Dropping Stuff from Fast Planes is Tougher than you Think

[Jonfliesgoats]

In KSP, we can drop an object from a pawn flying at Mach 3 with little difficulty.  In real life, when you go fast, separating an object from an airplane is really, really difficult.  Aerodynamic interference, shockwaves and paving centers of pressure frequently cause separating objects to behave unpredictably.

It was only in the last decade that we really got supersonic bomb release down.

If  we consider releasing spaceplanes from supersonic and hypersonic motherships, the engineering challenges become really daunting.  So, with current technology, we have a soft limit on the speed at which we can release a spaceplane.  All foreseeable motherships have to be subsonic, not just because that are big, draggy, subsonic airplanes but because the feasibility of releasing anything massive at high air speeds is questionable right now. 

The link is to an unclassified announcement from 2007 when we could finally drop a Mk.82 at Mach 2.

http://m.slashdot.org/story/93831

[OrbitalBuzzsaw]

Let's not start the "realism" versus "making the game better" thing again.

[monstah]

So, what's the suggestion here?

That Squad ditch Unity for a fluid-simulating supercomputer framework?

[Jonfliesgoats]

Nah.  I am just pointing out that dropping stuff from fast planes is difficult.  That's why it's in the science/spaceflight forum.  The ease of spaceplanes in KSP can let folks think HOTOL and other programs are closer o fruition than they are.  Same with stratolaunch, etc.

[monstah]

Just now, Jonfliesgoats said:

That's why it's in the science/spaceflight forum.

It isn't tho  Want me to move it?

[Jonfliesgoats]

Oh!  Yes please!   My mistake.

[monstah]

Heh. Done.

[Jonfliesgoats]

Cheers

[James Kerman]

The harder it is to drop bombs on people the better IMHO.

[YNM]

Well, we release strap-on booster stages just fine, so... Maybe some pusher SRBs would always work ?

 

4 hours ago, monstah said:

Squad ditch Unity for a fluid-simulating supercomputer framework?

Would be appreciable, but that would crash most computers XD

Edited December 3, 2016 by YNM

[Jonfliesgoats]

Pusher SRBs are a good way to go.  I think we ditch boosters at relatively low q, by being pretty high, though.  I may well be wrong about this.  

Do you or does anyone else have insight on aerodynamic loads during SRB separation on various craft?

 

[kerbiloid]

KSP also has SSTO spaceplanes (still none IRL).
Just one more piece of magic.
Or well lubricated b*mbs.

[Motokid600]

 I dont know ive had some very... destructive experiences trying to air-drop stuff in KSP.

[NSEP]

I think you could do just fine if you had them attached outside your plane and not in a cargo bay. But maybe not? I have not tried it yet.

[p1t1o]

I know that if you look closely at an F-18, any variant, the hardpoints on the wings are angled outwards a few degrees. This does increase drag slightly, but the way it affects airflow increases the "cleanliness" of dropped weapons. Quite noticeable in this pic:

Pictured below is the bomb-bay of a B-2, the vertical, perforated panels are drop-down "fences" that alter airflow for more stable weapons release.

 

It is relevant to know that every combat aircraft has to test every weapon release under every desired release envelope, you cant just attach a new weapon to an aircraft and deploy it, unless you want unexpected and severe accidents to occur in combat.

There's footage around somewhere (couldnt find it just now) of an F-111 conducting a test drop of a Mk-84 iron bomb, just after release (in subsonic, level flight) it flips out completely and in its tumble, sweeps its nose through the tail/engine section of the aircraft, causing catastrophic damage and loss of the aircraft. I am unaware of the fate of the crew.

***

Dont know much about SRB seperation, but I know that it is quite a common (and unwanted) phenomenon that discarded stages can "draft" behind the upper stages (ie: slow to seperate) which can cause significant damage and possibly failure, on ignition of the next stage. The sprint missile suffered from this during development (and you can read about how they solved it) and it is a factor effecting the accuracy of ICBMs (as the drafting stage can affect airflow patterns and thus have an impact on the trajectory of the weapon.)

Edited December 16, 2016 by p1t1o

[wumpus]

On 12/2/2016 at 2:41 PM, monstah said:

On 12/2/2016 at 7:08 PM, YNM said:

Well, we release strap-on booster stages just fine, so... Maybe some pusher SRBs would always work ?

 

Would be appreciable, but that would crash most computers XD

So, what's the suggestion here?

That Squad ditch Unity for a fluid-simulating supercomputer framework?

Modern GPUs might as well be thought of as a supercomputer framework (although they are relatively tied to unit-stride vectors).

The big catch is that like unity, they are [mostly] tied to single point precision.  In one of his videos, Scott Manley mentions that Spacex described their rentry-computing process at a GPU-computing conference (and mentioned KSP in the process), but I suspect they had double precision GPUs available.

[sevenperforce]

7 hours ago, p1t1o said:

Dont know much about SRB seperation, but I know that it is quite a common (and unwanted) phenomenon that discarded stages can "draft" behind the upper stages (ie: slow to seperate) which can cause significant damage and possibly failure, on ignition of the next stage. The sprint missile suffered from this during development (and you can read about how they solved it) and it is a factor effecting the accuracy of ICBMs (as the drafting stage can affect airflow patterns and thus have an impact on the trajectory of the weapon.)

Presumably this is one of the reasons for the open-strut pattern on many Soviet LVs, where the upper stage ignites prior to separation.

[PB666]

On 12/2/2016 at 1:17 PM, Jonfliesgoats said:

In KSP, we can drop an object from a pawn flying at Mach 3 with little difficulty.  In real life, when you go fast, separating an object from an airplane is really, really difficult.  Aerodynamic interference, shockwaves and paving centers of pressure frequently cause separating objects to behave unpredictably.

It was only in the last decade that we really got supersonic bomb release down.

If  we consider releasing spaceplanes from supersonic and hypersonic motherships, the engineering challenges become really daunting.  So, with current technology, we have a soft limit on the speed at which we can release a spaceplane.  All foreseeable motherships have to be subsonic, not just because that are big, draggy, subsonic airplanes but because the feasibility of releasing anything massive at high air speeds is questionable right now. 

The link is to an unclassified announcement from 2007 when we could finally drop a Mk.82 at Mach 2.

http://m.slashdot.org/story/93831

Well if you can build a big version of the F22, or any aircraft that has a positive TWR thrust about 20,000 feet you turn to vertical can basically head up at an 80 angle and allowing g to take away your v. When V is low enough fire, go into a dead mans spiral, pull out and reignite your engines, and hope you can stabilize the craft before you land. You could have something like the SR71 with a tank fed aired engine that keeps the engines spooling after critical air pressure falls off on the engine all maybe a few pop out turbines to steer the craft into an anti-stall drive and try to re-ignite the engines.

Another is you use something like an F35 and provide 'verticle' thrust after rocket separation and then ignite the rocket.

The problem if you study AC is that the heavier the aircraft the harder it is to achieve TWR at high altitude. This is because engines run on air and air thins, as you climb you are essentially loosing the ability to climb faster and eventually. A B777 can without retooling operate to about 48000 feet but without  a load, the wings can be modified to fly at a lower IAS (boeing design the plane with a short runway takeoff capability which affords better relative lift a lower airspeeds at the cost of its glide ratio relative to the 747). So it is possible to get some commercial heavies at sub-mach speeds. IIRC at 45,000 feet the coffin corner IAS is something like 220 kts and at FL480 it would be like 190 kts, which puts you in the flap zone with non-fuel loadings. So these speeds (dynamic pressure along the AoA) are not too bad.

Theoretically, if we grant horizontal flight, you could push up to maybe 55,000 feet (17km ,at sub Mach speed, still have enough air flow on the air control surfaces) This is going to take away 90% of the atmospheric drag that kills the climb of smaller rockets. The problem is marginal utility of gain. At 30,000 feet 2/3rds of the atmosphere is out of your way. If you launch vertically on Mount Everest by the  altitude in vertical flight were Mach is acheived, its not really worth worrying about because static air pressure is so low. If you then go to say 45,000 feet only 1/6th of the atmosphere remains above. Whats the point in going the additional 10,000 feet to gain a .05 ATM reduction when you have allready reduced the ATM by 0.8333. If you are traveling at .0833 Mach, the plane releases and separates the rocket slows down it also needs to turn so that by the time it reaches 55,000 it will probably just be crossing the sound barrier. The marginal utility of gain on altitude starts to drop at 10 kM.

I made this point a while back, for smaller rockets where drag really clobbers their ability to reach space. The perfect place to launch is between 3000 and 6000 meter elevation on an equitorial mountain (say in Ecuador). The idea here is that you would have less to slow down at maxQ because maxQ would be at a higher velocity at a higher elevation. Because you have less slow down your rocket operates at its optimal reducing the time and therefore gravities effect. While you gain a little advantage in altitude, the major advantage comes from the reduction of drag losses between 0.8 and 1.3 Mach and remedial measures.

 

 

[Jonfliesgoats]

One of the things I like about KSP is that it gets a lot of like-minded people discussing these technical details.  Great input and reference, guys!