Oh, a sticky valve and then shaky on the stick. My ( Armchair Rocket Scientist ! ) guess is heat cycling. That end of the rocket really goes from zero to hero at launch and then cools dramatically at throttle off, then up again on boostback, off again and then on again for terminal descent.
I think they should land it into a large mound of marshmallows. Success or not you'd have a great party afterwards.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
A much higher resolution video of the most recent failed barge landing has been posted on youtube.
Aside from showing much more detail, this one continues on long enough to show the toppling followed by fire that ensued.
Somewhere else, I saw a news story asserting that the SpaceX President (not Elon) said their next landing attempt would be on land. She specifically observed that would be quite a lot easier, and that safety was not a big problem since there was far less fuel aboard than at launch, and that few hesitations would apply to a range safety destruct if needed. (I doubt the range safety people appreciated her implication that they might hesitate on the launch end of needed).
What is interesting in this video is that there seems to be horizontal thrusters at the top of the 1st stage that attempt to maintain a vertical orientation.
The UK tabloid the Daily Mail has posted on their web site a video they describe as leaked and not verified that appears to be a pretty high resolution video shot from the surface of the barge of the last very little bit of the descent and then the topple.
It appears that this youtube video is the one their page repurposes as proprietary content.
This view gives a bit more detail on the thrusters robl commented on.
While the Daily Mail calls this unverified, it appears to be the real thing to me.
The UK tabloid the Daily Mail has posted on their web site a video they describe as leaked and not verified that appears to be a pretty high resolution video shot from the surface of the barge of the last very little bit of the descent and then the topple.
It appears that this youtube video is the one their page repurposes as proprietary content.
This view gives a bit more detail on the thrusters robl commented on.
While the Daily Mail calls this unverified, it appears to be the real thing to me.
Great video. It was so close to being successful. I had never noticed those thrusters until archae86 posted that first video. They must have been added for the "real" landings. They definitely don't show up on the prototype/test launches.
A news story mentioned SpaceX as officially posting a longer color corrected higher resolution video than the first one posted on social media, but on review it appears to be the same one to which I posted a link as "higher resolution, including topple and fire" a few posts back.
In watching it again my attention was drawn to how very late in the process the landing legs deployed. As the vehicle enters the frame they are fully stowed, start moving at about 0:01, are about straight out at 0:06 don't stop moving until about 3 seconds before landing/impact.
This seems bold. If the deployment induces any asymmetric forces at all this is very late in the game to be correcting for them.
A news story mentioned SpaceX as officially posting a longer color corrected higher resolution video than the first one posted on social media, but on review it appears to be the same one to which I posted a link as "higher resolution, including topple and fire" a few posts back.
In watching it again my attention was drawn to how very late in the process the landing legs deployed. As the vehicle enters the frame they are fully stowed, start moving at about 0:01, are about straight out at 0:06 don't stop moving until about 3 seconds before landing/impact.
This seems bold. If the deployment induces any asymmetric forces at all this is very late in the game to be correcting for them.
My guess is that deploying the "legs" earlier would further complicate the steering to pad so they are deployed at the last moment. Also this last minute deployment could be necessitated by landing on a floating platform. A land based landing might allow for more wiggle room and therefore earlier deployment. I still think they have more tweaking in the rate of descent. It just seems excessively high to me. Their prototypes were slow and graceful but then again only from ~1000 feet or so.
Another Elon Musk tweet now 22 hours old said "Cause of hard rocket landing confirmed as due to slower than expected throttle valve response. Next attempt in 2 months."
His next tweek posted a link to this wonderful video of the first ground pad up-down flight test of a Falcon 9R by way of showing how it should work if the throttling were working as intended.
By the way, on our discussion of late landing leg deployment, this old test video shows one dramatic reason not to deploy them too early--they get quite a toasting from the rocket plume.
Also, as I think robl pointed out earlier, in this test the side thrusters visibly active in the actual hard landing video don't show at all, nor do the "waffle" aerodynamic surfaces.
Yeah, I've just worked out what 'biprop valve' means : bipropellant valve. If you are going to throttle a rocket you have to control the flows of both reactants simultaneously. Certainly it looks like they came in hot compared to other landings, which is explained by a delay in opening to full thrust. That gave trouble in all dimensions. The wee thrusters are there to point the big one. You will note their use after stage separation to re-orient for boost-back.
Of interest therefore is also the SuperDraco rockets for the Dragon module to be used in coming down from orbit. The intention is to have pilot manual control through the full range of thrust ability when desired. Redundancy in rocket number is the safety feature there.
Certainly the success of 1st stage landings is key to the long term cost saving. Spacex has a good record for the present in terms of payload to orbit, for a given price. They now have 20+ successes in that. They are obviously learning and pretty transparent AFAIK for a commercial venture. Let's not forget that the Dragon modules have performed well, so the whole 'package' is coming together.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
Some days ago someone posted a picture on reddit which is said to be the target barge tied up in port after the most recent landing attempt but before much repair or cleanup.
The image is biggish, so I post a link to it instead of coding to display inline with this post.
It appears that the barge is tough enough. (This is after the most recent attempt--as the new name is visible, which was added during refurb from the previous crash).
Regarding holding the rocket in place after landing, someone posting as Elon Musk asserted that they intended to weld steel shoes over the legs. Unless some fancy robotics is involved, that sounds like waiting long enough for attending staff to reach the barge after landing, so more a means to make secure for higher sea state during transit to port than for immediate post-landing stabilization.
Oh, a sticky valve and then
)
Oh, a sticky valve and then shaky on the stick. My ( Armchair Rocket Scientist ! ) guess is heat cycling. That end of the rocket really goes from zero to hero at launch and then cools dramatically at throttle off, then up again on boostback, off again and then on again for terminal descent.
I think they should land it into a large mound of marshmallows. Success or not you'd have a great party afterwards.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
A much higher resolution
)
A much higher resolution video of the most recent failed barge landing has been posted on youtube.
Aside from showing much more detail, this one continues on long enough to show the toppling followed by fire that ensued.
Somewhere else, I saw a news story asserting that the SpaceX President (not Elon) said their next landing attempt would be on land. She specifically observed that would be quite a lot easier, and that safety was not a big problem since there was far less fuel aboard than at launch, and that few hesitations would apply to a range safety destruct if needed. (I doubt the range safety people appreciated her implication that they might hesitate on the launch end of needed).
What is interesting in this
)
What is interesting in this video is that there seems to be horizontal thrusters at the top of the 1st stage that attempt to maintain a vertical orientation.
The UK tabloid the Daily Mail
)
The UK tabloid the Daily Mail has posted on their web site a video they describe as leaked and not verified that appears to be a pretty high resolution video shot from the surface of the barge of the last very little bit of the descent and then the topple.
It appears that this youtube video is the one their page repurposes as proprietary content.
This view gives a bit more detail on the thrusters robl commented on.
While the Daily Mail calls this unverified, it appears to be the real thing to me.
RE: The UK tabloid the
)
Great video. It was so close to being successful. I had never noticed those thrusters until archae86 posted that first video. They must have been added for the "real" landings. They definitely don't show up on the prototype/test launches.
A news story mentioned SpaceX
)
A news story mentioned SpaceX as officially posting a longer color corrected higher resolution video than the first one posted on social media, but on review it appears to be the same one to which I posted a link as "higher resolution, including topple and fire" a few posts back.
In watching it again my attention was drawn to how very late in the process the landing legs deployed. As the vehicle enters the frame they are fully stowed, start moving at about 0:01, are about straight out at 0:06 don't stop moving until about 3 seconds before landing/impact.
This seems bold. If the deployment induces any asymmetric forces at all this is very late in the game to be correcting for them.
RE: A news story mentioned
)
My guess is that deploying the "legs" earlier would further complicate the steering to pad so they are deployed at the last moment. Also this last minute deployment could be necessitated by landing on a floating platform. A land based landing might allow for more wiggle room and therefore earlier deployment. I still think they have more tweaking in the rate of descent. It just seems excessively high to me. Their prototypes were slow and graceful but then again only from ~1000 feet or so.
Another Elon Musk tweet now
)
Another Elon Musk tweet now 22 hours old said "Cause of hard rocket landing confirmed as due to slower than expected throttle valve response. Next attempt in 2 months."
His next tweek posted a link to this wonderful video of the first ground pad up-down flight test of a Falcon 9R by way of showing how it should work if the throttling were working as intended.
By the way, on our discussion of late landing leg deployment, this old test video shows one dramatic reason not to deploy them too early--they get quite a toasting from the rocket plume.
Also, as I think robl pointed out earlier, in this test the side thrusters visibly active in the actual hard landing video don't show at all, nor do the "waffle" aerodynamic surfaces.
Yeah, I've just worked out
)
Yeah, I've just worked out what 'biprop valve' means : bipropellant valve. If you are going to throttle a rocket you have to control the flows of both reactants simultaneously. Certainly it looks like they came in hot compared to other landings, which is explained by a delay in opening to full thrust. That gave trouble in all dimensions. The wee thrusters are there to point the big one. You will note their use after stage separation to re-orient for boost-back.
Of interest therefore is also the SuperDraco rockets for the Dragon module to be used in coming down from orbit. The intention is to have pilot manual control through the full range of thrust ability when desired. Redundancy in rocket number is the safety feature there.
Certainly the success of 1st stage landings is key to the long term cost saving. Spacex has a good record for the present in terms of payload to orbit, for a given price. They now have 20+ successes in that. They are obviously learning and pretty transparent AFAIK for a commercial venture. Let's not forget that the Dragon modules have performed well, so the whole 'package' is coming together.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
Some days ago someone posted
)
Some days ago someone posted a picture on reddit which is said to be the target barge tied up in port after the most recent landing attempt but before much repair or cleanup.
The image is biggish, so I post a link to it instead of coding to display inline with this post.
image of damaged landing barge posted to reddit
It appears that the barge is tough enough. (This is after the most recent attempt--as the new name is visible, which was added during refurb from the previous crash).
Regarding holding the rocket in place after landing, someone posting as Elon Musk asserted that they intended to weld steel shoes over the legs. Unless some fancy robotics is involved, that sounds like waiting long enough for attending staff to reach the barge after landing, so more a means to make secure for higher sea state during transit to port than for immediate post-landing stabilization.