SpaceX Falcon 9 : CRS-16 (Dragon SpX-16) : December 5, 2018 - DISCUSSION

[hootowls]

Well, they didn't let them do an RTLS at Vandenberg earlier this week in case something happens, I wonder how the Air Force looks at that decision after this failure.

On the one hand if the exact same failure happens, seems like there's no danger to Vandenberg at all, the stage just lands in the Pacific (that recovery would be worse though, seeing as the West Coast SpaceX fleet is based further away than "just pop down the road"-Cape Canaveral).

On the other hand, if something happens a couple of minutes later during the landing burn or something else just doesn't work in general...sure shows there'll always be some danger to this whole thing.

Actually now that I think about it, having your landing site be directly next to your launch pad is probably convenient but if something goes wrong...especially at your only West Coast pad compared to the two Florida pads you have...well, I hope all their failures end as smoothly and entertainingly as this one.

We know the answer to your first question.

The stage doesn't shift the IIP onto shore until the main engine relight happens and the engine is confirmed good.

We can also assume that any other health checks that fail also result in an offshore landing.

They can also walk the IIP along a safe corridor as it approaches the landing pad.  There's no reason that the post-relight path must be collinear with the pre-relight path.

-----
ABCD: Always Be Counting Down

The IIP shift to land logic is not univesally true - look at the SAOCOM-1A mission.

[Johnnyhinbos]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

[meekGee]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

Computers, man.  They're the future!

-----
ABCD: Always Be Counting Down

[whitelancer64]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

The RCS thrusters.

[meekGee]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

The RCS thrusters.

True, but I don't think they do much compared to what the fins are doing, and they have limited stored impulse. They're mostly for use in vacuum when disturbing forces are low.  (Hence the grid fins)


 Can calculate though...  Later today...

-----
ABCD: Always Be Counting Down

[Kabloona]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

The RCS thrusters.

Yes, if you watch Scott Manley's commentary video, he observes that towards the end of the landing burn, as the stage slows, the grid fins exert less roll torque on the stage, and the RCS thrusters manage to null the roll, and even over-correct to the point that you can see an RCS thruster firing to counter the over-correction.

[Kabloona]

[

The RCS thrusters.

True, but I don't think they do much compared to what the fins are doing, and they have limited stored impulse. They're mostly for use in vacuum when disturbing forces are low.  (Hence the grid fins)


 Can calculate though...  Later today...

-----
ABCD: Always Be Counting Down

See above post. Watch Scott Manley's video. You can see the RCS thrusters gaining control of the roll in the last few seconds as the stage slows and the grid fins lose their mojo. The thrusters even manage to over-correct, then fire one last burst in the opposite direction to null the over-correction.

See at about 4:30 into the video.

[Lars-J]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

The RCS thrusters.

Yes, if you watch Scott Manley's commentary video, he observes that towards the end of the landing burn, as the stage slows, the grid fins exert less roll torque on the stage, and the RCS thrusters manage to null the roll, and even over-correct to the point that you can see an RCS thruster firing to counter the over-correction.

I think Scott misinterpreted the last part... The RCS never completely cancelled out the rotation completely. The last switch in RCS thruster direction was to prevent the booster from falling over.

[rocketguy101]

For conventions sake: Would apply to the entire stack including booster return.

From the horse's mouth...https://www.spacex.com/sites/spacex/files/falcon_9_users_guide_rev_2.0.pdf

[meekGee]

Here's are some quick numbers for the legs.

The stage weighs 25 tons empty.  good approx. Is that the mass is at 1.7 m radius.  MR2 = 72.25

After leg deploy, 2 tons moved out about 2.5 m (based on c.g. of the legs being about 1/3 of the way out) so now

23 tons at 1.7 m and 2 tons at 4.2 m = 66.5+35.3 = 112.

So the ratio of moments of inertia is 112/75 = 1.5. 

That's all there is.  I ball-parked it at 2.0 previously.

So leg deployment reduces the rate of rotation by 1.5x or maybe 2.0x if I got the c.g. a little bit wrong.

I'll see what the RCS can do later on, but I want to watch the video again to see if they were on after leg deployment.

[Kabloona]

I'll see what the RCS can do later on, but I want to watch the video again to see if they were on after leg deployment.

Yes, they were. Again, watch the video above. You can see around 4:30 one of the RCS thrusters is on almost coninuously to counter the roll just before splashdown.

[Kabloona]

the engine was able to recover a wildly spinning ship .

Not that incorrect statement again

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

The RCS thrusters.

Yes, if you watch Scott Manley's commentary video, he observes that towards the end of the landing burn, as the stage slows, the grid fins exert less roll torque on the stage, and the RCS thrusters manage to null the roll, and even over-correct to the point that you can see an RCS thruster firing to counter the over-correction.

I think Scott misinterpreted the last part... The RCS never completely cancelled out the rotation completely. The last switch in RCS thruster direction was to prevent the booster from falling over.

That's a good point about trying to stay upright vs. countering the roll, but it does look to me like the RCS nulled the rate pretty close to zero by touchdown.

[Rocket Science]

For conventions sake: Would apply to the entire stack including booster return.

From the horse's mouth...https://www.spacex.com/sites/spacex/files/falcon_9_users_guide_rev_2.0.pdf

Standard coordinate conventions...

[Kabloona]

This also begs the question of how F9 managed not to run out of GN2 for the RCS thrusters. From the video it's evident that the thrusters are only intermittently trying to counter the roll, not thrusting continuously, so there must be some control logic that is designed to conserve GN2 even in the case of a hard-over grid fin.

Maybe someone thought through failure scenarios enough to realize that, in the case of a stuck grid fin causing roll, the grid fins would eventually lose aerodynamic control authority just before landing, and the RCS thrusters would be able to recover as long as they had enough GN2 left.

In which case the priority becomes conserving GN2 on the way down vs. blowing it all out trying futilely to counter the grid fins.

[Rocket Science]

This also begs the question of how F9 managed not to run out of GN2 for the RCS thrusters. From the video it's evident that the thrusters are not "full on" all the time trying to counter the roll, so there must be some control logic that is designed to conserve GN2 even in the case of a hard-over grid fin.

Maybe someone thought through failure scenarios enough to realize that, in the case of a stuck grid fin causing roll, the grid fins would eventually lose aerodynamic control authority just before landing, and the RCS thrusters would be able to recover as long as they had enough GN2 left.

In which case the priority becomes conserving GN2 on the way down vs. blowing it all out trying futilely to counter the grid fins.

Throw in the conservation of angular momentum with leg extension and you're golden...

[whitelancer64]

This also begs the question of how F9 managed not to run out of GN2 for the RCS thrusters. From the video it's evident that the thrusters are only intermittently trying to counter the roll, not thrusting continuously, so there must be some control logic that is designed to conserve GN2 even in the case of a hard-over grid fin.

Maybe someone thought through failure scenarios enough to realize that, in the case of a stuck grid fin causing roll, the grid fins would eventually lose aerodynamic control authority just before landing, and the RCS thrusters would be able to recover as long as they had enough GN2 left.

In which case the priority becomes conserving GN2 on the way down vs. blowing it all out trying futilely to counter the grid fins.

Grid fins have superior control authority for the vast majority of the descent through the atmosphere. The RCS thrusters would only be useful when the grid fins can't control the booster, which would be prior to reentry and in the last couple seconds before touchdown.

Would be simple to program the booster to not bother to use RCS at all until it is useful, rather than for a failure.

[Rocket Science]

Don't underestimate the torque being imparted in the final moments during the rotating leg extension through the CoG...

[Kabloona]

This also begs the question of how F9 managed not to run out of GN2 for the RCS thrusters. From the video it's evident that the thrusters are only intermittently trying to counter the roll, not thrusting continuously, so there must be some control logic that is designed to conserve GN2 even in the case of a hard-over grid fin.

Maybe someone thought through failure scenarios enough to realize that, in the case of a stuck grid fin causing roll, the grid fins would eventually lose aerodynamic control authority just before landing, and the RCS thrusters would be able to recover as long as they had enough GN2 left.

In which case the priority becomes conserving GN2 on the way down vs. blowing it all out trying futilely to counter the grid fins.

Grid fins have superior control authority for the vast majority of the descent through the atmosphere. The RCS thrusters would only be useful when the grid fins can't control the booster, which would be prior to reentry and in the last couple seconds before touchdown.

Would be simple to program the booster to not bother to use RCS at all until it is useful, rather than for a failure.

It does appear in the video that the RCS is inhibited until just before the landing burn. Then the RCS thrusters come on intermittently trying to null the roll. But even then, they're not on continuously. The one visible in the camera field of view operates only intermittently, until just before splashdown when it fires for several seconds in a last, apparently successful, attempt to stop the roll.

So yes, it does make sense to inhibit RCS most of the way down, but even when the RCS is re-enabled and tried to stop the roll, the thrusters aren't running continuously.

Which intermittence is either control logic whose purpose escapes me, or possible conservation logic to preserve GN2.

[Jim]

So this leaves the engine. 

Wrong, there is the RCS.  Which does roll control on single engine burns above the atmosphere and during slow speeds in the atmosphere.

[Pete]

What I can’t conceive of is how there was any kind of control authority with the landing burn. The stage is rotating - the center engine is gimbaling to balance the booster during descent, but the rotation the changes the relative angle of thrust so the control algorithm has to gimbal in a new angle, but the booster rotates, so it has to gimbal again to correct, but the booster rotates... you get the idea. This looks like it would quickly lead to a runaway instability situation as the mechanical actuators fall behind the control inputs. How they kept that thing under control is pretty impressive.

Obviously, the reaction time for the engine gimbal is quicker than the rate of change imparted by that spin.
Which, yes, means that is one ridiculously nimble engine!
I expect they did this to make the propulsive landing possible in the first place, you do not want a large actuator lag for gimbal control on top of all the other control delays in the last few milliseconds of a landing.

Does anyone here have actual info on the max gimbal speed/rate of the center engine? For example, from first command to position achieved, how long for the engine to gimbal 4 degrees?
For that matter, just what is the maximum deflection achievable on that engine?