[1.12.x] Heat Control - More radiators! (August 18, 2024)

[Entropius]

Thanks for the answers Captain Sierra & RoverDude.

In regards to this:

The temperature at which a radiator begins pulling from a part is configurable at the part level. It's a property of the part itself (the Nerv being the only stock part with this property right now).

Does that mean all other parts besides the NERV have some default-threshold they use to by virtue of not having an explicit threshold in their config? (if so, what is the default-threshold?) Or that radiators just ignore all other parts?

[RoverDude]

Thanks for the answers Captain Sierra & RoverDude.

In regards to this:

Does that mean all other parts besides the NERV have some default-threshold they use to by virtue of not having an explicit threshold in their config? (if so, what is the default-threshold?) Or that radiators just ignore all other parts?

All other parts have a default threshold that is a function of their thermal mass and their max temp.

[hoojiwana]

Looking in the LV-N config there is this:

radiatorMax = 0.35 //Default = 0.25 but nuke engines are meant to run hot

[RoverDude]

Side note - we're aware that technically nukes should run cold, but be really hot when idle

[mikegarrison]

Side note - we're aware that technically nukes should run cold, but be really hot when idle

So is there any way, in KSP, to model that? Can exhaust mass carry off heat?

But I doubt they actually "run cold". Just a less hot than they run with the throttle open. Running cool would imply a really unlikely thermal efficiency.

[Angelo Kerman]

So is there any way, in KSP, to model that? Can exhaust mass carry off heat?

But I doubt they actually "run cold". Just a less hot than they run with the throttle open. Running cool would imply a really unlikely thermal efficiency.

One way to look at it is that the engine heat production represents what isn't carried away by the exhaust. The NASA Discovery II, for instance, shed most of the heat that the engine generated through its exhaust, but it still had to deal with about a gigawatt of waste heat.

[RoverDude]

Well yeah but you get the idea - i.e. they should generate more heat when they are not at full throttle.

[DerekL1963]

B1) Instead of hiding, rework stock radiators to work with HC/NFE modules

I'm a fan of this option.

[billkerbinsky]

Well yeah but you get the idea - i.e. they should generate more heat when they are not at full throttle.

Hmm. A reactor has a throttle (neutron-absorbing control rods) , but something like 5-10% of steady-state power is generated by the spontaneous decay of fission products. when the reactor throttle goes to zero, the heat production should drop to about 5-10% of where it had been and then decay exponentially from there. That level of heat production is going to still require active cooling for some time after the engine is shut down. Unclear if it would be greater or less than the cooling required at full thrust (seems like it would be tricky to get 95% of the reactor heat into the propellant but maybe I'm missing something...) but it's going to require significant cooling in any event.

one of these days I might try to climb the modding learning curve and build a.. more interesting reactor model that has a plausible decay heat model.

[Chyort]

I figure i might as well throw this out there first.

My previous post was never meant to throw wood on the fire, so to speak. It was simply about my confusion over how the stock radiators were released/handled. No more and no less.

And my question remains as far as i can tell. If you go with B or B1 as it seems most people in the thread are requesting, How will that affect other mods that make stock-alike radiators or reactors that use stock heating?

Rebalancing 1 stock engine to be more in line with your much larger pack is one thing, and it causes a fair degree of confusion by itself from what i have seen. What happens if you do the same to multiple radiators and, from the sounds of it, the stock heating system?

Not trying to take a side at the moment, just looking for a little clarification i guess.

[Streetwind]

@Chyort: Heat Control uses the stock heat system all the way. It was made to add parts that interact with the stock system, because stock KSP did not offer such parts at the time. The only thing that was unusual about it is the fact that the radiators and heatpipes circumvent conduction by stealing the heat directly... and now it is no longer unusual, because the new stock radiators do the same thing.

The proposed changes here wouldn't influence the stock heat system at all, or break Heat Control's compatibility with it. But yes, there would be an implicit conceptual incompatibility with other mods that decide to leverage the stock radiator functionality for their own parts, just like there is a conceptual incompatibility with the stock radiators themselves. That is, the parts would run perfectly fine together - it just wouldn't make logical sense to have both installed at the same time, because the presence of vessel-global heat management makes Heat Control's parts pointless.

In that scenario, the player would make the conscious choice to either not use Heat Control and go with the stock parts and others like it, or use Heat Control and not use stock parts and others like it. The decision would be based on whether or not the player would want to have their interaction with the heat system be detailed or abstract.

The thing that has me hesitating is the implication for Nertea's other mods, though. Heat Control was split off from NF Electrical. In fact, NFE still has a hard dependency on Heat Control's plugin component, and most players who install NFE also install the full Heat Control pack due to the additional radiator options and other useful parts. So would NF Electrical remain in lockstep with Heat Control, and also live in an environment where stock radiator functionality doesn't exist... in which case it, and with it the rest of the Near Future suite, takes a decisive step away from being stockalike? Or does NFE move away from Heat Control, change its radiators over to the stock system? Because technically, NFE doesn't have this conceptual incompatibility with the stock radiators. Provided you can get enough area for adequate cooling performance at low temperatures, they should work perfectly fine (if making the ships look slightly derpy, due to a massive amount of radiators**). Also RoverDude is going to take a dependency in the NFE plugin component to make reactors for his mods. And those will definitely use the stock radiators, because RoverDude likes them the way they are (else he wouldn't have made them). That in turn implies to a player that NFE and Roverdude's mods would be fully compatible with each other. But that falls flat if NFE does not switch to the stock system...

...it's a bit of a mess, as you can see, and it'll take a some fiddling to untangle it.

My suggestion for The Great Untangling:

- NF Electrical switches to stock type vessel-global radiators, preserving the stockalike status of the Near Future suite and ensures smooth play together with USI suite mods.

- Heat Control gets reworked as a mod specifically dedicated to replacing all instances of vessel-global heat handling with the more detailed part-based heat handling that Heat Control's fans have come to enjoy. It becomes an optional install for those who want that sort of thing, as opposed to the quasi-no-brainer sister mod to NF Electrical that it currently is. It would either conflict with or have the ability to override other mods that ship vessel-global type heat management solutions.

Upside: Everyone can have their cake and heat it, too. </rimshot>

Downside: I have no idea how much this increases the scope of Heat Control, and how much extra work this generates for Nertea.

** Bonus suggestion: massive amount of radiators may be reduced by allowing the reactors to grow hotter, which in turn allows the radiators to get hotter. Not sure if that won't cook the rest of the ship, but food for thought anyway.

Edited June 24, 2015 by Streetwind

[Psycho_zs]

Streetwind, that seems to be a great suggestion. Another thought: insulators can come handy even with stock mechanics (I doubt that conductors would have any use in that case). So switching global/local may be required even for HC.

[TheSaint]

Another solution for NFE would be to remove heat management from it entirely, make it completely neutral. Here is your reactor, it generates X amount of waste heat that must be removed in order for it to function. If you want to remove it with stock radiators, knock yourself out. If you want to remove it with Nertea's system, install Heat Control. If you want to remove it with somebody else's system, go for it.

[KAO]

Another solution for NFE would be to remove heat management from it entirely, make it completely neutral. Here is your reactor, it generates X amount of waste heat that must be removed in order for it to function. If you want to remove it with stock radiators, knock yourself out. If you want to remove it with Nertea's system, install Heat Control. If you want to remove it with somebody else's system, go for it.

But this presents a balance issue. If the stock radiators use global heating system, then there would be no upsides to using Nertea's Heat Control. You can't have two versions of physics being present at the same time, it breaks the gameplay.

[Randazzo]

Don't the Heat Control radiators actively get rid of heat?

The stock radiators don't do that. They do actively suck heat into themselves but then they just passively radiate it away. (Unless I missed them doing something with it. Didn't appear to be going anywhere. )

[Streetwind]

Don't the Heat Control radiators actively get rid of heat?

No, they don't, and never have They radiate heat using perfectly normal stock mechanics. They simply steal heat from the parent part at a fixed rate instead of relying on conduction, that's all there is to it.

However, because they are not limited by the parent part's temperature, this means that as the radiator continues to steal more and more heat, it becomes hotter and hotter. And blackbody radiation scales exponentially with temperature. So eventually the Near Future / Heat Control radiators simply get very hot, all the way to the point where the radiation flux becomes equal to the amount of heat that the radiator steals from the parent part. I think the XR-1100 gets the hottest, and can go as high as 1600 K under full load or something, last I checked. But when it gets there, it is in equilibrium, and sucks exactly as much heat as it loses to normal stock radiation mechanics.

The new stock radiators work exactly the same, except for two key differences: 1.) they can steal heat from all parts of the vessel, not just the parent; and 2.) they cannot steal heat from something that's cooler than they are themselves.

[futrtrubl]

Don't the Heat Control radiators actively get rid of heat?

The stock radiators don't do that. They do actively suck heat into themselves but then they just passively radiate it away. (Unless I missed them doing something with it. Didn't appear to be going anywhere. )

Actually the opposite. HC radiators actively pump heat into radiators (radiators can be hotter than source parts) but passively radiate. Stock parts passively suck heat from source parts (though over long distances) and passively radiate heat.

[Randazzo]

Actually the opposite. HC radiators actively pump heat into radiators (radiators can be hotter than source parts) but passively radiate. Stock parts passively suck heat from source parts (though over long distances) and passively radiate heat.

It's not passive, when a threshold is passed, they pull the heat. There's an active module specifically for it. I know the HC radiators actively pull the heat, I just thought they actively dumped it as well. I apparently made an incorrect assumption based on the heat equilibrium.

[TheSaint]

But this presents a balance issue. If the stock radiators use global heating system, then there would be no upsides to using Nertea's Heat Control. You can't have two versions of physics being present at the same time, it breaks the gameplay.

The "upside" is that some folks don't want to take the easy way of the stock radiators and Heat Control offers them that option. Balance doesn't really come into it.

If you read Steetwind's post above, I'm suggesting the same thing that he is, I'm just saying that they might want to take it a step further and remove all heat management components from Near Future Electrical entirely, leaving it completely neutral from a heat management standpoint. Just give the reactors a heat rejection requirement and leave it up to the end user to decide which system they want to use to meet it. Saying, "You're not allowed to use NFE reactors unless you use Heat Control," seems a bit extreme to me, but again, that's up to Nertea to decide.

[futrtrubl]

It's not passive, when a threshold is passed, they pull the heat. There's an active module specifically for it. I know the HC radiators actively pull the heat, I just thought they actively dumped it as well. I apparently made an incorrect assumption based on the heat equilibrium.

I'd argue that that isn't active heat transport but high conduction passive transport with a limit.

[Captain Sierra]

I'd argue that that isn't active heat transport but high conduction passive transport with a limit.

THe code disagrees with you. WHen the radiators start taking heat, they are doing so via internal flux, applying a negative value to the target part, and a positive value to themselves. This is the same function as Nertea's radiators (just with an on/off threshold) and can be observed in the thermal debug data.

[Randazzo]

I'd argue that that isn't active heat transport but high conduction passive transport with a limit.

If it turns on and off, then by definition it is not passive, but it's not really relevant since the premise of my point was incorrect to begin with.

[futrtrubl]

If it turns on and off, then by definition it is not passive, but it's not really relevant since the premise of my point was incorrect to begin with.

I think we may have been arguing different things with different definitions of "active transport".

Mine is transporting heat against a thermal gradient.

Your's is controlably transporting heat.

Captain Siera's is requiring a module to transport heat.

[mikegarrison]

OK, so here's how I see it, after thinking about it a little.

The problem is that Nertea doesn't really want to cause a fork. FAR is a fork. It completely redoes part of the game, and other mods need to either account for it or it needs to account for them, or else they risk becoming incompatible. Deadly Re-entry is a fork, and we saw how it was difficult to get Heat Control to work with DRE. Also, maintaining a fork becomes harder and harder as the stock game continues to develop in a different direction. So it's easier for everyone if a mod doesn't result in a fork.

So does 1.04 really require Heat Control to fork? Let's consider if we ignore the new stock radiators. Is Heat Control compatible with the rest of the new stock system? If not, then the choice becomes pretty stark: change/abandon Heat Control or become a fork.

However, if Heat Control would work fine with 1.04 but is simply unbalanced with the new radiators, then it doesn't really cause a fork. In that case, Option B becomes quite feasible. Just delete the new stock radiators as part of the mod. Let people know up front that if you select this mod it will delete the stock radiators.

Or, you could rebalance/rework the stock radiators to make them fit into the Heat Control system. Personally I would vote against that, because it leads to an unnecessary duplication of parts. But it would solve the issue of people who already have a bunch of ships with the stock radiators complaining about Heat Control breaking their saves. (FAR doesn't worry about that -- people select FAR because they want it to break their saves. They want the fork.)

Or there is a third option -- make it a user choice. Give them two cfg files, one of which deletes the stock radiators and one of which converts them to Heat Control. Make one default, and the other a selectable extra.

The big sticking point here is not really Heat Control. It can be advertised like FAR or DRE or Remote Tech -- choose this if you want the fun of complicating your life. The big problem is NFE, because of the reactors. If you make Heat Control mandatory for NFE, then that means players have to choose between using the stock heat control and adding the reactors.

Personally, I would be fine with that. In a sense, it was already like that anyway. Under 1.02 if you chose to install the NFE reactors you essentially had to install Heat Control too. So this really wouldn't change much.

Which means, in the end, that if the stock heat system (minus the radiators) works with Heat Control, then I think one or the other version of Option B is the way to go. If the stock heat system is untimately incompatible with Heat Control, then the choice is harder. In that case, Nertea needs to choose to fork or not fork.

[Fraz86]

I'm increasingly convinced that NFE needs to conform to stock radiator behavior. Any other option involves too much potential for confusion and conflicts with other mods (or future stock content) that rely on the stock behavior. I'm not thrilled with the stock radiators, but I think accepting it and adapting is the best way forward.

As noted by Streetwind, the greatest difficulty in adapting NFE to the stock behavior would be the rather silly surface area of radiators required to cool reactors. Streetwind suggested possibly increasing reactors' nominal temps as one means of adapting. Along similar lines, reactors' heat production could be decreased. The problem with these solutions is that - if these variables are tweaked to the point that Nertea's conformal radiators are capable of cooling their matching reactors using stock radiator behavior - it would become trivially easy to cool reactors using stock radiators (due to their much larger surface areas). Correspondingly, the mass & cost of Nertea's radiators would need to be drastically decreased to match stock radiators, thus disrupting the balance of NFE.

In light of the above conundrum, is there a means by which Nertea's radiators can be converted to use the stock radiator module, but also remain set apart from the stock radiators as "advanced radiators" with a higher heat rejection to surface area ratio? Perhaps by increasing their emissivity? I am aware that emissivity >1 is not technically physically possible, but I think it could be rationalized as an abstraction representing some form of "active heat dissipation". This could also justify maintaining the higher mass and cost of Nertea's radiators relative to stock, and thematically explain the different aesthetic appearance.

Edited June 24, 2015 by Fraz86