Preventing ablator burning up when mining?

[Foxster]

Here's my Mun biome hopper...

Seems that no matter how many radiators I add, it still cooks off the ablator when using the Convert-o-Tron .

What am I doing wrong? 

Edited January 14, 2017 by Foxster

[fourfa]

As a sideways non-answer, seems like you can re-enter a ship like that from the Mun with no heatshield at all, and definitely with a heatshield taken down to 0 ablator before launch.  Empty heatshield still has a 3300K temp rating.

But as to the actual question, I have seen that over several versions and don't know how to solve it.  Those rads apply cooling over the whole ship - have you tried placing a pair or two other places?  Like on the ascent stage tank, on the final decoupler, on the pod itself?

[Benjamin Kerman]

I remember seeing somewhere else on the forum that the mini convert-o-torn produces way too much heat and just sucks overall. I know this isn't an answer to your question, but just saying it for the next design! You could try bigger radiators...

The second post by Snark, with 2 votes. 

Edited January 14, 2017 by Benjamin Kerman

[ForScience6686]

Well there is a max cooling for parts, and each radiator can only absorb so much heat f from an item.  Are you over temp on your converter or drills?  If so I'd add more radiators or convert less.  If not, I'm not sure how to help with that.

[bewing]

Sounds like a bug to me. I'll see if I can duplicate it with a simpler setup.

[fourfa]

It's not specific to this drilling setup - it can happen with a shield near a hot nuke engine, or on ships inside the orbit of Moho

[Foxster]

I have tried a shed-load of both types of radiator and it still does it - albeit slower with more radiators. I could usually live with losing a few points of ablator but this craft is designed to refuel a number of times and I'll have none left. 

I think it is a bug. Ablation is not about just getting hot, it's about friction and there is none in this scenario. 

[linuxgurugamer]

42 minutes ago, Foxster said:

I have tried a shed-load of both types of radiator and it still does it - albeit slower with more radiators. I could usually live with losing a few points of ablator but this craft is designed to refuel a number of times and I'll have none left. 

I think it is a bug. Ablation is not about just getting hot, it's about friction and there is none in this scenario. 

It's Not friction.  Ablator vaporizer with the heat generated by friction.  So it doesn't matter as long as there is heat

[Foxster]

7 hours ago, linuxgurugamer said:

It's Not friction.  Ablator vaporizer with the heat generated by friction.  So it doesn't matter as long as there is heat

Are you talking KSP or RL? In RL I don't think the ablator on a craft would cook off before the crew or rest of the craft is damaged in non re-entry situations. 

Edited January 15, 2017 by Foxster

[linuxgurugamer]

1 hour ago, Foxster said:

Are you talking KSP or RL? In RL I don't think the ablator on a craft would cook off before the crew or rest of the craft is damaged in non re-entry situations. 

Quoting Wikipedia:

Quote

In a basic sense, ablative material is designed to slowly burn away in a controlled manner, so that heat can be carried away from the spacecraft by the gases generated by the ablative process while the remaining solid material insulates the craft from superheated gases. There

Thesaurus.com:

Quote

Aerospace. erosion of the protective outer surface (ablator) of aspacecraft or missile due to the aerodynamic heating caused by travelat hypersonic speed during reentry through the atmosphere.

and here is a link to an article on space.com, talking about the heatshield for the Orion spacecraft: http://www.space.com/22046-nasa-orion-spacecraft-heat-shield.html

While in all cases the heat shield is designed to be heated by atmospheric friction, heat is heat, and if it gets heated other ways, it will cook off.

Now, in KSP, it's quite possible that too much heat is being generated.

[Spricigo]

1 hour ago, Foxster said:

Are you talking KSP or RL? In RL I don't think the ablator on a craft would cook off before the crew or rest of the craft is damaged in non re-entry situations. 

That's exactly the point of ablator.  It absorb heat to the point it start to burn,  the burning portion become fragile and is teared off by the friction.

[bewing]

IRL it's the interaction of the ablative with plasma that causes the pyrolysis effect and starts the burnoff. Just getting the heatshield warm IRL will never do it.

But that's the question in my head here -- how does KSP model it? In any case, it shouldn't even start shedding ablative until the heatshield temp goes way over 800C, and in all these test cases I can't see it getting anywhere near even a measly 800C.

 

[Spricigo]

17 minutes ago, bewing said:

IRL it's the interaction of the ablative with plasma that causes the pyrolysis effect and starts the burnoff. Just getting the heatshield warm IRL will never do it.

And this interaction is in great part heat transfer,  warming the ablator if you prefer.  Also noticed that ablator is,  in real life,  a broad term also used for material that don't need interaction with plasma to erode. You don't manage reentry level of heat with Styrofoam,  but it's good enough hot coffee. 

48 minutes ago, bewing said:

But that's the question in my head here -- how does KSP model it? In any case, it shouldn't even start shedding ablative until the heatshield temp goes way over 800C, and in all these test cases I can't see it getting anywhere near even a measly 800C.

Well,  maybe KSC actually have Styrofoam heatshields.  

[bewing]

2 minutes ago, Spricigo said:

And this interaction is in great part heat transfer,  warming the ablator if you prefer.

True, but there's a big difference between warming the entire thickness of the ablative (all of the heatshield), and just transferring energy to a microscopic surface layer of it -- while leaving the rest of the heatshield cold.

[Benjamin Kerman]

Just an idea for a craft that might work better (meaning having no ablator being burned off) try having the heat shield on the opposite end of the vessel, at the top. You can have more radiators on the craft, and the heat will transfer more slowly to the heat shield. 

4 hours ago, Spricigo said:

That's exactly the point of ablator.  It absorb heat to the point it start to burn,  the burning portion become fragile and is teared off by the friction.

It is the same with a wood fire. The wood is turned into ash, and the ash is blown off by wind. 

[Foxster]

I get that ablator could possibly be reduced by heat other than high speed atmosphere friction - well not really but maybe.

But we are talking about something specifically designed to resist very high temperatures of 100s of degrees, someone suggested 800°+, to protect the rest of a craft from this kind of heat. Seems pretty unlikely then that any kind of heat transferred internally to an ablator sufficient to start removing material from it is not going to have caused the vessel to melt or explode first. I certainly don't see the crew surviving this. 

But not in the case of a KSP craft it seems. Here the crew happily get on with their inane grinning and the ship functions perfectly, despite temperatures being conducted through the craft enough to melt ablator. 

Nuh, uh. This is a bug.   

Update: Just checked and with the craft above running the Drill-o-Matic Junior and the Convert-o-Tron 125, the temperature of the craft climbs as you'd expect. Once the temp of the heatshield rises to around 560K (287C) the amount of ablator starts to fall. 

That means the ablator has a melting point quite a bit less than lead. It isn't far off that of tin. So I suspect that the ablator the Kerbals are relying on to get them safely home is, rather than a modern composite layer of ceramics, a layer of solder. 

Edited January 15, 2017 by Foxster

[Benjamin Kerman]

Agreed, 600 K should not even burn wood (I think), much less ceramics. 

[bewing]

5 hours ago, Foxster said:

So I suspect that the ablator the Kerbals are relying on to get them safely home is, rather than a modern composite layer of ceramics, a layer of solder. 

Lol! So that's why my kerbals always die when I try to "protect" them with a heatshield.

 

[Benjamin Kerman]

Maybe the kerbals just have a death wish...

[OHara]

@Foxster there is a parameter ablationTempThresh=500 in the part-configuration file for each heat-shield, so you might try changing that entry,
or if you have ModuleManager installed putting something like this in a *.cfg.

@PART[*]:HAS[@MODULE[ModuleAblator]] {
	@MODULE[ModuleAblator] {
		%ablationTempThresh = 1200
	}
}

I'm trying this myself, as well.       [Edit: Increasing ablationTempThresh to 1200K worked well for me.  I checked for unwanted side-effects using a quick-saved Eve re-entry and found no change (<0.5%) in ablation rate or temperature.  The threshold changes behaviour only below the threshold temperature according to this thread, which would explain why I saw no side-effects.

The equation for ablation rate is just the Arrhenius, rate = constant×exp[-E/T], which is physically reasonable for something like evaporation, so I wondered why the threshold was even necessary to prevent noticeable evaporation at 500K.  The constant E=7500K=62kJ/mol is about 3 times too low to be a reasonable vaporization energy, assuming KSPs temperature unit should correspond to a real-world Kelvin.  E should be 3× larger and the prefactor constant larger so that the evaporation rate changes more steeply with temperature.]

I hear the word 'ablation' mostly to mean sandblasting with molecules.  People use ablation to describe cleaning things in a vacuum chamber by bombardment, for example, whereas simply heating a part to clean off any residue has different words, like burning-off,  evaporation, or sublimation.   I don't expect KSP to simulate on a molecular level, of course, and hope that having the 'ablator' start to evaporate away when the shield itself reaches 1200K or so might be a satisfactory simulation.

Edited March 5, 2017 by OHara
Suggested fix at existing bug report http://bugs.kerbalspaceprogram.com/issues/6336

[Spricigo]

 

5 hours ago, Foxster said:

I get that ablator could possibly be reduced by heat other than high speed atmosphere friction - well not really but maybe.

But we are talking about something specifically designed to resist very high temperatures of 100s of degrees, someone suggested 800°+, to protect the rest of a craft from this kind of heat. Seems pretty unlikely then that any kind of heat transferred internally to an ablator sufficient to start removing material from it is not going to have caused the vessel to melt or explode first. I certainly don't see the crew surviving this. 

But not in the case of a KSP craft it seems. Here the crew happily get on with their inane grinning and the ship functions perfectly, despite temperatures being conducted through the craft enough to melt ablator. 

Nuh, uh. This is a bug.   

Update: Just checked and with the craft above running the Drill-o-Matic Junior and the Convert-o-Tron 125, the temperature of the craft climbs as you'd expect. Once the temp of the heatshield rises to around 560K (287C) the amount of ablator starts to fall. 

That means the ablator has a melting point quite a bit less than lead. It isn't far off that of tin. So I suspect that the ablator the Kerbals are relying on to get them safely home is, rather than a modern composite layer of ceramics, a layer of solder.

I still suspect is Styrofoam.  Made sense with the cardboard winglets. 

Anyway, the odd thing is that the ablator don't need to resist high temperatures to do its job.  Wich is to remove heat from a system (in your case the vessel). 

[Carl]

3 hours ago, bewing said:

Lol! So that's why my kerbals always die when I try to "protect" them with a heatshield.

 

If you have the right way of using it anything can be a heat shield, NASA studied but never flight tested a design that sprayed the underside of the craft with water to function as a heat shield.

[Streetwind]

10 hours ago, OHara said:

@Foxster there is a parameter ablationTempThresh=500 in the part-configuration file for each heat-shield, so you might try changing that entry,
or if you have ModuleManager installed putting something like this in a *.cfg.

@PART[*]:HAS[@MODULE[ModuleAblator]] {
	@MODULE[ModuleAblator] {
		%ablationTempThresh = 1200
	}
}

I'm trying this myself, as well.       [Edit: Increasing ablationTempThresh to 1200K worked well for me.  I checked for unwanted side-effects using a quick-saved Eve re-entry and found no change (<0.5%) in ablation rate or temperature.  The threshold changes behaviour only below the threshold temperature according to this thread, which would explain why I saw no side-effects.

The equation for ablation rate is just the Arrhenius, rate = constant×exp[-E/T], which is physically reasonable for something like evaporation, so I wondered why the threshold was even necessary to prevent noticeable evaporation at 500K.  The constant E=7500K=62kJ/mol is about 3 times too low to be a reasonable vaporization energy, assuming KSPs temperature unit should correspond to a real-world Kelvin.  E should be 3× larger and the prefactor constant larger so that the evaporation rate changes more steeply with temperature.]

I hear the word 'ablation' mostly to mean sandblasting with molecules.  People use ablation to describe cleaning things in a vacuum chamber by bombardment, for example, whereas simply heating a part to clean off any residue has different words, like burning-off,  evaporation, or sublimation.   I don't expect KSP to simulate on a molecular level, of course, and hope that having the 'ablator' start to evaporate away when the shield itself reaches 1200K or so might be a satisfactory simulation.

The evaporation energy constant is probably fudged intentionally by Squad, to account for the fact that reentries on Kerbin happen at so much lower speeds than IRL. If Squad didn't fudge the reentry physics on all sides, we probably wouldn't even see a plasma trail or get more than moderately warm on the way down

The one side effect of increasing ablationTempThresh is the fact that it doesn't affect the pyrolysis calculation in any way... it's just a hard cutoff. You will always get the same ablation rate at the same temperature, regardless of what value you set for ablationTempThresh, with the exception that if the temperature is lower than what you set, ablation rate gets nulled. As a result, if you set a high value for ablationTempThresh, your ablation rate will instantly jump from zero to a high value as you cross that threshold. With the stock setting, the cutoff threshold is so low that you can't tell the difference between being above or below the threshold.

 

As far as the initial question of this thread: I have to wonder why the spacecraft even gets that warm in the first place, and I have to agree with @Foxster: this might well be a bug, or at least unintended behavior. Here's why.

The drills and the refinery both produce "core heat". This is distinctly different from "normal heat", which is the thing that can eat away at the heat shield. Core heat is almost entirely internal to the PartModule that generates it. In other words, if the ISRU is running then the ore conversion process normailzes at 1000 K if given sufficient cooling. But only the ore conversion process. Not the ISRU itself.

Cooling "core heat" happens via the "core transfer" statistic of a radiator that can be looked up in the VAB. If there is enough (or more) core transfer available on connected radiators, then the PartModules that generate core heat stabilize at their target temperature. If there is less than enough, the PartModules will overheat. But again, not the part itself. Just the process on the part.

The one interaction between "core heat" and "normal heat" is that a hot core slowly leaks heat into the part internals. This is supposed to represent the fact that no insulation will ever be 100% effective. As a result, if the process that generates core heat is operating at full load for a long while, the part will slowly begin to accrue heat. Real, normal heat that can melt a heatshield (in KSP, at least).

However! There are still radiators on the craft. In the screenshot, there are thermal control systems. Which have this certain feature called vessel-wide cooling. And their threshold temperature is 400K. In other words, any part on the vessel in the screenshot that exceeds 400K internal or skin temperature at any time should be receiving active cooling from the thermal control systems. Now, unfortunately the relationship between core heat load and normal heat load on a radiator is an arcane topic. Core transfer represents only a small fraction of a radiator's total cooling capacity, and yet, utilizing core transfer shares radiator capacity with handling of normal heat in some fashion. I've tested and confirmed this over and over again. But in what fashion exactly? I can't tell you, and I've never seen anyone anywhere on the forums or reddit to even attempt an explanation. The radiator load indicator the game gives you seems to have no discernible relation to what the radiator is actually doing, and core heat handling is not displayed in the thermal debug readout, so there is no way for us to reverse-engineer what is happening internally. But even if we assume the absolute worst case, in which a fully utilized core transfer means that there is zero handling capacity left for normal heat (which I'm 99% sure is wrong), then that still would mean that adding additional radiators beyond the core transfer required by the drills and ISRU would provide additional capacity for handling normal heat. Even a single small thermal control system has so much normal heat handling capacity - way, way more than it has core transfer - that the minor leakage from the hot cores of the drills and ISRU should barely even register.

And yet, Foxster said that adding additional radiators does not keep his heatshield from melting. So why is that happening? Why is the heatshield - and, for that matter, every single part in between the heatshield and the ISRU, as well as the ISRU itself - allowed to reach and exceed 500K normal heat regardless of sufficient radiators being present, when said radiators should keep the entire ship strictly at 400K at all times? That is the thing that needs testing here. This should not be happening with the heat system as it is designed.

 

@Foxster: as an idea, try mounting a pair of small panel radiators on the FL-T100 where the arms of your engine modules connect. The panel radiators don't provide vessel-wide cooling, but rather only target the part they are attached to, plus any part directly attached to that (two decouplers and two struts). Since they are not in range of the ISRU and drills, their core transfer will not be utilized at all, and they will be 100% dedicated to removing normal heat that seeps towards the heatshield from below or from the engines. If there is a bug with radiators that cannot properly cool normal heat while utilizing core transfer even partially, such a setup should provide a solution.

If that still results in your heatshield (or the fuel tank itself) exceeding 500K, then you can definitely make a bug report.

Edited January 16, 2017 by Streetwind

[Sharpy]

I wonder about one thing: How much of a difference does ablator make?

I'd make a setup of MK1 capsule and heatshield (and various extras like more heatshields, spaced multiple heatshields) and try hyperediting it into various reentry trajectories.

I found that if I reenter too fast (something like 5000m/s) into atmosphere below some 24km, the craft explodes completely regardless of *anything*; no amount of cooling, ablator or heatshields would help, and it would happen even if the craft was fairly cool at that time, a hardcoded destruction.

And the craft *would* be fairly cool at that time. E.g. using the "spaced heatshields" where three shields separated by COStruts would explode in sequence as they overheat... well, the first would get pretty hot but not overheat. Two subsequent, and the capsule, were quite mild. Then 24km, boom.

Regardless, I haven't found a setup where a heatshield *without ablator* would be insufficient to protect the MK1 capsule, as long as the insta-death threshold was not breached. Ablator seems like just extra mass and unneeded formality - the heatshields are so durable to extreme heat, and conduct heat to other parts so poorly, I still need to find a scenario where ablator would make a difference. (...and actually, I can see a scenario where it's actively harmful. It's HEAVY. So, a light capsule with heatshield enters at a dangerous speed. Without ablator, drag will slow it to safe speeds before the insta-death altitude. With ablator, it will continue plunging to its demise.)

 

In conclusion: Let that ablator evaporate and never worry about it.

Edited January 16, 2017 by Sharpy

[Carl]

1 hour ago, Sharpy said:

I wonder about one thing: How much of a difference does ablator make?

I'd make a setup of MK1 capsule and heatshield (and various extras like more heatshields, spaced multiple heatshields) and try hyperediting it into various reentry trajectories.

I found that if I reenter too fast (something like 5000m/s) into atmosphere below some 24km, the craft explodes completely regardless of *anything*; no amount of cooling, ablator or heatshields would help, and it would happen even if the craft was fairly cool at that time, a hardcoded destruction.

And the craft *would* be fairly cool at that time. E.g. using the "spaced heatshields" where three shields separated by COStruts would explode in sequence as they overheat... well, the first would get pretty hot but not overheat. Two subsequent, and the capsule, were quite mild. Then 24km, boom.

Regardless, I haven't found a setup where a heatshield *without ablator* would be insufficient to protect the MK1 capsule, as long as the insta-death threshold was not breached. Ablator seems like just extra mass and unneeded formality - the heatshields are so durable to extreme heat, and conduct heat to other parts so poorly, I still need to find a scenario where ablator would make a difference. (...and actually, I can see a scenario where it's actively harmful. It's HEAVY. So, a light capsule with heatshield enters at a dangerous speed. Without ablator, drag will slow it to safe speeds before the insta-death altitude. With ablator, it will continue plunging to its demise.)

 

In conclusion: Let that ablator evaporate and never worry about it.

TLDR of why that is happening is that aero drag is exceeding the parts impact tolerance force resulting in the part imploding from stress. Go fast enough and thick enough atmosphere becomes solid rock for all intents and purposes.