[1.12.x] SystemHeat - a replacement for the CoreHeat system (September 1)

[RoverDude]

39 minutes ago, Nertea said:

• Technical: all MKS modules have their own implementation in software. As I don't expect MKS to take SH integration into its own codebase, you would need to build 'mirrors' of the modules in SH that include features like converter slots and swapping.  That's a lot of work and very hard to maintain, as well as not likely not reverse-compatible with existing MKS saves. 
• Gameplay: MKS modules have their own 'way' of working. They operate like stock converters in that they have a pattern where they modulate their effectiveness according to temperature and don't have SH styles of hard cutoffs. There's also a whole set of efficiency and general MKS mechanics you'd have to look at . Overall, you'd have to develop a technical SH implementation of the MKS converters, as per the first bullet, and validate that works with the rest of MKS gameplay, which is nontrivial. That would include busy work like assigning relevant fluxes and temperatures to a whole swath of MKS parts. Now you basically own support for all of MKS' converters.  

 

FYI - MKS modules use stock thermal (for those that even use heat), and all converters derive from stock modules (ModuleResourceConverter/ModuleResourceHarvester)  -  we just include an option to swap out the inputs and outputs, which does not affect thermal)  - so the core guts can be accounted for in the same way stock converters / drills are, provided you're looking for things that implement the base class.   If there's something I'm doing weird with the stock modules that's causing a headache, by all means let me know if it opens the path to better interop.

[Nertea]

2 minutes ago, RoverDude said:

FYI - MKS modules use stock thermal (for those that even use heat), and all converters derive from stock modules (ModuleResourceConverter/ModuleResourceHarvester)  -  we just include an option to swap out the inputs and outputs, which does not affect thermal)  - so the core guts can be accounted for in the same way stock converters / drills are, provided you're looking for things that implement the base class.   If there's something I'm doing weird with the stock modules that's causing a headache, by all means let me know if it opens the path to better interop.

Yes, that is what would cause the challenge. You derive your converters from stock modules, I also derive the SystemHeat converters from stock values (and replace the thermal interface). To support your derived converters, I would need to layer on top of your derived classes, which poses some implementation challenges.  However, I'm more concerned about solving gameplay before solving that, interoperation is just software challenges, that do need to be traded against the time available for modding.

I don't think I could create an appropriate SH-driven user experience for MKS without a lot of work to match up the gameplay concepts, basically, which doesn't particularly motivate me to solve the software bits. Fundamentally MKS's use cases don't hugely benefit from the problems SH solves anyways. I suppose that could change, and probably would change with high demand. 

[dave1904]

Why would I want to increase the system volume?? I think I understand what it does but have not found a practical way of using it. At first I thought it might be useful for keeping heat in atomic engines for more consistent ISP when doing more smaller burns but the temperature still drops fast.  

Edited January 15, 2021 by dave1904

[Nertea]

On 1/10/2021 at 4:39 PM, danfarnsy said:

I caused problems for myself when I docked two ships with loops on the same loop ID. Even after undocking, changing loop IDs, and redocking, they were put back on the same loop. I haven't tested this rigorously.

There's some gnarly bugs here that are resulting in me needing to rewrite a lot of code but should be fixed in the next update. 

5 hours ago, dave1904 said:

Why would I want to increase the system volume?? I think I understand what it does but have not found a practical way of using it. At first I thought it might be useful for keeping heat in atomic engines for more consistent ISP when doing more smaller burns but the temperature still drops fast.  

Best use is to stop things from heating up so fast so you can go longer without radiators and 'pulse' your engines. 

[vardicd]

@Nertea Forgive me for not understanding, and not being able to find a clear answer on my own, but if I want to add the extras from system heat to my install, do I just drag the whole folder from extras into gamedata folder, example:

Kerbal Space Program\GameData\SystemHeatHarvesters       Or would I unpack everything contained in the SystemHeatHarvesters folder into gamedata, or somewhere else?  The readme says: 

""To install them, drop the correct folder from the Extras folder in this archive into your KSP GameData Folder.""  I'm not sure if this is meant to say drop the correct folder "from" this archive into gamedata, or if it's telling me to do something else, and I'm too dense to understand? Sorry to be a bother, and thanks for any help/clarity. 

Edited January 18, 2021 by vardicd

[Nertea]

1 hour ago, vardicd said:

@Nertea Forgive me for not understanding, and not being able to find a clear answer on my own, but if I want to add the extras from system heat to my install, do I just drag the whole folder from extras into gamedata folder, example:

Kerbal Space Program\GameData\SystemHeatHarvesters       Or would I unpack everything contained in the SystemHeatHarvesters folder into gamedata, or somewhere else?  The readme says: 

""To install them, drop the correct folder from the Extras folder in this archive into your KSP GameData Folder.""  I'm not sure if this is meant to say drop the correct folder "from" this archive into gamedata, or if it's telling me to do something else, and I'm too dense to understand? Sorry to be a bother, and thanks for any help/clarity. 

For these patches, it technically doesn't matter. However, my preferred way is the path you have there. 

[werdman2]

Issue with system heat; Nuclear engine constantly running away

Hi I have been enjoying your mods for some time now but have recently had some trouble with the heat control aspect of the mods, for some reason the craft that I created has some real trouble with heat, whenever I us the engine or the reactor it runs away almost immediately, as if the heating/and or cooling system is not doing anything, just wanted to make sure it wasn't anything obvious. And on another note I feel that that these mods, although they are some of the best I have played could really do with a wiki of some sort, and some example ships bundled in the download. 

Here is an imgur link hopefully highlighting the issue: https://imgur.com/a/8O6GdW5

* I installed everything on ckan and have all the optional config files *

Edited January 19, 2021 by werdman2
adding ckan details

[Nertea]

6 hours ago, werdman2 said:

Issue with system heat; Nuclear engine constantly running away

Hi I have been enjoying your mods for some time now but have recently had some trouble with the heat control aspect of the mods, for some reason the craft that I created has some real trouble with heat, whenever I us the engine or the reactor it runs away almost immediately, as if the heating/and or cooling system is not doing anything, just wanted to make sure it wasn't anything obvious. And on another note I feel that that these mods, although they are some of the best I have played could really do with a wiki of some sort, and some example ships bundled in the download. 

Here is an imgur link hopefully highlighting the issue: https://imgur.com/a/8O6GdW5

* I installed everything on ckan and have all the optional config files *

Looks like not enough radiators. What does the VAB planning UI tell you?

Also, there is a wiki: https://github.com/post-kerbin-mining-corporation/SystemHeat/wiki

[werdman2]

I took your advice and checked the VAB and this is what it said:

https://imgur.com/a/KTWSJhj

After I found this to be ok I added a load more radiators and tested again, its like the system isn't even interacting, as if the nuclear engine is not dissipating heat using the loops. I got this ship to work once, immediately after I had docked them together it seemed to work, but now whenever I load this save up I am having this issue.  Also if its any help whenever I have this issue, the reactor starts to churn out a load of heat and there is nothing in the heat ui monitor saying that there is any rejected, then all at once as soon as it hits max threshold and the reactor scrams it all appears in the rejected all at once, like 2.5k rejected and then falls off, but this might be a feature, I don't know this mod to well

Oh and thank you for the wiki link

[Nertea]

3 hours ago, werdman2 said:

I took your advice and checked the VAB and this is what it said:

https://imgur.com/a/KTWSJhj

After I found this to be ok I added a load more radiators and tested again, its like the system isn't even interacting, as if the nuclear engine is not dissipating heat using the loops. I got this ship to work once, immediately after I had docked them together it seemed to work, but now whenever I load this save up I am having this issue.  Also if its any help whenever I have this issue, the reactor starts to churn out a load of heat and there is nothing in the heat ui monitor saying that there is any rejected, then all at once as soon as it hits max threshold and the reactor scrams it all appears in the rejected all at once, like 2.5k rejected and then falls off, but this might be a feature, I don't know this mod to well

Oh and thank you for the wiki link

Ahh interesting I think I see what is going on overall. I think there might be a problem related to the reactor shutdown temperature being reset somehow. The auto-shutdown value in the UI is automatically set to the reactor's critical temperature when the game starts, but maybe on subsequent loads this isn't happening. I see in all your images the auto-shutdown temperature is set to 1000K, the operating temp of that part is 1600K. So the reactor tries to heat up to 1600K and auto-shutdowns before it gets there so no cooling ever happens. 

[werdman2]

Awesome this immediately fixed the issue, and I will have to keep that slider in mind in the future. Finally to have some fun with these mods in the endless boredom and monotony of this lockdown! 

[Nertea]

1 hour ago, werdman2 said:

Awesome this immediately fixed the issue, and I will have to keep that slider in mind in the future. Finally to have some fun with these mods in the endless boredom and monotony of this lockdown! 

Glad that worked! I ran a few tests but didn't see any actual bugs - let me know if you can reproduce the situation where that slider gets reset to 1000K. 

[Nertea]

SystemHeat 0.3.5

• Fixed an issue where the Overlay panels were not reporting radiator performance properly
• Fixed an issue with fission reactors firing a harmless NRE on launch
• Fixed an issue where loop temperature could trend below local space temperature sometimes
• Adjusted radiator patches for HeatControl 0.6.0
• Rewrote signficiant pieces of the UI backend
• Improved behaviour of in-flight loop incrementing, should handle more edge cases, handle reducing the loop number, correctly update the UI
• Merging loops when docking now works again
• UI no longer violently explodes when undocking
• Fixed missing storage container on Scylla
• Tuned heat generation of NTR patches to be more aligned with FFT-style heat balance

[Kielm]

Oooh, I was contemplating what mods to add into a new career game (something broke unity in my last one so I'm starting over fresh) but it looks like Nertea is updating a bunch of mods, maybe I'll hang fire for a bit

[ussdefiant]

 

So, uh, i'm getting spammed with this in my logs when i have a bog-standard starter rocket with no radiators, nukes, or other fancy things to throw heat around. Any advice on suppressing it?

[computercat04]

38 minutes ago, ussdefiant said:

 

So, uh, i'm getting spammed with this in my logs when i have a bog-standard starter rocket with no radiators, nukes, or other fancy things to throw heat around. Any advice on suppressing it?

I think you have to go to the Settings.cfg of SytemHeat and set the DebugXYZ lines to false.

[Nertea]

Known issue in last version, it's an issue on GitHub, will resolve with next version. 

[Nertea]

SystemHeat 0.3.6

• Fixed some debug settings being on by default
• Implemented improved log levels, better handling for log messages overall
• Fixed fission reactors not updating their electricity generation in the UI or in the NF systems manager when disabled
• Generic converter patch now ignores converters that don't specify ConverterName

[jinks]

Just to get my head around this, is this mod supposed to massively increase part count?

 

I tested the NFE UX-0 KerboPower with vanilla and with SH and my radiator count went from 2 "YF-15 Volta" from HeatControl up to 10 of the same to keep the reactor cool.

[Nertea]

6 hours ago, jinks said:

Just to get my head around this, is this mod supposed to massively increase part count?

 

I tested the NFE UX-0 KerboPower with vanilla and with SH and my radiator count went from 2 "YF-15 Volta" from HeatControl up to 10 of the same to keep the reactor cool.

In some cases this is going to happen. The problem with the stock radiators is that they are very, very good for their size. Like violate the laws of thermodynamics repeatedly. To balance the old radiators with stock I had to use that as reference, here I can throw that all out the window and try for a little more area realism.

Unfortunately this does mean that some of the parts I made for HC just aren't useful anymore, like the small fixed radiators. 

[WarriorSabe]

4 hours ago, Nertea said:

In some cases this is going to happen. The problem with the stock radiators is that they are very, very good for their size. Like violate the laws of thermodynamics repeatedly. To balance the old radiators with stock I had to use that as reference, here I can throw that all out the window and try for a little more area realism.

Unfortunately this does mean that some of the parts I made for HC just aren't useful anymore, like the small fixed radiators. 

That just gave me an idea for a part - a subsurface heatsink. Basically, a part that functions like a radiator but actually dumps the heat into the ground with some sort of drill or something, like a reverse geothermal generator. Maybe a version using ocean too.

[jinks]

7 hours ago, Nertea said:

Unfortunately this does mean that some of the parts I made for HC just aren't useful anymore, like the small fixed radiators. 

As far as I can see in my (early midgame) career save, HC got a "nerf" of a factor of about 3x to 5x, Squad parts are down by a factor of 20x to 30x (50kW to 2kW or 150kW to 5kW for example), which feels a bit too much.

While I like the concept of heat loops, some other aspects make me think that SH, in its current iteration, is not for me.

I'm not sure if it's helpful, but here's a few things I disliked during testing (in no particular order):

• The weak radiators give the impression that they encourage part spam for no particular reason.
• The removal of the NFE reactor control panel. I really like that one.
• The fact that I will have to deal with two (or more) competing heat systems for an extended period of time (I have a lot of mods, it's unlikely many of them will be integrated any time soon).

I think for now I will stick to vanilla and use SH in FFT only mode (which is a ways off either way at the rate my Kerbals are advancing).

[Nertea]

I mean, you're of course welcome to do what you want, but I do want to dig into this a bit more for the betterment of all. 

On 1/22/2021 at 5:12 PM, jinks said:

As far as I can see in my (early midgame) career save, HC got a "nerf" of a factor of about 3x to 5x, Squad parts are down by a factor of 20x to 30x (50kW to 2kW or 150kW to 5kW for example), which feels a bit too much.

I don't think this is quite correct, the Squad ones are more in the 5x range. For example I've moved the deployables to 200/40/10, which is down from 1000/200/50. This is still really generous vs reality, the large TCS-alike items that the ISS mounts give 75 kW together,  in SystemHeat that gives you a crazy 400 kW. 

In addition, things that make heat got 'nerfed' too! A standard ISRU in stock makes 200 kW, I've cut that down to 50 kW. I did this specifically so that things that used the low-end of the heating system that is present in stock (ISRUs, drills) would feel pretty similar in terms of part count, and make the 'feel' changes limited to my other mods where I have more control. 

Taking a few samples from Heat Control too, things didn't change as much as you're describing in most cases. Universal deployables for example at 150, 1000 and 1250 kW went to 175, 500 and 600 kW, so maybe halved at worst. I think you might see some more drastic stuff in the smaller tiny fixed radiators but yeah, those are kinda useless, which brings me to point 2:

On 1/22/2021 at 5:12 PM, jinks said:

• The weak radiators give the impression that they encourage part spam for no particular reason.

I mean, one of the key things with this mod is to provide a somewhat better model of how much radiator you need, compared to stock's... effectively zero need.  Functionally that *should* mean you need more radiator area than in the previous model.

Fundamentally that needs bigger radiator models though to not increase part count, and I don't really have those for the 'high temperature' category - the microchannel ones at the high tech range are fine for area, but the mid-range ones I made for HC are meh and all over the place, and when they are area-corrected, as you pointed out, you often need a huge amount of them to get anything done. 

The only real solution to this is to actually remove the pointless parts  from Heat Control, which I probably won't do. I will however probably take a swipe at a better, more unified set of high temp radiators in HC eventually. 

In the (near) future, I will be implementing radiator convective flow though, which should massively increase the utility of the fixed tiny radiators for moving ships, which will at least give them a small purpose. 

On 1/22/2021 at 5:12 PM, jinks said:

• The removal of the NFE reactor control panel. I really like that one.

Can you tell me what is not sufficient about the new panel? I did a small focus group during its design and there seemed to be a general agreement that it was better. 

On 1/22/2021 at 5:12 PM, jinks said:

• The fact that I will have to deal with two (or more) competing heat systems for an extended period of time (I have a lot of mods, it's unlikely many of them will be integrated any time soon).

Ah, guess I can't really help there, though I can ask 'what mods?'

[mikegarrison]

OK, I have 0.3.6 installed. I have read the wiki. I have a few questions.

1) What is part EBR-1000 for? It is labeled as a heat exchanger, but it does not function the same as part PX-1F, which is also labeled a heat exchanger. I think it predates System Heat (in fact, I remember a similar part going way back to the original Heat Control mod many years ago). The text associated with it mentions something about engine heat.

2) I have a ship with three loops. The highest temp loop has the radiators. The others use the PX-1F to dump their heat (with a temperature adjustment) to the hot loop. But in the VAB simulator, the two heat exchangers have the red over-temp icon. Nothing else does. According to the simulator, all three loops are running at their expected temperature and the heat rejection is fine. Is this overtemp icon showing a real problem? If so, how would I fix it?

3) Engines don't seem to have system heat loops. What happens to the heat from them? (I haven't been using the "Far Future" engines yet.)

Edited January 24, 2021 by mikegarrison

[Nertea]

On 1/24/2021 at 7:10 AM, mikegarrison said:

3) Engines don't seem to have system heat loops. What happens to the heat from them? (I haven't been using the "Far Future" engines yet.)

There is an assumption that chemical engines have been designed to cool themselves independently (having a sufficient mass flow to carry away the heat). This is generally a decent assumption. 

On 1/24/2021 at 7:10 AM, mikegarrison said:

1) What is part EBR-1000 for? It is labeled as a heat exchanger, but it does not function the same as part PX-1F, which is also labeled a heat exchanger. I think it predates System Heat (in fact, I remember a similar part going way back to the original Heat Control mod many years ago). The text associated with it mentions something about engine heat.

It's not really a functional part in this mod. Also the model is bad. 

On 1/24/2021 at 7:10 AM, mikegarrison said:

2) I have a ship with three loops. The highest temp loop has the radiators. The others use the PX-1F to dump their heat (with a temperature adjustment) to the hot loop. But in the VAB simulator, the two heat exchangers have the red over-temp icon. Nothing else does. According to the simulator, all three loops are running at their expected temperature and the heat rejection is fine. Is this overtemp icon showing a real problem? If so, how would I fix it?

 

I'd test it in operation. That icon is probably a red herring, the cross-loop transfer is a bit tacked on now, I will look into it.