How much can the falling speed of starship approximately be reduced by the bellyflop fall with fins? Are we talking about numbers around 5% or 50% here?

Wouldn't it make sense to put holes where the fins are attached to the body and blow out the air at the end of the fin through channels to "increase" the fin size with air?

  • 1
    $\begingroup$ If you watch a Flacon 9 booster landing and compare it to Starship SN8, it looks like Starship is bellyfloping at less than 10% of the speed. Starship comes in really slowly. I'm sure someone can find the videos and do a proper analysis. $\endgroup$
    – UEFI
    Dec 10, 2020 at 17:29

2 Answers 2


Let's look at the ballistic coefficients of a bunch of reentry vehicles. This is the mass divided by the cross section area. The drag coefficients of all these blunt bodies will drop from very roughly 1.0 while hypersonic to 0.5 near terminal velocity, except for the Pershing II MARV which being a sharp cone is smaller.

Dry Mass Wet mass Length Diameter Xsec area Ballistic coefficient
Falcon 9 S1 end-on 24 mt 49 mt 3.7 m 10.2 m^2 4803 kg/m^2
Pershing II MARV 0.64 mt 0.5 m 0.2 m^2 3260 kg/m^2
Starship end-on 120 mt 150 mt 50 m 9 m 63.6 m^2 2358 kg/m^2
Dragon 2 9.5 mt 12.5 mt 4.0 m 12.6 m^2 992 kg/m^2
Apollo CM 5.5 mt 3.9 m 11.9 m^2 465 kg/m^2
Shuttle 78 mt 37.2 m 23.8 m 250 m^2 312 kg/m^2
Starship sideways 120 mt 150 mt 50 m 9 m 520 m^2 288 kg/m^2
Shuttle SRB 91 mt 590 mt 45.5 m 3.7 m 603 m^2 151 kg/m^2

By descending sideways, Starship decreases the ballistic coefficient by 8.2x, which should decrease the terminal velocity of the vehicle by 2.8x with the fins fully outstretched.

The vehicle will not actually get down to terminal velocity.

The decrease in velocity when the landing burn starts will decrease the amount of propellant needed during that burn, approximately linearly.

On Falcon 9, every 7 kg of propellant used for the landing burn reduces the potential payload into low Earth orbit by 1 kg. The landing propellant is approximately equal to the stage dry mass.

On Starship, it's 1 for 1, so they are far more sensitive to landing propellant usage. The landing propellant is 20% of the stage dry mass. I'm not sure what accounts for the propellant savings beyond lower terminal velocity, but the most obvious candidate would be greater deceleration which reduces gravity losses.

  • $\begingroup$ +1 for that chart. Need drag coefficients to "ace" it. Interesting that the numbers should work well for re-entry, but for landing, see Shuttle (glider) vs Starship sideways (you know). Need Starship end on with 1 Shuttle booster parachute for +20! $\endgroup$ Mar 16, 2021 at 15:30

Can't speak to Starship. With the shuttle, the belly of the craft is the primary drag energy dissipation device. In the initial stages the essential goal is to keep the force of drag matched to the centre of mass so that the ship doesn't tumble.

(I think this is why missing a tile or so is a big problem. A single tile will have the air fill the space and following air will essentially flow around it. (think dragging a cup of milk through a pool of water very evenly) Lose either a bunch of tiles in a row, or tiles on the leading edge then there is no cushion effect and now you have turbulence which makes drag; irregular asymmetry, which which distorts the drag, and the potential for burn through at the unprotected part of the ship.)

Initial reentry attitude corrections are done with the steering rockets. Craft has to slow down and get into thicker air before the wing's control surfaces work.

Remember that a lot of this is happening in VERY thin air and very high speed. I'm not sure how you would go about taking in air that's coming in at Mach 15 or so, then blowing it out through slots fast enough to make a difference. Seems like a lot of equipment and huge engineering effort for a fairly small gain. Just make the fins larger.

Slowing down faster increases the acceleration by definition. Ship has to be stronger to afford the additional stress. But you may ask, "It has to be strong for launch!" True, but coming in on its belly it has to be strong enough for stress in a different direction.

(The Atlas rocket, designed for 10+ G's vertically, couldn't support it's own tanks lying on it's side. They had to keep the tanks pressurized with inert gas to keep them from collapsing.)

  • 2
    $\begingroup$ The shuttle has been out of service since 2011. Do you have any references to back up your assertions? $\endgroup$ Nov 27, 2020 at 3:44
  • 1
    $\begingroup$ I'd have to do serious digging. Was an article in something like Pop. Sci or Discover or Scientific American. Possibly Analog. $\endgroup$ Nov 27, 2020 at 15:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.