Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[BFEL]

Pics:

Spoiler (click to show/hide)

[Aseaheru]

Oh gods... Guys, listen to this :THIS IS AWESOME.
Also, BFEL, what mod?

[Sean Mirrsen]

Pics:

Spoiler (click to show/hide)

Okay, you... have a control problem, to put it mildly.

First of all, add large control surfaces to the main wings, and swap out the tailfins for Standard Canards.

Then, using the right-click menu: disable ALL control for the advanced canards you have in the middle pointing forward, as they will serve no practical purpose in that orientation. For the standard canards up front, disable everything but pitch. For the large control surfaces you added, do the same - pitch only. The standard canards that will replace your tailfins should be set to Yaw only.

This configuration should ensure that your plane is somewhat aerodynamically stable. If you have problems getting enough roll authority, rearrange the large and small control surfaces so that the small roll-only surfaces are on the wingtips, and the large pitch-only surfaces are closer to center.

I don't know if the thermal thruster has vectoring, but from what I see your problem is a horrible aerodynamic disbalance. The moment you try to oversteer this thing, it will go into a tumble. Less of a problem in thicker atmosphere - it probably has enough lift to make it relatively unnoticeable during takeoff, and you probably don't try to do aerobatics with it, but when you're in the final stage of leaving atmosphere, you will generally try to point your plane's nose away from your movement vector, upwards, and when you do that you will immediately lose all manner of control over your pitch.

Plus, by the way, that thing you put behind the inline docking port is an air intake. Make sure to close it when you switch to rocket mode.

[jocan2003]

Oh gods... Guys, listen to this :THIS IS AWESOME.
Also, BFEL, what mod?

Am i the only one who feel a tad of BattleStar Galactica vibe in that music?

[Aseaheru]

Dunno, but I did hear some Firefly vibes.

[BFEL]

Pics:

Spoiler (click to show/hide)

Okay, you... have a control problem, to put it mildly.

First of all, add large control surfaces to the main wings, and swap out the tailfins for Standard Canards.

Then, using the right-click menu: disable ALL control for the advanced canards you have in the middle pointing forward, as they will serve no practical purpose in that orientation. For the standard canards up front, disable everything but pitch. For the large control surfaces you added, do the same - pitch only. The standard canards that will replace your tailfins should be set to Yaw only.

This configuration should ensure that your plane is somewhat aerodynamically stable. If you have problems getting enough roll authority, rearrange the large and small control surfaces so that the small roll-only surfaces are on the wingtips, and the large pitch-only surfaces are closer to center.

I don't know if the thermal thruster has vectoring, but from what I see your problem is a horrible aerodynamic disbalance. The moment you try to oversteer this thing, it will go into a tumble. Less of a problem in thicker atmosphere - it probably has enough lift to make it relatively unnoticeable during takeoff, and you probably don't try to do aerobatics with it, but when you're in the final stage of leaving atmosphere, you will generally try to point your plane's nose away from your movement vector, upwards, and when you do that you will immediately lose all manner of control over your pitch.

Plus, by the way, that thing you put behind the inline docking port is an air intake. Make sure to close it when you switch to rocket mode.

Will try these first chance I get, but what exactly does this accomplish? I mean how do I master aerodynamic balancing on my own? Right now the CoL is right inside the CoM ball as is, and I don't know what my Col/CoM/CoT SHOULD look like. ((I forgot to put up a pic with those in there))

I knew enough to disable the advanced canards beforehand at least , but the "air intake" behind the docking port was changed to a precooler in Interstellar. Which means it does still have some air intake, but it also has atmosphere intake and more importantly is basically required for higher altitude flight ((it messes with air intake to not have one, will have to do science with it closed))

[Sean Mirrsen]

Will try these first chance I get, but what exactly does this accomplish? I mean how do I master aerodynamic balancing on my own? Right now the CoL is right inside the CoM ball as is, and I don't know what my Col/CoM/CoT SHOULD look like. ((I forgot to put up a pic with those in there))

I knew enough to disable the advanced canards beforehand at least , but the "air intake" behind the docking port was changed to a precooler in Interstellar. Which means it does still have some air intake, but it also has atmosphere intake and more importantly is basically required for higher altitude flight ((it messes with air intake to not have one, will have to do science with it closed))

Basically, your center of lift should be behind your center of mass for stable flight. Center of lift close to center of mass is an "unstable" craft - it will fly, but it will do so with all the grace of a sheet of plywood in a hurricane - it will spin easily, and you will need concentrated effort to keep it flying where you need it to. Center of lift ahead of center of mass is what's normally called "unflyable" - it will always try to make the "ahead" become the "behind".

And it's always a good idea to keep at least two thirds of your control surfaces towards the back. As the plane empties its fuel, the center of mass will tend to shift backwards, and any control surfaces in the front will tend to start oversteering and flipping you out without some control authority towards the back to balance them out.

[BFEL]

Ok, so taking your advice helped, but !!SCIENCE!! revealed some new problems.

Specifically two facts emerged: 1. I didn't need that damned precooler at all ((it apparently only counts for normal intakeair using engines, not "atmosphere" burning thermal ones))
and
2. The thermal turbojet, while putting out a nice stable 120 newtons of thrust in atmospheric mode, puts out a dismal EIGHT newtons of thrust in closed-circuit liquid mode. And while it has 1.00 thrust vectoring, 8 thrust is too little to even put the now perfectly stable nose pointing upward, let alone reach orbit with it.

So apparently this design, while a great supersonic jet, is not going to go to space today, or probably ever.

Maybe if I retrofit it with a fusion reactor it will give some better thrust ((thermal stuff is dependent on the reactor you pair it with)) but that also means I will need some way to throw in a few Megawatts to jump start said reactor.

Well, at least I learned things today. Thanks Sean!

[BigD145]

Just pay attention to the thermal ratings of each reactor. You pay attention to megajoulles for transmitted power.

[QuakeIV]

Oh god I sound nasally as fuck why did I mumble while recording this.

Anyways I made a missile, heres a video.

Heres the guidance code, which is horrible and ugly since I have yet to get around to optimizing/beutifying it.

Code: (horrible spaghetti code) [Select]

stage.
lock steering to up + R(0,0,180).

print "Ascending.".

wait until ship:solidfuel < 1316.
stage.

wait until ship:solidfuel < 466.
stage.

print "Clearing spent booster due to collision hazard.".
print "Waiting for optimal boost altitude.".

wait until ship:altitude > 60000.

lock steering to target:velocity:orbit - ship:velocity:orbit.
wait 3.
stage.

print "Killing relative velocity.".

until target:velocity:orbit:mag - ship:velocity:orbit:mag < 50
{
lock steering to target:velocity:orbit - ship:velocity:orbit.
wait 1.
}

print "Accelerating towards target.".

lock steering to target:position.

wait until ship:solidfuel < 1.

stage.


set relvel to ship:velocity:orbit - target:velocity:orbit.

//set ship relative vectors
lock shipx to ship:facing * V(1,0,0).
lock shipy to ship:facing * V(0,1,0).
lock shipz to ship:facing * V(0,0,1).

lock relvel to ship:velocity:orbit - target:velocity:orbit.


set xvel to vdot(relvel,shipx).
set yvel to vdot(relvel,shipy).


//if it might not be able to kill the lateral velocity in time
//with just two rcs ports
lock latvel to vxcl(target:position,relvel).

if latvel:mag > 100
{
  print "Making rapid course corrections.".
  lock steering to vxcl(target:position,relvel).

  wait 3.
  set SAS to true.
  set RCS to true.

  set ship:control:fore to -1.
  until latvel:mag < 50
  {
    lock latvel to vxcl(target:position,relvel).

    set forward to ship:facing.
    if vang(forward:vector, latvel) > 5 //in degrees
    {
      //if shit is fucked then fire up the good ol autopilot
    set SAS to false.
    set RCS to false.
    }
     else
    {
     //if not then back to ye olde SAS and start linear translation again
     set SAS to true.
     set RCS to true.
    }
      wait 0.1.
   }
   set ship:control:fore to 0.
}

set SAS to false.
set RCS to false.

print "Making coarse course corrections.".

lock steering to target:position.

wait 2.

set SAS to true.
set RCS to true.

until target:position:mag < 20000
{
//lock steering to target:position.

set xvel to vdot(relvel,shipx).
set yvel to vdot(relvel,shipy).

  //keep velocity within certain tolerances
  if (xvel > 2)
  {
    set ship:control:starboard to -1.
  }
  else
  {
  if (xvel < -2)
  {
    set ship:control:starboard to 1.
  }
  else
  {
    //but avoid constantly running rcs at max
    set ship:control:starboard to 0.
  }
  }

  //keep velocity within certain tolerances
  if (yvel > 2)
  {
    set ship:control:top to -1.
  }
  else
  {
  if (yvel < -2)
  {
    set ship:control:top to 1.
  }
  else
  {
    //but avoid constantly running rcs at max
    set ship:control:top to 0.
  }
  }
  set forward to ship:facing.
  set forvel to vdot(relvel,forward:vector).

  //if too slow speed up, if too fast slow down
  if forvel < 500
  {
    set ship:control:fore to 1.
  }
  else
  {
  if forvel > 600
  {
    set ship:control:fore to -1.
  }
  else
  {
    set ship:control:fore to 0.
  }
  }

  //periodically adjust heading
  //this will end up being the heading used for fine adjustments
  lock steering to target:position.
  if vang(forward:vector, target:position) > 0.7 //in degrees
  {
    //if shit is fucked then fire up the good ol autopilot
    set SAS to false.
    set RCS to false.
  }
  else
  {
    //if not then back to ye olde SAS and start linear translation again
    set SAS to true.
    set RCS to true.
  }
  //if this convoluted shit isnt done then RCS will harmonically overcorrect
  //until craft spins out of control


  wait 0.1.
}


print "Making fine course corrections.".


//clean up after the coarse adjustments, dont want to back away from tgt or something
set SAS to true.
set RCS to true.
set ship:control:fore to 0.

until sqrt(xvel*xvel + yvel*yvel) < 0.1
{

//rock back and fourth over 0 at very low thrust, should be pretty accurate
//this is generally a lot lighter than coarse to make it run faster

  set latvel to vxcl(target:position,relvel).

//compute velocity components in direction of thrusters
  set xvel to vdot(latvel,shipx).
  set yvel to vdot(latvel,shipy).
  set zvel to vdot(latvel,shipz).

  print "x: " + xvel.
  print "y: " + yvel.
  print "z: " + zvel.
  if (xvel > 0)
  {
    set ship:control:starboard to -0.2.
  }
  else
  {
  if (xvel < 0)
  {
    set ship:control:starboard to 0.2.
  }
  }

  if (yvel > 0)
  {
    set ship:control:top to -0.2.
  }
  else
  {
  if (yvel < 0)
  {
    set ship:control:top to 0.2.
  }
  }

  if (zvel > 0)
  {
    set ship:control:fore to -0.2.
  }
  else
  {
  if (zvel < 0)
  {
    set ship:control:fore to 0.2.
  }
  }
  wait 0.1.
}


//ultra guidance
print "Making ultra-fine course corrections.".


until latvel:mag < 0.05
{

//rock back and fourth over 0 at very low thrust, should be pretty accurate
//this is generally a lot lighter than coarse to make it run faster

  set latvel to vxcl(target:position,relvel).

//compute velocity components in direction of thrusters
  set xvel to vdot(latvel,shipx).
  set yvel to vdot(latvel,shipy).
  set zvel to vdot(latvel,shipz).

  print "x: " + xvel.
  print "y: " + yvel.
  print "z: " + zvel.
  if (xvel > 0)
  {
    set ship:control:starboard to -0.1.
  }
  else
  {
  if (xvel < 0)
  {
    set ship:control:starboard to 0.1.
  }
  }

  if (yvel > 0)
  {
    set ship:control:top to -0.1.
  }
  else
  {
  if (yvel < 0)
  {
    set ship:control:top to 0.1.
  }
  }

  if (zvel > 0)
  {
    set ship:control:fore to -0.1.
  }
  else
  {
  if (zvel < 0)
  {
    set ship:control:fore to 0.1.
  }
  }
  wait 0.1.
}


//Uber guidance
print "Making uber-fine course corrections.".

until target:position:mag < 1
{
  set latvel to vxcl(target:position,relvel).

//compute velocity components in direction of thrusters
  set xvel to vdot(latvel,shipx).
  set yvel to vdot(latvel,shipy).
  set zvel to vdot(latvel,shipz).

  print "x: " + xvel.
  print "y: " + yvel.
  print "z: " + zvel.
  if (xvel > 0)
  {
    set ship:control:starboard to -0.03.
  }
  else
  {
  if (xvel < 0)
  {
    set ship:control:starboard to 0.03.
  }
  }

  if (yvel > 0)
  {
    set ship:control:top to -0.03.
  }
  else
  {
  if (yvel < 0)
  {
    set ship:control:top to 0.03.
  }
  }

    if (zvel > 0)
  {
    set ship:control:fore to -0.03.
  }
  else
  {
  if (zvel < 0)
  {
    set ship:control:fore to 0.03.
  }
  }
  wait 0.01.
}

[Aseaheru]

Kos!

[QuakeIV]

The video is finally up, in all its awkwardly worded glory.

[Aseaheru]

You meanie, you took an idea I had and never coded and coded it.

Although, you want the missile to be going as fast as it can when it hits the target, although that sorta breaks the game's physics.

[QuakeIV]

Yeah, it actually needs slo-mo to reliably collide.  I have had only the rear fuel tank hit the target in real time, otherwise it will kinda skip through it between simulation ticks since its going so fast.  Would work better with larger targets.

[miauw62]

kOS uses until instead of while?
Uuugh :/