Pet peeve: orientation during atmospheric re-entry.

[Vanamonde]

Why not put fins at the very top of your lander?

Causes nasty problems on ascent, and requires carrying otherwise useless parts all throughout the flight.

SAS should also help to stabilize it.

If you look at my pictures, SAS is on, and trying to hold attitude. It gets over-powered.

- detach the capsule and re enter with just that, as suggested before;

Viable option, if one is resigned to losing the recovery funds for the discarded parts.

Spam with SAS and enough batteries to last re-entry.

Carry multiple parts all through the entire flight just to help in the last few moments? Possible, but wasteful.

I think that you shoudn't grab and hold yourself into the past...

It's not solely about the past. The game has a mechanic which rewards the player financially for bringing back as much of the craft as possible. But this issue with the aerodynamics, however, penalizes one for bringing back additional parts. If it's the case now that we must discard everything but the capsules when landing on kerbin, they might as well scrap the parts recovery system as well, because then it would only meaningfully apply to aircraft and spaceplanes.

Add a drogue chute.

I'd love to. So far, however, it's been my experience that chutes opened at the altitudes/temps where this is happening are quickly destroyed.

As is so often the case on the internet, some of you are replying to things I never said. I do not want a step backwards in aerodynamics. I am not saying it's unrealistic for the parts to behave this way with their current attributes. I am not asking for a revision of physics. What I am saying is that the attributes of the parts are arbitrary and entirely under the control of the game's makers, and it would eliminate a problem I (and possibly others) are having if the part attributes were adjusted.

And seriously, guys, stop telling me I'm trying to force others to play my way. I have stated several times that this change would be my preference rather than an insistence, and the purpose of this thread is to gauge support for what I would like to see done about it. (It's not looking good so far, I must say.)

[pellinor]

I'd doubt that part attributes are as arbitrary as you think, they still need to fit into a consistent system. Have you tried playing with them? What changes would you propose?

About the general problem, I fully agree. Aero sort of acts as a black box at the moment: you put in your design and you get some strange result without any explanation why it behaves that way or how you could change it.

The "stable orientation" display someone suggested sounds like a good idea. I'm just not sure if such an orientation is unique, or is guaranteed to exist at all for an arbitrary craft.

[Vanamonde]

And for those complaining about the design of my lander, consider that this little thing also insists on falling front-first.

[JoCRaM]

Carry multiple parts all through the entire flight just to help in the last few moments? Possible, but wasteful.

Why bother with parachutes then?

The game has a mechanic which rewards the player financially for bringing back as much of the craft as possible. But this issue with the aerodynamics, however, penalizes one for bringing back additional parts. If it's the case now that we must discard everything but the capsules when landing on kerbin, they might as well scrap the parts recovery system as well, because then it would only meaningfully apply to aircraft and spaceplanes.

If you want to bring more parts back, you have to design around it. The extra cost of bringing parts back will often outweigh the savings - look at how little of Apollo was reusable. Look at how much payload Musk will lose when he boost backs his lower stage.

I'd love to. So far, however, it's been my experience that chutes opened at the altitudes/temps where this is happening are quickly destroyed.

Does that include drogues? I honestly don't know I've never had need to use them.

What I am saying is that the attributes of the parts are arbitrary and entirely under the control of the game's makers, and it would eliminate a problem I (and possibly others) are having if the part attributes were adjusted.

So you want the science bay and the fuel tanks to be made of lead, so they will lead during rentry?

Edited June 15, 2015 by JoCRaM

[tater]

And for those complaining about the design of my lander, consider that this little thing also insists on falling front-first.

http://i.imgur.com/Or2Eslq.png

As it should, the science bay masses nearly nothing.

The part description for the landercan says it cannot possibly survive reentry (a lie, obviously), so don't do that.

[Vanamonde]

Why bother with parachutes then?

Because they work properly and there is no effective alternative.

If you want to bring more parts back, you have to design around it.

I have designed around it. Now the conditions have changed, and I will have to lose more than I used to. This is not desirable.

Does that include drogues?

Yes.

So you want the science bay and the fuel tanks to be made of lead, so they will lead during rentry?

Wild mischaracterizations don't help the discussion. I was just spitballing, but a bit more drag and less mass for capsules was what I had in mind.

As it should, the science bay masses nearly nothing.

People were telling me I should accept having to trim more parts off of my landers to get them to fall right, so that was an example of a probe lander which has no extraneous parts, and still falls nose-first. You're telling me you're okay with that little thing being unable to orient its engines for braking burns, and snapping violently around when chutes open?

The part description for the landercan says it cannot possibly survive reentry (a lie, obviously), so don't do that.

I didn't do that. Its root part is an OKTO probe core.

Guys, you're free to disagree with me, but the save the smarmy condescension. It doesn't improve discussions.

[basic.syntax]

It's mildly funny that the first lander falls with the lighter part facing down, while the second with the Science Jr, falls with the heavy part facing down That's an oversimplification, however. The first is broad, with three 1.25m parts arranged radially, and should catch more air flow, while the second is narrow, dominated by a single 1.25m stack.

Both of these landers are too small to visually accommodate airbrakes, but, they are a useful tool for stowing - and later deploying - drag, where you want it. It might help you, on larger designs. For rocket reentry use, I disable all control axes in the VAB, remove them from the Brakes action group, and toggle them with a custom group.

[Tankman101]

The first picture has the fuel tanks acting essentially as fins, adding more drag farther back on the craft. I doubt any amount of mass changes will help that.

The second picture has what I assume to be a returning mun lander with a mostly empty fuel tank entering nose first due to the capsule moving the CoM forwards. Honestly it's stubby enough that I think that's fine. Keep in mind reducing the mass also would have an a(/e , I have no clue)ffect on the lander capsules (They would also need to be lighter).

As for the tiny lander with an octo and science junior, keep in mind the radial parachutes weigh a lot. There's also at least six shielded solar panels above the Sci Jr. The thing may be heavy, but not heavy enough to be a counterweight.

Overall I'm not convinced there is a problem. At least not yet.

[tater]

The root part doesn't matter. Even as a noob, not knowing that reentry wasn't actually a thing I never built things that looked like the LEM and tried to reenter them. You could, but only because KSP was silly and utterly unrealistic about reentry (still is).

The first craft is not a capsule any more than the entire rocket that launched it would be by virtue of having a capsule on top.

Again, if aero is realistic to zeroth or first order, then if it flips, then so be it, design better craft for reentry. If aero is unrealistic in such a way that your craft should behave otherwise, then aero needs fixing. If aero needs fixing in a way that makes your airless world lander even less able to reenter, then that is what it should change to. Arbitrarily changing parameters so that lumpy craft can fly backwards through the air at 2-3 km/s makes no sense at all.

BTW, how do we know the 1st craft is heavier at the bottom, the tanks are mostly empty.

- - - Updated - - -

I just built a copy of the first craft, with slightly more fuel than posted in that image (~40 liquid fuel, setting it in the VAB let me pick 13.5 or 9 per tank), the CM is right at the interface of the capsule and Science Jr. Drained slightly the CM is in the capsule. If it flips, what a surprise.

- - - Updated - - -

Reality check #2.

I just reentered my copy of the above craft. Set the OCTO to point retrograde. Retracted 2 panels until it started showing reentry effects, then retracted the last one. It never flipped, popped the chutes and it landed fine. The game is just as silly as one would expect. I was in LKO. It's a lot to ask that a munar lander type vehicle be able to reenter not "at all," but on a direct reentry from the Mun. Mine didn't even wobble at 2300 m/s.

So the craft doesn't flip with SAS on. I can only assume the problem is that it should.

The fact that stuff like this can survive reentry does merit a suggestion thread---that reentry needs to be far nastier than it is.

Edited June 16, 2015 by tater

[kellpossible]

And seriously, guys, stop telling me I'm trying to force others to play my way. I have stated several times that this change would be my preference rather than an insistence, and the purpose of this thread is to gauge support for what I would like to see done about it. (It's not looking good so far, I must say.)

Fair enough, and brave of you to show your designs here in this thread, knowing that there would be those who would criticise them. I hope you're still okay with this thread serving a secondary purpose of receiving feedback and through this allowing yourself or others to learn more about how the physics works in the game and in real life (and thus how to design better craft), which is clearly how many of the responders here are taking it.

Causes nasty problems on ascent, and requires carrying otherwise useless parts all throughout the flight.

Carry multiple parts all through the entire flight just to help in the last few moments? Possible, but wasteful.

As pointed out before, sometimes parts placed on your craft will serve only once purpose. Science modules tend to be fairly useless when they aren't gathering science, and they make the craft heavier with less delta-v...

Wings are pretty useless on SSTO's once they get into space. In fact why should an ssto need wings at all? Actually I guess it might be possible to design an ssto craft without wings if you didn't want to have them, just like it's possible to design a re-entry craft without stabilizing fins, but as always in KSP, design compromises must be made.

It's not solely about the past. The game has a mechanic which rewards the player financially for bringing back as much of the craft as possible. But this issue with the aerodynamics, however, penalizes one for bringing back additional parts. If it's the case now that we must discard everything but the capsules when landing on kerbin, they might as well scrap the parts recovery system as well, because then it would only meaningfully apply to aircraft and spaceplanes.

...

As is so often the case on the internet, some of you are replying to things I never said. I do not want a step backwards in aerodynamics. I am not saying it's unrealistic for the parts to behave this way with their current attributes. I am not asking for a revision of physics. What I am saying is that the attributes of the parts are arbitrary and entirely under the control of the game's makers, and it would eliminate a problem I (and possibly others) are having if the part attributes were adjusted.

From what I can see, you personally have a strong attachment to the part recovery mechanic in the game which has carried over across versions? Perhaps more so than a lot of us here, who I'm guessing, at least early on in our KSP designing careers with FAR or the new 1.0 aero, write it off as a lost cause, after observing real life space programs, make the observation that the holy grail of re-usable spacecraft is a huge design hurdle, not to be approached lightly.

I think one of the beautiful things about KSP erring on the side of realism in this case is that those who have even a simple understanding of aerodynamics, like that of experimenting with paper planes, can take their knowledge gained through that experience and apply directly to the game. And vice versa, I like to think (and perhaps this is somewhat misguided, but given I'm studying engineering, all the reading I've been doing lately on the topics inspired by KSP seems to confirm that idea.) that after playing a decent amount of KSP I should have a better chance at designing my own aircraft or rocket in real life. From my point of view, (and I'm not saying yours is "wrong" or anything), you're suggesting that it would be better for the game to allow paper planes to fly nicely backwards when thrown, when all real life experience seems to suggest otherwise, it would be equally frustrating for those who have experience with making paper planes as it would be for you if it wasn't this way. I'd also suggest that you'd be finding the process of making paper planes just as frustrating if you applied the same principals.

[Vanamonde]

There are two arguments going on in this thread.

1) That assembly is falling the way physics says it should. Great, Dandy. I never said the physics should be otherwise. I am not part of that argument. I don't care about it. Please, take it somewhere else, because it has nothing to do with what I'm talking about.

2) This is undesirable behavior for that assembly and the attributes of the parts could be adjusted so that this does not happen in realistic physics. THAT is my suggestion and request.

How many times do I have to say that?

Even as a noob, not knowing that reentry wasn't actually a thing I never built things that looked like the LEM and tried to reenter them.

Well, that's too bad for you, because it used to work, and recovered more of the cost of the craft. Just because you chose not to utilize this doesn't mean others shouldn't.

Arbitrarily changing parameters so that lumpy craft can fly backwards through the air at 2-3 km/s makes no sense at all.

Arbitrarily changing parameters for gameplay reasons is what game balancing is all about. Are you suggesting that the way parts currently behave shouldn't be adjusted, and all players must learn to work around their flaws rather than fix the parts?

Drained slightly the CM is in the capsule. If it flips, what a surprise.

Applies to argument #1, which I don't care about.

... you're suggesting that it would be better for the game to allow paper planes to fly nicely backwards when thrown...

Not even close to what I'm saying, because this is also part of argument #1.

From what I can see, you personally have a strong attachment to the part recovery mechanic in the game which has carried over across versions?

Perhaps. Just like some people are strongly attached to rovers, others are strongly attached to building and let Mechjeb fly for them, etc. If the recovery system is included as a game feature, why is it wrong of me to point out that another aspect of the game diminishes its utility?

Okay, so we've mostly heard from people who insist that the attributes of the parts should not be changed no matter what contortions realistic physics forces them to perform. Is there ANYONE who would like to discuss what I'm actually suggesting, which is that an adjustment of the attributes of the parts would eliminate this undesirable behavior of the parts within realistic physics?

[regex]

2) This is undesirable behavior for that assembly and the attributes of the parts could be adjusted so that this does not happen in realistic physics. THAT is my suggestion and request.

It's an unreasonable request. The center of mass and the center of drag of the craft determines how it will reenter (at least, that has been my experience since 1.0.x came out) and that is the way it should be. Now that you know, you can design your craft around that. If you need help with how to design craft using the new system you can always ask in the appropriate place, but I generally find refusing to adapt, learn, or use the appropriate cheat options is entirely unreasonable, especially when you're asking the devs to penalize players who have already done those things.

[tater]

We disagree that there is anything undesirable happening at all. One, the airless world lander you posted reenters just fine, I even posted a picture for you. Two, you are right that something undesirable is happening, that lander should burn. Good gameplay is having to make design choices with consequences. You apparently think pretty much any craft should reenter and survive. You keep claiming you are making an argument you don't seem to be making.

you said recovering less than you used to is undesirable. Why? Is it also undesirable if you need to add weight and cost via a fairing at launch? How are the two at all different?

Edited June 16, 2015 by tater

[KSK]

Okay, so we've mostly heard from people who insist that the attributes of the parts should not be changed no matter what contortions realistic physics forces them to perform. Is there ANYONE who would like to discuss what I'm actually suggesting, which is that an adjustment of the attributes of the parts would eliminate this undesirable behavior of the parts within realistic physics?

Like reducing the capsule weight or tweaking drag on some other parts? No modification of the physics models but something to make the parts a bit more forgiving for certain uses?

Edit: I'm fine with this in principle - there's no particular reason why, for example, the Mk1 capsule has the mass it currently does.

The problem I see with this is that you could end up going round in ever decreasing circles. Tweak the parts so one set of designs becomes more feasible and you might break another set of designs. At some point, I think one has to just suck it up and design around whatever limitations the part attributes impose.

Edited June 16, 2015 by KSK

[Vanamonde]

The center of mass and the center of drag of the craft determines how it will reenter...

And changing the mass of the parts would move the center of mass. What is unclear about that? You are telling me I am refusing to change, while taking the stance that the current attributes of the parts can't be altered. What is perfect and holy and inviolable about those attributes? They are arbitrary, and can and should be changed when gameplay reasons call for it.

You apparently think pretty much any craft should reenter and survive.

As I said before, wild exaggerations of my position do not help the discussion.

We disagree that there is anything undesirable happening at all.

Okay. You're happy with having to discard additional parts before re-entering, sticking fins on the top or spamming it with SAS to avoid a problem which does not necessarily have to arise in the first place, go on insisting that the current attributes of the parts be set in stone and never change despite the fact that they are entirely arbitrary, and congratulate yourself for not caring about a problem which vexes someone else. Consider yourself the victor of an argument I was not having, and please, move on.

Capsules could be made a bit lighter and engines a bit heavier, and then landers would no longer descend with their engines aimed at the sky while their parachutes yank them violently right-side-up again, and no violations of physics would be required. Is there anyone who does not consider this heresy?

[regex]

And changing the mass of the parts would move the center of mass. What is unclear about that? You are telling me I am refusing to change, while taking the stance that the current attributes of the parts can't be altered. What is perfect and holy and inviolable about those attributes? They are arbitrary, and can and should be changed when gameplay reasons call for it.

I see no compelling gameplay reason for the part masses to change to adapt to your style of play. Part of KSP is figuring out how to do things, rocket science as it were, and this is an interesting physics puzzle for you to solve. Recovering parts for cash should never be an easy proposition, it should take a decent engineering challenge that at least reflects in some small way the challenge of doing it in real life. The game is much more satisfying when you figure out how to do something rather than have it handed to you on a platter, no?

[tater]

Then you'd be on here complaining that your rockets are flipping because the top is too light.

you have the smallest engine on your craft, and you expect it to be a lot heavier? Then all should be heavier, too, then the rocket that sent your airless world lander to the target would likely get bigger. If the capsules change mass (lighter, apparently, then virtually every part needs a new mass. This is all on top of the fact that your craft doesn't need to be spammed with sas, I showed a pic of it after reentry, it never budged from retrograde.

The game doesn't require bringing back any science parts, take the data, store in capsule.

[Laie]

Why would anyone expect a lander like that to reenter in one piece?

Because it has to. When returning to Kerbin, one can claim that the lander should be abandoned and a proper return vehicle be used. But what if you want to land on Eve? The way KSP Lego works, we have little choice but to expose a lot of things to the air.

However, by the same token I consider it perfectly acceptable that parachutes can deploy below the payload and flip it around without causing damage.

I don't think it would be a good idea to make (presumably empty) tanks so heavy, and pods so lightweight, that this lander would enter tail-first.

Edited June 16, 2015 by Laie

[Vanamonde]

I see no compelling gameplay reason for the part masses to change to adapt to your style of play.

I see in another thread that you're unhappy about "wobbily rockets." This is not a problem that has ever bothered me. So, I see no compelling gameplay reason for the part rigidities to change to adapt to your style of play. Why is my point invalid and yours valid?

The game is much more satisfying when you figure out how to do something rather than have it handed to you on a platter, no?

Such as figuring out how to avoid rocket wobble?

tater, I've already conceded that you've won the argument you're having. Since it's not the discussion I'm having, I'm going to ignore you from now on.

But what if you want to land on Eve?

Hey, someone who actually sees the problem I'm pointing out. Even if you don't agree with me about the solution, it's refreshing.

[basic.syntax]

Parachute deploy time can be adjusted with one of Claw's fixes, to reduce the G-shock.

Changing part masses is a complex balancing job, with ripple effects. How much heavier would the engines have to be, to allow that particular design to fall engines-down? Laie brings up Eve, indirectly a concern for different atmospheric densities. Would one mass number work for the original 3-symmetry design, but not be enough to prevent reentry flipping, in a 6-symmetry design? Would the decreased TWR be acceptable, in liftoff scenarios? Would thrust be increased to maintain TWR? Would that cause different unintended balance problems? (I don't have answers.)

[regex]

I see in another thread that you're unhappy about "wobbily rockets." This is not a problem that has ever bothered me. So, I see no compelling gameplay reason for the part rigidities to change to adapt to your style of play. Why is my point invalid and yours valid?

Neither point is valid or invalid, I'm stating my preference here while also stating that your opinion isn't what's right for the game. It's a dissenting opinion and it happens a lot around here. I see no compelling gameplay reason why your suggestion should be adopted by the devs, the parts work pretty well as-is and engineering a workable "reuseable" craft that can reenter properly isn't a huge challenge once you understand how the new system works.

Such as figuring out how to avoid rocket wobble?

I figured out long ago how to avoid wobbly rockets and I don't have any real problems with them now, but it would be nice to have procedural parts with better joint strength. Do I expect that to happen? No, but that shouldn't stop me from expressing my opinion, nor should it stop other people from dissenting and saying so in a public forum.

[DundraL]

Anyway... back on topic....

As usual, I suck with words. Here's what I meant:

I like this way because it can handle 30 degree slopes without tipping, and its easy to get in and out on kerbin once landed.

It is also like to point the right direction during ascent AND descent. Its stable but not so stable you cant turn it to steer where you need it.

Another idea that might work is for some pods to have their CoM adjusted towards their floors. That would help it during descent but hurt it during ascent, though you would have alot of room for stabilizing during ascent. It wouldnt do much overall but it would help with tiny capsule reentry.

Edited June 16, 2015 by DundraL

[ferram4]

Vanamonde, is it wrong to state that your entire argument boils down to, "not being able to reenter some previously workable designs is bad gameplay?" If that's true, then justify why it is bad gameplay. There are many situations in this game where designs that are excellent for specific situations will fail catastrophically in others (as examples, Munar lander on Tylo, Laythe airbreather on Eve, highly efficient Kerbin ejection stage on Moho insertion, spaceplanes pretty much anywhere without an atmosphere); why is this, in your opinion, different enough to qualify for "bad gameplay" rather than "incorrect solution attempt" and deserving of changes to the parts while the others do not? What about the unintended consequences of changing the balance of all of these parts as well?

[Laie]

What about the unintended consequences of changing the balance of all of these parts as well?

That's what I assumed to be self-evident in my previous post: making the tanks so heavy that this lander would enter tail-first can't possibly end well. You have to lift these tanks (and many more) into orbit first. Making the tanks any heavier has ramifications throughout the entire game. Ditto with the engine.

The command pod could, arguably, weigh less. But I don't think there's enough room for weight reduction to make this thing come down tail-first without, again, having to recast all engine efficiencies and fuel tank fractions.

Shifting the capsule's CoM toward it's bottom would actually be a good idea. Not only because that's where it should realistically be: it would also have prevented the pod-flipping in 1.0.0, it could possibly help with your design, and it would make little-to-no difference while the pod is sitting on top of a tall rocket.

Edited June 16, 2015 by Laie

[tater]

Don't forget that his craft reenters just fine, and doesn't actually flip. Build one and try it.