Problem with manned re-usable launch vehicle

[tent405]

This is my first time posting here for a problem I've been having with KSP, so excuse me for lack of conciseness.

I have developed a craft which uses twin ram-jets launched from the VAB, to deposit a satellite into orbit. To save money. To save DV on polar/reverse orbits.Because rockets are wasteful. , etc.

The problem I'm having is that, despite the existence of somewhat unreliable autopilot mechanisms which I don't fully know how to use it is still very dangerous to launch this craft to APO, burn to orbit, and return to fly the launch vehicle back down to solid land.

Manned craft on a re-entry trajectory randomly get killed. This happens when I'm doing an initial orbital insertion burn flying a probe. After the launch vehicle detaches. Usually within two minutes. The re-entering manned vehicle is still well above 50k, but it's either not there in map view, not there in center, dead or bugged stuck still.

So I have three questions, all of which can be answered by someone more experienced with these types of problems:

• How can I reliably instruct a probe to execute a maneuver and return to control my launch vehicle back down to the groundd
  
• In what ways is it possible to control two vehicles at once autonomously per program (need to wait for unity upgrade/engine problems?)
  
• How can I tell the engine not to kill craft on a re-entry trajectory if they contain Kerbals?
  

Here's a craft file link

ay

And an image of the problem before I burn the probe for insert and lose the craft

Edited July 31, 2015 by tent405

[Logris]

You can only control one vessel at a time, all other vessels outside the physics range of the active vessel (25km I think) are put "on rails".

This means that their trajectory is fixed on a pre-calculated orbit.

There are no physics calculations done on them, therefore they can't execute a burn at some future time.

If a vessel is outside the physics range, and inside the atmosphere, it is assumed to crash or burn up and is destroyed.

This will be what you are experiencing, while you're controlling the satellite the launcher gets >25km away and enters the atmosphere (70km).

Maybe what you can do is raise the Ap of the satellite to say 150km to give you time to fly the launch vehicle all the way back to land, and then switch back to the satellite in time to circularize it.

[steve_v]

Yeah, you'll need a mod for that

[Lelitu]

Logris has the cause right.

Easiest solution here is to not undock the satellite until the whole thing is in a stable orbit. If the spaceplane is in a stable orbit (above 70km) then it will still be there, no matter how much later you choose to come back to it.

[Sharpy]

You can only control one vessel at a time, all other vessels outside the physics range of the active vessel (25km I think) are put "on rails".

[snip]

This will be what you are experiencing, while you're controlling the satellite the launcher gets >25km away and enters the atmosphere (70km).

About 2.5km. 2.2 is the entering range, around 2.5 is the departing range.

Just skimming the atmosphere doesn't do anything to a vehicle "on rails". You need some percentage of the pressure - on Kerbin that's altitude of about 23km where the craft "on rails" is destroyed. (if you have a craft without power, with too low periapsis, say, in range of 40km, to save it from falling switch away to KSC or other craft before it enters the atmosphere, and it will circle Kerbin as if there was no atmosphere there).

[Warzouz]

Can you tweak your launcher to do the insertion burn on its own THEN release the payload ? That way, the launch stage would be on a stable orbit and you could deorbit it when ever you want.

I'm using this technique with my SSTO launcher series

The other advantage of this solution is the payback value : you can tweak your landing zone to get 96%-98% of the dry cost, instead of 20%-50% if you keep on a suborbital trajectory (you'll end far from KSC). My average return value is 97,5% of dry cost (less than 100km of KSC). KSC landing is 98%. Launchpad is 100%.

Rocket SSTO launchers are very cheap and easy to design. They are scalable.

To get the launch cost to LKO

- detach payload : that gives you "stage wet price"

- remove fuel from all tanks : that gives you "stage dry price"

- LKO price = Stage wet price - 97% * stage dry price" (not including payload cost)

[tent405]

The launcher is just a very light, very fast jet that doesn't work in space (still waiting on R.A.P.I.E.R.'s ).

I will either adjust my mission parameters to compensate or use the FMRS plugin

Thank you for the help answering my question about the physics engine.

[Warzouz]

The launcher is just a very light, very fast jet that doesn't work in space (still waiting on R.A.P.I.E.R.'s ).

I will either adjust my mission parameters to compensate or use the FMRS plugin

Thank you for the help answering my question about the physics engine.

Try to add rocket engines on your assembly. Don't use any decoupler (except for the payload when in orbit). You should get a more expensive launcher, but at 97% (of dry cost) free. It's better than a cheaper launcher you get 20% or 0% back.

But then, as you use if for launching satellite, the return value might not be worth the effort. I usually don't use SSTO rocket for payloads under 30tons.

Few minutes ago, I just sent to LKO a 100T payload ready to depart for Dres (space station, lander, return vehicle and refueller)