Unguided missile sharpshooter challenge

[Pds314]

The goal of this challenge is to launch an uncontrolled vehicle from the KSC at a target vehicle on the dessert airfield, with something you can suitably call a warhead onboard. Take a screenshot of how close (meters or kilometers) you get to the target.

You may not use any sort of direct control. If you desire, you may use KAL-1000 to execute instructions at pre-set times like staging or maneuvers.

You may only launch one vehicle, but you may use multiple "warheads" if you so desire. If you do, then the closest one counts.

A note: land vehicles are not missiles. It's ok to go for cruise missiles or ballistic missiles, but please don't make an automated rover warhead delivery system.

 

Clarifying what is uncontrolled: you may use KAL or SAS but not, e.g. directly touching the keyboard in flight. No using external mods like KSPRC or KOS for control, since these might as well be manual, closed loop, guided control.

Edited May 13, 2020 by Pds314

[sevenperforce]

I'll give this a shot....

[QF9E]

I got to within 4 kilometers north of the target. This is the limit of accuracy for my unguided missile: If I turn the missile by the minimum amount in the VAB it will miss the target 5.4 km to the south.

The only commands I am giving it are to launch and to separate the warhead once above the atmosphere. I don't give any other inputs and leave SAS off.

My missile on the launch pad. Its warhead consists of 5 tons of napalm rocket fuel inside a protective fairing

My warhead about to impact close to the Dessert Airfield

Edited May 13, 2020 by QF9E

[sevenperforce]

By "unguided" I took @Pds314 to mean that this needs to have no control input from me beyond staging. It does use SAS but it does not deviate from its predetermined settings prior to launch.

SRB has extremely specific thrust limiter and fuel quantity to fine-tune ascent. Even things like the rotation of the antennae and the translation of the cubic struts were tweaked by trial and error to get the desired trajectory. The warhead is a set of four radial ore tanks, filled to the brim with plutonium in a classic "pretending-to-be-a-nuke" fashion, with a probe core, battery, and reaction wheel inside underneath a heat shield.

Warhead descending, moments before impact:

Spoiler

Sole control input is pressing spacebar to fire the Thoroughbred and release the launch clamps.

Everything after that is just a spectator sport.

Moment of ignition:

Using the Thoroughbred with its TVC tied to fixed SAS helps prevent random drift at takeoff.

Ascent looking like a Sprint ABM:

Of course the Sprint ABM got off the ground much, much faster than this. I'm at Mach 3 in 36 seconds; it hit Mach 10 in 5.

Burnout with a rather high apoapsis:

And then begins the long descent back toward fire and fury.

That's a good-looking trajectory if I do say so myself:

Atmospheric heating starts around 50 km. It enters tail-first on purpose (also because with no input I couldn't change it to "hold prograde"):

Prepare for explosive decoupling of the booster:

As designed, the destruction of the booster immediately flips the detached payload compartment around prograde:

The payload fairing promptly overheats, releasing the warhead:

And this is the impact point:

So, I guess my score is...zero meters?

 

[QF9E]

I put a spacecraft on the Dessert airfield launch pad to get a precise measurement. How did you get the launch azimuth as accurate as that?

[sevenperforce]

17 minutes ago, QF9E said:

I put a spacecraft on the Dessert airfield launch pad to get a precise measurement. How did you get the launch azimuth as accurate as that?

I was targeting the runway specifically so putting something on the pad wouldn't have helped.

Getting the right angle wasn't as hard as I thought it would be. I was using an extremely lofted trajectory and I turned off angle snap to make fine adjustments.

Having SAS on, even just to steady it, was key. That way the SRB thrust vectoring maintained my precise angle until I had a lot of speed built up. Otherwise I would have had more uncontrolled pitchover at ignition, amplifying fine attitude presets.

[Pds314]

4 hours ago, sevenperforce said:

By "unguided" I took @Pds314 to mean that this needs to have no control input from me beyond staging. It does use SAS but it does not deviate from its predetermined settings prior to launch.

SRB has extremely specific thrust limiter and fuel quantity to fine-tune ascent. Even things like the rotation of the antennae and the translation of the cubic struts were tweaked by trial and error to get the desired trajectory. The warhead is a set of four radial ore tanks, filled to the brim with plutonium in a classic "pretending-to-be-a-nuke" fashion, with a probe core, battery, and reaction wheel inside underneath a heat shield.

Warhead descending, moments before impact:

  Hide contents

Sole control input is pressing spacebar to fire the Thoroughbred and release the launch clamps.

Everything after that is just a spectator sport.

Moment of ignition:

Using the Thoroughbred with its TVC tied to fixed SAS helps prevent random drift at takeoff.

Ascent looking like a Sprint ABM:

Of course the Sprint ABM got off the ground much, much faster than this. I'm at Mach 3 in 36 seconds; it hit Mach 10 in 5.

Burnout with a rather high apoapsis:

And then begins the long descent back toward fire and fury.

That's a good-looking trajectory if I do say so myself:

Atmospheric heating starts around 50 km. It enters tail-first on purpose (also because with no input I couldn't change it to "hold prograde"):

Prepare for explosive decoupling of the booster:

As designed, the destruction of the booster immediately flips the detached payload compartment around prograde:

The payload fairing promptly overheats, releasing the warhead:

And this is the impact point:

So, I guess my score is...zero meters?

 

That's really impressive. Especialy since you didn't use KAL. It would be nice to see a craft placed on the runway so I can score the entry for distance though.

[sevenperforce]

3 minutes ago, Pds314 said:

That's really impressive. Especialy since you didn't use KAL.

Thanks!

3 minutes ago, Pds314 said:

It would be nice to see a craft placed on the runway so I can score the entry for distance though.

If I put a craft on the runway I would have hit it.

[Pds314]

21 minutes ago, sevenperforce said:

Thanks!

If I put a craft on the runway I would have hit it.

So hit a craft on the runway then?

[QF9E]

8 hours ago, sevenperforce said:

If I put a craft on the runway I would have hit it.

Hitting an object several kilometers long is not the same as hitting an object a few meters in diameter, as you should be well aware. I suppose you could spawn a rover on the runway, drive the rover to the known impact point of a previous attempt and then launch a missile at it, but that would go against the spirit of the challenge. In my opinion this challenge should be to spawn a capsule on the runway or the launchpad and try to hit that.

[sevenperforce]

12 hours ago, Pds314 said:

So hit a craft on the runway then?

Maybe.

4 hours ago, QF9E said:

Hitting an object several kilometers long is not the same as hitting an object a few meters in diameter, as you should be well aware. I suppose you could spawn a rover on the runway, drive the rover to the known impact point of a previous attempt and then launch a missile at it, but that would go against the spirit of the challenge. In my opinion this challenge should be to spawn a capsule on the runway or the launchpad and try to hit that.

Aiming at a capsule is very different from aiming at the runway. I designed my rocket to have a runway-crossing trajectory and then built in small range adjustments so as to hit the runway. If I was going to aim at a capsule I would need a start from a clean-sheet design that had capability for finer lateral adjustment to the final impact point.

Accuracy at present is an ellipse around 30 meters wide and 75 meters in length, due to error induced during explosive decoupling. Actually hitting the runway took a half-dozen attempts.

What I could do is place a rover in the center of the runway and then see how close I get. But with the involved error it's just a numbers game.