Piggyback Design Tips (Like the Shuttle)

[Glaran K'erman]

Hi there, been a bit busy lately and haven't been playing as much as I'd like the past couple weeks but decided to jump back in the other night with trying to build my first shuttle design and have been having a lot of trouble. Tried searching the forums a bit but couldn't find anything new or useful for what I need (if it exists my apologies, just link me there).

So what are the fundamentals/most important factors when designing something like the shuttle where it's piggybacked on it's boosters?

Main issues are CoM shifting before separation during ascent causing havoc, better way of calculating thrust levels for the different engines than just guess and check. I'd include a picture or two but I have tried a bunch of designs with the same problems so I'll leave it more general for now.

Edited August 15, 2015 by Glaran K'erman

[Alshain]

I wouldn't call that a piggyback, the Shuttle on top of it's Boeing 747 carrier is a piggyback. The shuttle's launch assembly is just a vertical launch spaceplane.

Experimentation is going to be a large part of it, even for the experts. Scott Manley did a video on it, even he crashed many times but he does offer quite a few pointers. Note this video was using FAR and pre-1.0.

[Glaran K'erman]

Ah ok, well then I need help with vertically launched spaceplanes!

As for the video that is my problem, all advice is using outdated aero and/or using FAR. I don't use FAR....Thank you however!

[GoSlash27]

It's a fact of life that the CoM is going to shift. The trick is to design it "crooked" to account for that.

The SRBs are aligned to thrust through the CoM when full.

For the engines, you plot 2 points: 1 when the tank is full and 1 when the tank is empty. Draw an imaginary line between these 2 points and align the engines so that they thrust through the center of this line.

You design the orbiter first, then the stack. Shuttles are difficult, but a lot of fun.

Best,

-Slashy

[Glaran K'erman]

Thanks for the CoM tip Slashy, I'll start putting that to use.

[Slam_Jones]

The best "piggybck" I've done was a fun little 2STO (2-stage to orbit) that launches horizontally, and can make a smooth, gliding landing back at KSC. Intended for Space Tourism, til I made a better SSTO.

Doesn't really compare to a shuttle, but it's an option to keep in mind.

I built the jet portion of it first, making sure everything was aligned, then attached it (and some wings) to the SRB. The idea was that the empty SRB would be able to make a safe, gliding landing somewhere to be recovered.

The little ship then gets to orbit, lines up with KSC, then de-orbits to the KSC runway.

The only real issue I've run into, is that when the SRB is about ~2 seconds from dry, it starts pitching up pretty sharply until released.

[Pecan]

...So what are the fundamentals/most important factors when designing something like the shuttle where it's piggybacked on it's boosters?...

The most fundamental factor in 'shuttle' design for KSP is "Just about any other way is easier, and most of them are more efficient too."

[GoSlash27]

The most fundamental factor in 'shuttle' design for KSP is "Just about any other way is easier, and most of them are more efficient too."

This is true. Any other way is going to work out cheaper and easier, especially ssto spaceplanes. But it's the very fact that shuttles are so difficult that makes them rewarding.

Best,

-Slashy

[Pecan]

...it's the very fact that shuttles are so difficult that makes them rewarding...

Surely not; "We choose to do these things ... not because they are easy ... but because they are hard"? *grin*

[GoSlash27]

I'm picturing John F. Kerman giving that speech backwards in Spanish...

Best,

-Slashy

[FlyingPete]

One of the best ideas I've seen for a vertical TSTO spaceplane used a drop tank mounted ahead of the plane. That way, you CoL is behind the CoM at launch and you can actually fly the thing without flipping over. The CoT stays in line throughout so there's no issues with off-axis thrust. It's also slightly more mass-efficient as you use the same engine (mounted on the spaceplane) for the whole launch.

When the tank is empty, you dump it and complete the orbital insertion on the plane's internal fuel supply. The only complication is that you have to avoid crashing into the tank you just jettisoned, but strategic use of sepratrons can deal with that. I guess if you added a probe core and some parachutes to the tank you could recover it as well.

[Edax]

Keep the piggyback ship as close as you can to the hull to reduce drag (so you maybe have to attach the radial mount/docking port on a slight slope. You'll also want the ship placed near the front and have it loaded with fuel to act as ballast so that your center of mass is ahead of your center of lift (but keep at least one tank inactive so that you don't accidentally jettison it without fuel). You'll also want the piggyback ship held down with struts, even if they cause drag, cause you DO NOT WANT the piggyback ship bouncing as it'll make flying horribly difficult and dangerous. It also helps if the piggyback ship can produce it's own thrust to help the mothership SSTO out to compensate for the extra drag.

Edited August 14, 2015 by Edax

[ajburges]

One of the best ideas I've seen for a vertical TSTO spaceplane used a drop tank mounted ahead of the plane. That way, you CoL is behind the CoM at launch and you can actually fly the thing without flipping over. The CoT stays in line throughout so there's no issues with off-axis thrust. It's also slightly more mass-efficient as you use the same engine (mounted on the spaceplane) for the whole launch.

When the tank is empty, you dump it and complete the orbital insertion on the plane's internal fuel supply. The only complication is that you have to avoid crashing into the tank you just jettisoned, but strategic use of sepratrons can deal with that. I guess if you added a probe core and some parachutes to the tank you could recover it as well.

That was clever in .90, in 1.x that could be a nightmare. Putting your launch fuel ahead of CoM makes balancing CoP more difficult.

The best advice I can give for vertical launch space planes is to maintain symmetry. A single fuel tank is hard to balance CoT and CoM. Adding drag to that makes it even harder.

Also, unless you want to model STS or Buran, consider why you are doing VTHL. Faster ascent, increased final stage recovery fraction, or higher payload capacity? Each of these possibilities guides you to different design paths.

[DancesWithSquirrels]

Gah. The thrust imbalances in a US style shuttle are really painful.

My first microshuttle with Mk2 parts was center-thrust - it had four KD25 solids in an X arrangement, with one set of opposing boosters burning faster and emptying sooner than the others, all liquid fuel in the main shuttle. Worked OK... but a bit hard to control even with RCS on the solids - just not enough control authority. Might be interesting to repeat in 1.x aerodynamics, or with vernors and extra probe cores for reaction wheels. From behind:

[FONT=courier new]   B1 | B2
  ----S----  
   B2   B1[/FONT]

What I did with my Mk3 shuttle attempt (in 0.90) was to have two size large drop tanks, one on either side of the orbiter above the wings and even with the fuselage - they extended past the shuttle nose. Solid boosters were clusters under the drop tanks. That way all thrust was through the C0G no matter where you were in the flight. Sorry, no pictures at the moment, but with booster==B, tank==T, shuttle==S, from the side:

[FONT=courier new] T   T
 T S T
 B S B
 B S B
 B S B
[/FONT]

From the base:

[FONT=courier new]       |
       S
  B/T SSS T/B
______SSS______
[/FONT]

That would be challenging with LoX/Hydrogen, though, due to the size of the tanks - they would have to be much wider than thick so they didn't interfere with the wings. Worked OK with the LoX/Kerosene KSP equivalents for me, though.

Eventually I've come to the conclusion that for my style of KSP what works better is reusable VTOL boosters with lots of chutes. I find landing a shuttle to be an interesting exercise, but (for me) kinda boring on a repeated basis. Faster to drop a VTOL in directly.

Edited August 17, 2015 by DancesWithSquirrels
Edit: Wrapped ASCII in 'Code' HTNL to keep spacing