Heat shield mass added to pod, heatshield physicsless

[Dunbaratu]

Drag chute to stabilise the craft during re-entry worked a treat, Val is safe and sound

Fixing one bug by exploiting another doesn't appeal to me. And yes, he fact that parachutes live through heat just fine that tends to explode the main craft is certainly buggy. And if that's the way they wanted it then it's buggy by design rather than buggy by code, but still buggy. Broken as intended.

- - - Updated - - -

It is intended that the sciencejr and heat-shield add to its parent parts weight without changing the center of mass, making the capsule top heavy? It's not a bug?

The science jr doesn't have the bug. It's mass is being added properly and moving the CoM properly. For it, it's actually *correct* that it tends to throw off the balance by being a less dense part on the bottom. What's incorrect is how the heat shield does so even MORE than the science Jr does, because it's even less dense, with its mass of zero, and is thus acting more like an airbrake or a sail.

[omnomtom]

Presumably this is for when coming back from Moho/Eeloo and you're really booking it, although I have yet to get that far out.

Actually, the lack of ablation is because of using the chute to maintain orientation - the parachute has enormous drag and absorbs almost all the energy of re-entry, leaving very little required for the heat shield. Try a munar aerobraking return without using the chute (with extra reaction wheels and batteries to maintain orientation as long as possible) - it's fallen off stable orientation a bit too much each time for me (it seems to have to be exactly oriented retro for the HS to protect all the parts, but I didn't have a 'Lock Retro' capable SAS aboard when I tried this), but before doing that the ablative shield has lost 20-30 units.

Had 3 failed attempts to aerobrake return from munar orbit because it's not aerodynamically stable, then tried a direct return with chutes staged before hitting atmo - the chute deployed and the craft pulled 93.6g when it popped (at least according to engineer). No problems, and that time the shield only lost 4 units instead of 30.

[furrykef]

I think this should be tagged as "Bug". I was a little cheesed when I saw this marked "Not a Bug" until I was reassured later in the thread that it was, in fact, acknowledged by the devs as a bug.

Is this going to be hotfixed? Heat shields being borderline useless is a pretty darn serious issue, in my opinion.

[furrykef]

I thank the moderators for finally marking this issue as a bug.

[Superluminaut]

Fixed!!

[furrykef]

This issue has been hotfixed and I thank the devs greatly for this -- but still, was it the right fix?

The changelog says "Fixed an issue where the physicsSignificance flag was set to 1 for heat shields", but as noted in the thread, the real problem, as I understand it, wasn't the part being marked physicsless, but that all physicsless parts behave incorrectly because they are still accounted for in drag equations when they shouldn't be. So now heat shields work correctly because they're no longer physicsless, but other parts such as batteries and landing gear will still misbehave.

I understand of course that the fix we got was much easier and quicker to implement and test in a short amount of time, and the issue was serious enough that a quick fix was needed -- so I'm not complaining; I think you guys did the right thing for now. I'm just noting that the fix is probably addressing the symptom and not the cause, so there is still an issue that needs to be fixed.

[Claw]

I think, generally, the intent is for the smallest parts to be physicsless. That basically applies their drag to the parent part (and is supposed to do the same for mass). This way, you can design things that are slightly asymmetric in terms of drag, but it doesn't cause a bunch of annoying little stability issues. (Like a ladder on the side of an airplane.)

The problem with the Chute-Pod-Heatshield combo is that the chute pulled the CoM up, but the heat shield didn't pull the CoM down but did offset some of the drag, as applied to the capsule. The heat shield was probably way to heavy and to big for this sort of thing (unlike some of the small science bits) to be doing this sort of thing. Turning the physics back on for the heat shield should now have things acting more or less correct.

Cheers,

~Claw

[Leozaur]

Yep, that physicsSignificance config line fixed it correctly even before the patch, and now it's official. No issues with reentry anymore, thanks guys!

And fairings are fixed too, hurray

[POllik]

Hmmm.

I tried to update, was offered a dialogue box to login with email and password, which I did with my KSP login (not my forum login) and I get a message:

Preparation

Detecting if OS-resident rsync is present...

rsync is not already installed, so making use of bin\rsync.exe.

Authentication

Getting login token...

Unable to authenticate with the server: Received malformed data from the server: File not found.

The update doesn't seem to have come down.

So, instead I downloaded a new install of 1-0-2 (my login worked fine, there), but if I move my save files over, I get told the science tree has not been unlocked.

Edit: I am stupider than I look. The science tree issue was my failure to move Mechjeb over. Sorry

Edited May 3, 2015 by POllik