Overheat meter

[kungfufishstick]

This is strange to me. I have never known about overheating (except entry heating when coming back from orbit). Nothing was wrong with the launches I did yesterday and then all of a sudden I get an overheat meter when I launch a rocket.

Has this always been a thing? I never once had an engine overheat issue and then all of a sudden the rocket I have used for a while in career won't even make it to orbit because of overheating.

My only thought is that a mod that I updated on first launch of KSP today made this an issue. Any ideas?

[DarkGravity]

Is there any chance that you recently started overclocking your computer?

[T-Bouw]

It sounds strange indeed, but it would be helpful if you could mention the name of the mod in question?

Perhaps some folks here, who have more experience with mods, have encountered this already.

[kungfufishstick]

Is there any chance that you recently started overclocking your computer?

I don't overclock my pc. I never got into doing that since I don't understand it and couldn't afford an immediate replacement for any part that fails if something happens to go wrong.

It sounds strange indeed, but it would be helpful if you could mention the name of the mod in question?

Perhaps some folks here, who have more experience with mods, have encountered this already.

Let me see if I can recognize what was updated. I have a terrible memory.

I know for sure contract configurator, contract pack: kerbin space station, and Interstellar fuel switch were updated. I recently did a refresh of CKAN and saw a lot of boxes to check for updates so I went through and updated everything I could.

Others that I think were updated were EVAManager, Kerbal Alarm Clock, Module Manager, ModuleRCSFX, Near Future (not sure exactly which ones updated), RasterPropMonitor Core, and Toadicus Tools.

I want to say there were a couple of others but I can't recall and CKAN closed recently so I have no way of looking at what was updated.

[Alshain]

Heating has been in the game for a very long time. Engines have been able to over heat since at least 0.24 (I really don't know the exact version) and since 1.0 they can affect the parts around them through heat convection. That is to say, they transfer their heat to nearby components. There are important logic rules, which I myself don't know from memory, that govern how and when heat is transferred. I often see control fins get hot when ascending, but usually not enough before I'm ready to stage. The heat gauges themselves were not added till 1.0.2.

Edited May 26, 2015 by Alshain

[kungfufishstick]

Heating has been in the game for a very long time. Engines have been able to over heat since at least 0.24 (I really don't know the exact version) and since 1.0 they can affect the parts around them through heat convection. That is to say, they transfer their heat to nearby components. There are important logic rules, which I myself don't know from memory, that govern how and when heat is transferred. I often see control fins get hot when ascending, but usually not enough before I'm ready to stage. The heat gauges themselves were not added till 1.0.2.

That is good to know. That's a reason I love this game. So much has gone into it since I played the demo years ago and I am discovering things about KSP the hard way.

I just decided to roll with it. The redesign I did for my satellite launcher seems to be far more efficient and less costly than the original, so I guess running into engine overheat was a blessing in disguise.

This game is so epic. My only complaint is the memory issue. I hope squad gets a fix so I can get rid of the need for ramrush.

[Alshain]

Yep, those gauges are killer. I hope we see 1.0.3 soon to fix that.

[Commander Jebidiah]

Is it possible you pressed f10? press it again to get rid of temperature guages on ships.

Edit: Thanks Varsi for spotting that.

Edited May 26, 2015 by Commander Jebidiah

[Varsi]

Is it possible you pressed f12? press it again to get rid of temperature guages on ships.

Isn't F10 for gauges. If my memory hasn't completely failed the F12 is for aerodynamicthingies.

[Murph]

One important thing to realise with overheat, is that it can often be entirely ignored. It only matters if you reach 100% overheat. If your launcher get to 99.999% overheat by the end of launch, you're actually ok (but have almost no error margin, so should probably try to improve that). 100% == KABOOM! Less than 100% == no immediate consequences.

- - - Updated - - -

This game is so epic. My only complaint is the memory issue. I hope squad gets a fix so I can get rid of the need for ramrush.

Try getting rid of "ramrush". A quick bit of searching has me thinking that it's about 99% likely to be completely unnecessary snake oil. Tools like that have been obsolete for about 15 years now, with modern operating systems. Sometimes these cleaners or optimisers can actually do more harm than good, and even the ones that don't do any real harm are little more than a placebo or snake oil while consuming more resources than not using them at all.

One thing that is absolutely certain, is that "ramrush" cannot make any difference to problems with KSP crashing due to running out of memory.

[magnemoe]

Heating has been in the game for a very long time. Engines have been able to over heat since at least 0.24 (I really don't know the exact version) and since 1.0 they can affect the parts around them through heat convection. That is to say, they transfer their heat to nearby components. There are important logic rules, which I myself don't know from memory, that govern how and when heat is transferred. I often see control fins get hot when ascending, but usually not enough before I'm ready to stage. The heat gauges themselves were not added till 1.0.2.

Heating moves trough the ship from engines, this effect continues after the engines has been turned off, I once had static solar panels explode after engine shutdown, they was mounted on the radially mounted nosecones who continued to heat up after the LV-N was off.

radial batteries also don't like heat, same with RTG and 0.625 meter probes.

[Geschosskopf]

Heating has been in the game for a very long time. Engines have been able to over heat since at least 0.24 (I really don't know the exact version) and since 1.0 they can affect the parts around them through heat convection. That is to say, they transfer their heat to nearby components. There are important logic rules, which I myself don't know from memory, that govern how and when heat is transferred. I often see control fins get hot when ascending, but usually not enough before I'm ready to stage. The heat gauges themselves were not added till 1.0.2.

Actually, a hot part affecting another has been around since forever. Putting a Mainsail directly beneath an orange tank always made it go BOOM, so you'd stick a pancake tank in between. Also used to be, there'd be these overheat bars in the staging list added on to the fuel bars. The Mainsail always had one of those and would itself explode if you filled the bar up, even if you'd prevented it from exploding the orange tank. The bottom line was, you had to throttle Mainsails down to like 90% or they'd explode during launch. Since 1.0, however, Mainsails haven't had any overheat problems at all. This is something nobody seems to talk about in all the arguments about LV-N overheating .

And to be a bit pedantic, parts transfer heat between each other via conduction, not convection. Conduction is the flow of heat through direct physical contact. Convection is the relative movement of bodies of fluids with different temperatures and AFAIK the game only really uses convection as a cooling mechanism, allowing parts to cool down faster if they're surrounded by air or water. I've heard some folks say the nose of a rocket can heat up air, which then blows on and heats up other parts of the rocket, but that's really more conduction than convection. The game also appears only to deal with radiant heat transfer as a cooling mechanism, the default rate at which parts cool in vacuum. It ignores radiation for making things hotter, so that the flames of radially mounted engines have no effect on nearby parts.

[katateochi]

This game is so epic. My only complaint is the memory issue. I hope squad gets a fix so I can get rid of the need for ramrush.

Until Squad fixes it you can use this to disable it by default - DMagic's Temperature Guage Killer mod

[Wanderfound]

Another overheating point that is often overlooked: gimbal.

Gimballed rockets are good from a stability POV (especially for FAR), but they do have a drawback: if you mount the gimballed rockets in the wrong place, at extreme gimbal they'll cook their neighbours within seconds. Be careful with where your thrust is going.

[kungfufishstick]

One important thing to realise with overheat, is that it can often be entirely ignored. It only matters if you reach 100% overheat. If your launcher get to 99.999% overheat by the end of launch, you're actually ok (but have almost no error margin, so should probably try to improve that). 100% == KABOOM! Less than 100% == no immediate consequences.

- - - Updated - - -

Try getting rid of "ramrush". A quick bit of searching has me thinking that it's about 99% likely to be completely unnecessary snake oil. Tools like that have been obsolete for about 15 years now, with modern operating systems. Sometimes these cleaners or optimisers can actually do more harm than good, and even the ones that don't do any real harm are little more than a placebo or snake oil while consuming more resources than not using them at all.

One thing that is absolutely certain, is that "ramrush" cannot make any difference to problems with KSP crashing due to running out of memory.

The original satellite launcher I had built probably weighed too much for the engines I had for the launch stages. I just wasn't sure what caused it to get overheat all of a sudden. The new model I designed works great and doesn't reach a critical overheat point. The old version was no where near orbit and was at least 75% on the overheat. (ap was around 50k)

I disagree with you about ramrush. I alt tab to check the process monitor and when it reaches around 2 GB I hit the keyboard shortcut and it reduces KSP down to 500k and then it raises from there. Rinse and repeat. I haven't had a crash since using it regularly.

Until Squad fixes it you can use this to disable it by default - DMagic's Temperature Guage Killer mod

I have this mod installed and KSP still gains memory until it crashes without ramrush.

[Murph]

I disagree with you about ramrush. I alt tab to check the process monitor and when it reaches around 2 GB I hit the keyboard shortcut and it reduces KSP down to 500k and then it raises from there. Rinse and repeat. I haven't had a crash since using it regularly.

All that is doing is slowing KSP down by forcing the operating system to swap out less active pages of memory. It does absolutely nothing to help with the amount of per-process address space used by KSP. No tool of that nature can ever help with that, it's simply impossible. KSP will crash when total virtual address space usage hits 4GB (minus the Windows overhead), whether it has 500kB, 500MB, or 3.5GB in physical memory at the time. Your actions are actually harming the performance of KSP, as you are forcing it out to virtual memory (on disk), which means that when it needs to access a portion of code or data that wasn't active at the time of your last purge, there will be a big pause and delay as the operating system loads the needed pages of memory from disk.

A relevant degree followed by decades of experience tells me that "ramrush" is mostly placebo or snake oil, with most of its claimed features being dubious at best, or outright false claims or misrepresentation at worst. The claim it makes to "remove memory leaks" is 100% snake oil, for example. The leaked memory is still allocated to the process that leaked it, it's just swapped out to disk, which is something the operating system does automatically when it is actually required and useful to do so (and absolutely does not need snake oil tools to help it).