20
$\begingroup$

In the Dragon v2 abort test yesterday, it lifted off with its unpressurized trunk in tow, then ejected the trunk at apogee. Is this the real abort profile? Wouldn't it be easier to pull away without the trunk?

If it's the real profile, why spend the fuel mass to save the cargo?

If not, how would the (impressive) acceleration of the initial test compare with a realistic scenario, where the engines are fighting drag but pulling less mass?

Improve this question
$\endgroup$
3
  • 3
    $\begingroup$ I would imagine that the trunk provides a bit of stability in the abort, so it might actually be easier to abort with the trunk than without it. The engines are designed for propulsive landing, so they may provide excess thrust anyway and then the trunk's weight is not an issue. Just speculation on my part $\endgroup$
    – neelsg
    Commented May 7, 2015 at 10:48
  • $\begingroup$ @neelsg Abort is what sets the maximum thrust requirement of the engines. They need to pull the fully-fueled capsule + trunk at 6 G's, compared to landing an empty-ish capsule alone at about 1 G. $\endgroup$
    – Potatoswatter
    Commented May 7, 2015 at 11:30
  • $\begingroup$ Part of the weight penalty of the trunk depends on the mass of the cargo in the trunk at the time of the abort. I don't imagine the had the trunk at max gross during the abort? $\endgroup$
    – Bear Lemley
    Commented Oct 14, 2018 at 3:53

3 Answers 3

Reset to default
30
$\begingroup$

It's an aerodynamics problem. SpaceX has designed both of their Dragon capsules to be inherently stable for reentry; should the capsule not be able to orient itself actively due to a Draco engine failure - it will passively orient itself during reentry with the blunt heatshield pointing forward, into the velocity vector. This minimizes risk and ensures the heatshield is always facing into the atmosphere for descent. This is an inherent property of the truncated cone reentry body.

This design has one drawback however: if you have a scenario where Falcon 9 encounters a failure mode that risks the lives of the astronauts on board the Dragon, and you need to escape from the rocket, you want to do so as quickly as possible and encounter the least aerodynamic resistance as you can. This means you need to point nosecone forward, and heatshield back, which is precisely opposite to the aerodynamic design of Dragon.

The solution SpaceX chose was to keep the trunk on during the abort. Note that the trunk itself does not keep the capsule pointing forwards, but rather the fins on the trunk do. They sculpt the airflow around the capsule in such a way that it is stable nosecone forward into the velocity vector.

Dragon Pad Abort Dragon immediately begins to rotate post-trunk deploy, prior to drogue chute jettison.

You can see all of this during the Pad Abort video. As soon as the trunk deploys and separates from the vehicle, the capsule immediately begins to rotate, trying to stabilize with the heatshield forward.

Additionally, the trunk is incredibly lightweight. Dragon weighs in at around 8000-9000kg. The trunk is less than 1000kg. A minor drawback, sure, but who cares when you're saving the lives of the crew inside.

Improve this answer
$\endgroup$
11
  • 2
    $\begingroup$ Thanks! I found a reference, but I still wonder: 1. Why not just attach the fins to the capsule, shuttlecock-style — or deploy some other kind of drag-generating device? 2. The trunk is only light when it's empty, does this constrain its maximum load? $\endgroup$
    – Potatoswatter
    Commented May 7, 2015 at 11:23
  • 1
    $\begingroup$ @Potatoswatter If the fins are what makes the capsule go pointy-end-first, you don't want to put them on the capsule. Remember, the idea is to be re-entry safe in case of failure. Retractable fins are more mechanically complex than explosive bolts, and you really don't want to end up with the capsule stuck configured to go point-end-first when it's trying to re-enter. Leaving the fins on the trunk lets you shed a little extra weight (you don't need to take the fins to space; they're no use after launch) and it makes sure that the capsule re-enters heatshield-first, even in case of failure. $\endgroup$
    – anaximander
    Commented May 7, 2015 at 11:46
  • 4
    $\begingroup$ @Potatoswatter ...it does, yes. Somehow I forgot that part! I think in the end it's down to the principle of keeping things simple. Multiple fins each with their own decouplers would add complexity. Having some kind of separate fin assembly, large enough to carry suitably-sized fins and able to be jettisoned separately to the trunk, would just add weight and complexity (and potentially wasted internal spaces) for not much benefit. In almost every case where you need the fins, you need the trunk (or at least the penalty to having the rest of the trunk as well as the fins is minimal). $\endgroup$
    – anaximander
    Commented May 7, 2015 at 12:43
  • 4
    $\begingroup$ "Previous spacecraft capsules have all required active control to maintain a stable orientation during reentry." I do not believe this is true. In fact, I believe the opposite is true. $\endgroup$
    – Erik
    Commented May 8, 2015 at 3:09
  • 2
    $\begingroup$ I don't really need to since it is not my answer and truncated cones are well known to be aerodynamically stable with their blunt end forward. $\endgroup$
    – Erik
    Commented May 8, 2015 at 3:32
1
$\begingroup$

I would also think that in a failure where the rocket either explodes or must be detonated, having the trunk attached might act as shielding from penetrating debris. There would only be a short window where that might be needed, and it would seem to coincide with the duration that the trunk remains attached.

Improve this answer
$\endgroup$
1
$\begingroup$

I figured keeping the truck attached was an effort to put CG ahead of CP so Dragon is naturally stable flying forward under power. On it's own, Dragon wants to orient heatshield first. Note Boeing Starliner also has pusher abort system, but does not have a long trunk...

Improve this answer
$\endgroup$

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.