SpaceX Falcon 9 v1.1 - SpX-6/CRS-6 DRAGON - Discussion Thread

[Jim]

Nah, it sounds like you might need something to generate a small short explosive burst at just the right moment, almost like a firecracker.

Huh?  Do you understand what thruster is?

[dorkmo]

Is it possible to actuate the legs to dampen or absorb some of the lateral landing momentum?

Dampening of the legs did occur shortly after they slid off the deck.

i agree that it appears some dampening happened

i think a problem is that only one (maybe 2) of the legs is doing the absorbing in a case where the rocket is rotating to correct the lateral motion.

once the leg piston bottoms out im assuming that the top pierces the stage and it loses its integrity and pops.

[Space Ghost 1962]

@ID_AA_Carmack Looks like the issue was stiction in the biprop throttle valve, resulting in control system phase lag. Should be easy to fix.

Source Tweet

OK, so that's Elon's leading explanation for the phase lag in the control system. Is that something that should have been caught in a test or is that just bad luck on this flight?

Should have been caught in stage test / static fire / preflight check.

Add:
You'd ramp the valve and observe flow rate change and/or back emf.

Or perhaps it got "cooked" or frozen in flight? Then you'd test at altitude, ramp if sticky til free, and if can't free, adjust feedback terms in control software to compensate for flaw.

[meekGee]

2.  No, they are not.  Again, the rocket flies to a coordinate and not an object.

We haven't seen any evidence that there is or isn't a closed loop terminal guidance system.

Clearly most of the descent is towards an absolute coordinate, but we don't know about the last mile.

It is unarguable that active guidance (the rocket doesn't need to communicate with the barge for that) is more precise, since by using absolute coordinates you add the errors in barge and rocket station navigation.

The only question is whether SpaceX implemented a last mile terminal guidance system of some sort.

So far, all we have are opinions.

Speak for yourself

I do.   My opinions are opinions, unless I can find real data and show it.
That's why I'm ok being wrong sometimes - that how I learn.

Let's get back to terminal guidance now. 

Since SpaceX mentioned radar, we know that at least in Z, at some point the rocket transfers from GPS/INS knowledge to a close loop measurement-based terminal guidance.

So really, the open question is only in X/Y.

I'm fine either way, really...  It could be that SpaceX decided that "open-loop" accuracy is good enough.
I just wish we had direct data.

[meekGee]

@ID_AA_Carmack Looks like the issue was stiction in the biprop throttle valve, resulting in control system phase lag. Should be easy to fix.

Source Tweet

I didn't see that.

That's door #4 - a simple malfunction.  Easiest of all other doors to correct.    Kewl.

[sanman]

Huh?  Do you understand what thruster is?

Sure, I just meant SuperDraco sounds like overkill for that.

@ID_AA_Carmack Looks like the issue was stiction in the biprop throttle valve, resulting in control system phase lag. Should be easy to fix.

Source Tweet

Gee, so the valve was "sticking" (ie. getting stuck) - just a random malfunction, or was it a deficiency relative to this high-intensity landing application?

What's the standard remedy for valve sticking?

[Jim]

So really, the open question is only in X/Y.

Not an open question.  The vehicle flies to a specific coordinate point.  <<<<<<< direct data

[Kabloona]

What's the standard remedy for valve sticking?

A spritz of this...

[puhnitor]

Huh?  Do you understand what thruster is?

Sure, I just meant SuperDraco sounds like overkill for that.

@ID_AA_Carmack Looks like the issue was stiction in the biprop throttle valve, resulting in control system phase lag. Should be easy to fix.

Source Tweet

Gee, so the valve was "sticking" (ie. getting stuck) - just a random malfunction, or was it a deficiency relative to this high-intensity landing application?

What's the standard remedy for valve sticking?

In this case, it was stiction, i.e. static friction. The valve didn't start moving when it was told to, and so the rest of the control system lagged behind. The fix could be as simple as commanding the valve to open, perhaps not even fully, a second earlier to overcome stiction and be ready to respond to input immediately when received. Of course it could also be more complicated than that, but from Elon's tone, it shouldn't be a huge deal.

Of course, then they run into the next thing to go wrong...

[meekGee]

So really, the open question is only in X/Y.

Not an open question.  The vehicle flies to a specific coordinate point.  <<<<<<< direct data

Okidoke then.

[davey142]

These barge ladings are reminiscent of the Falcon 1 days. SpaceX is in unexplored territory and something tells me it'll be a while before they master reuse.

[sanman]

In this case, it was stiction, i.e. static friction. The valve didn't start moving when it was told to, and so the rest of the control system lagged behind. The fix could be as simple as commanding the valve to open, perhaps not even fully, a second earlier to overcome stiction and be ready to respond to input immediately when received. Of course it could also be more complicated than that, but from Elon's tone, it shouldn't be a huge deal.

Of course, then they run into the next thing to go wrong...

So is the valve sticking/stiction for this particular valve something that's manifested itself before? Otherwise, what bad time for it to suddenly/randomly make an appearance. Maybe it was the high intensity operation and demanding nature of this landing which caused the valve to underperform.

[Zach Swena]

     Without knowledge of the design, it is hard to know what the source of the sticky valve was.  It could have been a quality control issue.  Sometimes you don't know where to set your tolerance limits until you test all the possible situations.  This is really the first thing they have done that has required this level of response from that valve.  It could also be a design issue.  They probably know that part already, because test data should tell them how fast the valve is supposed to be.  I would imagine it could be a combination of both, because they could predict that they need a certain response time from the valve, but didn't have the need to require compliance from the production valves until now.

     Keep in mind, that they never did the F9R Dev2 testing at spaceport america where they would have encountered and solved issues like this.  It is cheaper to do high risk testing on vehicles that are already paid for and destined for destruction anyways though.  Overall, it is a smart move financially.

These barge ladings are reminiscent of the Falcon 1 days. SpaceX is in unexplored territory and something tells me it'll be a while before they master reuse.

Falcon 1 was them learning and mastering existing technology.  Basically learning the ropes and how to do their homework.  For reuse, they are writing the book, it could potentially take a lot longer.  One advantage though is that the testing is much less financial risk, as they are testing after the primary mission is complete.  Their test rate is much higher also, because they are flying a whole lot more rockets.  Because of the higher flight rate, and greater experience and competence overall, I think it should actually take a lot less time.  Each test can identify problems like this that are hard to pinpoint and fix without hindsight.

[llanitedave]

Huh?  Do you understand what thruster is?

Sure, I just meant SuperDraco sounds like overkill for that.

@ID_AA_Carmack Looks like the issue was stiction in the biprop throttle valve, resulting in control system phase lag. Should be easy to fix.

Source Tweet

Gee, so the valve was "sticking" (ie. getting stuck) - just a random malfunction, or was it a deficiency relative to this high-intensity landing application?

What's the standard remedy for valve sticking?

WD-40?

[Zach Swena]

So is the valve sticking/stiction for this particular valve something that's manifested itself before? Otherwise, what bad time for it to suddenly/randomly make an appearance. Maybe it was the high intensity operation and demanding nature of this landing which caused the valve to under perform.

I don't believe they have ever flown a flight profile where throttle lag like this was critical.  That said, it seems like if it was a fundamental design issue of the valve that caused the stiction, that they would have thought to test it and found the problem earlier.  It is most likely a combination of a valve that was on the edge of the allowable range and using a valve that may not have been designed exactly with response time in mind.  A biprop valve has a whole lot of other things for the design engineers to worry about after all.

[Pete]

I don't believe they have ever flown a flight profile where throttle lag like this was critical. 

*this* is exactly why they wanted to do multiple flights with the F9R test vehicles.
You have a vehicle having to perform to a whole new set of requirements, in a whole new environment.
You want to do your testing in a controlled test location, with several truckloads of monitoring tools at hand.
You want to be able to test when your test team is ready, not just when a launch is occurring for another customer.
You want to test with greater margins on fuel and timing.
You want to do your testing a bit more out of the public eye.
.
They will get it, eventually, using the current approach of piggybacking the tests on actual launches. But don't expect a fast-track development cycle until or unless they get test flights up and running again.

[deltaV]

Looking at the video (https://vine.co/v/euEpIVegiIx) I'm struck by the dramatic attitude changes. I wouldn't expect a problem with the throttle of the main engine to directly change the rocket's attitude and I don't see why the control system would respond to incorrect thrust by purposefully changing attitude. I'm therefore guessing that the malfunctioning valve is related to attitude control, i.e. either thrusters or the system that gimbals the main engine. But the thrusters are nitrogen gas (no biprop) and the main engine gimbaling is hydraulic (also not biprop) so what's the "biprop throttle valve"?

[rickl]

For reuse, they are writing the book, it could potentially take a lot longer.  One advantage though is that the testing is much less financial risk, as they are testing after the primary mission is complete.  Their test rate is much higher also, because they are flying a whole lot more rockets.  Because of the higher flight rate, and greater experience and competence overall, I think it should actually take a lot less time.  Each test can identify problems like this that are hard to pinpoint and fix without hindsight.

At this point, the only way that test vehicles like Grasshopper would be useful is if they can fly them high enough to shut off the engine, let it accelerate to terminal velocity, then restart the engine for landing.  Otherwise, they may as well just carry on with their "testing" with actual flight hardware in actual flight conditions.  The stages would be dropped into the ocean and destroyed anyway.

[cosmicvoid]

... I wouldn't expect a problem with the throttle of the main engine to directly change the rocket's attitude and I don't see why the control system would respond to incorrect thrust by purposefully changing attitude....

I'd think that too much (or too little) thrust while gimballed results in an attitude overshoot (undershoot).

[friendly3]

At this point, the only way that test vehicles like Grasshopper would be useful is if they can fly them high enough to shut off the engine, let it accelerate to terminal velocity, then restart the engine for landing.

The plan was to send F9R Dev1 to New Mexico to do just that.