 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
 |
Posted: |
Dec 10, 2020 - 10:09 AM
|
|
|
|
By: |
Grecchus
(Member)
|
According to Scott Manley's breakdown, the first engine cut off was probably planned for. The two remaining engines were then gimballed to take the rocket up in a crab to offset the necessity to gain slant range on the landing pad target so that when SN8 was in the belly-flop attitude it could sky-dive along the hypotenuse of a right triangle, or summit like that.
As for the attempted landing, again according to the Stateside observers, Mr Musk said there was a problem with a header tank - loss of pressure mebbe? They probably need some kind of heavy internal ullage push system to ensure the propellent gets to the engine bell as exhaust, especially with all that rotation around the CG of such a huge, rapidly moving structure. Could be the g-forces in there were whopping! A while back, before the F9 landings became commonplace and the problem was unsolved, I had the idea of rotating the whole rocket along its vertical, longitudinal axis like a skater with a roller skate-like undercarriage to absorb the rotational motion upon landing. After doing some quick reading, one of the many problems associated with such a technique was the g-forces generated would push the fuel centrifugally to the sides of the tank walls and hold it there so it can't move. That body-flip motion could have sent the fuel sloshing backwards and forwards with gusto so it couldn't go to where it was needed as and when. The Apollo Saturn V had quite severe problems with POGO. It took time to figure out how to damp down the unhelpful sloshing motion of the propellant. My guess is Starship might need a very complex baffle system with piston-like mechanical strength needed to keep the fuel flowing just for that flip.
Whether the last remaining engine burned out with the green copper glow due to overpressurisation or perhaps because of a programmed response, as I initially suggested, needs clarification.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: |
Dec 10, 2020 - 10:24 AM
|
|
|
|
By: |
Grecchus
(Member)
|
Blabber away, Chris. I've done it a darn sight more than you have!
Sol - that right-side-up routine is, I agree, something of a moment of terror, however, Mars' atmosphere is so much thinner than the Earth's. Not much they can do about that. I mean, they're the experts so that must be the optimal solution for a powered landing on the surface. In fact it's probably easier on Earth because of the thicker air. On Mars, the ship would be entering from outside the atmosphere. Quite what that means is harder to contemplate - I think they'd need those powerful pulse nitrogen cold thrusters to help rotate the airframe to get in step for the touchdown. Lots of questions remaining. Like, when landing on Mars or the Moon, a sighting of the most appropriate spot to set down requires some hover time, as with the Apollo LEMs. Also, the SpaceX moonship is different to the one we just witnessed. It doesn't have to negotiate with an atmosphere. Which of these two is the most difficult to perform, you tell me? My guess is the atmospheric entry overall due to all the heat beating the thing has to absorb. Landing on the Moon uses more fuel for the simple reason there's no atmosphere to act like a huge brake-pad.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: |
Dec 19, 2020 - 9:52 AM
|
|
|
|
By: |
Grecchus
(Member)
|
Today SpaceX launched NROL-108, a confidential payload atop a Falcon-9. They boosted the first stage back to base, but we did not see the deployment of the cargo package.
They've landed 70 times either on the off-shore sea-going platforms or on dry land. This is the 5th time that first stage has been used. The rates are interesting to watch. For some reason they did not keep them on-screen during the final few seconds all the way to touchdown. I last registered that the first stage was closing the ground at 636KM/H with 1.6KM to go at an elapsed flight time of 8 minutes. The booster touches down at about 08:18. So, at 8 minutes mission elapsed time the booster is closing with the ground at about 176.7 meters per second, which is about 395mph. It has to lose almost 400mph of radial velocity in 18 seconds! A simple linear breakdown (which is, strictly speaking, not the case) shows the stage has to slow down by about 22mph of velocity every second from 8 minutes elapsed time, or else it will impact with the ground.
That puts the rocket at an altitude of about 5249 feet at that point in time, with an instantaneous downwards velocity of about 579.6ft per second at 8 minutes into the flight.
At 6:31, the stage is closing with earth at 2956mph at an altitude of about 190,289 feet when it starts its entry burn. By the time the entry burn shuts down at 6:59, the first stage has slowed to 1400mph at an altitude of 100,394 feet. The entry burn shaves off something like 1556mph of speed in the 26 to 28 seconds of the burn covering about 89,895 ft! Even watching it in real time with your own eyes doesn't register the true scale of what is going on.
https://www.youtube.com/watch?v=9OeVwaFBkfE
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
 |
Posted: |
Dec 22, 2020 - 1:10 PM
|
|
|
|
By: |
Grecchus
(Member)
|
I think I figured out how SpaceX bring the F-9 first stage down to stick the landing.
I've seen the views out of the top of the 1st stage so many times without registering anything of note, it just stayed unregistered. Then on this last ride down of the first stage I noticed something I'd not seen before. It concerns that tripod structure you can see where each of the 3 'legs' join to a point and on top of that is what looks like a probe that extends outwards. All in all, it looks a bit like the docking probe on the Apollo CSM command module.
While the stage was flipping over to present it's rear end engines forward, you can see that probe device is vibrating, like it's absorbing transmitted forces along the length of the rocket by damping them out.
While I was reading Jeffrey Quill's book Spitfire, which he test piloted, one of the many problems the design team encountered was unwanted tail vibration, which was very dangerous to flight. They solved the problem by putting a probe with a horn on the end onto the vertical stabiliser, pointing forward. The unwanted forces were channelled through the probe where the rubber-like horn would vibrate away to it's heart's content, dissipating all those unwanted, structurally devisive forces.
Well, it looks like they've done the same sort of thing with the Falcon-9 first stage. That horn is flexing about in response to the arching moments the stack is performing while rotating around its centre of gravity. The horn flexes according to the amount of force it is experiencing as it rotates around. These forces are then channelled down each of those tripod arms/legs where sensors can detect how the end of the rocket is flexing about. It's a three-way split. Any one of those arm sensors will register the proportional forces generated, so the stage computer can figure out which way and how much it is toppling over. These inputs are then processed and the appropriate signals are sent to the central rocket engine so it can counter-gimbal to bring it back to a centred attitude where it is vertical to the ground. Depending on how the stage is falling over to a particular side, all 3 main sensors can determine where the rocket is going over and rectify it immediately. It is literally triangulating any vertical off-centre motions the stage is making in real time. If the arms are labelled 1, 2 and 3, then either the toppling motion is towards any of 1, 2, 3, or pairs such as, 1 & 2, 2 & 3 or 1 & 3. It basically works along the same lines as a force-feedback joystick. Instead of someone gripping the stick and moving it any which way, it is g-like forces the system reacts to.
This also explains why the Falcon-9 has such long and straight sides - why it is so pencil-like. The cross-sectional area of the rocket has to be equal across its entire length so that it is highly evenly balanced along its entire length. Extreme arching motion at the top is then translated to a proportional gimbal management of the rocket engine at the exact opposite end. That has to be something close to how it works in principle. Unlike the Spitfire, where the randomly vibrating motions did not matter just so long as it absorbed all the unwanted energy, the Falcon-9 detects precise motions of the mass balance horn and uses them to help guide it down to a vertical landing. The mass balance is being subjected to inertial forces during the entire flight. Maybe this is already known, but I'm damned if I've come across it anywhere. If say, Scott Manley has already outlined something like this because I can remember him discussing that tripod-like arrangement in one of his previous podcasts, then it can rest. Alternatively, SpaceX might have relayed the information themselves.
If I'm wrong it matters not a jot. At least I keep mulling things over. Anyone have anything to add?
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: |
Dec 24, 2020 - 1:12 PM
|
|
|
|
By: |
Grecchus
(Member)
|
More thoughts on observations from what we can see looking forward/up from the top/front of the 1st stage. When the 2nd stage separates during staging, the 2nd stage engine bell pretty much fills the open compartmental space we're seeing as events unfold. The tripod probe actually appears to be well inside the 2nd stage engine bell while the two stages are joined throughout the ride up to MECO. It may serve two purposes - 1) as a damper for forces transmitted throughout the joined stack following liftoff as already suggested and 2) as a sort of ramrod that prevents any unwanted contact of the engine bell with the sides at the top of the 1st stage during separation. Any contact scraping between the 'tripod probe' and the engine bell can't be a good thing but may well be the lesser of the evils should unwanted contact actually occur during stage separation. I don't think it is a literal ramroad in the sense it spring releases and pushes the 2nd stage away because that would seem to be taking risks damaging the combustion chamber innards on the 2nd stage as it moves away. Also, the internal resonances generated within the tube forming the body of the rocket may require damping at the level where the engine bell on the 2nd stage is located. They can't afford the coolant tubing in the 2nd stage engine bell to be damaged prior to use. The longer length of the tube while the stage 1 and stage 2 pieces of the stack are physically joined may well require that probe to absorb vibrational forces where it is located. I seem to remember Hurley and Behnken described their ascent as bucking like a bronco, or similar.
There are several other interesting observable features of the open top end of the 1st stage. There are a pair of loose, equal length free floating cables (with a third strand strangely doing something completely different to the other two) you can see flopping around while in free fall, but which are pulled back when the engines under thrust are causing the stack to endure constant acceleration. While thrusting under rocket power, you can see those gossamer strands pulled back but they are swaying in opposite directions rhythmically to one another. It's like some kind of corkscrew force is rapidly running back and forth inside the tube under acceleration. Like the rifling you can see in the James Bond gun barrel opening animation. There are two pairs of these free floating ant-like antennae flopping around in there. I'm really quite curious as to what they are all about. They might just be there for simple visual confirmation of the stage being under acceleration or not.
Another thing is the probe at the end of the tripod structure was actually slowly retracted, so that it was not as far forward as it was when the 2nd stage separated from the 1st stage. This was a puzzle at first, but it makes sense when you realise the entire rocket length was shortened during staging and, therefore, the resonant internal forces would have a different vibrational frequency. It may still have some degree of freedom to displace off-center even when inside the tube casing enclosing it.
In theory, there's not much wrong with the mass balance idea. On further reflection, though, piezo-electronic accelerometer sensors are probably weighing up how the balance of forces inside the rocket's tube-like structure are being distributed in real time, so that's probably how the internal computing machinery gets its inputs. Looking at the rocket's cross section, how many sensors you'd need to be equally spaced around its circumference is academic.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: |
Feb 5, 2021 - 7:45 PM
|
|
|
|
By: |
Grecchus
(Member)
|
Some milestones for the last Falcon-9 Starlink launch. Note that at the re-entry burn timestamped at 6:26 into the mission, the first stage booster is clocked coming down to Earth at 5000mph. That is almost unbelievable.
https://www.youtube.com/watch?v=fe6HBw1y6bA
Max Q ------- 01:21 - velocity: 1805km/h, 0501m/s, 1122mph, 1645ft/s, altitude: 14.5km, 9 miles, 047,572ft
Check ------- 02:00 - velocity: 4225km/h, 1174m/s, 2626mph, 3852ft/s, altitude: 33.5km, 20.8 miles, 109,908ft
MECO ------- 02:37 - velocity: 7942km/h, 2206m/s, 4935mph, 7238ft/s, altitude: 63.4km, 39.4 miles, 208,005ft
Max Alt ------ 04:27 - velocity: 7100km/h, 1972m/s, 4411mph, 6470ft/s, altitude: 115km, 71.5 miles, 377,297ft
Re Ent ------ 06:26 - velocity: 8047km/h, 2235m/s, 5000mph, 7333ft/s, altitude: 54.4km, 33.8 miles, 178,478ft
Sht Dwn ---- 06:48 - velocity: 5755km/h, 1599m/s, 3577mph, 5246ft/s, altitude: 35.1km, 21.8 miles, 115,157ft
Pre Land ---- 08:02 - velocity: 0894km/h, 0248m/s, 0556mph, 0813ft/s, altitude: 2.2km, 1.37 miles, 7218ft
Land Brn ---- 08:08 - velocity: 0639km/h, 177.5m/s, 0397mph, 0582ft/s, altitude: 1.1km, 0.68 miles, 3609ft
Land Brn ---- 08:08 - velocity: 0615km/h, 171.0m/s, 0382mph, 0561ft/s, altitude: 1.0km, 0.62 miles, 3281ft
Land Brn ---- 08:12 - velocity: 0441km/h, 122.5m/s, 0274mph, 0402ft/s, altitude: 0.5km, 0.31 miles, 1640ft
Land Brn ---- 08:14 - velocity: 0374km/h, 104.0m/s, 0233mph, 0341ft/s, altitude: 0.4km, 0.25 miles, 1312ft
Land Brn ---- 08:16 - velocity: 0303km/h, 084.0m/s, 0188mph, 0276ft/s, altitude: 0.2km, 0.12 miles, 0656ft
Landing between 08:28/08:29. From 08:16 loses about 15.67mph each second linearly interpolated.
After the re-entry 3-engine firing shutdown, it appears atmospheric drag has slowed the stage from 3577mph to 397mph. Incredible.
Unlike the NROL-108 flightpath, which was a boostback to base, this was a sub-orbital parabolic flight from the launch pad to the drone ship stationed in the Atlantic. The parameters for each mission flight profile are more varied than I'd have thought at first glance.
Edit: there is actually one more crucial item of information that could be put up on the screen to satisfy the most ardent of fans. A 'G meter' would be very handy, especially when stage separation takes place, as well as the braking effects that occur when the 1st stage activates 3, then 1 engine for landing. Also, the acceleration rate would be most convenient to see, but the screen could be quite clogged if this is put in for both stages. Worth thinking about, though.
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
