eventually. It just doesn't feel right that it goes up a few thousand degree too easy and too quickly.

5. At first it was very confusing about System Heat because there is Near Future System Manager and Legacy Heat parameters. Took some time to understand Legacy heat parameters (xFer value and max cooling) are not used. And with optional modules and patches, whenever something does go not as I expected, I had to try different option/patch modules thinking my setup was not right. It is good to have many choices. Just these combined, things were not straightforward.

Well, I am not complaining. I am very happy user of @Nertea's eco system and spending a few days now working on MK4 cargo shuttle. Much appreciations to @Nertea and just wish my suggestions be helpful.

Edited March 29, 2021 by yjchun

[Spaceman.Spiff]

44 minutes ago, yjchun said:

There are not enough aero dynamic radiators which makes atomic motors and reactors practically not usable in atmosphere. There is only one small YF series for each fuselage size but you need 100ish such radiators to cool 1500kW Stubber Atomic Motor for example.

You need Nertea's Heat Control mod which adds many powerful radiators.

https://forum.kerbalspaceprogram.com/index.php?/topic/112027-19x-heat-control-more-radiators-february-20th-2020/

[Nertea]

22 hours ago, Gallert3 said:

Hello, I'm struggling with making a interplanetary spaceship using FFT and SH. I built the ship in the VAB, then  separated it to launch it. Now, I cannot get the engine and the radiators on the same cooling loop. (if i put the engine on loop 1, the radiators cannot use loop 1) I looked through here and the wiki, but couldn't find out what my problem is. Does anyone know how to solve this?

Part loop assignments cannot be changed in flight, you can only change the ID of the loop. If you want to connect two loops (say, in your case), then you should put them on the same loop in separate ships. Docking the ships will them merge the loops. 

@yjchun Glad you enjoy the mod, happy to try to answer a few of these comments...

3 hours ago, yjchun said:

1. There are not enough aero dynamic radiators which makes atomic motors and reactors practically not usable in atmosphere. There is only one small YF series for each fuselage size but you need 100ish such radiators to cool 1500kW Stubber Atomic Motor for example.

You can use pretty much any static radiator in the atmosphere, don't limit to the conformal ones. However the stubber has nearly full-flow cooling as it is (relatively) low-power, and only needs something like 30 kW when running at full power. 

3 hours ago, yjchun said:

2. There is no thermal consumers other than radiators. I wish there is thermal electricity generator to make use of the heat.

I don't really intend to add this anytime soon. The simulator is not designed to handle things that would 'step down' heat, do effectively deal with multiple partial consumers a lot of work would need to be done to 'order' consumers correctly. 

4 hours ago, yjchun said:

3. Atomic reactor's efficiency is too poor. Here, atomic reactor includes thermal electricity generator which is its main role. But it seems most of reactor's heat is wasted instead of generating electricity. Or generating too much heat for its generated electricity. Convert-O-Tron, a small chemical factory, generate mere 40kW heat but a compact fission reactor is wasting a thousand kW heat after generating electricity. In RL, Nuclear submarines are (almost) closed system as well and I've read, discharged cooling seawater is not even boiling. Most heat is used by power generator.

Nah it's a bit too efficient. In a nuclear submarine, you have the entire ocean to use as a heat sink. Sub reactors are somewhere around 30% efficient, and they transfer their heat to the surrounding ocean. This is why in some cases you can detect a nuclear submarine through space-based thermal imaging techniques which is kinda cool. Space reactors have the same problem but worse because radiation sucks. I've modeled the efficiency as dependent on the technology in the reactor, some power conversion techniques are very bad, like thermionics and stirling engines (10-20%). Stepping up to the larger reactors I use some approximation of full-cycle Brayton turbines at ~30%. The fanciest reactors have up to 45% (iirc) efficiency and are modeled as futuristic MHD generators. No matter what though, you have vast amounts of waste heat from any thermal cycle. Compared to naval reactors (and commercial reactors) I've specced the space reactors as very high temperature (better for radiation), PWRs don't really exceed 600K but the stuff I've cribbed from achieves 800-1000K. 

4 hours ago, yjchun said:

4. It seems Heat energy is just added and converted to temperature increase linearly. I suggest heat increment should be inverse exponential because there is no perfect heat loop. I mean 0.4 version's new convection should be more evident, even in space and even more in atmosphere, and exponentially as temperature go up, and temperature and convection reach Equilibrium eventually. It just doesn't feel right that it goes up a few thousand degree too easy and too quickly.

I'm not completely sure what you mean here, can you rephrase? A couple comments based on what I understood.

• This mod is specifically designed to be deterministic rather than equilibrium based, for increased player predictability when planning missions
• Convection is simulated in a similar fashion to stock aerodynamic convection, for simplicity and for better feedback (e.g. when looking at re-entry effects and such). It should definitely not be evident in space ;).
• Simulation of heating rates is fairly correct in the sense that it uses basic thermal equations and assumes loops are made of sodium-potassium coolant (a good or a bad assumption depending on what part you're referring to). The thing is with the nuclear stuff, for space tech you're often dealing with massive thermal powers in tiny spaces with limited amounts of coolant volume. It will heat up very fast.
• One thing that is not well captured is how long it takes a reactor (or chemical process) to warm up - even accounting for very high power densities the rates I've used are probably 10x too high. This is very easy to change in config, but seems to cause a lot of player frustration so I've kept it quite forgiving.

4 hours ago, yjchun said:

5. At first it was very confusing about System Heat because there is Near Future System Manager and Legacy Heat parameters. Took some time to understand Legacy heat parameters (xFer value and max cooling) are not used. And with optional modules and patches, whenever something does go not as I expected, I had to try different option/patch modules thinking my setup was not right. It is good to have many choices. Just these combined, things were not straightforward.

 

Yeah unfortunately I don't think there's a world where I can replace everything that uses core heat so I've kept those UI things around to help with compatibility with such mods. Don't really have a better option, sadly ;( .

[yjchun]

5 hours ago, Spaceman.Spiff said:

You need Nertea's Heat Control mod which adds many powerful radiators.

https://forum.kerbalspaceprogram.com/index.php?/topic/112027-19x-heat-control-more-radiators-february-20th-2020/

@Spaceman.Spiff Yes and the only aerodynamic model in Heat Control is those small wraparound types and cooling capacity is too small to use for most atomic engines. To cool even a low tier atomic engine 'Neptune' 2000kw model in a 2.5m fuselage then you need more than 100 of them.

2 hours ago, Nertea said:

You can use pretty much any static radiator in the atmosphere, don't limit to the conformal ones. However the stubber has nearly full-flow cooling as it is (relatively) low-power, and only needs something like 30 kW when running at full power. 

Definitely cannot be running the engine at full throttle all the time and you don't need many seconds to melt down most atomic engines.@Nertea Thank you for explanations.

3 hours ago, Nertea said:

I don't really intend to add this anytime soon. The simulator is not designed to handle things that would 'step down' heat, do effectively deal with multiple partial consumers a lot of work would need to be done to 'order' consumers correctly. 

I feel sorry a lot of heat is just wasted away. If I were a space engineer, it would be my highest priority to develop something useful with those wasted energy. Btw, do 'order' of components matter? I found no difference of radiators' location.

3 hours ago, Nertea said:

• Convection is simulated in a similar fashion to stock aerodynamic convection, for simplicity and for better feedback (e.g. when looking at re-entry effects and such). It should definitely not be evident in space ;).
• Simulation of heating rates is fairly correct in the sense that it uses basic thermal equations and assumes loops are made of sodium-potassium coolant (a good or a bad assumption depending on what part you're referring to). The thing is with the nuclear stuff, for space tech you're often dealing with massive thermal powers in tiny spaces with limited amounts of coolant volume. It will heat up very fast.

I think the heat loop and atomic reactor itself should radiate some heat naturally, same way radiators work and also melt down process itself would use a lot of energy. Natural radiation, heating the ship itself, and melting down reactor, it would be difficult to design all these correctly, we could just say heat is 'convect'ed away in some way. Umm but it could be very small percentage and neglegible. It is difficult to argue technically but feel like things are fragile and lightweight when a few tones of huge space system melt down in just a few seconds.

4 hours ago, Nertea said:

Yeah unfortunately I don't think there's a world where I can replace everything that uses core heat so I've kept those UI things around to help with compatibility with such mods. Don't really have a better option, sadly ;( .

My beautiful ship is just covered with radiators. System Heat forces use of lots of radiators but radiator choice is farely limited. I wish to have more aerodynamic radiators and thermal generators.

 

 

[panarchist]

On 3/26/2021 at 9:42 AM, Nertea said:

It just wasn't fun, honestly. I tried playing with it and refuelling a tiny reactor - fine. Refuelling anything larger needed so many trips and clicks that it made me feel like I was babystitting MPLs or something equally mind-numbing. 

Ah - yeah, I only did it with a small reactor, so I can see how that would be a pain.

2 hours ago, yjchun said:

I feel sorry a lot of heat is just wasted away. If I were a space engineer, it would be my highest priority to develop something useful with those wasted energy. Btw, do 'order' of components matter? I found no difference of radiators' location.

That's actually one of the most complex issues in space travel, and a serious problem for any ship with a high energy density powerplant or propulsion system. There's just too much heat being produced that can't do anything really useful. The exception being electrical power generation from the waste heat of the propulsion system - but that only extracts 1/3 of the energy and the other 2/3 remains lower temperature waste heat that needs to be gotten rid of. Unfortunately in space you can't shed heat from the vessel via conduction or convection - only radiation, and space is the universe's best insulator. (hence why we "put" vacuum in our thermos bottles)