SpaceX News

[Maui]

There is one other factor that I do not believe they are accounting for, and that is the transition turbulence from water to a solid platform. This was a huge factor that the pilots had to deal with when doing the crossover maneuver for the landing on LHA. I suspect they have not enough experience with that part of the transition.
AV8B harrier is what I am talking about.

Oh, I think they very much have that in mind (whether or not they fully account for it). They recently upgraded the barge's thrusters from 300 hp to 1000 hp. Whether or not that's enough to keep the platform perfectly still in rolling seas, I can't say, but certainly addressing this issue via the barge vs via the rocket makes much more sense. There is a long history of companies and technologies dealing with keeping vessels steady in the sea while what SpaceX is doing is a first. Best to leave this factor to be mitigated by existing solutions.

[ladajo]

Well, if anything, they are proving that the barge is really durable.

On another note, I have wondered about the after. Once its sets down, (and by the way placement was amazing for the second try: The CEP was impressive by my book for first timers.) what then? How stable is the booster standing on its legs at sea on a rolling platform? Do they have a means to secure it? Must, one would think, but I have seen no evidence to date to suggest it. Things that make you go, "hmmmm?"

All in all, good showing, and I expect try three to work based on what I have seen so far.

[mvanwink5]

Translation, it seems to me, is the trouble and it seems with a larger target, perhaps double the barge target radius, there would have been a successful landing. So, perhaps if there was enough room for cows they would have a used rocket for relaunch.

[paperburn1]

Translation is the problem, or at least that is what I think as well. The turbulent buffet characteristics from the exhaust is different over water than over a solid platform. When crossing over onto a hard surface the plume will change buffet characteristics and manifest itself by rebounding upward instead of being absorbed by the water . This will create a asymmetric disruption of the exhaust flow putting some forces that will push the base around until it is fully on the platform. Because the flow is highly non-linear in the transonic regime of the rocket exhaust , It will be dominated by moving shock waves as it is crossing on to the solid surface of the platform. Without knowing real numbers I can only assume it will place some major forces on the lower regions of the rocket. This would be something they did not encounter in their testing as I am assuming(once again ) that all landing tests were done over water or dry land. Both surfaces that would be of the same constancy and stability. I did not see any testing done where it transitioned from one surface to another.
As a thought experiment remember the water spray at the base of the rockets fired by NASA. Jets of water released by the sound suppression systems installed on the pads to protect orbiters from being damaged by acoustical energy, reflected from the platform during the liftoff stage of a rocket launch. this could very well be the same for landing the rocket.
http://www.nasa.gov/missions/shuttle/f_watertest.html

[paperburn1]

As for what happens after landing I understand they are going to rush onto the platform and weld the legs down before moving the barge. And as far as stability goes there are many dampers that can make it steady as a rock. cruise ships use wings under the water and there is also the cone that can be deployed on barges to reduce pitching. And if you really wanted to get high tech you could mount one of those sway stabilizers they use in skyscrapers to stop sway from winds and earthquakes.
http://en.wikipedia.org/wiki/Tuned_mass_damper
http://gizmodo.com/5019046/how-a-730-to ... earthquake#

[D Tibbets]

Remember that an open ocean barge landing is at best a temporary stop gap for testing. Space X has obtained a landing pad at Cape Canaveral (I assume they still have to convince NASA that it is safe), and may have plans to use east Texas. I don't know how these sites effect the first stage recovery dynamics but it eliminates much of the problems and cost of a barge landing.

Dan Tibbets

[mvanwink5]

This would be something they did not encounter in their testing as I am assuming

Perhaps the surface transition is the dominant problem for a slight off target approach at sea (that makes sense to me FWIW), and if sea landing is indeed untested due to logistics and budget then landing might be difficult without a dead center drop from XX feet above the barge. In the end, the relatively small barge radius might be too aggressive for a water landing.

If a dead on landing from XX altitude is not a consistently achievable goal (and needed), then it might be prudent to mount a sea barge test landing program using a test rocket? It might be possible, just brainstorming here, with a separate fuel barge.

Anyway, it will be interesting to see what unravels from this point...

[GIThruster]

SpaceX has a "strict no-assholes" policy. Time index 32M:

https://www.youtube.com/watch?v=MJHTY0gWOGw

[paperburn1]

After spending a day thinking about this crossover problem it really is quite easy to solve. a larger platform for the easy solution and just center punch the landing. I think they may be landing off center in case there is a control failure they do not "center punch" the barge. And maybe a few tweaks to the control feedback is all that is needed as well. In any case I am thinking the third time might be the charm.
as far as testing they could just make a large pond next to the landing pad in the cow field and get all the data they need from the grasshopper.

[mvanwink5]

From Musk:

Cause of hard rocket landing confirmed as due to slower than expected throttle valve response. Next attempt in 2 months.

[Skipjack]

From Musk:

Cause of hard rocket landing confirmed as due to slower than expected throttle valve response. Next attempt in 2 months.

Yeah, from what I understand, the "sticktion" of the valve was higher than the software expected, causing a delayed reaction and subsequent overcompensation by the software. I guess they will simply adjust the software instead of making changes to the actual hardware, but I might be wrong.

[hanelyp]

A delayed reaction to a thrust control caused a delayed reaction and overcompensation in directional control? The longer distance video clearly shows a delay in gimbal action and directional control.

[D Tibbets]

Speculating, yes a sticky flow valve could lead to gimbal excess. Say the computer tells the engine to increase thrust 20% with a 10 degree deflection from vertical, but the sticky valve retards/ delays this adjustment in vectored thrust- intensity, not direction. The yaw and/ or translational movement is inadequate in the given period of time. Now the rocket needs additional more aggressive gimbal movement and thrust to make up for the increased fall rate and similar or worse yaw attitude. With the flow valve lagging, the situation compounds to the point that the computer tries maximum allowable thrust and gimbal variations to try to catch up. Good sensor and programing might correct the out of bounds response, given enough time and fuel. Apparently the programing and/ or sensor interactions were inadequate. Weather the sensers, programing, valve stiction, mechanical gimboling apparatus, or a combination needs to be addressed is the question.

Dan Tibbets

[hanelyp]

Looking at the high res video from a distant viewpoint, looking at the direction of the flame from the engine, the flame is still tilted for faster rotation until shortly after the rocket passes vertical. Then it snaps over to the other side.

[DeltaV]

It has long been a standard control systems technique to use intentional dither to reduce the bad effects of valve stiction and other nonlinearities.

http://people.mech.kuleuven.be/~farid/t ... dither.htm



There are potential negatives, such as exciting unwanted resonances, valve wear, etc.

If SpaceX decides to go that route, they will likely restrict it to just before touchdown.

One of my early mentors told me of a nonlinear control experiment that he conducted, right after reading a text on nonlinear control. He drove his Volkswagen down a pothole-saturated road while dithering the steering wheel back-and-forth and found that he had better directional control than without dithering.