One way stock 1.0 heat shielding is a bit unfair versus old DRE mod's heat sheilding.

[Dunbaratu]

KSP 0.9 DRE mod's heat shielding: As long as you don't make your return capsule too tall, the areodynamics will naturally keep your heat shield pointed forward, in both stock and FAR areo models.

KSP 1.0's heat shielding: Your return capsule wants to flip nose-first and requires powerful torque to keep the heat sheild in front.

I just lost Valentina to this effect - the torque of a single capsule wasn't strong enough to keep a vessel composed of capsule, science jr module, and heat shield, pointed butt-first on re-entry.

[tater]

Yeah, the whole point of capsules is that they are stable with the shield down, 1.0 requires constant control input. How could testers not see this?

[Empiro]

I think that's more of an aerodynamic issue than anything else. I've noticed the instability of many rockets on the streams over the weekend, even when they're pointed prograde with stabilizing fins at the back.

[Jouni]

The heat shields are physicsless parts, so they change the craft's shape, but don't move its center of mass.

[Robotengineer]

I just lost a mission because I don't start with Ablator.

[Daid]

Tried a simple parachute to keep it stable? I noticed that chutes are a lot less likely to rip your ship into a thousand pieces. And just deployed a single chute for the above reason while I was still going at 1000m/s at 10km.

[mikegarrison]

I haven't come in at anything but sub-orbital speeds yet, but so far what has worked is to trigger the small chute (Mk 16?) in space. About the time the capsule really wants to teakettle, the magic chute pops out and a 20+g force pulls from the top of the capsule. No more flipping.

KSP 0.9 DRE mod's heat shielding: As long as you don't make your return capsule too tall, the areodynamics will naturally keep your heat shield pointed forward, in both stock and FAR areo models.

KSP 1.0's heat shielding: Your return capsule wants to flip nose-first and requires powerful torque to keep the heat sheild in front.

I just lost Valentina to this effect - the torque of a single capsule wasn't strong enough to keep a vessel composed of capsule, science jr module, and heat shield, pointed butt-first on re-entry.

[FenrirWolf]

The heat shields are physicsless parts, so they change the craft's shape, but don't move its center of mass.

That seems kind of odd. Would changing that with a module manager config resolve the issue?

[tater]

In DRE/FAR (haven't tried stock 0.90 since it first came out) a mk1 as is, or a mk1-2 with a heat shield and nothing else but antenna/chute is 100% stable no SAS on reentry---as it should be. In 1.0 it is not. The entire point of capsules is that they are stable blunt-side down. If they are not, something is broken.

[Leszek]

I am not sure why they did this. The basic capsule still has the shakes and kills your battery quickly, meaning manual entry sometimes.

I thought that perhaps they didn't want to add weight to the rocket but the new aero makes it easier to get to space anyway so it should more than balance out.

Perhaps some ballast parts should be made. (That way you can have off center balance, or not.) These parts should ideally be sandwiched between the capsule and the heat shield.

[Basilo1146]

I've noticed this as well, and it greatly annoys me, as this does not happen realistically. It seems almost like the air resistance being applied to the sides of objects when they start to tilt is way, way less powerful than it should be, so the capsule is trying to flip to the pointy side down.

[John FX]

Sounds wrong to me. You should be able to have no input and for the pod and HS to be stable.

[Dunbaratu]

Tried a simple parachute to keep it stable? I noticed that chutes are a lot less likely to rip your ship into a thousand pieces. And just deployed a single chute for the above reason while I was still going at 1000m/s at 10km.

Really? Wouldn't a chute deployment just instantly rip apart in the heat?

- - - Updated - - -

I'm surprised this didn't show up during all that experimentals testing. I mean - it's obvious the very FIRST time you try to de-orbit.

[DJK]

In DRE/FAR (haven't tried stock 0.90 since it first came out) a mk1 as is, or a mk1-2 with a heat shield and nothing else but antenna/chute is 100% stable no SAS on reentry---as it should be. In 1.0 it is not. The entire point of capsules is that they are stable blunt-side down. If they are not, something is broken.

Mk1 pod + heat shield was perfectly stable for me once it hit the higher air densities. Required a bit of input during the first 15km, but after that it was as stable as it gets. Even tried to purposely flip it over and couldn't.

[ola]

To be perfectly honest, I find stock re-entry broken to the point of rage quitting. I cannot see any way that a new player would appreciate the way this is implemented. You need to hit an extremely narrow entry corridor and fine-tune your craft all the way down, with teeny tiny control inputs, otherwise it'll flip around and heat-kill your parachute. DRE + FAR did this way, way better. How the heck did this go past QA?

[Coga19000]

Huh, that's odd. The Enterelysium video showed nothing like that.

[ola]

And I totally agree with regard to chutes, any chute opening above mach one deserves to be ripped to shreds.

[linarn]

If I remember right the heat shields on all USA crew pods had off set center of grav to keep the craft at the correct angle so there was no need for the crew to keep making changes, just incase they were not able to make siad changes.

Can not see why the Kebrals would not have come up with the same idea. then again they are Kerbals they like doing it the hard way.

[Johnny Wishbone]

earlier, i launched a single kerbal capsule with nothing but a chute on a 75k sub orbital trajectory. once it hit apoapsis, i turned off SAS and let Jeb ride it down ballistically. the capsule oriented itself blunt end first once it started hitting thick enough atmosphere and stayed on the retrograde marker the whole way down with no input. i was surprised it survived without a heat shield (which prompted my question in my thread) and i popped the chute about 3000m up and was surprised again by how much an undeployed chute slowed the capsule (down to like 25m/s). splashed down safely with no ill effects on Jeb.

[ola]

earlier, i launched a single kerbal capsule with nothing but a chute on a 75k sub orbital trajectory. once it hit apoapsis, i turned off SAS and let Jeb ride it down ballistically. the capsule oriented itself blunt end first once it started hitting thick enough atmosphere and stayed on the retrograde marker the whole way down with no input. i was surprised it survived without a heat shield (which prompted my question in my thread) and i popped the chute about 3000m up and was surprised again by how much an undeployed chute slowed the capsule (down to like 25m/s). splashed down safely with no ill effects on Jeb.

Yeah, I noticed that too. Try a 800k sub orbital and see what happens. Or a Mun pass...

[Jarin]

Yeah, not seeing any of the problems on my initial tests either. Going to need more information on the situations people are in when they're seeing this flipping behavior. Did you do a clean install? Is it possible you have remnants of 0.90 lying around borking things up?

[X-SR71]

Yeah, I noticed that too. Try a 800k sub orbital and see what happens. Or a Mun pass...

Well, perhaps we're supposed to properly deorbit and reduce speed first before entering kerbin, instead of only worrying about getting captured by the atmosphere straight from the mun..

[Johnny Wishbone]

Yeah, I noticed that too. Try a 800k sub orbital and see what happens. Or a Mun pass...

Oh, I'm sure higher re-entry velocities will burn up an unshielded capsule, but even at 75k apoapsis, I expected to need a heat shield. Maybe my expectations were too high because of all the hype and nonsense over this release.

[Radam]

I had a capsule, like 12 smallest tanks and a first liquid engine. I failed to get to proper orbit but still had 2100ms at apoapsis. The craft orientated itself retrograde by itself and no part overheated during reentry... Well the engine did smash itself at 10ms landing, otherwise it was fine.

[ola]

Well, perhaps we're supposed to properly deorbit and reduce speed first before entering kerbin, instead of only worrying about getting captured by the atmosphere straight from the mun..

Ie, opposite of how NASA did the moon landings?