SpaceX launch abort test

[chris_leyson]

SpaceX crew dragon launch abort test in about an hour, 15.30 UTC. Good luck SpaceX.

[Kilrah]

Watching now, T-8

[rdl]

I bet there's a lot of happy people at NASA and SpaceX today.

[German_EE]

Job done, everything appeared to work as it should.

Now we need Boeing to do the same test, only fair!

[Homer J Simpson]

Edited video of test.

[donotdespisethesnake]

Do we know if the booster self destructed, or if Range Safety Officer pressed the destruct button?

[rdl]

I'm pretty sure they just cut off the main engines to simulate a total failure and let the automatic systems take over from there.

In the Post Test Media Conference, Musk said "... the booster actually exploded, as expected."

[Neilm]

SpaceX have been saying they expected the pressure of the air to destroy the the vehicle.

Boeing are going down the alternate route for abort qualification that requires overdesign and a lot of certified paperwork. I hope they don't get the group that did the 747Max paperwork.

[chris_leyson]

Just seen the post launch conference and Tim Dodd (Everyday Astronaut) asked the same question, what was the sequence of events. AFAIK they command a main engine shutdown or cut off, MECO, and the launch abort system sees an anomalous event like an unscheduled main engine cut off. Elon said "they set the trigger thresholds very low". And that just got me wondering what the hell triggers an escape sequence. Edit: Booster exploded after crew dragon had departed.

[Monkeh]

Do we know if the booster self destructed, or if Range Safety Officer pressed the destruct button?

Massive structural failure occured due to atmospheric forces. ie. it crumpled and went boom.

It turned from a nice dart to a mostly flap-topped tube at supersonic velocities and tumbled.

[rdl]

Yeah, the Falcon booster became very non-aerodynamic after the Dragon departed and it was in max q region of flight. The second stage wasn't going to fire so it had no engines, but it had a full load of fuel and the first stage was still pretty full, so it was a big boom.

Apparently a large piece of the booster did survive and fell back into the ocean.

Range Safety normally won't resort to destruct unless the rocket deviates significantly from the planned flight path.

[Homer J Simpson]

[TerraHertz]

Watched the video last night, and it's very entertaining.

One thing bothers me. The abort sequence was very controlled, and required a short interval of foreknowledge. Apparently they initiated startup sequence on the escape pod rockets, _then_ shut down the primary booster engines. The capsule separated, and the now un-powered main booster and 2nd stage slowly skewed, crumpled and blew up. By that time the escape capsule was well clear.

The 2nd stage separated intact from the exploding 1st stage, fell to the ocean and impacted at like mach 1.5 with full fuel tanks, making another spectacular explosion.

But... what if the 1st or second stage blows up without warning?  This has happened before.
What is the reaction time from something going wrong, to the escape capsule boosting away?

[wraper]

Apparently they initiated startup sequence on the escape pod rockets, _then_ shut down the primary booster engines. The capsule separated, and the now un-powered main booster and 2nd stage slowly skewed, crumpled and blew up. By that time the escape capsule was well clear.

You are wrong here. Capsule started escaping after detecting anomaly due to engine shutdown. FYI engines do not shut down instantly. There was no command for capsule escape as such.

[wraper]

But... what if the 1st or second stage blows up without warning?  This has happened before.
What is the reaction time from something going wrong, to the escape capsule boosting away?

Watch press conference instead of speculating. Reaction time is milliseconds. Also it was claimed that it is designed to escape even during explosion, which was not really an explosion, just a big fire.

[iMo]

But... what if the 1st or second stage blows up without warning?  This has happened before.
What is the reaction time from something going wrong, to the escape capsule boosting away?

The capsule may boost away in zero altitude and zero speed. There is a video with that test too.
A good question how fast - when for example the second stage explodes while still at launching pad.
It still will be XXXX times faster than with Space Shuttle, imho.



Orion test:

[iMo]

Soyuz is using the same system for ages and it works well, it seems

[rdl]

About the only similarity between the Soyuz system and what the the SpaceX Dragon has is that they're both for escaping emergencies.

[iMo]

Is the liquid propelant in Dragon abort boosters the best choice? Solid one would be more reliable, imho (Orion/Soyuz/Apollo/Mercury).

[ajb]

Is the liquid propelant in Dragon abort boosters the best choice? Solid one would be more reliable, imho (Orion/Soyuz/Apollo/Mercury).

Originally Dragon was intended to land propulsively using the same thrusters, so the level of control in terms of throttle and possibly multiple starts led them to go with a liquid fueled system.  This was actually part of the reason for the catastrophic failure during a prior test, when some propellant leaked past a check valve--if it hadn't been the need for multiple starts they would have used a burst disc instead of a check valve, and in fact that's what they switched to after discovering the cause of the failure.  Since the Superdraco thrusters use pressure-fed hypergolic propellants they're about as simple as liquid fueled engines get, although I'm not sure how that translates into reliability as compared to solid fuel systems (which are admittedly much simpler in terms of mechanics, but this is, after all, rocket science).

[bw2341]

Just seen the post launch conference and Tim Dodd (Everyday Astronaut) asked the same question, what was the sequence of events. AFAIK they command a main engine shutdown or cut off, MECO, and the launch abort system sees an anomalous event like an unscheduled main engine cut off. Elon said "they set the trigger thresholds very low". And that just got me wondering what the hell triggers an escape sequence. Edit: Booster exploded after crew dragon had departed.

Elon is a bit tongue tied in his explanation.

I've read speculation elsewhere on how this sequence was triggered. The abort system is triggered automatically if any one of many different parameters exceed a trigger threshold. If you want the abort to trigger at a specific precalcuated max drag velocity, all you have to do is insert a rule to reduce the trigger threshold at the target velocity.

For example:

if v>=vmaxdrag then reduce acceleration threshold from 0.01g to 0.000000000001g

If the threshold is unreasonably low, the intrinsic noise of the sensor input would be enough to exceed the trigger threshold and the abort will be triggered 100% of the time.

If this is how they did it, they chose a clever solution. This would allow them to use unmodified software on the Falcon first stage. The only change in the Dragon software would be to insert this rule. This would test the abort software logic in the Dragon in the same way as a real fault.

[bw2341]

Watched the video last night, and it's very entertaining.

One thing bothers me. The abort sequence was very controlled, and required a short interval of foreknowledge. Apparently they initiated startup sequence on the escape pod rockets, _then_ shut down the primary booster engines. The capsule separated, and the now un-powered main booster and 2nd stage slowly skewed, crumpled and blew up. By that time the escape capsule was well clear.

The 2nd stage separated intact from the exploding 1st stage, fell to the ocean and impacted at like mach 1.5 with full fuel tanks, making another spectacular explosion.

But... what if the 1st or second stage blows up without warning?  This has happened before.
What is the reaction time from something going wrong, to the escape capsule boosting away?

Apparently, it takes 700 milliseconds. Is that fast enough?

I took a quick look at the video for the AMOS-6 accident. The payload was hung up on the tower for 9 or 10 seconds before it fell over in one piece.

For CRS-7, the second stage was leaking for about 9 or 10 seconds before the rocket broke up.

For Cygnus CRS Orb-3, it looks like the top of the rocket was above the burning fuel for about 3 seconds.

So I guess kerosene fuelled rockets blow up in a big fireball rather than explode like a stick of dynamite. Is 700 milliseconds fast enough? From a quickie web free fall calculator, an object will fall less than 3 metres in 700 milliseconds.

[iMo]

Another "on-the-job" abort training (took pretty long to find the right button..)