Ballistic Missiles

Started by
17 comments, last by TerranFury 21 years ago
quote:Original post by Anonymous Poster
However the formula above is a _model_ of the behavour in a vicous fluid (=air). Another model is to use v^2 instead of v.
And probably something interesting happens around the speed of sound...


Actually, v2 is more appropriate for a missile, which can move quite fast. The "v" version is only a good approximation at very low speeds.

And....since you added "..." I assume you want some information on what happens around the speed of sound (Mach 1), . Since I happen to know just a wee little bit about that subject....

At something less than the speed of sound (say, Mach 0.8), shock waves begin to form on the surface of a body. The shock waves occur before the object reaches the speed of sound since the flow accelerates around the body. And when the object is around Mach 0.8 or so, the accelerated air around the body exceeds Mach 1. The air just in front of the shock wave is supersonic. The air just behind the shock is subsonic. There are associated massive changes in air properties such as air pressure across the thickness of the shock. (A strong shock is very thin, actually only a few mean-free-paths between molecules in thickness, almost a true physical discontinuity. But lets not talk about the shock waves that occur in nebulae and galactic systems!)

The emergence of these shock waves, and corresponding large changes in air properties across the shock thickness, leads to a phenomenon known as "drag rise," which is a large increase in drag. The drag rise occurs within the transonic flight regime, which begins at the Mach 0.8 and extends to something beyond Mach 1, say Mach 1.2. Within this transonic regime, the shock waves emanate from the object surface. As the speed increases to the high end of the transonic regime, the shocks move rearward until eventually they move as far back as they can. The shock waves are then emanating from the most rearward edges of the object, and this is the limit of the transonic regime. The flow over the surface at this limit, the transonic/supersonic boundary, is entirely supersonic (Mach > 1). But the flow immediately behind the object, immediately behind the shock, is subsonic (Mach < 1). For example, an airplane wing in transonic flight would have a shock wave on the top and bottom surfaces, starting somewhere between the leading and trailing edge and extend upward above and below the wing until the shock is dissipated by the viscous and turbulent air above and below the wing. (The shock kind of smears out away from the wing to become merely a strong pressure wave, and this is what causes sonic boom.) But, I digress. As the wing moves at higher and higher speeds, ultimately the shock will move backwards until it reaches the trailing edge of the wing. When this happens, but not before, the flow over the surface of the wing is entirely supersonic. When the flow over the surface is entirely supersonic, you''re out of the transonic regime. When the flow becomes entirely supersonic, the shocks no longer interact with the flow over the wing in the same way. And, though it may seem strange, drag will begin to decrease from the maximum drag seen during drag rise. But the drag will once again increase as the speed continues to increase beyond the transonic/supersonic boundary.

Cool, n''est ce pas?

Graham Rhodes
Senior Scientist
Applied Research Associates, Inc.
Graham Rhodes Moderator, Math & Physics forum @ gamedev.net
Advertisement
Graham:
That was an awesome post!

Will
------------------http://www.nentari.com
Heh heh. Shock waves ARE awesome! You can actually see the shock waves in certain circumstances. There is a wind tunnel technique known as "Schlieren" or "shadowgraph" imaging that allows you to visually see the changes in air density. Here is a link to a photo:

http://aerodyn.org/Courses/aerodyn.html

And another, of the space shuttle. (In this case, the shuttle is moving supersonic relative to the world, but since the flow behind the main shock up front is subsonic, the airflow that actually surrounds the shuttle is almost entirely subsonic. There is another shock in back, which indicates that the flow does eventually reach sonic speed somewhere on the wing.)

http://www.eng.vt.edu/fluids/msc/gallery/gall.htm

That airfoil is in the transonic flow regime, and you can see that in this case there are multiple shock fronts on the top and bottom surfaces. You can see the main shocks shooting off above and below the wing as I described.

And a link discussing how you can sometimes see this in the wild:

yarchive.net/air/airliners/shock_waves.html

What''s really cool is that you can write quite simple fluid dynamics simulations that will automatically "capture" these shocks around arbitrary shaped bodies just because you''ve picked the right numerical method (so-called "upwind" numerical methods capture shocks naturally without any special coding or conditional statements). The calculations can take a long time though.

But for things that are made of straight lines and sharp corners, shock wave properties can be calculated using fairly simple analytic formulas, as this Java applet demonstrates:

http://www.grc.nasa.gov/WWW/K-12/airplane/shock.html

My challenge to you. Figure out how this can be useful in a game!

Graham Rhodes
Senior Scientist
Applied Research Associates, Inc.
Graham Rhodes Moderator, Math & Physics forum @ gamedev.net
That really is some interesting stuff, grhodes. Now you ask how all this discussion of shock waves could be useful in gamedev? Easy. It should affect the handling of the fighter jet in the dogfighting game. With some refraction effects (due to the different air densities), it would also be an impressive graphical gimmick. I''m afraid my ballistic missile problem is hard enough though without adding CFD to the mix!!

Anyway, getting a little more back on track, I''ve so far been unable to find a way to solve that system of differential equations that cedricl and I were working on.

cedricl: I''m not 100% positive, but I''m not sure that trigonometric substitution get us much closer, because there are still vs and dvs in the equations. I know you can eliminate them with inverse trigonometric functions, but the results are even uglier than the original cartesian equations.

Can anyone solve this? I''m completely stumped.
quote:Original post by TerranFury
cedricl: I''m not 100% positive, but I''m not sure that trigonometric substitution get us much closer, because there are still v s and dv s in the equations. I know you can eliminate them with inverse trigonometric functions, but the results are even uglier than the original cartesian equations.

Yeah, that''s what I thought after posting them . To be honest, I haven''t given this problem much thought afterwards. Since I don''t know the general approach to solving differential equations, I can only guess the solution, which is pretty tedious. And so far, my guesses haven''t worked very well.

A Google search could probably give you an answer.

Cédric
quote:Original post by cedricl
A Google search could probably give you an answer.


Come on, give me a little more credit! That was the first ting I did before posting. Just about everything is about Star-Wars-esque money-wasting missile defense schemes, North Korea, or some US or Israeli military stats. Occasionally you run across a way-over-my-head paper from NASA on missile aerodynamics. That''s it, I''m afraid!
Alright, I made a search and couldn''t find anything either, except that non-linear differential equations seem really hard to solve. I''ll try my hand at it once again, but I''m not very optimistic.

Cédric
quote:Original post by TerranFury
With some refraction effects (due to the different air densities), it would also be an impressive graphical gimmick.


Yeah, yeah, definitely! There are actually closed form solutions to find 2D shock wave angles, and this could be tied to a graphical effect quite easily. It would be super cool to see translucent shock waves emanating from the leading edge of a fighter jet or from the forebody of a ballistic missile flying at transonic speed, and changing angle as angle of attack changes! VERY cool, I would think.

quote:Original post by TerranFury
I''m afraid my ballistic missile problem is hard enough though without adding CFD to the mix!!


Actually, you don''t really need CFD to get a good simulation of this. The CFD really is only useful for low level design of the shape. In practice, flight simulation is done using rigid body simulation not unlike the usual gaming rigid body simulation. Aerodynamic forces and moments/torques are calculated using derivatives known as "stability derivatives" such as dCL/dalpha (change in lift coefficient / change in angle of attack). I don''t know where you''d get those for a missile, since this is all most likely super secret stuff.


Graham Rhodes
Senior Scientist
Applied Research Associates, Inc.
Graham Rhodes Moderator, Math & Physics forum @ gamedev.net
Why not feed the equation to Maple or some other math program to solve?
.

This topic is closed to new replies.

Advertisement