Of Radiators, Reentry, and Random Explosions

[Geschosskopf]

So I put a couple small, fixed radiators on a ship just for kicks, to see just how well they worked at eliminating reentry heat. I noticed no difference there. But what they did do was blow up the part they were attached to after using time-warp out in space. Despite the ship not being hot at all.

In this case, the radiators were attached to the 1.25m SAS unit about in the middle of the ship. Also attached to this SAS unit were 2 1x6 solar panels (the ones that look vertical in the left part of the image). The left image shows the ship immediately after leaving Kerbin's atmosphere after an aerobraking pass at 40km. The entire ship is the "cool" color, no thermal gauges showing. At this point, I warped around orbit (Ap about 5Mm) for another pass. I stopped warp at about 120km, before hitting the atmosphere again, to get lined up properly retrograde, retract the solar panels, etc. As soon as the warp stopped, the SAS unit blew up, tearing the ship in half as shown in the right part of the pic.

The plan had been to lower the Ap down to a couple hundred km, circularize and then deorbit with the engine, stage that, and land the pod and the science stuff and SAS unit. The heat shield was on the rear side of the SAS unit, so when the ship was blown in half, no more heat shield. There being nothing for it, the decapitated head of the ship came in for a landing sans heat shield, leading with a 1.25m service bay from a 5Mm Ap. Reentry caused no damage at all to any of the remaining parts. This was with 100% reentry heat.

So, something seems terribly wrong here. Radiators don't help, they hurt. Furthermore, the fact that hte SAS unit exploded reveals a flaw in the model, with heat flowing into the SAS unit faster than the attached radiators could get rid of it. It doesn't work that way.

[Gamer025]

Are you seriously trying to deal with reentry heat by putting radiators on your ships?

[windows_x_seven]

Heat is so glitchy right now, clipped parts overheat and explode as well.

[Geschosskopf]

Are you seriously trying to deal with reentry heat by putting radiators on your ships?

No. I had just heard people were doing that so wanted to see if it worked. It shouldn't. If the ship is getting hot from the air, then the air is hotter than the ship. Because heat only flows downhill, radiators should not work during reentry. Anybody saying they do is pointing out a flaw in the system. So I tried it just to see for myself. I saw no difference in heating with or without radiators, and an explosion with radiators that doesn't happen without them. So there's a bug there, anyway.

Heat is so glitchy right now, clipped parts overheat and explode as well.

No clipped parts here. This was just a part getting hot for no reason. Glitchy, yes.

[Rhomphaia]

There is a bug where stowed parts can cause rapid overheating.

There is another bug where parts not in a service bay think they are stowed

maybe you have hit a combo.

[Scotius]

Can you and the other players replicate it? It can be a one-time glitch, not necessarily inherent bug of heat system.

[Megadeath]

"If the ship is getting hot from the air, then the air is hotter than the ship." This statement doesn't match my understanding of the situation. I'm of the belief that most heat from reentry is caused mostly by compressive, or friction heating (no idea which) rather than ambient air temperature. In either case, wouldn't there be a wind shadow behind the vessel at lower temperature than the leading face? I thought the reality issue with radiating reentry heat was that it's skin heating, not the core.

I've also had heating glitches with time warp. Often with service bays, but not always. At least once it was enough to explode a station. It looked spectacular.

[Geschosskopf]

Can you and the other players replicate it? It can be a one-time glitch, not necessarily inherent bug of heat system.

With this ship, yes, I could replicate it. Come back from Mun, make a pass at Kerbin at 40km and all is well. Once out of the air, warp to any point in the ensuing orbit prior to hitting the air again. The SAS unit would explode immediately upon coming out of warp, no matter where you did this. If you didn't warp, then no problem, other than having to spend hours of real time on the landing.

"If the ship is getting hot from the air, then the air is hotter than the ship." This statement doesn't match my understanding of the situation. I'm of the belief that most heat from reentry is caused mostly by compressive, or friction heating (no idea which) rather than ambient air temperature. In either case, wouldn't there be a wind shadow behind the vessel at lower temperature than the leading face? I thought the reality issue with radiating reentry heat was that it's skin heating, not the core.

As explained towards the end of this thread, apparently the ship is surrounded by an envelope of hightly heated air/plasma, the heat from which enters the ship. This is called "forced convection" although it operates more like direct conduction.

Anyway, heat only flows downhill, from hotter things into colder things. Thus, for the ship to get hot at all, it must be surrounded by hotter air. Therefore, radiators should be counterproductive during reentry, increasing the rate of ship temperature rise by increasing the surface area exposed to the surrounding hot air.

I've also had heating glitches with time warp. Often with service bays, but not always. At least once it was enough to explode a station. It looked spectacular.

This ship had both a 1.25m service bay (containing an OKTO core and 8x small radial batteries) and a 4x Universal Storage thing with DMagic Orbital Science wedges (2 Goo, 1 Materials, 1 Orbital Telescope). US has been updated to 1.0.4 but DMOS is still 1.0.2 and its wedges tend to get hot during reentry if air can flow over their outer surfaces (IOW, using a US thing the same diameter as the ship). That was the case with the above ship, although at 40km their thermal gauges remained green. Interestingly, the thermal gauges vanished instantly upon leaving the atmosphere. The F11 heat color code thingy, however, showed everything nice and cool, as shown in the pic.

Anyway, because DMOS isn't for 1.0.4, the data here is of course questionable. But I thought it was interesting enough to post up and see what others were experiencing.

[MalfunctionM1Ke]

I had a satelite in Orbit of Minmus and it exploded completely during timewarp.

First I thought it might have accumulated heat because i kept it in the Sun at all time but I tried that with another Satelite and it was fine.

Weird glitch.

[Goddess Bhavani]

Are you seriously trying to deal with reentry heat by putting radiators on your ships?

I use a radiator to channel heat away from hot parts and towards a 'cold' part of the ship during re-entry. Kinda works.

The idea is to simply draw heat away from the leading edge of the vehicle where the cockpit section, without heatshield, is taking the brunt of re-entry heat.

[PB666]

So here is one heat anomaly.

I have a Munar lander. Its designed to go back and forth from a fuel station while it picks up little tidbits of low-hanging munar fruit. (under the optimization strategy that time is as valuable as money or reputation).

So the little engine that this lander has causes the heat gui to turn green and sometimes orange during the long ascent and landing burns.

The craft has 3 of these engines and 3 small heat radiators. The gui on many parts, mostly science, goes orange eventually subsides to green.

However_ after 9 days on the Mun, the last of which were on the dark side of the Mun, the gui is still green. During the period I turned the game off, shut down the computer and restarted.

Having filled the vessel with science I then returned it to Kerban, still gui showing green on 30 or so parts.

I set the reentry ellipse from the moon to 50k Pe and when the crafts (retrograde) erupted in "reentry plasma light-show" the GUIs all disappeared.

Apparently the heat model for the game does not know how to deal with radiative heating on a light-less, vacuum-less planet.

It does know how to dissipate heat purely radiatively.

I did a partial retro-burn partial aerobraking descent to Kerbin, during 4x physics warp one part, I don't know which exploded although no heat bars were showing.