ISRU Converter overheating

[Pixel of Life]

The ISRU Converter on my asteroid tug is overheating for no apparent reason (or maybe I'm just stupid). I've tried shutting down the engines, retracting solar panels and even turning the asteroid so that the craft is hidden behind it and should be in complete darkness, but it still refuses to refine any of the ore.

What do?

[purpletarget]

An Engineer will make ISRU and Drills work better.

Also extended solar panels will act as passive radiators, so if you can leave them extended, you'll be better off.

Tanks on either side of the ISRU should act as heat sinks and pull the heat away, make sure you don't have any heat shields or service bays in the way.

They are also power hungry...make sure you have something to power them. If not solar, then take some fuel cells to keep running in the dark.

[zzubnik]

I'm having a similar problem with my ISRU converter. I landed on the moon, converted and filled my tanks successfully, but now when I am in space and want to convert the remaining ore, I am told that it is overheated. No amount of waiting seems to cure this. I have many solar panels deployed, and it has previously converted successfully. What am I doing wrong?

[kendfrey]

The overheating behaviour on ISRU converters is pretty erratic. This is what it looks like to me so far:

• At 1000x time warp, overheating goes quickly to 0. This is the only way I found to efficiently process Ore.
  
• Once 100% overheat and "zero efficiency" is reached, there is no going back. No amount of timewarping or spacecraft cooling will cool the converter. Not even an Engineer can fix it, as far as I can tell.
  
• At 100x time warp and below, the behaviour is unpredictable. Overheating seems to tend toward 100%. It will quickly rise, but stop in a random spot. It will sometimes change, but that also appears to be random.
  
• Overheat is not the same thing as the normal temperature of a part.
  

Can someone confirm this, especially the first two? They seem like bugs to me.

[THX1138]

The overheating behaviour on ISRU converters is pretty erratic. This is what it looks like to me so far:

• At 1000x time warp, overheating goes quickly to 0. This is the only way I found to efficiently process Ore.
  
• Once 100% overheat and "zero efficiency" is reached, there is no going back. No amount of timewarping or spacecraft cooling will cool the converter. Not even an Engineer can fix it, as far as I can tell.
  
• At 100x time warp and below, the behaviour is unpredictable. Overheating seems to tend toward 100%. It will quickly rise, but stop in a random spot. It will sometimes change, but that also appears to be random.
  
• Overheat is not the same thing as the normal temperature of a part.
  

Can someone confirm this, especially the first two? They seem like bugs to me.

The second one might not be a bug. Maybe if it reaches 100% overheating, it ceases to function forever (most things in KSP seem to explode when this happens).

Overheating in seconds, despite being fairly well shielded from the sun by the large solar panels however, must be bug.

[basic.syntax]

Sandwiching between fuel tanks does help slow down exhaustion due to heat buildup, but yeah.

I'm also looking for & experimenting with heatsink strategies to keep one running full time... without help from an Engineer.

[KerikBalm]

The second one might not be a bug. Maybe if it reaches 100% overheating, it ceases to function forever (most things in KSP seem to explode when this happens).

Overheating in seconds, despite being fairly well shielded from the sun by the large solar panels however, must be bug.

I think you misunderstand him.... I can replicate his problem... but in my case, if I stop timewarp, shut down the ISRU, then it cools down, and I can restart it, and it extracts ore at a good rate, until it overheats again.

I can time warp at 1,000x, and it will sit there at 100% heat, not generating any ore... but if I start -> timewarp until overheat -> stop -> cools down without timewarp needed -> timewarp again/repeat, I can extract ore...

... obviously this is not how it is intended to work.

It seems to be some bug where it keep generating heat, but it doesn't keep generating ore, so it stays overheated, producing nothing.

I haven't seen what happens if I'm focused on another vessel while timewarping.

[kendfrey]

I've done some more investigation with the debug menu and by viewing the save file. Overheating in an ISRU is linked to the part's heat, but not in a way you would expect. This is the behaviour as it appears to me:

• When the temperature goes above 500K, the ISRU is throttled down until the temperature goes below 500K.
  
• When the temperature goes below 475K, the ISRU is throttled up until the temperature goes above 475K.
  
• How fast the throttle moves depends on how low the throttle is, i.e. it moves exponentially. If it's running at 1% efficiency, it will take about 50x as long to reach 2% as it would from 50% to 51%.
  
• The "overheat" display is nothing but 100% minus the real throttle value. So really 100% overheat means the throttle is at 0.
  
• During timewarp, the temperature can overshoot by a long way before the converter can throttle down. If you're timewarping, the temperature could hit 510K before the throttle kicks in, in which case it will take some time to cool down to below 500K, and the converter will be throttling down exponentially the entire time, even though the converter is cooling.
  
• "Nerv" engines tend to produce temperatures significantly above 500K, so firing them can heat any nearby ISRU units to the point where they begin throttling down, thinking they're overworked, even though they're not processing any ore.
  
• Due to the nature of the exponential throttling, the longer it spends unnecessarily throttling down, the slower it will throttle back up.
  
• Due to the nature of the floating-point math used, there is a certain point where the throttle becomes so low that rounding errors take over, and it becomes stuck at a very very small value, and it can't throttle back up.
  

If you're not a geek and don't care about all that, basically all you need to know is to avoid overheating these units for extended periods of time. Using "Nerv" engines to do long burns is a good way to make sure you won't be able to process any ore when you arrive at your destination.

[THX1138]

My ISRU was reaching over 99% (but less than 100%) overheating in a couple of seconds though. I used it on Kerbin, and mostly shielded from the sun with large solar arrays. I didn't have any Nerv engines on there.

[RoverDude]

Heya - first, be sure to have an engineer.

Second - I'm looking at these numbers currently and running tests.

[Radam]

Does heat transfer trough the claw? As in that case it would be simple to "ground" the temperature of the device...

[Ryder]

I'm left wondering why *mining equipment* can't handle heat.

The bug here is that they simply aren't robust enough to be used for what they are intended to be used for.

I suppose a mining strategy, given the limitations of the tool, that landing on the dark side of the body might be tried. I landed on the sunny side of the Mun, and everything is 100% overheated all the time.

When I land (or wait) for the darkness to come... planting the drills in the ground then waiting caused them to cool... even 100% overheated drills. But the ISRU is staying overheated... even on the dark side of the Mun.

This is all very unfortunate... I hope they fix this soon. (and the BSODs that are suddenly rather common)

[kendfrey]

When I land (or wait) for the darkness to come... planting the drills in the ground then waiting caused them to cool... even 100% overheated drills. But the ISRU is staying overheated... even on the dark side of the Mun.

This looks like a bug in the math of KSP. One possible workaround (use at your own risk, I haven't tried it) is to open your save file in a text editor and find the ISRU part that's overheating. Each ISRU "MODULE" section should have a setting called "heat throttle" (I forget the exact label). Change this value to 1 on all the modules. You may also need to change the "average heat throttle" value to 1.

[Lycake]

I'm left wondering why *mining equipment* can't handle heat.

The bug here is that they simply aren't robust enough to be used for what they are intended to be used for.

Most real life mining equipment has to be cooled aswell in one way or the other. The problem in KSP however is, that we have no proper dedicated tools to cool something down like radiators, water cooling and so on. Using parts like solar arrays and especially small wings work, but they seem hacky and it can't really be intended to attach 16 wings to an ISRU so it works without overheating (works really well btw).

[RuBisCO]

Ok so here is my mining station on minmus. Ore concentration is 7.46%. To not get it to overheat I can only run one drill at a time, I have to warp right way to X10k and get at best ~23 units of ore per second, drill is at ~83% overheated. I can run the ISRU as well but only if Ore tanks are empty and the ISRU is sucking off the drills directly, if there is excess ore the ISRU goes to fast and overheats to 100%. Often if a drill or ISRU reaches 100% it become bricked and I have to go into the quicksave and change the extremely tiny numbers for "HeatThrottle" and "avgHeatThrottle" to "1" to unbrick them.

I've tried attaching things to "increase heat sink mass" like a mobile harvester/fuel truck, habitat section, and even fuel shuttle atop: no effect.

It takes a year to fill 3 Jumbs, which I guess is kind of realistic.

[THX1138]

Most real life mining equipment has to be cooled aswell in one way or the other. The problem in KSP however is, that we have no proper dedicated tools to cool something down like radiators, water cooling and so on. Using parts like solar arrays and especially small wings work, but they seem hacky and it can't really be intended to attach 16 wings to an ISRU so it works without overheating (works really well btw).

The wings might be hacky but the part descriptions now specifically state that the large solar arrays and the small solar arrays with white housing have radiators on the backs of the panels.

- - - Updated - - -

It takes a year to fill 3 Jumbs, which I guess is kind of realistic.

Do you have to remain focussed on it for it to work or does it work while you do stuff with other craft far outside physics range?

[Bill Zarr]

Exactly how many Gigantors used as radiators are needed to stop the ISRU from overheating? 16 doesn't (& that's with an unobtanium engineer)

How many should be needed for it to be a balanced game mechanic?

[RuBisCO]

So on Kerbin, on the pad, it manages to extract ore at 0.0397971/s or 75.96/s at 1000X, it can extract and convert at 68.37/s for liquid fuel and 83.57/s for oxidizer at 1000X: in short it works 100 times better on kerbin than on mimus. Ore level is 3.8%

Do you have to remain focused on it for it to work or does it work while you do stuff with other craft far outside physics range?

Yes it requires you remain focused, it does not work when I go away from it.

[RuBisCO]

With the addition of 3 engineers, 1 having a star and the other two newbes, performance increases dramatically. With just one drill I am able to get ~75/s at 1000X on Minmus, and able to get 1/4-1/2 the conversion rates I got Kerbin.

[Ryder]

Most real life mining equipment has to be cooled aswell in one way or the other.

But that's the point, isn't it... real mining equipment, INCLUDES provisions for cooling... in other words... they are engineered to work as intended.

[basic.syntax]

But that's the point, isn't it... real mining equipment, INCLUDES provisions for cooling... in other words... they are engineered to work as intended.

Do you have Engineer Kerbals helping? They give a big bonus, as RuBisCo is reporting. The system was not really meant to be run "unmanned," tho I certainly tried it that way, to poor results.

... - I'm reviewing the heating mechanics - expect it to iterate as I observe how players use it and fine tune the system.

[Seleukus]

I didn't see it mentioned anywhere that drills required Engineers. I brought 2 drills and an ISRU along to my science mission to Ike with a scientist onboard. Poor ....... was counting on turning ore to fuel to return. He's staying there for a long time