Wobbly rockets - Now 1.09 - 2 metre parts, verniers, payload fairings, MUCH more

[molo]

Yeah, that's what I thought. Which was why I was confused by his statement about keeping his core booster filled for so long.

[jrsteensen]

Hey sunday, awesome packs! Can you post your example .craft files so we can reference the best design practices for your parts? Thanks.

[NovaSilisko]

I'm really having trouble building things with these parts... They all seem to just shake themselves apart very easily.

[nebfer]

I really need some advice on stopping my designs from tipping over off the pad. Seems that if I build something using 2 and/or 3 meter parts the whole thing slowly starts tilting after launch and cannot be brought under control. No matter how much SAS I use, or where I put it, I can't stop these larger rockets from falling over.

Wobbly Rockets and the new version of Kerbal are amazing fun, but it's a bit annoying that my most grand designs aren't viable as they just fall over all the time.

Regards,

Pete.

This is the problem I'm having...

had a rocket with 3x3 meter long tanks and the 7x connector plate with 7 engines, and a second stage of a 2x meter long. and the thing would tilt over after lift off, no mater how many SAS I put on (at one point I had like 30 of thoughs new SAS modules that came with the pack...).

[Pete-Dakota]

This is the problem I'm having...

had a rocket with 3x3 meter long tanks and the 7x connector plate with 7 engines, and a second stage of a 2x meter long. and the thing would tilt over after lift off, no mater how many SAS I put on (at one point I had like 30 of thoughs new SAS modules that came with the pack...).

Yeah same. Seems that rockets over a certain weight/height will always tilt over. I'm trying to recreate the Saturn V; pretty sure I've got the design and staging right, but can't keep it pointing upright, even though it's symmetrical.

[Andras]

I'm really having trouble building things with these parts... They all seem to just shake themselves apart very easily.

I noticed that too. I've increased the mass of the multiple engine adapters to (# engine) slots and so far it seems to be working much better on my test shots.

[DrLucky]

@DrLucky, what order do you need to add the boosters to create the Falcon Heavy scenario? I haven't been able to get it working properly, as the core gets emptied and then I'm carrying dead weight. Pretty weird.

I'm testing mine out some more to see what I'm doing differently.

My configuration is a core of 2x5000 fuel 3m tanks with 5 SPS-10; the boosters are 6 x (aeros with 3x1000 fuel 1m tanks and the big M-50 at the bottom)

I kickstart the thing with 12 solids, and start all the liquids at the same time, but throttled back. As the fuel burns, it gets light enough to keep going after staging.

When I stage down to just the core, it's going ~650m/s and loses a little speed until it gets higher/burns more fuel, then goes to 4.9kps at burnout. The top stage is 2x1000 fuel tanks and a K2-X engine, for another batch of kps.

On further inspection, it's not doing what I thought - it's consuming from the 6 booster tanks and the central tank simultaneously. So, for some reason it's not cross-tanking, but the central 5 engines only burn 10 fuel a second combined, whereas the outer ones burn 22 each from 30% as much tankage. So when the staging occurs I've burned a bit off the top of the central tank, I just mentally registered it as 'full'.

I'll edit my post to correct it.

[Darkfox190]

I'm not sure if this has already been suggested, or is already in the works, but I've noticed a slight gap in the supply of parts - There are all these fancy engines for late stage, but not for early stage. In other words - I'd love to see some two and three meter solid rocket engines, as well as the liquids we currently have!

[DrLucky]

This is the problem I'm having...

had a rocket with 3x3 meter long tanks and the 7x connector plate with 7 engines, and a second stage of a 2x meter long. and the thing would tilt over after lift off, no mater how many SAS I put on (at one point I had like 30 of thoughs new SAS modules that came with the pack...).

Okay, this is still a little bit voodoo for me, but I can usually build a stable rocket. On reading your post, I threw together something similar to what you described, I think:

From the Top:

Top Stage:

Command module

3 x FLS-100 tank (long 1m)

LV-T30 Liquid Engine (stock 1m)

QS-91 Stack Decoupler (Shrouded 1m)

Core Stage:

RCS Module

SAS Module

1x3 Module Adapter Shroud (1m to 3m adapter, non-decoupling)

2x FL-T3000 tank (3m long - the big one)

3M7X1 Adapter Plate (3m to 7x1m)

7x K2-X Liquid Engine (One of the lighter 1m ones)

Boosters (3x, symmetric, each with:)

Aero Decoupler

2 x FLS-100 tank (long 1m)

M-50 Liquid Engine (Large Liquid 1m)

1 strut attaching bottom of lower fuel tank of each booster to central tank

• 
  [li]I positioned the boosters so that their engines ended just short of the central group.[/li]
  [li]I moved the 7 central engines into the same stage as the boosters so they all light right away.[/li]
  

At the pad, I press 'T', hold shift to ramp power to max, then press space. No need to fly it at all, and that's a pretty tall (a fair bit more than the launch gantry) and moderately heavy rocket using only two extra RCS/SAS pieces. (You'll need to throttle down after a bit to avoid cooking the engines, mind you). It's a little slow, but mostly I wanted to point out it can be done.

Here are the key points:

• 
  [li]Don't overpower the 7 engine part. There are problems inherent to KSP right now that prevent you putting large thrusts (ie: 7x M-50 or even LV-T30s) on there. If you do, it'll vibrate until an engine or two breaks off. This is fun, in the Dwarf Fortress 'Losing is fun!' sense.[/li]
  [li]Don't rest a reeeeeally huge rocket on a 5- or 7- engine part; they sometimes seem to crack under the weight, so you launch with fewer engines, leading to instability and fun. Stick some boosters or something down there.[/li]
  [li]Don't put the RCS too close to the center of mass. It's more effective out toward the end. I only use RCS on medium sized or larger rockets, but I've never needed more than one.[/li]
  [li]I sometimes use extra SAS modules, but I'll do something like put them on boosters which I jettison - as the rocket gets lighter, I don't need them or the mass they represent. I have a 17,500 m/s rocket which uses only 7 of them - one just like the sample rocket I gave, and 6 on boosters which I drop around 30 km up.[/li]
  [li]Make your rocket rigid with sparing use of struts. If your upper stage tends to fold around a certain engine or decoupler, turn on 3-way symmetry and tie the upper tank to the one below the engine.[/li]
  [li]Many SAS modules on parts that are wobbly compared to each other often fight or overcorrect repeatedly, effectively paralyzing their effectiveness. Use as few as you can.[/li]
  [li]More often than not, the uncontrollable tipping is due to one engine in you cluster being broken - try turning your engines off, shifting the camera and looking to see if that's the case.[/li]
  [li]Particularly floppy assemblies may twist under the stress of thrust and skew one way or the other. If this is uneven, it will uncontrollably steer your rocket. This happens a lot with SRBs on radial decouplers.[/li]
  [li]If your thrust to mass is super high, the rocket becomes hard to control - this often happens just before booster burnout, when lots of fuel is gone. Try to always have a liquid component to your thrust, and turn it down a bit if you're losing control[/li]
  

I'm off to see how tall a rocket I can still fly.

Let me know if this helps. If it doesn't, assume I'm wrong; it happens a lot.

Edit: Tried flying this one a few times; it gets squirrely toward the end of the boosters/core stage as the centre of mass moves forward toward the RCS. I usually fly a 2xlong fuel tank as my last stage instead of 3x, and this helps a lot. But this thing will still easily put you into orbit. Just dial that thrust down as the boosters empty.

[rdfox]

Do decouplers also sever strut connections between stages? (Seriously, I've not played around with them yet.)

I've actually managed a pretty good Saturn V with Wobbly 1.09, using 5xM50 in the first stage, 5xK-2X (with the 'narrow' 3m-to-5x1m plate) in the second stage, and 1xK-2X in the third stage. It *usually* manages to stay upright, and so long as I throttle the first and second stages down to 70% at about halfway through their burn time (thus simulating the early CECO that both the S-IC and S-II stages used in real life to control G loading), it doesn't suffer structural failure in flight.

Of course, it also is able to put the spacecraft (with Yawmaster, small fuel tank, and Sunday Punch's small LFE that I can never remember the name of) and third stage into orbit on the first two stages alone... I think I need some sort of LM mass simulator to make it more accurate!

[DrLucky]

Yup, couplers sever. I've never had one hang up on me. I have had them torn apart by ill-advised rockets.

Speaking of ill-advised, I decided to try a stupidly tall rocket using many, many of the new Wobbly parts:

Astute observers will note I'm not flying with SAS... too unstable.

This design isn't efficient, or terribly easy to fly, or even a good idea in any sense, but it is tall. And flyable.

It reached orbit with the last stage (lost a lot of ground on the 2m stage - it was wayyy too heavy for the 2 engines, so I had about 1.5 tanks too many and the engines were overheat prone and shaking violently.) but I had plenty of dV left to deorbit and parachute land.

It's a little over 2x the height of the launch gantry, accomplished by building the 3m stage and putting it to one side, and adding it on at the end. It's substantially taller than the VAB.

[Rocket Surgeon]

So, in my ignorance I have to ask for instructions on how to use those multi-panel shrouds. What part do they connect to and how to I have to put it all together?

[DrLucky]

So, in my ignorance I have to ask for instructions on how to use those multi-panel shrouds. What part do they connect to and how to I have to put it all together?

They're a little sketchy, but with some care they can be made to work like this:

They're designed to attach kinda like radial couplers - that is, they attach to the exterior surface of the rocket. Because of the gap, it doesn't look like it, but that's what's happening. They're sized to work on 1m parts, and they match a 2m diameter.

For practice, whip up the following stack:

Parachute

Command Module

Light Decoupler

Small Liquid Fuel Tank

SPS-10 Liquid Engine

Light Decoupler

2m to 1m shroud (yes, the one that narrows, not widens)

2x long 1m fuel tanks

M-50 Engine

Then pick up the side panel, set 4x symmetry, and bring it in close to the 1m section above the 2m to 1m adapter. You should see them snap into place, and with a little fiddling you can make them look good. Add another set of the cylindrical panels, and that the nose caps.

The only area of the part that physically exists afaik is the bottom centre of the piece, so if you need to pick them up to adjust, aim between the two small circles in the texture. Ignore the blue attach point, I think that's an artifact of the way KSP designates parts.

When the part stages, the panel blows clear; it's effectively a decoupler. So add a stage to your rocket at some point when you'll be exoatmospheric, and dramatically blow the shroud at that point 8)

[Rocket Surgeon]

Ah, I see now. I also wasn't aware that if you want to use the rounded fairing, you need the flat one, too. Thanks.

[Baker Easy]

Aero decoupler is great, finally I make my Proton.

But I dont understand fuel flows through them or not?

Heehee, I made a Proton myself but it looks like you beat me to posting it. Looks almost exactly like that too, though I've got a longer variant for satellite launches.

We need some angled sorts of tanks so we can make an R-7. I tried making one with the GG-M4 engines but it wasn't quite right. I know someone on the SA forums was working on a serious one though.

Also, just remembered I can use those vernier rockets to make an Atlas. Time to get on that. ;D

[Jay]

I noticed that the 2m 2x adapter was pretty heavily unbalanced, as compared to most anything else.

And that you didn't seem to include any 2m engines.

Ignore the blue attach point, I think that's an artifact of the way KSP designates parts.

The attach point is for the matching fake-nosecone parts to attach to.

[rdfox]

Heehee, I made a Proton myself but it looks like you beat me to posting it. Looks almost exactly like that too, though I've got a longer variant for satellite launches.

We need some angled sorts of tanks so we can make an R-7. I tried making one with the GG-M4 engines but it wasn't quite right. I know someone on the SA forums was working on a serious one though.

Also, just remembered I can use those vernier rockets to make an Atlas. Time to get on that. ;D

You'd be better off using the liquid-fuel radial boosters for an Atlas, since the verniers don't consume fuel or display any exhaust. And even then, it's not really accurate, since the side-mount booster engines generated the majority of the thrust on the Atlas, with the center sustainer being a relatively low-power engine; the side-mount boosters were actually jettisonable in flight, too.

Per Wiki's article on the Atlas:

'Powerplant: 1 × Rocketdyne LR105 rocket engine with 57,000 lbf (254 kN) thrust, 2 × Rocketdyne LR89 rocket engines with 150,000 lbf (670 kN) thrust, 2 × Rocketdyne LR101 vernier rocket engines with 1,000 lbf (4.4 kN) of thrust'

[Baker Easy]

You'd be better off using the liquid-fuel radial boosters for an Atlas, since the verniers don't consume fuel or display any exhaust. And even then, it's not really accurate, since the side-mount booster engines generated the majority of the thrust on the Atlas, with the center sustainer being a relatively low-power engine; the side-mount boosters were actually jettisonable in flight, too.

Per Wiki's article on the Atlas:

'Powerplant: 1 × Rocketdyne LR105 rocket engine with 57,000 lbf (254 kN) thrust, 2 × Rocketdyne LR89 rocket engines with 150,000 lbf (670 kN) thrust, 2 × Rocketdyne LR101 vernier rocket engines with 1,000 lbf (4.4 kN) of thrust'

True; rechecking reference photos (ie Wikipedia) the vernier engines have roughly the right profile but are severely undersized.

[molo]

Looks like you could use the aero decoupler with some M-50s, with the central stage running something smaller and more efficient as the sustainer.

[Sunday Punch]

I noticed that the 2m 2x adapter was pretty heavily unbalanced, as compared to most anything else.

And that you didn't seem to include any 2m engines.

The attach point is for the matching fake-nosecone parts to attach to.

What do you mean by unbalanced? Do you mean in the game design sense or the 'makes rocket tumble out of control' sense? I plan to make more 2 and 3 metre parts, including engines, but like I said before I just wanted to get this update out with the parts I had completed since it was way overdue already.

[Salda007]

Wheeee! Love the update! Love the whole package, really! (Well, aside from the increased 3m tank masses, but you can't win them all!)

I did notice a couple bugs, though; not sure if they'd be directed at you or at HarvestR, though.

The 3m->7x and 3m->5x couplers seem to want to force 2x symmetry, going to opposite sides of the coupler. This may be an engine bug, but it's really something of a PITA, not being able to go to 6x or 3x symmetry on the former or 4x on the latter. *shrug*

The parachute->1M nosecone (great idea, btw!) and the 1->2x coupler both 'wiggle' in flight. I can't really describe it and I don't have any way of taking a video of it, but they bobble around like a kid's playground toy. I don't think it's affecting performance much, but TBH the rockets I've run them on have had so many other problems, I couldn't say for certain. Looks bizarre, though...

Keep up the ridiculously awesome work! 8)

[Sunday Punch]

Yeah those couplers are set to 2X symmetry, because anything greater doesn't work. It just resets to no symmetry when you try and attach a part if the stackSymmetry variable is anything other than 1 (2X symmetry). You can still attach any arrangements of parts you like by turning off symmetry, turning it on just cuts the time it takes to attach the parts.

The increased fuel tank masses are part of an effort to make all of the tanks more balanced, there's now a direct correlation between fuel tank mass and fuel content. The 3 metre tanks were too light for the amount of fuel they carried before.

And the wiggliness is almost certainly caused by the mass of that part being too low, I'll try and fix that for the next release.

[Causeless]

The Retro Backfire rockets seem a bit useless, because they can't turn them on while the part is disconnected, and if you turn them on and then disconnect a stage, the rocket loses velocity.

[DrLucky]

Yeah those couplers are set to 2X symmetry, because anything greater doesn't work. It just resets to no symmetry when you try and attach a part if the stackSymmetry variable is anything other than 1 (2X symmetry). You can still attach any arrangements of parts you like by turning off symmetry, turning it on just cuts the time it takes to attach the parts.

The increased fuel tank masses are part of an effort to make all of the tanks more balanced, there's now a direct correlation between fuel tank mass and fuel content. The 3 metre tanks were too light for the amount of fuel they carried before.

And the wiggliness is almost certainly caused by the mass of that part being too low, I'll try and fix that for the next release.

I played a little with this, and 3x symmetry works (on the 3m to 1mx7), but 4x on the 5 and 6x on the 7 both fail. I think this one's back in HarvesteR's court. Shame, because the new stage parts grouping makes design a lot easier for grouped parts.

[Rocket Surgeon]

The Retro Backfire rockets seem a bit useless, because they can't turn them on while the part is disconnected, and if you turn them on and then disconnect a stage, the rocket loses velocity.

I played with that, too. It's a neato idea and it does work somewhat if you stack the retro and then the decoupler in the same stage (or it used to when the part was first included during KSP 0.8). You still lose velocity, but it kinda works. If anyone knows how to make it work flawlessly, please spill.