Shuttle problems

[SlabGizor117]

So, my shuttle, when reentering had some problems.  The CoM and CoL were fine, because I put fins clipped into the cabin at the front to move the CoL forward, but after I slowed down and tried to land, my shuttle would literally just float down horizontally at 80-ish m/s, and I couldn't get the nose down, no matter how hard I tried.  If I move the CoL back though, it'll have bad maneuverability.  What should I do?  Would pics help at all? I don't really know what else you would need to know..

[A_name]

Post a pic with the COM and COL indicators in the Spaceplane Hangar. Nosing up uncontrollably definitely sounds like your COL is too far forwards. Are you accounting for the shift in COM by consuming fuel and/or deploying your payload? My suggestion at this point is to move the COL backwards and add more control surfaces to make sure you're still maneuverable.

[ForScience6686]

With a shuttle, I don't think you want it to be highly maneuverable.  I would just focus on enough lift for a good glide

[InterCity]

CoM too far behind/not ahead enough of CoL. 

If you can't pitch down as it is, I doubt that you'll have any manoeuvrability issues even after fixing the problem. On the contrary, I think it still could be quite wild, and as ForScience6686 has said, you don't need that manoeuvrability anyway - it's a shuttle, not a fighter jet.

Happy lithobrakings landings!

[SlabGizor117]

Well, I know for a fact that the CoL and CoM are fine, but even with the CoL back far enough with no payload or fuel(or little fuel) I still have the same problem.  And with maneuverability, it's either very maneuverable(but not too much so) or almost immovable when I move the CoL back further to try to let it nose down.  I already said in the first part that I knew the CoL and CoM were fine.  I know it's something else.

Edited December 22, 2015 by SlabGizor117

[Temstar]

We really need a picture of your shuttle to probably diagnose the problem.

But if I were to guess, your shuttle probably have a lot of engines at the back, something like three Vectors or a Mammoth. And you don't have much weight at the front of your shuttle, maybe you don't even have a mark 3 cockpit in front and this is instead a probe controlled shuttle? And what you are doing is entering the atmosphere at high AoA to reduce velocity higher up in the atmosphere and spread out the heat load over a wider area, but once you hit denser air you're having trouble dropping the nose.

Are my guesses correct?

[SlabGizor117]

Yes, my shuttle is just the same as the real one, 3 Vectors and two Thumper radial engines for OMS.  I guess I'll post pics, but I didn't think that was so important.  The reentry you described is exactly how I did it, by the way.

Album:

 

Edited December 22, 2015 by SlabGizor117

[SlabGizor117]

Ok, so I tried a lower AoA and my shuttle was just completely uncontrollable.  What the heck is going on?  I've done so many flight tests with full fuel(which should've made it more unstable) taking off from the runway with my SSMEs and now all the sudden they just completely flip the shuttle all over the place.  What the crap is the deal?!

EDIT:  After trying it some more, it also seems like it just has very little lift, I had a hard time landing in flight tests at lower speeds even with no fuel left.

Edited December 22, 2015 by SlabGizor117

[swjr-swis]

On 12/20/2015 at 9:10 PM, SlabGizor117 said:

after I slowed down and tried to land, my shuttle would literally just float down horizontally at 80-ish m/s, and I couldn't get the nose down, no matter how hard I tried.

Do you cut throttle completely on approach? With the amount of lift of those wings, and the lifting body effect, the KSP shuttle needs very little speed to keep gliding. On top of that the angle of the engines makes any bit of thrust work as a lever that will want to raise the nose - more reason to cut throttle. You pretty much need to stall it over the runway to make it set down.

If you haven't done so (not obvious from the pictures): consider adding slanted mirrored elevons to the tail fin, then add those and the wing elevons to the Break action group to further slow you down as you glide over the runway. Could also add a chute at the rear.

[SlabGizor117]

I always cut the throttle, but as for the engines I've taken off from the runway for flight tests with them before, both SSMEs and OMS, so I don't know why they're lifting the nose so much now.  The problem seems to be that it's just constantly stalling, whether floating horizontally or, like my more recent update(which it seems you haven't read) just flipping out of control.  It doesn't seem at all to be a very good glider, and I don't understand why.  I did make a tail fin brake but never thought to use it in pointing the nose down, I'll try that next time, though I doubt it'll have much effect.  And I also do have two full size radial chutes, but they're usually used to land on my nose when I can't actually glide down.  I don't understand what the deal is here, but it's really frustrating me..

[swjr-swis]

At the risk of being Mr Obvious: did you check whether all elevons are deployed in the correct orientation? Do they all react correctly to control input?

Not sure what else to make out from the pics. You do mention you clipped in some more parts, but we can't see what effect that has just from the pics.

You could post the craft file and let others test it, might give more result. Or at least we could confirm the problem.

[SlabGizor117]

Well I know none of the control surfaces have any problems.  As I'm testing it now, I remember someone clipped in upside down vertical stabilizers under the engines, I'm gonna try that.  The only other thing I have clipped in with the current version I'm testing is two tail fins in the cabin to move the CoL forward a bit, because I took all the overkill fins and wings I clipped in there out before, and then it had no maneuverability.  Is there any place I could put the craft for download that I don't have to make an account for?  I don't have any mediafire or dropbox accounts and I can't make one because my parents are kinda strict about email accounts because then they have to get notifications for all the emails I get from them.

[swjr-swis]

7 minutes ago, SlabGizor117 said:

Is there any place I could put the craft for download that I don't have to make an account for?  I don't have any mediafire or dropbox accounts and I can't make one because my parents are kinda strict about email accounts because then they have to get notifications for all the emails I get from them.

I use kerbalx.com myself, but that too needs an account.

[SnakyLeVrai]

1 hour ago, SlabGizor117 said:

I always cut the throttle, but as for the engines I've taken off from the runway for flight tests with them before, both SSMEs and OMS, so I don't know why they're lifting the nose so much now.

Taking off from the runway, with your SSME configuration will always flip the nose up. I suggest you install RCS Build Aid and look at the "engine" tab. The yellow arrows will show you the thrust vector of each engine and, in the RCS Build Aid panel, you'll see there is a HUUUUUGE torque applied to your craft (it will be shown by a big, red, circle-shaped arrow, showing the direction of the torque.

My shuttle can use the SSME for taking off the runway, only if I disable the top engine (I don't use the OMS either). For this, I have made sure the 2 bottom engines are centered and lined up with the CoM and DCoM (Dry CoM = CoM when fuel tanks are empty). This is why I don't have OMS: the 2 bottom SSME engines are my OMS. Not realistic but effective.

Now to reentry problems:

Well, space shuttles fly as well as bricks. So it's only normal that you'll stall pretty quickly, because they don't produce enough lift for their weight.

Your CoM and CoL look quite acceptable but a little too far back. I only replaced your Hitchiker module by a rockomax tank (the 720 units of fuel version), so I could balance the CoM to make the shuttle stable at an angle of 40° to 50° above prograde vector during reentry. Then I pour every bit of fuel, monoprop and stuff at the front to make sure it won't flip.

Although, your OMS shows some intriguing caracteristics. Its angle may generate a downwards drag that could tend to pull the aft down. Also, it could weight a lot, pulling the CoM back too much. By choosing lighter engines, empty tanks and removing the adapter, you'll have a much lighter OMS. You may then be able to push your wings front a little so the CoL is not too far back, re-balancing the whole thing. That may help solving some problems there.

Look how far front are my shuttle's wings. CoM is farther to the front too, due to a heavy Cupola module at the front of the hold, and a tank. -> http://forum.kerbalspaceprogram.com/index.php?/topic/128220-satellite-retrieval-with-space-shuttle/

[SlabGizor117]

Well first, I do use OMS engines because I want the realism too.  3 Vectors and 2 Thuds(not thumpers, my bad)  for OMS.  I don't care about whether the SSMEs flip the shuttle, because I only use them when the EFT is still there.  What I care about are the OMS engines, although that doesn't concern me too much either because I wouldn't want to use them in atmosphere like reentry.  The only thing I said about them flipping me around was that they did so much more than they do normally.  Also, I don't know what I could take off that wouldn't sacrifice the realism, as far as engines or the adapter.  But I'm kind of getting confused with what you're suggesting, so, with realism in mind, that being that I don't wanna remove anything that would take away from the look, what do you recommend I do?

[SnakyLeVrai]

You can replace the Thuds by Terriers, that'll help shifting the CoM forward a little and you'll be able to pitch the nose down a bit. Remove the FL-A5 adapters (I suppose it's the tiny FL-A5, if not it's the rockomax brand adpater 2, i don't remember). And if the whole OMS structure is made of tanks, empty them to save the weight... maybe you can even replace them with parts that weight less (empty FL-T400 weight more than the Structural Fuselage). See if that helps.

[SlabGizor117]

Ok, first, what rockomax adapter are you talking about?  Where is it?  Also, it would help to know the general goal of removing all that stuff, too.  Is it just to move the CoM forward?  If so, why not just move the CoL back?

Edited December 22, 2015 by SlabGizor117

[SnakyLeVrai]

2 hours ago, SlabGizor117 said:

Ok, first, what rockomax adapter are you talking about?

FL-A5 adapters

2 hours ago, SlabGizor117 said:

Is it just to move the CoM forward?

yes

 

2 hours ago, SlabGizor117 said:

If so, why not just move the CoL back?

Could work, but that's ugly  What will happen to the whole balance when you will have a payload in the hold? Best is to keep CoM as close to the hold's center as possible.

[SlabGizor117]

I meant where on the shuttle.  I know what adapter you're talking about since you already posted the link.  Where is it?  Also, That makes sense to have the CoM closer to the cargo bay, thanks

[SnakyLeVrai]

4 minutes ago, SlabGizor117 said:

I meant where on the shuttle

oh. Look at the last pic on your imgur gallery (the one showing your shuttle from the back) : on the OMS, it's the rearmost part you've connected to the FL T400 tanks, the bell of the thud engines passes through it. You don't need that part except for aesthetics I presume  

[Temstar]

Imagine your average empty rocket stage, if you drop it down from orbit which end tends to fall first? The engine end right because the engines are heavy and empty fuel tanks are basically big metal balloons, so the stage naturally wants to orientate itself heavy end down as it encounters air.

So why doesn't SSTO spaceplanes flip during reentry so their engine end fall first? Three reason:

1. SSTO spaceplanes tend to have less engine mass relative to the rest of the craft, since they use lift in the lower atmosphere they can get away with a lower TWR at launch
2. SSTO spaceplanes tend to have some heavy stuff like cockpit at the front to counterbalance the engines
3. SSTO spaceplanes have wings and control surfaces which give them control authority in atmosphere to keep them pointed right way forward

Now consider your shuttle:

1. You have an extreme amount of engine mass relative to the rest of the craft, as Vectors are large engines in a deceptively small physical package. If we forget about the SRBs and ET for a second and instead replace your cargo hold with fuel tanks and try to launch that then you will have on your hands a winged rocket with excessively high TWR
2. Give point 1 above, your cockpit and hitchhiker are insufficiently heavy to counterbalance the weight of the engines, which is why you see a CoM that is clearly towards the back of the craft
3. Given point 2 above, you need a lot of lift at the back which means big delta wings at the back. This then results in control surfaces mounted in a position that makes them hard to control pitch

Point 1 and 2 are easy so let's talk a bit more about point 3. If you imagine your fuselage as a lever with the CoM as the fulcrum, the control surfaces is then the torque that can be applied to this lever to make it rotate about its fulcrum. With say your average subsonic passenger jet the CoM is somewhere close to the middle of the fuselage. This means the control surfaces on the tail have a very large amount of mechanical leverage so only a small amount of torque needs to be applied to make the plane pitch up or down.

In your case the CoM is towards the back and your control surfaces are also at the back of the plane attached to the delta wings. This puts the torque very close to the fulcrum with a very small leverage, which means huge amount of force must be applied by your control surfaces to pitch up or down. So all else equal this design doesn't have much pitch control authority.

Now when you use large AoA during reentry what you're doing is deliberate put the wings in a stalled state to generate drag. Whilst this is good idea for slowing down and spreading out the heat load it also means you're further reducing your pitch authority since your control surfaces will also be in a stalled state along with the wings. So once it's time to transition to level flight they generate very little torque to pitch the nose down, hence you continue to fall belly first.

The solution to this is canards. Canards won't help you with point 1 or point 2 (well they do actually, but not enough to matter), but they do help with point 3 since they will be at the very front of the craft a very long distance away from the CoM so they enjoy a very large mechanical leverage for pitch control. Having canards at the front also means you are now generating more lift at the front of the plane during level flight, which means you could move your main wings and the big control surfaces further back on the fuselage if you want to maintain the current CoL, which then further increases their leverage and pitch control. If you don't want to ruin the shuttle look clip your canards inside your cockpit.

I had to struggle long and hard with this same "CoM on the ass end" problem recently with an even more extreme example than your shuttle:


So even more engine than your shuttle and even less weight at the front than you as it's unmanned. Finally managed to make it fly nose first in lower atmosphere by:

1. Making the front heavier by using the SLS tank at the front. SLS tank is 9 tons when empty, much heavier than an empty mark 3 tank
2. Attach canards all the way at the very tip of the craft for extreme leverage
3. CoM quite far in front of CoL, making it excessively stable. Without control input this thing flies like a lawn dart - it always automatically try to return to prograde when there's no control input.

Edited December 23, 2015 by Temstar

[SlabGizor117]

Ok, so if I attach canards at the front for more control(which by the way would be clipped as I believe they still work like that and I really wanna keep the realism), that'll move the CoL forward.  So what could I do to move the CoM forward as well, without adding the complication of fuel to have to pump around or something like that?  Just clip full unused fuel tanks in the cabin also?  Again, would canards in the front still do their job as control surfaces?  Or do they only count for lift?

 

Great reply by the way, thanks for your help!

Edited December 23, 2015 by SlabGizor117

[Temstar]

Yes canards in front do work as control surfaces, in fact if you want to use canards behind the CoM you have to reverse its controls to make it work as canards are explicitly designed as control surfaces that go in front of the CoM.

Ideally if you could figure out a way to move the CoM forward just by moving existing parts around then go for it. Don't sweat it though - I suspect canards will be enough and in any case adding deadweight to your plane just to move the CoM around seems like quite an inelegant solution to me.

To keep the CoL well behind the CoM just nudge the main wings back slightly - the canards only contribute a little bit of lift at the front so a tiny shift by the main wings could easy compensate for that.

[SlabGizor117]

Thanks, but I meant, would it work the same if it was clipped into the cabin?

[Temstar]

As I understand it, anything radially attached (and not inside fairing/cargo bay) is treated by the game as in free airstream. It's a pain in the ass for rocket design but in your case clipped canard should work.

Only one way to be sure though