I will be the first to admit that I am not an expert on this topic, but as my father-in-law described it to me, (his credentials are that he worked for Convair (Consolidated Vultee) in the late 50's and helped in the design of the airframe of the B-58 Hustler) the idea for the coke bottle design is to keep the cross sectional area at a constant along the length of the airframe. If you will notice, where the wings join the fuselage, the body narrows down. This is where your coke bottle begins and it normally ends after the trailing edge of the wings. By taking into account the wing intersections with the fuselage, the body needs to get smaller so that cross section stays consistent. This was first done on the F-102, the Delta Dart, ( and was part of a redsign when the plane could not reach it's performace specifications) then on the F-106 Delta Dagger, and also on the B-58 Hustler. I do not believe it is really necessary on rockets since the fins are not normally mounted mid-way along the body. Also since rockets have such a short flight time compared to possibly hours on an aircraft, it is not really necessary. Possibly on a cruise missile, and other aircraft that will be within the transonic and supersonic envelopes, but not on rockets. I hope that helps.

Bear is referring to what's known as the "area rule" for high sub-sonic and supersonic aircraft:

http://en.wikipedia.org/wiki/Area_rule

This principle should apply to rockets as well, especially ones that make a significant part of their flights in the Mach 0.8-1.2 range. However, the benefits of an overall smaller profile (min diameter) may very well outweigh the benefits of following the area rule (and therefore needing a wider profile near the middle).

NCR altitude junkies: your thoughts/experiences?

FWIW, I suspect the single most significant aerodynamic factor for these types of rockets to be the frontal area and the associated drag. Consider a 3" rocket vs a 4" rocket on the same 75mm motor. All other things being equal, the rocket with the smaller frontal area will go higher. Remember, pie are square. The more subtle things come into play when the diameters of the two designs are the same.

Coke bottle shape? You must be from the Pepsi Generation?

But why is it that we tend to build "minimum diameter" instead of having a more optimized shape?

Supersonic flight is usually a consequence, not a goal in hobby rocketry.

When the Century Series jets were being developed, the jet engines were not powerful enough. So the aerodynamics were really important in that the jet could not use bulk thrust to force it's way thru the sound barrier, the engines were just not powerful enough. Then as NACA came out with the "Area Rule" it was found that with the right shape, you could slide thru the sound barrier with much less effort and you could increase the range of the aircraft by lowering fuel consumption. In our rockets, I do not believe that we are normally attempting to see how fast we can go, but we might be interested in altitude. This is where tail cones sometimes come in. So if you are building a minimum diameter rocket, why would you want to increase the cross section, thereby adding surface area and air resistance, so as to be able to incorporate the area rule? You are only going to do it where the fins join the body. You can start to see this built into rockets with the second generation of the Honest John, at one time known as the MGR-1B/ MGR-31/MGR-50. Smaller fin diameter, more powerful motor, a foot shorter, lighter weight and a boat tail; it doubled the range. If you are going for altitude, then you know you want the shortest length, the smallest diameter, the shortest fin span and the thinnest fin thickness, thereby reducing the surface area/wetted area as much as possible. The nose cone is shaped by the expected velocity, (such as a round nose or a Von Karman) so you can increase the altitude also. Then you start to figure out the lightest weight you can go, and then calculate the optimum weight to see where you stand. Get a super shiny finish on it and fly it to see how it compares to the calculations so you can see where you might improve. I know I am not telling you anything new, but reminding you of what you already know.

Ken,

One minor correction;

Remember, pie are square.

Actually, Pie are round, cornbread are square.

(somebody had to respond)

Wait, wait, WAIT! My cast iron cornbread pan bakes my cornbread in the shape of ears of corn. What am I supposed to do with that? (Besides making it jalapeno cornbread with butter and honey.) And how is that going to affect the airflow on my rocket?