NASASpaceFlight.com Forum

SpaceX Vehicles and Missions => SpaceX Reusability => Topic started by: meekGee on 01/25/2017 02:04 pm

Title: BFR landing cradle discussion and updates
Post by: meekGee on 01/25/2017 02:04 pm
In the grand reveal of a few months ago, Musk showed the booster landing in a  cradle, co-located with the launch mount.

Some have called this a fantasy. Some think it's feasible and quite clever.

This thread's for you.

Speculation welcome.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/25/2017 02:08 pm
 To start out, Masten Space Systems had this idea before SpaceX did.

They've recently even developed a vehicle which will be used to develop this technology:
http://masten.aero/vehicles-2/xaerob/
Title: Re: BFR landing cradle discussion and updates
Post by: John Alan on 01/25/2017 02:12 pm
I speculate SpaceX will come up with an add on lower thruster package they can add to RTLS bound F9 S1's...
The will paint 4 leg pad outlines on the pad and see how well they can hit them...

Assuming that goes well...

I then speculate SpaceX will build a mobile rig they can wheel out to the RTLS pad and build a few lightweight F9 S1's with fin like structures to use in continuing research using RTLS bound stages...

This may all be BS hogwash... but a guy can speculate...  ;)
Title: Re: BFR landing cradle discussion and updates
Post by: woods170 on 01/25/2017 02:13 pm
Let me put it this way. Outright dismissal of the whole concept, by certain members here, without having insight into the technical details of said concept, is premature.

Outright statements, by certain members here, that the whole concept WILL work, without having insight into the technical details of said concept, is equally premature.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/25/2017 02:14 pm
To start out, Masten Space Systems had this idea before SpaceX did.
Noted.  Kistler was showing it too, and I've seen a helicopter pilot load up his MD500 onto a trailer by landing on it.

A cradle is just a vertical runway. Think of an aircraft carrier. You have to nail it just so, or crash.

The concept is not new. The execution in rocketry, and scale - that's new.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 01/25/2017 02:15 pm
The issue is not a landing cradle.  The issue is a landing cradle that is also a launch mount.
Title: Re: BFR landing cradle discussion and updates
Post by: woods170 on 01/25/2017 02:17 pm
The issue is not a landing cradle.  The issue is a landing cradle that is also a launch mount.
No doubt Jim. But issues only exist to be transformed into non-issues by providing solutions.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/25/2017 02:18 pm

"Hovering is for humans"

If flight control computers could choose a bumper sticker, that'd be it.

When you hover you just give winds and other unknowns more time to influence​ your position.

The best way to increase X-y accuracy is to hit the pad at higher velocity.  Less time for unknown forces to act and unlike a human, a computer doesn't need to hover nearby to estimate the IIP.

The only problem is that a you try to hit the pad at higher decelerations, you increase your Z uncertainty.

The remedy for this is increase Z travel on capture.  Which means a heavier mechanism.

This is the gain of the cradle - you don't have to carry the landing gear with you, you leave it on the ground.  Make it as tall and heavy as you want, to absorb as much residual vertical velocity as you need.

My bet is that we'll see more slam, and less hover, but that because of the increased physical size, it'll look about the same.


Basically, it's a smaller step then showing that they can land an F9.

And they will be able to tune F9 landings further, and also transfer that knowledge to BFR landing modeling.

If they say today that it can be done, they are certainly in a position of authority to do so.

Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/25/2017 02:21 pm
The issue is not a landing cradle.  The issue is a landing cradle that is also a launch mount.
Isnt that what legs are on a vehicle like Grasshopper or the NGLLC type vehicles?
Title: Re: BFR landing cradle discussion and updates
Post by: Basto on 01/25/2017 02:24 pm
A common concern I am seeing repeated about the cradle mount has been lining up the booster in heavy cross winds. Personally I think that this would be a non issue as rockets don't usually launch with heavy wind conditions.

It is often an issue when landing on an ASDS because the landing site is out at sea.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/25/2017 02:27 pm
The issue is not a landing cradle.  The issue is a landing cradle that is also a launch mount.
So if landing in a cradle, in and of itself, is not an issue, then "fantasy" has nothing to do with it.

You have a simple trade off between two feasible solutions:

Co-locate the two, and risk damaging the launch pad if the landing fails, or have two separate facilities and some kind of transporter.

But since a failed landing will certainly damage the landing mount, and this is a reusable vehicle, what value is the launch pad without the landing pad?

If you want redundancy, build two launch/land facilities, so each one will perform both functions, and if one is damaged, you don't care that the other was too.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/25/2017 02:36 pm
I always kind of thought the landing cradle could be co-located with the launch mount, and the launch mount attaches to the vehicle after landing. It doesn't have to literally be the same mechanism, only has to be concentric.
Title: Re: BFR landing cradle discussion and updates
Post by: Jimmy Murdok on 01/25/2017 02:37 pm
A common concern I am seeing repeated about the cradle mount has been lining up the booster in heavy cross winds. Personally I think that this would be a non issue as rockets don't usually launch with heavy wind conditions.

It is often an issue when landing on an ASDS because the landing site is out at sea.

You need to add gas (or more powerful) thrusters in the bottom to ensure good alignment. I would forget about this concept on a small ASDS, it's been shown on ground because you require stability.
http://forum.nasaspaceflight.com/index.php?topic=41249.msg1592281#msg1592281
Title: Re: BFR landing cradle discussion and updates
Post by: Basto on 01/25/2017 02:55 pm
A common concern I am seeing repeated about the cradle mount has been lining up the booster in heavy cross winds. Personally I think that this would be a non issue as rockets don't usually launch with heavy wind conditions.

It is often an issue when landing on an ASDS because the landing site is out at sea.

You need to add gas (or more powerful) thrusters in the bottom to ensure good alignment. I would forget about this concept on a small ASDS, it's been shown on ground because you require stability.
http://forum.nasaspaceflight.com/index.php?topic=41249.msg1592281#msg1592281

I never claimed otherwise. Just pointing out that the argument of cross winds affecting landing precision should be a non issue because you just wouldn't launch under conditions with heavy cross winds.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/25/2017 02:59 pm
A common concern I am seeing repeated about the cradle mount has been lining up the booster in heavy cross winds. Personally I think that this would be a non issue as rockets don't usually launch with heavy wind conditions.

It is often an issue when landing on an ASDS because the landing site is out at sea.

You need to add gas (or more powerful) thrusters in the bottom to ensure good alignment. I would forget about this concept on a small ASDS, it's been shown on ground because you require stability.
http://forum.nasaspaceflight.com/index.php?topic=41249.msg1592281#msg1592281
Definitely not on an ASDS - the barge location is too unpredictable.

I predict either a test site in New Mexico, or a whole new landing pad for this purpose.

If, that is, F9 (being smaller) can achieve the required level of accuracy.

How accurate must it be?

Imagine that there are guide slideways, and the final cone of acceptance has a half angle of about 15 degrees.

Assume the slam is set for 1g.

If the rocket was able to land within 1 m of ground zero, it would only engage the slides about 4 m above ground, at a vertical speed of 9 m/s, and this will move it at a horizontal speed of about 2 m/s.

Not so crazy.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/25/2017 04:28 pm
RB - I'm pretty sure there will be a mechanical guidance  structure, a set of rails or what have you, that will take care of final positioning.

There is no way the rocket will settle into hold-downs all on its own.

There will be some features on the rocket, in lieu of legs, that will interface with this centering structure, and as someone observed, they should be located at the height of the empty booster's CG.
Title: Re: BFR landing cradle discussion and updates
Post by: NotOnImpact on 01/25/2017 06:13 pm
RB - I'm pretty sure there will be a mechanical guidance  structure, a set of rails or what have you, that will take care of final positioning.

There is no way the rocket will settle into hold-downs all on its own.

There will be some features on the rocket, in lieu of legs, that will interface with this centering structure, and as someone observed, they should be located at the height of the empty booster's CG.

Musk stated in the Mars video that one of the purposes of the three fins on the 1st stage was to provide that last little bit of alignment on the pad for the landing.
Title: Re: BFR landing cradle discussion and updates
Post by: CeeJayDugan on 01/25/2017 07:41 pm
Given the amount of exhaust from all those engines maybe the landing cradle to would have to be hinged so it could fold up against the tower for launch and then lower down at landing? Or would there be a practical way of protecting it? Just mulling that over.
Title: Re: BFR landing cradle discussion and updates
Post by: Jdeshetler on 01/25/2017 07:49 pm
Drawn on a paper napkin (Stone Age MS Paint) during my lunch break. 

Pure Speculation.

The Cradle Ring and 8 hydro-shafts would need to be massive and blast proof.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 01/25/2017 08:54 pm
Drawn on a paper napkin (Stone Age MS Paint) during my lunch break. 

Pure Speculation.

The Cradle Ring and 8 hydro-shafts would need to be massive and blast proof.

I think you are overthinking it. With base thrusters and canted walls to guide the three base fins into the cradle, I don't see why a moving cradle is necessary.
Title: Re: BFR landing cradle discussion and updates
Post by: matthewkantar on 01/25/2017 08:58 pm
Reminds me of this:
https://www.youtube.com/watch?v=M1DHYJwXwuQ

Matthew
Title: Re: BFR landing cradle discussion and updates
Post by: Lar on 01/25/2017 08:59 pm
Drawn on a paper napkin (Stone Age MS Paint) during my lunch break. 

Pure Speculation.

The Cradle Ring and 8 hydro-shafts would need to be massive and blast proof.

I assume you used 4 lobes for easy mirroring in paint but the real one would have 3 (or multiples) of everything to  match the fins/alignment tabs, right?
Title: Re: BFR landing cradle discussion and updates
Post by: vanoord on 01/25/2017 09:04 pm
Imagine that there are guide slideways, and the final cone of acceptance has a half angle of about 15 degrees.

Assume the slam is set for 1g.

If the rocket was able to land within 1 m of ground zero, it would only engage the slides about 4 m above ground, at a vertical speed of 9 m/s, and this will move it at a horizontal speed of about 2 m/s.

Not so crazy.

Aye.

Landing into something which acts as a funnel / guide, allowing a tolerance of say +/- 5m, is about the only way this is going to work.

Suspect it may have to be relatively tall, to deal with the lean of the stage if it's guided sideways to the maximum possible, but that could be designed out of accuracy were improved.

Matched thrusters top and bottom may enable some sort of horizontal translation in the final few metres.

Essentially, guiding the stage down to somewhere that it can rest until the launch clamps are re-engaged (and the legs retract) is the critical bit, ie dealing with any potential horizontal displacement - presumably caused by wind and targeting errors.

The fix is probably as much in software as it is in physical design - the experience with F9 will be vital.
Title: Re: BFR landing cradle discussion and updates
Post by: Req on 01/25/2017 09:20 pm
I picture something like a very tall king's crown with 3 triangles/points.  Possibly wider diameter at the top than the bottom.  Some kind of low friction material(and/or lubrication) along the leading edges and the fins.  Maybe the whole thing is 3 interlocking pieces that retract vertically/horizontally or even swing up and down like the TEL.

Edit:  Attached an mspaint image.  I'm afraid you'll have to use your imagination a bit, but this is a very very rough sketch of the idea shown directly from the side.  The red line indicates a seam between two of the segments, and you're seeing around half each of 2 of the 3 segments(third segment in the background omitted).  From the top it would just look like a circle.  In this example the diameter does not increase towards the top, but I wouldn't be surprised if that was a desirable trait.  Maybe not necessary with so much room to be off-nominal at the top already.  I'm picturing fairly thin material here - maybe an inch or two(or less) as viewed from the top, just enough to handle the landing environment and structural loads.  It likely wouldn't be solid/sheeted but it's easier to think of it and depict it that way.

I don't go into the mechanism to securely clamp it down once it's there, but I imagine you'd want that to be separate so you can retract these guides once it's secure.  Ideally these guides would get the rocket directly where it needs to be to use the same hold-downs that it uses for launch with little or no dancing cradle/mount equipment.

Edit again - This idea has a fairly slammy landing in mind.  If they plan on bringing it down nice and slow, then something like this is probably the completely wrong idea.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/25/2017 09:44 pm
You can imagine the launch cradle like a giant docking port. Put a huge Stewart platform on it (like the NDS/iLIDS docking port) to help catch the stage.

This Stewart platform can handle a 10 ton payload:
https://www.youtube.com/watch?v=AhNnjXeDt9k

 The ITS booster is 275 tons dry, so you're talking 30x greater.

Attached is another 10 ton Stewart platform. The amount of play needed in absolute terms is probably similar, but not in proportional terms. So you could probably gang up like 30 actuators each 6x stronger around the perimeter instead of the usual 6.

I also attached a picture of the NDS docking port which is based on the 6 degrees of freedom Stewart platform.
http://www.dynamic-concepts.com/projects/nds.html
Title: Re: BFR landing cradle discussion and updates
Post by: matthewkantar on 01/25/2017 10:08 pm
One thing to remember that will be helpful, the landing engine are 19 feet or so away from the tips of the bosses shown in the IAC presentation. This should make exhaust impingement more manageable.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/26/2017 02:19 am
I'll give it a try too.

A couple of things about the sketch:

- The size of the "Flippers" shown is arbitrary.  It depends on what is the 3-sigma(?) confidence in getting the booster within a certain deviation of the zero position.  I drew them way too large probably.

- The flippers are shown on the rocket since it's conceptually easier to understand, but they can (and probably would) be put on the cradle (and so the "cone" will be on the rocket.)

- The flippers are set at a wider angle than the receiving cone.  So during the mating, the rocket gets centered, and the flippers get compressed inward, gradually.

- The flipper suspension is a spring-shock, so it both centers the vehicle and dissipates lateral motion energy.  It's important not to transmit sideways shock, since the rocket is very weak in that direction.

- The flipper's anchor points straddle the CG of the empty stage, so they move it laterally with minimal tilting.

- Final capture happens after the rocket centers at the bottom of the cone, by a different mechanism at the bottom of the rocket, not shown.

Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/26/2017 04:05 am
And this is the "shocks on the ground" version.

The difference in angles is the same, to guarantee that first contact is made at the height of the C.G., and then the flippers center the stage and cushion the sideways forces.

The choice is not clear, and depends on the ability to control the clocking of the stage, even after the first "flipper" makes contact.  If clocking is guaranteed, then both sides of the mechanism can be only 3 planar members.

If clocking is not guaranteed, then the "cone" needs to really be a cone.  In that case, maybe it's easier to put the flippers on the rocket and the cone on the ground, like in the first picture.

In the picture below, the "hole" is larger, since it has to fit rocket+fins.   In the picture above, it had only to fit rocket plus retracted (flush to the body) flippers.
Title: Re: BFR landing cradle discussion and updates
Post by: drzerg on 01/26/2017 06:38 am
it is obvious that moving the cradle is easier then moving the rocket. in final moments any wind gust can ruin everything. and cradle is only need to hold 300 tons. for example some bridge cranes could move 750 ton things. after catching the booster it could be positioned/transported (if you have enough long rails you could catch it 300-400m off the pad)  to launch hold downs.
Title: Re: BFR landing cradle discussion and updates
Post by: Semmel on 01/26/2017 07:13 am
And this is the "shocks on the ground" version.

The difference in angles is the same, to guarantee that first contact is made at the height of the C.G., and then the flippers center the stage and cushion the sideways forces.

The choice is not clear, and depends on the ability to control the clocking of the stage, even after the first "flipper" makes contact.  If clocking is guaranteed, then both sides of the mechanism can be only 3 planar members.

If clocking is not guaranteed, then the "cone" needs to really be a cone.  In that case, maybe it's easier to put the flippers on the rocket and the cone on the ground, like in the first picture.

In the picture below, the "hole" is larger, since it has to fit rocket+fins.   In the picture above, it had only to fit rocket plus retracted (flush to the body) flippers.

I like this version better, meekGee. You can make the cone V-shaped that aligns the rockets rotation first and then hits the cone part where it slides into the middle.

It would certainly help if the rocket would be able to hover. Not for hovering as such, but for canceling the gravity of earth. The flight profile would be like breaking to 10m/s or so just before the cradle and then slide down with constant velocity. That might increase the burn time by 1 to 2 seconds but would limit the force when touching the cradle and gives more time to align the rocket.
Title: Re: BFR landing cradle discussion and updates
Post by: JamesH65 on 01/26/2017 11:06 am
it is obvious that moving the cradle is easier then moving the rocket. in final moments any wind gust can ruin everything. and cradle is only need to hold 300 tons. for example some bridge cranes could move 750 ton things. after catching the booster it could be positioned/transported (if you have enough long rails you could catch it 300-400m off the pad)  to launch hold downs.

Been pondering how much of an effect the wind can have. The ITS weighs quite a lot (cannot find figures, but 100t?), so therefor has a lot of momentum. It will take quite a gust to cause excessive movement.
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/26/2017 01:40 pm
it is obvious that moving the cradle is easier then moving the rocket. in final moments any wind gust can ruin everything. and cradle is only need to hold 300 tons. for example some bridge cranes could move 750 ton things. after catching the booster it could be positioned/transported (if you have enough long rails you could catch it 300-400m off the pad)  to launch hold downs.

Been pondering how much of an effect the wind can have. The ITS weighs quite a lot (cannot find figures, but 100t?), so therefor has a lot of momentum. It will take quite a gust to cause excessive movement.

Booster mass is 275 tonnes per the IAC presentation.

A calculation assuming a Cd of 1.0 and a wind speed of 10 m/s gives a wind force of 58 kN on the 12m x 80m booster. Handy (but simplistic) calculator here: http://www.engineeringtoolbox.com/wind-load-d_1775.html

If the terminal control thrusters produce 100 kN and have an adequately responsive control scheme they should have little trouble countering wind gusts in moderate weather.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/26/2017 01:56 pm
it is obvious that moving the cradle is easier then moving the rocket. in final moments any wind gust can ruin everything. and cradle is only need to hold 300 tons. for example some bridge cranes could move 750 ton things. after catching the booster it could be positioned/transported (if you have enough long rails you could catch it 300-400m off the pad)  to launch hold downs.

Been pondering how much of an effect the wind can have. The ITS weighs quite a lot (cannot find figures, but 100t?), so therefor has a lot of momentum. It will take quite a gust to cause excessive movement.

Booster mass is 275 tonnes per the IAC presentation.

A calculation assuming a Cd of 1.0 and a wind speed of 10 m/s gives a wind force of 58 kN on the 12m x 80m booster. Handy (but simplistic) calculator here: http://www.engineeringtoolbox.com/wind-load-d_1775.html

If the terminal control thrusters produce 100 kN and have an adequately responsive control scheme they should have little trouble countering wind gusts in moderate weather.
That's no cold gas thruster...  that's a superdraco...
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/26/2017 02:11 pm
That's no cold gas thruster...  that's a superdraco...

AIUI Musk stated at IAC that the ITS booster will use 10 tonne gaseous methalox thrusters for terminal landing guidance. Grid fins are useless for terminal guidance and cold gas thrusters are far too small.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 02:13 pm
You can build cold gas thrusters as big as you want. They're not "too small." They're too heavy for the total impulse needed.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 02:15 pm
By the way, 10 ton methane thrusters is the same order of magnitude as the typical (8 ton) superdraco.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/26/2017 02:19 pm
That's no cold gas thruster...  that's a superdraco...

AIUI Musk stated at IAC that the ITS booster will use 10 tonne gaseous methalox thrusters for terminal landing guidance. Grid fins are useless for terminal guidance and cold gas thrusters are far too small.
I meant "SD-sized" - for a sense of scale...

Yes, if it's designed for rapid turn around, working with methane makes more sense of course.

By gasous I take it that it takes gas into the combustion chamber.  A big blow torch.  Does anyone know if the that's correct?
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/26/2017 02:26 pm
You can build cold gas thrusters as big as you want. They're not "too small." They're too heavy for the total impulse needed.

For a reasonable mass budget to meet the ITS booster thrust requirements they are far too small.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 02:28 pm
Nah, you're still not getting it. This has to do with total impulse, not that the thrusters are "too small." I promise the cold gas thrusters are pretty dang powerful on the F9 booster.
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/26/2017 02:31 pm
That's no cold gas thruster...  that's a superdraco...

AIUI Musk stated at IAC that the ITS booster will use 10 tonne gaseous methalox thrusters for terminal landing guidance. Grid fins are useless for terminal guidance and cold gas thrusters are far too small.
I meant "SD-sized" - for a sense of scale...

Yes, if it's designed for rapid turn around, working with methane makes more sense of course.

By gasous I take it that it takes gas into the combustion chamber.  A big blow torch.  Does anyone know if the that's correct?

Yes. Mostly likely pressure fed, either from high-pressure bottles of O2 and CH4 gas, or possibly directly from the autogenous pressurant system for the main landing tanks.
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/26/2017 03:01 pm
Nah, you're still not getting it. This has to do with total impulse, not that the thrusters are "too small." I promise the cold gas thrusters are pretty dang powerful on the F9 booster.

Same thing, for various definitions of "small". Sure you can build a huge N2 thruster, but the prop it needs will mass 3x a SuperDraco and 4x a methalox thruster. Put the other way, for the same mass budget a N2 thruster will either be 3 to 4x underpowered or run out of propellant 3 to 4x sooner.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 03:21 pm
You've got it! It has nothing to do with the thruster being "too small," but the tanks being too heavy. I've made you think about it so now you're changing the definition of "small" so it's the opposite of what you originally intended. My work here is done. :)
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/26/2017 03:39 pm
Or, they could make the tanks the right mass and feed the thrusters long enough, but the thrusters would be underpowered. You can pick two of the three (mass, impulse, thrust), but that's it. :D
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 03:55 pm
Or, they could make the tanks the right mass and feed the thrusters long enough, but the thrusters would be underpowered. You can pick two of the three (mass, impulse, thrust), but that's it. :D
As I said, my work here is done.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 01/26/2017 04:43 pm
it is obvious that moving the cradle is easier then moving the rocket. in final moments any wind gust can ruin everything. and cradle is only need to hold 300 tons. for example some bridge cranes could move 750 ton things. after catching the booster it could be positioned/transported (if you have enough long rails you could catch it 300-400m off the pad)  to launch hold downs.

No, it is not obvious. And the rocket is already moving! :) Adding a 2nd moving piece is just adding a bunch of complexity and failure modes.

The rocket will not launch during adverse weather conditions, similar constraints for the landing. So there will be no landing during heavy winds. Powerful thrusters at the base and top of the rocket should be sufficient for lateral accuracy.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 05:15 pm
Parts of the cradle will need to move simply to absorb the energy of impact. Just like a docking port or landing legs.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/26/2017 05:20 pm
We know that... To your method, the main engines can be cold thrusters too.  Granted. Semantically.

Meanwhile, in practical land, cold gas thrusters have very low ISP, and high dry tank mass.  Lose-lose. 

They win on simplicity, when the penalty is small.

Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 01/26/2017 05:30 pm
Parts of the cradle will need to move simply to absorb the energy of impact. Just like a docking port or landing legs.

Obviously. But dampening motion is not the same has having an actively moving landing cradle.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 06:45 pm
Parts of the cradle will need to move simply to absorb the energy of impact. Just like a docking port or landing legs.

Obviously. But dampening motion is not the same has having an actively moving landing cradle.
But the mechanism will have to extend and retract anyway, and being actively guided means you can better absorb the energy without exceeding your limits. It also could help position the booster after capture to mate up to the launch hold-downs (or launch mount posts, in case hold-downs aren't used) prior to refilling.

Again, the NDS docking system is actively positioned, too, and I think you might as well actively guide the cradle since it already needs everything else (pistons with significant travel) anyway.

Again, I think it's helpful to think of the cradle as a giant docking port.
Title: Re: BFR landing cradle discussion and updates
Post by: matthewkantar on 01/26/2017 07:46 pm
The equipment on the ground can be as heavy as it needs to be. The idea is all about shedding weight from the booster. Adding X tons to the docking device is worth it if it takes X pounds off of the booster. Step one is to calculate how much error there will be, step two is design the catcher to account for it.

Some sort of sensor needs to be able to precisely locate the base of the rocket while flames and smoke are present, then the catcher has to twist and tip to catch the hold downs, then drop to absorb the impact while straightening out out any tilt.

Hydraulics can be arbitrarily large and can be extremely agile. The booster will do its best until it touches the cradle and then surrender its fate to the equipment on the ground. I think this is doable.

Matthew
Title: Re: BFR landing cradle discussion and updates
Post by: Doesitfloat on 01/26/2017 08:17 pm
I'm on the boat that believes the only thing that moves on the landing platform is the do-dad that locks it down after it lands. The booster will have the precision to find it's landing pad from space; the center of the landing pad from the start of the landing burn and the ability to adjust to the center and land gently from 2 meters up.
Without those abilities just put legs on it.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 08:21 pm
The only way to gently land is with some kind of springiness. On F9, that's the legs and the cylinders of the legs. You're not going to have two hard surfaces (one stiffly connected to the rocket and the other to the ground) collide without some kind of damage. Just like docking in space, you're going to need something to absorb that energy. Basic physics.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 01/26/2017 10:16 pm
The booster will have the precision to find it's landing pad from space; the center of the landing pad from the start of the landing burn and the ability to adjust to the center and land gently from 2 meters up.


not possible.  It may find it, but hitting is a different matter.  Show me an airplane on autoland hitting the exact same spot.
Title: Re: BFR landing cradle discussion and updates
Post by: Space Ghost 1962 on 01/26/2017 10:30 pm
The booster will have the precision to find it's landing pad from space; the center of the landing pad from the start of the landing burn and the ability to adjust to the center and land gently from 2 meters up.


not possible.  It may find it, but hitting is a different matter.  Show me an airplane on autoland hitting the exact same spot.

Interesting. I've actually witnessed attempts at this. The scatter was about a yard with a modified MLS with doppler. In best conditions. A far smaller craft than a F9 first stage.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 01/26/2017 10:47 pm
The only way to gently land is with some kind of springiness. On F9, that's the legs and the cylinders of the legs. You're not going to have two hard surfaces (one stiffly connected to the rocket and the other to the ground) collide without some kind of damage. Just like docking in space, you're going to need something to absorb that energy. Basic physics.

Nobody disputes this, so I'm not sure why you keep writing it. But it could be as simple as heavy duty springs or something with a crush zone.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/26/2017 10:52 pm
No one disputes this? The guy above mine just said "the one thing that moves is the do-dad that locks it down..." Also: Crush needs to be replaced, and in the F9 is placed in cylinders in the legs. Either way, not amenable to turnarounds of just hours without removing the rocket from the pad. Springs also no good by themselves, must be damped, i.e. By cylinders.

You'll end up with cylinders of some sort whether passive or active. Since you need positioning of the rocket and since you could improve performance by actively catching the rocket, I think you might as well make the cylinders actively actuate.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 12:02 am
Hey - opinions vary.

I don't think hitting the release mechanism (mm accuracy) is feasible, and even if it were, it's a waste of resources on the rocekt.

I don't think an active catcher for such a large body makes sense either.

So I have a passive or half-active "final guidance" rails that will bring the rocket in from ~1-2 m accuracy to final capture.

This will remain open for debate till we see what SpaceX has in mind, and even then, it'll be open to debate as to how they should have done it.

Title: Re: BFR landing cradle discussion and updates
Post by: Space Ghost 1962 on 01/27/2017 12:14 am
Hey - opinions vary.

I don't think hitting the release mechanism (mm accuracy) is feasible, and even if it were, it's a waste of resources on the rocekt.

I don't think an active catcher for such a large body makes sense either.

So I have a passive or half-active "final guidance" rails that will bring the rocket in from ~1-2 m accuracy to final capture.

This will remain open for debate till we see what SpaceX has in mind, and even then, it'll be open to debate as to how they should have done it.


Agreed.

There is so much presumption here about landing systems and approaching BFR level RTLS.

First off, Falcon 9 is neither built like BFR nor scales like BFR. So it is unlikely to actually land like BFR.

What could you do to bring it close enough? Precision landing like BFR. You'd want, as Jim pointed out above, reproducible within feet/inches, which would be hard if not impossible.

That you could do with the current design, forget about the mechanism for the moment. Because neither is the vehicle suitable structurally nor is the landing platform (it moves!).

So what would be structurally and operationally suitable? Well, the vehicle accepts landing loads from where the legs attach, as well as being stablized with. So remove the legs from the LV and have a mechanism from the barge that performs the same functions when the LV has completed its precision approach.

Then you can move forward to BFR landings. Don't get hung up on the need to use the exact same mechanism. That would be suboptimal.

Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 01/27/2017 12:18 am
What could you do to bring it close enough? Precision landing like BFR. You'd want, as Jim pointed out above, reproducible within feet/inches, which would be hard if not impossible.

That you could do with the current design, forget about the mechanism for the moment. Because neither is the vehicle suitable structurally nor is the landing platform (it moves!).

So what would be structurally and operationally suitable? Well, the vehicle accepts landing loads from where the legs attach, as well as being stablized with. So remove the legs from the LV and have a mechanism from the barge that performs the same functions when the LV has completed its precision approach.

Appreciate the gradual approach, and learning with F9 first. But do so on land; don't bother with the barge.
Title: Re: BFR landing cradle discussion and updates
Post by: Space Ghost 1962 on 01/27/2017 12:56 am
What could you do to bring it close enough? Precision landing like BFR. You'd want, as Jim pointed out above, reproducible within feet/inches, which would be hard if not impossible.

That you could do with the current design, forget about the mechanism for the moment. Because neither is the vehicle suitable structurally nor is the landing platform (it moves!).

So what would be structurally and operationally suitable? Well, the vehicle accepts landing loads from where the legs attach, as well as being stablized with. So remove the legs from the LV and have a mechanism from the barge that performs the same functions when the LV has completed its precision approach.

Appreciate the gradual approach, and learning with F9 first. But do so on land; don't bother with the barge.

I agree it would be easier. Thus the "it moves!" reference. However, now you have other considerations too.

First, they'll want to have an operational tempo with RTLS to become commonplace to build a launch skill set and expectations also for BFR - you might not want to interfere with that.

Barge recoveries always will be chancy and slow by their nature. Also, since the barge recoveries are downrange, you get an immediate benefit for the mass reduction of missing the legs. Plus, high performance FH core recoveries will doubtlessly push to the max any attempts to recover a usable stage, so it will remain experimental for some time to come, if even tried at all.

Another advantage of NOT modifying the LV is that you could continue to push reuse either way until you decided which to use at the last moment for a given mission.

Finally, I think SX would rather have a "boom" on a barge than on land.

But again, you're right land is much easier.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 01:05 am
You can't and shouldn't try to mimic BFR with F9.  As always, you need to match model A to reality A, then transfer model A to model B, and thus gain insight into reality B.

So I'd add thruster to the bottom of an F9 (if that's indeed the plan) and learn how to land with high accuracy into virtual cradles, making sure you can hit the acceptance parameters.

Then design a cradle that takes you the rest of the way.  That can be tested in a hangar somewhere.

Putting the two things together with real F9s is secondary, but will still be useful.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/27/2017 01:20 am
They should make a BFR or BFS or subscale and just test the system that way, ala Grasshopper/F9Rdev1. They want to test those vehicles out anyway. Can start with a battleship design like Grasshopper if they need to to save time/money.
Title: Re: BFR landing cradle discussion and updates
Post by: Space Ghost 1962 on 01/27/2017 02:36 am
They should make a BFR or BFS or subscale and just test the system that way, ala Grasshopper/F9Rdev1. They want to test those vehicles out anyway. Can start with a battleship design like Grasshopper if they need to to save time/money.

You can't ever let that idea die, can you? Even if Musk says "no".  ::)
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/27/2017 04:40 am
They should make a BFR or BFS or subscale and just test the system that way, ala Grasshopper/F9Rdev1. They want to test those vehicles out anyway. Can start with a battleship design like Grasshopper if they need to to save time/money.

You can't ever let that idea die, can you? Even if Musk says "no".  ::)
What in the world are you referring to? I'm talking about experimental ITS development, most certainly not an operational vehicle.
And, um, ahem: http://forum.nasaspaceflight.com/index.php?topic=41385.msg1608795#msg1608795
Title: Re: BFR landing cradle discussion and updates
Post by: jpo234 on 01/27/2017 05:38 am
The booster will have the precision to find it's landing pad from space; the center of the landing pad from the start of the landing burn and the ability to adjust to the center and land gently from 2 meters up.


not possible.  It may find it, but hitting is a different matter.  Show me an airplane on autoland hitting the exact same spot.
http://www.dodbuzz.com/2016/08/18/navy-f-35c-landed-so-precisely-it-tore-up-a-runway/
Title: Re: BFR landing cradle discussion and updates
Post by: JamesH65 on 01/27/2017 08:08 am
The booster will have the precision to find it's landing pad from space; the center of the landing pad from the start of the landing burn and the ability to adjust to the center and land gently from 2 meters up.


not possible.  It may find it, but hitting is a different matter.  Show me an airplane on autoland hitting the exact same spot.

Planes, accurate to a few meters, helicopters, accurate to less than 1m.  I suggest helicopters are a more accurate comparison. I see no reason why, with modern guidance systems, the booster cannot land within 1m of intended target. 1m, proportionally, should be good enough for a landing cradle, the slack can be taken up by adjustable arms/clamps. Nothing beyond the ability of a decent mechanical engineer to design.

Just of the top of my head, I'd have a set of legs on the ground, hydraulically activated to rise up vertically giving the booster something to land on. They would handle shock absorption. Only extend the legs that match up with the booster attachment pads ie you have a pin cushion of legs, only use the ones you need.

Or a ring cradle that can move in X,Y and Z slightly to cope with slight misalignments.

I think there are number of fairly obvious ways this could be made to work. There are also plenty of non-obvious ways I suspect.
Title: Re: BFR landing cradle discussion and updates
Post by: LouScheffer on 01/27/2017 12:13 pm
The booster will have the precision to find it's landing pad from space; the center of the landing pad from the start of the landing burn and the ability to adjust to the center and land gently from 2 meters up.


not possible.  It may find it, but hitting is a different matter.  Show me an airplane on autoland hitting the exact same spot.
Active landing can be extremely accurate.  In  parachute accuracy competitions (https://en.wikipedia.org/wiki/Accuracy_landing), the target is 2 cm in diameter, and a good competitor will hit it on most tries, with a cm or two miss on the rest. (They had to decrease the diameter from 3 cm since they were getting too many ties.)    They don't even bother measuring past 16 cm.  And this with a parachute that is quite sensitive to winds.  This shows what can be achieved with active targeting right up to touchdown.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 01/27/2017 01:23 pm
  In parachute accuracy competitions,[

Not even relevant to this discussion.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/27/2017 01:31 pm
Masten Space Systems landed their rocket within about 6 inches (multiple times) of a designated spot as part of the NGLLC.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 01/27/2017 01:43 pm

Planes, accurate to a few meters, helicopters, accurate to less than 1m.  I suggest helicopters are a more accurate comparison.

Not relevant.  See the mass and size of the vehicle


I see no reason why, with modern guidance systems, the booster cannot land within 1m of intended target.

There are many, one being limited propellant.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 01/27/2017 01:53 pm
What ever device catches the vehicle also has to support 26,0000,000 lbs and be able to hold down 6,000,000 lb of thrust (ITS total thrust minus ITS total mass)
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/27/2017 02:00 pm
What ever device catches the vehicle also has to support 26,0000,000 lbs and be able to hold down 6,000,000 lb of thrust (ITS total thrust minus ITS total mass)
Doesnt have to be the same device, just needs to be co-located.
Title: Re: BFR landing cradle discussion and updates
Post by: robert_d on 01/27/2017 02:01 pm
What ever device catches the vehicle also has to support 26,0000,000 lbs and be able to hold down 6,000,000 lb of thrust (ITS total thrust minus ITS total mass)
Doesnt have to be the same device, just needs to be co-located.
Distinction without a difference?
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 02:02 pm
So to recap, "landing a rocket is easy,  easier than docking even."

The demonstrated accuracy of F9, which is a few meters, is clearly included in that easiness.

But "1 m is fantasy", because it's impossible to do better than a few meters  which  F9 is doing without side thrusters even.

--------

The catchment slideways don't have to hold any weight. They just add side forces during the last few meters of descent. With the right shaped fins, they don't even have to be slideways, just passive nubs or rollers.

The rocket then settles on touchdown supports, which are clearly not impossible, since the rocket was sitting on something WHILE FUELED before it took off.

----

Sorry - I just don't see any argument for "impossibility" other than that it can't be done because it isn't being done.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/27/2017 02:28 pm
What ever device catches the vehicle also has to support 26,0000,000 lbs and be able to hold down 6,000,000 lb of thrust (ITS total thrust minus ITS total mass)
Doesnt have to be the same device, just needs to be co-located.
Distinction without a difference?
Not if you're actually building one.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 02:36 pm
What ever device catches the vehicle also has to support 26,0000,000 lbs and be able to hold down 6,000,000 lb of thrust (ITS total thrust minus ITS total mass)
Doesnt have to be the same device, just needs to be co-located.
Distinction without a difference?
Not if you're actually building one.
Co-located means that the physical structure accepting the (empty) rocket is different from the physical mechanism holding it down during launch, but the that the latter can engage after the rocket landed on the former.

So the distinction makes sense.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/27/2017 02:38 pm
Precisely.
Title: Re: BFR landing cradle discussion and updates
Post by: JamesH65 on 01/27/2017 03:01 pm
  In parachute accuracy competitions,[

Not even relevant to this discussion.

Neither are aeroplanes, which you brought up earlier.
Title: Re: BFR landing cradle discussion and updates
Post by: JamesH65 on 01/27/2017 03:06 pm
So to recap, "landing a rocket is easy,  easier than docking even."

The demonstrated accuracy of F9, which is a few meters, is clearly included in that easiness.

But "1 m is fantasy", because it's impossible to do better than a few meters  which  F9 is doing without side thrusters even.

--------

The catchment slideways don't have to hold any weight. They just add side forces during the last few meters of descent. With the right shaped fins, they don't even have to be slideways, just passive nubs or rollers.

The rocket then settles on touchdown supports, which are clearly not impossible, since the rocket was sitting on something WHILE FUELED before it took off.

----

Sorry - I just don't see any argument for "impossibility" other than that it can't be done because it isn't being done.

Certain things are only impossible until someone actually does it. The world of full of heavier than air craft, contrary to Lord Kelvin's proclamation.

I think to claim landing on the launch mount is impossible is wildly pessimistic.  It's just an engineering and materials issue. The human race has been quite good at figuring those out.
Title: Re: BFR landing cradle discussion and updates
Post by: ZachF on 01/27/2017 03:29 pm
How about something like this:

There are (3) large catcher plates covered in either wheels or one large tank-tread type device. There is a hinge on the middle for it to rotate and a cam that pushes the hinge closer when the rocket falls down past the hinge closer to the cradle bottom. There is also a roller or tread covered funnel device to direct the three "fins" into a shock absorber. The three main treads/rollers would also be able to rotate along with the rocket when the fins are falling into them.
Title: Re: BFR landing cradle discussion and updates
Post by: ZachF on 01/27/2017 03:33 pm
The rocket landing in my diagram is purposely off-center and rotates slightly to show how the mechanism works.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 03:44 pm
The rocket landing in my diagram is purposely off-center and rotates slightly to show how the mechanism works.
Good man!   :)
Title: Re: BFR landing cradle discussion and updates
Post by: alang on 01/27/2017 04:12 pm
Has it been discussed that in one sense landing the BFR should be easier: there should be less influence of wind on landing because of the larger mass per unit area ?
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 04:14 pm
Has it been discussed that in one sense landing the BFR should be easier: there should be less influence of wind on landing because of the larger mass per unit area ?
Yup, upthread, though the numbers are of course tentative.
Title: Re: BFR landing cradle discussion and updates
Post by: ZachF on 01/27/2017 04:19 pm
Has it been discussed that in one sense landing the BFR should be easier: there should be less influence of wind on landing because of the larger mass per unit area ?

Yep. A Falcon 9 empty stage weighs about 25,000 kg and is 3.7m wide and 42m tall,

3.7x42 = 155m^2 side area

25,000/155 = 161 kg per m^2

BFR/ITS is 275,000 kg empty and 12m wide and 78m tall

12x78 =936m^2 side area

275,000/936 = 294 kg per m^2

BFR/ITS has a side sectional density about 83% higher than the Falcon 9.

wind effects should be roughly 45% less than Falcon 9.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 04:38 pm
Overnight thinking, cartoon #3, similar to the track above.

Rollers at ends of the spring-shocks, that match the fins in the rocket.

The shocks are perpendicular to the fin slope, to reduce bending moments

When the rockets is centered, all the shocks are fully extended.  Otherwise, they try to center it.

The rollers are V-shaped (shown below) so that they also help with clocking the rocket.

Not shown are fixed "V" rails that help in the case of gross misalignment.

----

Title: Re: BFR landing cradle discussion and updates
Post by: ZachF on 01/27/2017 04:58 pm
Overnight thinking, cartoon #3, similar to the track above.

Rollers at ends of the spring-shocks, that match the fins in the rocket.

The shocks are perpendicular to the fin slope, to reduce bending moments

When the rockets is centered, all the shocks are fully extended.  Otherwise, they try to center it.

The rollers are V-shaped (shown below) so that they also help with clocking the rocket.

Not shown are fixed "V" rails that help in the case of gross misalignment.

----

The reason mine has long(ish) rollers is to also cancel out any angles that may occur. Remember when you push against the rocket one way on the bottom, it will push in the opposite direction against the top, maybe with a much larger magnitude since the CoG is closer to the bottom. The long roller arms with a hinge and a cam on mine are so that the weight of the ITS booster passively corrects any alignment problems, and it "slide" down into it's mount from gravity.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/27/2017 07:06 pm
Overnight thinking, cartoon #3, similar to the track above.

Rollers at ends of the spring-shocks, that match the fins in the rocket.

The shocks are perpendicular to the fin slope, to reduce bending moments

When the rockets is centered, all the shocks are fully extended.  Otherwise, they try to center it.

The rollers are V-shaped (shown below) so that they also help with clocking the rocket.

Not shown are fixed "V" rails that help in the case of gross misalignment.

----

The reason mine has long(ish) rollers is to also cancel out any angles that may occur. Remember when you push against the rocket one way on the bottom, it will push in the opposite direction against the top, maybe with a much larger magnitude since the CoG is closer to the bottom. The long roller arms with a hinge and a cam on mine are so that the weight of the ITS booster passively corrects any alignment problems, and it "slide" down into it's mount from gravity.
The bottom push is very much near the center of mass of the empty rocket.

If it isn't, you need large torques to counter it while engaging only a short stretch - easier to do it from the top.

Alternatively, you can extend the parallel (top) part of the fins, and use two rollers (starting to approximate your track)

Overall though, I think the rockets come down vertically.  When  we see one leg hitting first, it's because the barge is tilting.

The only other reason for tilt is crabbing, but with side thrusters, even that goes away.

Title: Re: BFR landing cradle discussion and updates
Post by: MikeAtkinson on 01/27/2017 07:37 pm
I wonder about using non-contact methods of alignment, using electromagnets, or perhaps strong fans. I may be possible to blow enough air upwards to keep the whole 275 tonnes suspended.
Title: Re: BFR landing cradle discussion and updates
Post by: envy887 on 01/27/2017 07:41 pm
I wonder about using non-contact methods of alignment, using electromagnets, or perhaps strong fans. I may be possible to blow enough air upwards to keep the whole 275 tonnes suspended.

Assuming a Cd of 1, the booster falling end-on like a pencil has a terminal velocity of ~175 m/s. Those are some strong fans.
Title: Re: BFR landing cradle discussion and updates
Post by: MikeAtkinson on 01/27/2017 07:44 pm
I wonder about using non-contact methods of alignment, using electromagnets, or perhaps strong fans. I may be possible to blow enough air upwards to keep the whole 275 tonnes suspended.

Assuming a Cd of 1, the booster falling end-on like a pencil has a terminal velocity of ~175 m/s. Those are some strong fans.

Sure, but not impossible I think, they only need to be active a a few seconds.
Title: Re: BFR landing cradle discussion and updates
Post by: Space Ghost 1962 on 01/27/2017 09:29 pm
You're overthinking things. Look, they've already solved the thrust/shock acceptance to the Falcon vehicle, as well as the stabilization.

Part of the leg design. And we know the legs work.

All you have to do is have "leg like" "arms" that attach at the right time/points.

Why make it any more difficult than that?

In this arrangement, the hardest part would be absorbing the additional loads/load paths created by an off center arrival. The point of the precision landing would be to reduce the scope of that, and the "overbuilding" of the "arms" would handle the rest.
Title: Re: BFR landing cradle discussion and updates
Post by: drzerg on 01/27/2017 10:44 pm
i do not understand you guys. if robots can do this what a problem to catch a rocket? imagine just strong and fast enough X Y frame on rails. its just 300 tons. cmon. why you even need this weight cost system on the booster if you can put it on land.

https://www.youtube.com/watch?v=cyN-CRNrb3E
Title: Re: BFR landing cradle discussion and updates
Post by: Jdeshetler on 01/27/2017 11:18 pm
Same goes for this.
https://www.youtube.com/watch?v=XxFZ-VStApo

The cradle ring can be design to moved into correct position in almost real time after sensing the 1st stage's incoming position.

I believed the 7m Mini-ITS (same height size as F9's but same proportional as ITS's) would be a perfect platform to start with before moving up to a full scale.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 01:32 am
You're overthinking things. Look, they've already solved the thrust/shock acceptance to the Falcon vehicle, as well as the stabilization.

Part of the leg design. And we know the legs work.

All you have to do is have "leg like" "arms" that attach at the right time/points.

Why make it any more difficult than that?

In this arrangement, the hardest part would be absorbing the additional loads/load paths created by an off center arrival. The point of the precision landing would be to reduce the scope of that, and the "overbuilding" of the "arms" would handle the rest.

The legs are optimized for a different circumstance.  My first sketch up thread has "stumps", but I think it's more elegant to keep the mass of the shocks on the ground.

If the mechanism allows more Z travel (which can only be done with ground mounted shocks) then it's easier to increase X-Y accuracy...

So I think that's the way they'll go, and the video is in agreement, FWIW.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 01:35 am
The two videos are cute.

Whether something like this is implemented on the cradle is not a matter of feasibility but of practicality, when compared to the other options, which are also feasible.

Title: Re: BFR landing cradle discussion and updates
Post by: Karlman on 01/28/2017 02:02 am
Couple of thoughts..
At least initially.. would you want to expose your landing cradle to the rigours of a launch with 42 Merlin engines, and hope it works perfectly 10m later to catch the returning BFR? Lot less risky if your landing cradle isn't part of your launch apparatus.

How do you launch more than one BFR form a certain site? If you cycling through 2, 3, or more BFR's at one launch pad, then they need to be moved around anyway.. so the landing cradle doesn't have to be launch cradle.

If the landing cradle has to having moving parts to guide the landing, or some type of shock absorbing role, do you want the cradle to then sit for weeks/months with the weight of BFR on it before the next launch? How is the BFR stored between launches?

Will BFR be vertical all the time.. from transport to / from the launch pad, and 2nd stage /BFS integration? or will be like falcon 9 / falcon heavy transported on its side, and integrated on it's side?

While launch and land at the same spot seems like a good goal to enable daily type launches.. something less "eggs all in 1 basket" seems to have lots of benefit as the initial steps on the path to that goal.

(my apologies if some of these are already known answers outside of L2).
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 02:05 am
No, the assumption is the BFR booster is made for truly rapid reuse, meaning no refurb between launches. No cycling through boosters. Cycling through boosters would also be inconsistent with the ITS video.

The whole point of the launch cradle is to minimize having to process/move around the biggest piece of equipment of the whole stack, plus maybe some weight reduction.

The idea is the mechanism catches the rocket, then places it in a mount or hold-down clamps for storage or to prepare for launch (stacking and fueling). It's not carrying the load of the (empty) rocket for more than a minute or so.

And if you want more than one BFR per site, then build another pad. LC-39a, 39b, 39c, etc...
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 02:25 am
Couple of thoughts..
At least initially.. would you want to expose your landing cradle to the rigours of a launch with 42 Merlin engines, and hope it works perfectly 10m later to catch the returning BFR? Lot less risky if your landing cradle isn't part of your launch apparatus.

How do you launch more than one BFR form a certain site? If you cycling through 2, 3, or more BFR's at one launch pad, then they need to be moved around anyway.. so the landing cradle doesn't have to be launch cradle.

If the landing cradle has to having moving parts to guide the landing, or some type of shock absorbing role, do you want the cradle to then sit for weeks/months with the weight of BFR on it before the next launch? How is the BFR stored between launches?

Will BFR be vertical all the time.. from transport to / from the launch pad, and 2nd stage /BFS integration? or will be like falcon 9 / falcon heavy transported on its side, and integrated on it's side?

While launch and land at the same spot seems like a good goal to enable daily type launches.. something less "eggs all in 1 basket" seems to have lots of benefit as the initial steps on the path to that goal.

(my apologies if some of these are already known answers outside of L2).
Based on your questions, it's clear you haven't seen this awesome video which answers many of them:
https://www.youtube.com/watch?v=0qo78R_yYFA
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 01/28/2017 02:33 am
The video shows a single booster lofting multiple vehicles, and we are not lead to believe the booster will be cycled frequently.

But of course, a single booster won't be joined to a given pad for all time; how that the booster is moved and stored is a valid question. Given Musk's past comments, I'll suspect they're inclined to fly it when it moves any significant distance. The implication being that any off-site testing/storage/repair location will need its own launch/landing pad as well.

If the landing cradle has to having moving parts to guide the landing, or some type of shock absorbing role, do you want the cradle to then sit for weeks/months with the weight of BFR on it before the next launch?

Sure, why not? The weight-bearing portion of the cradle need not be the 'guide' portion with shock absorbers. The ITS diagrams show a tapered ring surrounding the set of engine bells, about .25 meters in thickness. This is what will make contact with the cradle/pad.

Quote
How is the BFR stored between launches?

I like meekGee's notion that the booster remains vertical all all times. The tradeoff is you need more expensive structures to shelter it. Will be interesting to see what they do.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 02:38 am
I think they'll move it around on the ground sometimes. I doubt they'll literally build it vertically, for instance.
Title: Re: BFR landing cradle discussion and updates
Post by: Karlman on 01/28/2017 02:40 am

Based on your questions, it's clear you haven't seen this awesome video which answers many of them:

Oh I've seen that video more than a few times.. I just don't think that's meant to be a video of the initial launch.. it also doesn't answer questions about how the BFR got there in the first place... is the pad meant to be the only home for a BFR.. e.g. it's built there, lands there, is serviced there vertically always and never leaves the pad?

I haven't seen answers to those questions.. and those types of questions are interesting to think about for the initial BFR.

Edit:
Instead of the rocket being assembled and serviced in a building, and rolled out to the pad (like Saturn / Shuttle was), would SpaceX have a moving assembly / service building instead? No idea if this would even be workable, but's the sort of left field solution I'd almost expect from SpaceX.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 02:43 am
I definitely think the launch cradle is being designed for the first ITS.

How it'll get there? Probably built horizontally, moved by barge and multi wheel trailer, then erected by cranes. A lumbersome process, which is why they want the cradle in the first place.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 02:50 am
How will it get there?   It will fly there, obviously.

It will be built vertically somewhere else, where construction doesn't hold up a working pad.  It will have a test flight, and then a transfer flight as it joins the active fleet.

Unless it absolutely breaks down, then for regular service, it will fly to a service pad.

Why not?

EDIT: when they build airplanes, the assembly facility has an airstrip, even thought runways are hella expensive, right?   Cause how else do you transport something the size of a jetliner?
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 02:53 am
...because single stage point to point over non-trivial distances is nearly the same performance as SSTO, and the ITS booster doesn't have the TPS to be a SSTO RLV.

A once-in-1000-flights barge trip is fine. You should be able to fit two ITS boosters (side by side) on a barge the size of the ASDSes. And the STS/SLS's Pegasus barge is nearly the right size, too.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 03:10 am
...because single stage point to point over non-trivial distances is nearly the same performance as SSTO, and the ITS booster doesn't have the TPS to be a SSTO RLV.

A once-in-1000-flights barge trip is fine.
Why does it have to be substantial?

I'm not talking flying from Texas to the cape, but if you want to go 50 miles, why not?  Easier than figuring out road transport for a 12 m diameter cylinder.

BFRs are meant to be used hundreds of times.  So use them!

They'll be flying without a second stage.

Same btw for BFS.  It's a very big beast, but can fly just fine (partiallly fueled) to its own pad.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 03:18 am
50miles? Piece of cake with a wheeled vehicle. WAY easier than getting clearance to launch from wherever your factory is (probably close to a city and/or major highway).

If you're launching from your factory, why not build the factory on LC39a?
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 03:20 am
Remember, BFR is likely to have a single final factory and multiple launch sites (not just LC39a), so you're going to need long distance transport anyway. Not a problem for BFS, since it can go to orbit.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 03:23 am
50miles? Piece of cake with a wheeled vehicle. WAY easier than getting clearance to launch from wherever your factory is (probably close to a city and/or major highway).
12 m wide, 12 m tall, 50 m long - good luck with that cake.

All I'm saying, build your factory along the Florida coastline.  "Near the launch site".

But don't waste your time building it on the pad...

Remember that deal with Toray for composite fiber?  How many BFR/BFS will fit in that?   

You need a way to build them at point A, then get them into the operational fleet that's operating at point B, and I can't see a more natural way to do so then to fly them.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 03:36 am
ITS booster WOULD be a piece of cake compared to most of the things in this catalogue:
https://www.scheuerle.com/fileadmin/data_all/files/Self-Propelled_Transporters_EN.pdf

Wheeled vehicles like these Kamags move stuff weighing up to 15,000tons. ITS booster is just 275 tons, so that'd be over 50 ITS boosters.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 01/28/2017 03:39 am
Are you going to fly them from Florida to Kwaj? To Texas? How about from Orlando to the Cape? You have to vacate a launch site, so are you going to vacate the factory just because you're moving the ITS booster and your only way of moving it is to fly it??
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 03:56 am
ITS booster WOULD be a piece of cake compared to most of the things in this catalogue:
https://www.scheuerle.com/fileadmin/data_all/files/Self-Propelled_Transporters_EN.pdf

Wheeled vehicles like these Kamags move stuff weighing up to 15,000tons. ITS booster is just 275 tons, so that'd be over 50 ITS boosters.

Seeing something in a catalog doesn't make it a "piece of cake".

It's the size of the operation that matters.  When you are building a new power station or refinery, you get one of those transporters, and it's a giant operation that disrupts half the state.

But SpaceX want to build BFSs like 737s.  Twice a week eventually, was it? (Your quote below...)  But even if it's 4 times a year.   They are not easy to transport by vehicle.

And yes, if the factory has a launch pad, then you build the ship on the pad, then step away and launch it.

At $1000/kg (T1000 costs about $2000/kg, so I'm assuming some cost reduction), 30 tons of carbon fiber per BFS, and 100 new BFSes per year (needed to maintain 1000 BFS in use needed for ~80,000 colonists per year, since each BFS can only be reused about 12 times), that works out to about $3 billion worth of carbon fiber per year. Not counting cargo missions or the BFR booster first stage (which could be reused hundreds of times, so may actually end up a smaller portion of the carbon fiber demand).

Works out to about $30,000 worth of carbon fiber per round trip ticket. $30 million per BFS made.
Title: Re: BFR landing cradle discussion and updates
Post by: Nascent Ascent on 01/28/2017 04:24 am
Gee, might as well have a huge swarm (murmuration) of fairly large coordinated drones that fly up and attach to the falling booster.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 04:34 am
yeah, the conversation did get silly.

Back to the on-pad / cradle landing?
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 01/28/2017 06:15 am
What about clocking? The top of the booster isn't symmetrical; considering the need for a crew access arm it appears there will be but one workable orientation.

Some of the designs above show rollers/guides to clock via the fins. Alternatively, the cradle itself could rotate the booster following touch-down.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 06:37 am
What about clocking? The top of the booster isn't symmetrical; considering the need for a crew access arm it appears there will be but one workable orientation.

Some of the designs above show rollers/guides to clock via the fins. Alternatively, the cradle itself could rotate the booster following touch-down.
If I have time this weekend, I'll model it.

Clocking during flight is done easily with the fins, and there are only weak disturbing torques.

My concern is that when the first fin makes contact, it might apply a sudden torque.

I think it's solvable with a couple more guide rails.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 01/28/2017 06:54 am
My concern is that when the first fin makes contact, it might apply a sudden torque.

I think it's solvable with a couple more guide rails.

For that approach I think ZachF was on to something with his design, in that he corrects for position first and rotational orientation second. His upper roller surfaces aren't intended to impart torque, and the v-guides which clock the fins are at the very bottom of the cradle.

Or, again, don't bother with v-guides at all – just keep everything symmetrical, let the rocket do its best, and make any rotational adjustments post-lading. It's simpler.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 01/28/2017 12:32 pm
My concern is that when the first fin makes contact, it might apply a sudden torque.

I think it's solvable with a couple more guide rails.

For that approach I think ZachF was on to something with his design, in that he corrects for position first and rotational orientation second. His upper roller surfaces aren't intended to impart torque, and the v-guides which clock the fins are at the very bottom of the cradle.

Or, again, don't bother with v-guides at all – just keep everything symmetrical, let the rocket do its best, and make any rotational adjustments post-lading. It's simpler.
If I understand his design correctly, he  achieves something similar to what the v-shaped rollers do, centering each fin in its own groove.
Title: Re: BFR landing cradle discussion and updates
Post by: MP99 on 01/29/2017 07:06 pm
I am assuming that the landing adapter will be actively maneuvered under the stage as it lands.

This makes me wonder whether the adapter could catch a stage which is not vertical.

When an F9 is blown sideways as it lands, the stage must:
1) tip into the wind,
2) translate to get back on course,
3) tip the other way,
4) cancel any lateral dV,
5) then make itself vertical to land.

If the landing adapter can tip itself, it might catch the stage without needing (5) to happen?  Of course, the stage would need to be strong enough to survive the cantilever force of being canted over after capture.

If the adapter has active absorption of landing shocks, including a sideways element, maybe (4) is not needed either.

Cheers, Martin
Title: Re: BFR landing cradle discussion and updates
Post by: OneSpeed on 01/29/2017 07:29 pm
I am assuming that the landing adapter will be actively maneuvered under the stage as it lands.

This makes me wonder whether the adapter could catch a stage which is not vertical.

When an F9 is blown sideways as it lands, the stage must:
1) tip into the wind,
2) translate to get back on course,
3) tip the other way,
4) cancel any lateral dV,
5) then make itself vertical to land.

If the landing adapter can tip itself, it might catch the stage without needing (5) to happen?  Of course, the stage would need to be strong enough to survive the cantilever force of being canted over after capture.

If the adapter has active absorption of landing shocks, including a sideways element, maybe (4) is not needed either.

Cheers, Martin

If the booster has lateral thrusters at the top and the bottom of the fuselage, then it could simply translate horizontally without tipping.
Title: Re: BFR landing cradle discussion and updates
Post by: Bob Shaw on 01/29/2017 08:17 pm
Regarding mobile structures, they're hardly new - even big ones.

There's previous experience: For Saturn launches, LC39 was serviced by a mobile structure which was rolled up against the rocket on the crawler transporter (a bit difficult to do now that SpaceX has a building at the bottom of the ramp). For Shuttle, Vandenberg AFB SLC-6 had a mobile building which encapsulated the Shuttle stack then backed off. And, at KSC there was of course the Rotating Service Structure which covered most of the vehicle and then was pulled out of the way before launch. So, large and moving objects have been used before.

Perhaps this is why SpaceX is in no hurry to demolish 'their' RSS?
Title: Re: BFR landing cradle discussion and updates
Post by: CraigLieb on 01/31/2017 09:34 pm
Regarding mobile structures, they're hardly new - even big ones.
... the rocket on the crawler transporter

Seems like this was probably covered in discussion above, but not sure...
If the rocket lands ON a crawler/transporter this platform/device could make (in the days/hours after landing) slight adjustments for-relaunch including orientation (clocking discussions), and/or coordinate shift/offsets. This reduces the issues of mis-alignment back to the class of "landing on a sufficiently large and sturdy platform within a circle of error"  This platform doesn't have to do the delivery to the launch site from the hangar. It also doesn't have to make the vehicle or sections of the vehicle vertical like the current TEL.  The transport could have connections for fueling/defueling operations built-in as well as sound/flame suppression systems, or those could be in the areas surrounding the platform instead.

If the launch site has sufficient crane capacity as shown in the video, empty vehicle sections can be placed on the launch platform from TE vehicles off set from the site. Likewise, sections can be disassembled from the launch
platform and placed on mounts or TE devices for movement back to hangars for maintenance.

While all this complicates things somewhat for launch, it seems to solve the alignment issues nicely on landing and re-launch. Plus it takes advantage of learning going on with the ASDS, the methods being leveraged on returning boosters, etc.

Title: Re: BFR landing cradle discussion and updates
Post by: rakaydos on 02/09/2017 04:54 am
It occurs to me that controlling lateral thrusters to maintain a bullseye on the landing structure is basiclly the same thing as what the drone ship does to remain steady and stationary in heavy seas. The engineering end will be different but the programming side should have comparable input/output responces.
Title: Re: BFR landing cradle discussion and updates
Post by: CraigLieb on 02/09/2017 12:54 pm
...

I like the idea, but as I thought about it, I realize the movable, adjustable landing "plate" would have very little travel in any direction without hitting something else in order to realign the booster that just landed.  They'd have to have a "clean pad" design, and pad 39 isn't really the place for that.

Seems to me that if they can land an F9 in the landing ring every time using a hoverslam, they can guide an ITS onto a similar sized spot with precision if they have some additional loiter time to adjust before shutting down the motors (a la New Shephard).  KISS and no pad rebuild required...

So maybe it is built in to the launch site and all it can do is rotate to adjust minor clocking issues
Title: Re: BFR landing cradle discussion and updates
Post by: Johnnyhinbos on 02/09/2017 01:01 pm
I don't think the clocking of the booster is so important. I'd think putting V-shaped guides above each clamp on the pad would provide the fine rotational alignment as booster "touches down".
Title: Re: BFR landing cradle discussion and updates
Post by: CraigLieb on 02/09/2017 01:42 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.

Air liners typically don't pull into the gate with the door on the other side from the gang-way. 
If SpaceX can guarantee the alignment and clocking within ~30 degrees then maybe this isn't an issue.

I Suppose a mobile cart could pull up like a giant lift and be raised up to the door, or they could put multiple doors on the vehicle and use the one that makes sense for the landing alignment, but that adds weight to the vehicle.
Title: Re: BFR landing cradle discussion and updates
Post by: sghill on 02/09/2017 02:05 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.

Air liners typically don't pull into the gate with the door on the other side from the gang-way. 
If SpaceX can guarantee the alignment and clocking within ~30 degrees then maybe this isn't an issue.

I Suppose a mobile cart could pull up like a giant lift and be raised up to the door, or they could put multiple doors on the vehicle and use the one that makes sense for the landing alignment, but that adds weight to the vehicle.

Strawman argument.  The point is that it's a whole boatload easier to have things on the pad adjust to the needs of the returning vehicle than have the returning vehicle adjust to the needs of the pad.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/09/2017 02:21 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.

Air liners typically don't pull into the gate with the door on the other side from the gang-way. 
If SpaceX can guarantee the alignment and clocking within ~30 degrees then maybe this isn't an issue.

I Suppose a mobile cart could pull up like a giant lift and be raised up to the door, or they could put multiple doors on the vehicle and use the one that makes sense for the landing alignment, but that adds weight to the vehicle.

Strawman argument.  The point is that it's a whole boatload easier to have things on the pad adjust to the needs of the returning vehicle than have the returning vehicle adjust to the needs of the pad.
Regarding clocking, I think it's the easiest degree of freedom to control.
Title: Re: BFR landing cradle discussion and updates
Post by: Lar on 02/09/2017 02:43 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.
I thought we're talking about the booster?

For BFS if it also lands in a cradle rather than on its legs as Mars, then there are 3 orientations, 120 degrees apart... just use the right orientation... what am I missing?
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/09/2017 02:48 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.
I thought we're talking about the booster?

For BFS if it also lands in a cradle rather than on its legs as Mars, then there are 3 orientations, 120 degrees apart... just use the right orientation... what am I missing?

The mounting point at the top of the booster is not symmetrical. There's only one way BFS can orient itself.
Title: Re: BFR landing cradle discussion and updates
Post by: Lar on 02/09/2017 02:50 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.
I thought we're talking about the booster?

For BFS if it also lands in a cradle rather than on its legs as Mars, then there are 3 orientations, 120 degrees apart... just use the right orientation... what am I missing?

The mounting point at the top of the booster is not symmetrical. There's only one way BFS can orient itself.

Still confused, are you saying BFS is going to land on top of the booster? Not be craned up after landing as in the vid?

OK then both the booster and BFS have to come in the right way.... Still not seeing the issue.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/09/2017 03:05 pm
The mounting point at the top of the booster is not symmetrical. There's only one way BFS can orient itself.

Still confused, are you saying BFS is going to land on top of the booster? Not be craned up after landing as in the vid?

OK then both the booster and BFS have to come in the right way.... Still not seeing the issue.

When craned onto the booster there is but one working orientation for the BFS and tanker. Unless the crew access arm is surprisingly flexible (per sghill's comment) then there is but one working orientation for the entire stack.

There is no issue. But three usable orientations seems unlikely IMO.
Title: Re: BFR landing cradle discussion and updates
Post by: CraigLieb on 02/09/2017 06:40 pm
Just wondering how flexible the passenger disembarkation can be if the landing is 180 degrees away from the gang-way.
I thought we're talking about the booster?

For BFS if it also lands in a cradle rather than on its legs as Mars, then there are 3 orientations, 120 degrees apart... just use the right orientation... what am I missing?

Duh!!! (smacks forehead)  You're right! The booster!

I still assume for the booster landing, you need to orient it correctly so that when you stack the BFS on on top the BFR, it has to be oriented correctly to allow boarding and connections between BFR and BFS. But you are totally correct, I was thinking about the BFS.  The landed Rocket has nobody getting off it! 
Title: Re: BFR landing cradle discussion and updates
Post by: GabrielP on 02/10/2017 12:20 pm
Would this also be possible or is there some kind of unavoidable inflexibility in ground systems that would prevent it?

1. booster lands in whichever of the 3 possible orientations
2. landing cradle rotates the booster to the orientation required for BFS stacking
3. apply BFS to booster, now in the "one working orientation"
Title: Re: BFR landing cradle discussion and updates
Post by: Lar on 02/10/2017 01:20 pm
Would this also be possible or is there some kind of unavoidable inflexibility in ground systems that would prevent it?

1. booster lands in whichever of the 3 possible orientations
2. landing cradle rotates the booster to the orientation required for BFS stacking
3. apply BFS to booster, now in the "one working orientation"

We're assuming fill/drain of propellants in something analogous to the "tail service masts" I think.

So unless you have some very complex flexible or bendable plumbing (more than just whatever pivot you need for the tail service mast like thing to engage/release the fill/drain ports) I'm gonna go with no, it has to land in the right orientation. BUT maybe the holder could rotate before the tail service masts engaged? They'd have to swing back a fair bit. I would think the more flexible your plumbing needs to be, the more places it could leak?
Title: Re: BFR landing cradle discussion and updates
Post by: rakaydos on 02/10/2017 05:46 pm
We may all be overthinking this.

Assuming the CoG of the returning stage is below the top of the notches in the 3 "fins" of the booster. Assume the OUTER (lateral) EDGES of those notches are reinforced to bear the rocket's weight.
Edit: And assuming a "good enough" landing- >1 degree rotation error, >1m x/y error, >2m hoverslam Z error.

You just have 3 pairs of vertical shock abosorbers with several meters of play to them, one pair for each fin, each pair with something like a 50% margin for the center of the fin to slide between them. (that is, the gap is 150% as wide as the center of the fin) Within that 50% margin, the landing mount doesnt care about extra precision- the piping and boarding ramp has enough articulation to handle all remaining possible locations and orentations, and all loads are strictly vertical.
Title: Re: BFR landing cradle discussion and updates
Post by: Oersted on 02/11/2017 09:11 pm
Some considerations:

1) The KISS rule. The system really should be as simple as possible. Preferably completely passive and not relying on complex interactions.

2) Minimal shift of the rocket on capture is preferable. Forced lateral shifting of a meter or more sounds like a very dangerous thing which will put lots of stress on the craft.

My schematic idea is this:

(http://dalsgaard.eu/images/2017-02-11.BFR.Earth.landing.by.Soeren.Dalsgaard.800px.jpg)

- A very big horizontal grid, which I think would be a perfect capture device for three tapered column-like legs pointing straight down from the circumference of the booster. The same legs used for the Mars landing, but leaving the pads up on the rocket hull (the pads can be latched and unlatched from the tip of the legs in the system proposed here).

Imagine, if you will, these pencil-shaped legs pointing vertically down from the booster: the tips will slide into whichever holes in the grid they happen to descend on. Sideways movement will be minimal: half the width of any grid hole. As the taper reaches the size of the grid it will lock the rocket well in place.

Lateral movement on landing will be taken up by the pistons connected half-way up the legs (which served to extend the legs for the Mars landing).

Vertical forces on landing are partly taken up by the taper but primarily by the compression mechanism of the leg. I am thinking that a system similar to the Apollo Lunar Module landing leg struts will be used, containing a simple crushable honeycomb material dimensioned for two landings, Mars and Earth.

The landing pad grid happens to be a perfect configuration for receiving a rocket blast - which just blows through it - and at the same time being a supremely sturdy construction. I imagine it placed on four legs. After landing four crawlers (similar to the NASA crawler-transporters) can drive up to the legs, lock onto them, and then turn and shift the grid platform as needed, orienting it for optimal access from the access tower. Any damaged grids can be quickly swapped out by a team of welders.

The booster would have similar legs. Just straight up-and-down tapered poles, with a crushable honeycomb somewhere along the pole. They would be lowered vertically down through a cylinder along the outer skin of the booster and click into place at max. extension. I think it would be possible to have some cushioned "play" in the mounting at maximum extension allowing for a bit of lateral dampening movement from the poles upon landing.

This is the simplest and most flexible landing capture device I can imagine. No need for a totally pinpoint landing and an overall passive system without complex realtime interactions needed.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 02/12/2017 08:13 am
Some considerations:

1) The KISS rule. The system really should be as simple as possible. Preferably completely passive and not relying on complex interactions.

2) Minimal shift of the rocket on capture is preferable. Forced lateral shifting of a meter or more sounds like a very dangerous thing which will put lots of stress on the craft.

My schematic idea is this:

(http://dalsgaard.eu/images/2017-02-11.BFR.Earth.landing.by.Soeren.Dalsgaard.800px.jpg)

- A very big horizontal grid, which I think would be a perfect capture device for three tapered column-like legs pointing straight down from the circumference of the booster. The same legs used for the Mars landing, but leaving the pads up on the rocket hull (the pads can be latched and unlatched from the tip of the legs in the system proposed here).

Imagine, if you will, these pencil-shaped legs pointing vertically down from the booster: the tips will slide into whichever holes in the grid they happen to descend on. Sideways movement will be minimal: half the width of any grid hole. As the taper reaches the size of the grid it will lock the rocket well in place.

Lateral movement on landing will be taken up by the pistons connected half-way up the legs (which served to extend the legs for the Mars landing).

Vertical forces on landing are partly taken up by the taper but primarily by the compression mechanism of the leg. I am thinking that a system similar to the Apollo Lunar Module landing leg struts will be used, containing a simple crushable honeycomb material dimensioned for two landings, Mars and Earth.

The landing pad grid happens to be a perfect configuration for receiving a rocket blast - which just blows through it - and at the same time being a supremely sturdy construction. I imagine it placed on four legs. After landing four crawlers (similar to the NASA crawler-transporters) can drive up to the legs, lock onto them, and then turn and shift the grid platform as needed, orienting it for optimal access from the access tower. Any damaged grids can be quickly swapped out by a team of welders.

This is the simplest and most flexible landing capture device I can imagine. It allows for less precise landings while still minimizing the sideways movement on landing.   
You are solving the wrong problem. The issue is Booster (1st stage) landing capture. Not the ITS spaceship itself - it already has landing legs - it does not need landing capture.
Title: Re: BFR landing cradle discussion and updates
Post by: Oersted on 02/12/2017 12:13 pm
Sorry, I forgot to put the bit about the booster in my post. Regarding the booster landing it is basically the same system.
Title: Re: BFR landing cradle discussion and updates
Post by: Ludus on 02/22/2017 05:58 am
The video shows the booster landing back on the launch cradle while the tanker variant BFS is sitting next to the launch site where it can be moved by the launch tower crane on top of the Booster when it returns. It seems to imply that the Tanker (which like the BFS has legs) lands back on its pad which is within the radius of the tower crane along with the launch cradle.

The system might be modified to allow the Booster launch cradle to be separate from it's landing cradle. The launch tower crane would have several specialized installations within its radius. If the launch tower crane can lift the BFS(Tanker) and raise it from a pad within its radius to stack on the BFR, it might also be designed to lift the BFR (empty) off of it's landing cradle and put it on it's launch cradle. It doesn't seem less plausible to have a separate landing cradle within the operating radius of the tower crane, than it is to have a landing pad for the BFS(Tanker) within that radius.

This allows the launch and landing cradles to be separate designs. It also allows you to transport BFR/BFS/BFS(Tanker) to/from the pad on a crawler (that has a mount to hold BFR in place). The crawler track just has to be extended so it can approach within the radius of the tower crane, the elevation doesn't matter.

It seems within the spirit of rapid reusability for the launch tower crane to move the Booster from one mount to another, or move a BFS or boosters between a mount and a crawler. Boosters and ships do have to be moved around locally anyway.
Title: Re: BFR landing cradle discussion and updates
Post by: DanielW on 02/22/2017 03:37 pm
I had a crazy thought. They are going to need a big hammerhead crane to put the booster and ITS in place. Why not just over size it a bit and have it catch the booster.

Instead of putting big guide fins on the bottom they could add a bit of reinforcement to the top and grappling hook extending from the center. The crane would just have to be able to slew pretty quickly to snag it with a cable or cable net.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 02/22/2017 03:46 pm
While the video obviously simplifies the infrastructure around the pad, it is CLEAR on the intention of the booster landing on the launch mount itself. (and Musk has been clear on this too)

Now obviously an empty booster can be moved off or on the pad... it won't be built on the pad.
Title: Re: BFR landing cradle discussion and updates
Post by: Paul_G on 02/22/2017 04:00 pm
Do we know where the tanker would land. If each Mars shot needs 5 or so tankers to to top up the spaceship going to Mars, those tankers are going to have to land somewhere close to the pad where the BFR will already be, so that it can be picked up by the crane, put on top of the BFR and refueled ready to go again.

Paul
Title: Re: BFR landing cradle discussion and updates
Post by: DanielW on 02/22/2017 04:12 pm
While the video obviously simplifies the infrastructure around the pad, it is CLEAR on the intention of the booster landing on the launch mount itself. (and Musk has been clear on this too)

Now obviously an empty booster can be moved off or on the pad... it won't be built on the pad.

Yes, I agree that their intention is to land on the mount directly. I am just wondering if there aren't ways to simplify the problem. Catching with the crane adds steps and is not as elegant, but perhaps it could allow for a simpler launch mount, wider tolerances, and some flexibility on where the rocket needs to come down.

If they needed to avoid damage to the mount from the return exhaust they could just as easily catch the booster on some other side of the tower, perhaps even over a retaining pond full of water.

Just an idea if the ideal situation of just plunking it right back down where you want it does not pan out.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 02/22/2017 04:29 pm
While the video obviously simplifies the infrastructure around the pad, it is CLEAR on the intention of the booster landing on the launch mount itself. (and Musk has been clear on this too)

Now obviously an empty booster can be moved off or on the pad... it won't be built on the pad.

Yes, I agree that their intention is to land on the mount directly. I am just wondering if there aren't ways to simplify the problem. Catching with the crane adds steps and is not as elegant, but perhaps it could allow for a simpler launch mount, wider tolerances, and some flexibility on where the rocket needs to come down.

If they needed to avoid damage to the mount from the return exhaust they could just as easily catch the booster on some other side of the tower, perhaps even over a retaining pond full of water.

Just an idea if the ideal situation of just plunking it right back down where you want it does not pan out.

Catching with a crane is a non-starter. The forces involved would be incredible... No it will be tricky enough for a massive structure just on the ground.

I actually would not be surprised if they end up going with a separate landing mount - within the pad area - but that is not their stated intent at the moment, and we should take them at their word. No Rube Goldberg landing catchers need apply.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/22/2017 05:19 pm
While the video obviously simplifies the infrastructure around the pad, it is CLEAR on the intention of the booster landing on the launch mount itself. (and Musk has been clear on this too)

Now obviously an empty booster can be moved off or on the pad... it won't be built on the pad.

Yes, I agree that their intention is to land on the mount directly. I am just wondering if there aren't ways to simplify the problem. Catching with the crane adds steps and is not as elegant, but perhaps it could allow for a simpler launch mount, wider tolerances, and some flexibility on where the rocket needs to come down.

If they needed to avoid damage to the mount from the return exhaust they could just as easily catch the booster on some other side of the tower, perhaps even over a retaining pond full of water.

Just an idea if the ideal situation of just plunking it right back down where you want it does not pan out.

Catching with a crane is a non-starter. The forces involved would be incredible... No it will be tricky enough for a massive structure just on the ground.

I actually would not be surprised if they end up going with a separate landing mount - within the pad area - but that is not their stated intent at the moment, and we should take them at their word. No Rube Goldberg landing catchers need apply.

Since there are several minutes between launch and landing, the landing mechanism can be shielded during launch.  It is easier to move some armor plates away than it is to move the whole booster later.

I actually don't think the mechanism is that complex.

Think about it like a rolling landing gear, but attached to the ground, not the plane.  An A380 landing gear takes up some vertical velocity at touch down, on a comparably heavy body, and forces it to roll horizontally on the runway, right?

So all you'll have here are three landing-gear-scale mechanisms, with double-cone rollers that give the fins their final alignment, and that'd be it.

Travel on the suspension of the "landing gears" will exceed the worst-case landing error, and that'll be it.

Once the rocket is latched, the landing gears can retract out or be covered, and not be affected by the launch.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 02/22/2017 08:16 pm
Oh I agree it is a solvable problem.

But unless SpaceX only plans on ever building one ITS booster, there will be some rotation - perhaps even multiple pads. So having an additional/separate landing cradle as backup nearby would also be useful to simplify booster rotation.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/22/2017 08:31 pm
I dont.

I bet the launch mount will not be the same as the landing mount
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/22/2017 09:20 pm
Oh I agree it is a solvable problem.

But unless SpaceX only plans on ever building one ITS booster, there will be some rotation - perhaps even multiple pads. So having an additional/separate landing cradle as backup nearby would also be useful to simplify booster rotation.

You absolutely need multiple launch pads.  Consider there are people already on Mars - you can't be limited to a single pad.  So of course they'll start with one pad, but they'll need 2 or 3.

Since each booster has to fly tens of times per launch campaign (at least!), there is no advantage of having the landing pad separate from the launch pad, at least from the point of view of survivability.  If you "only" lose the landing pad, your ability to launch is equally screwed.

In fact, you need to consider which is the greater risk - launch or landing?  Today, arguably landing still is.  But once landing is no longer a "cross your fingers" maneuver, having the landing at the launch pad doesn't significantly increase the risk to the pad.

But handling the rocket, moving it from landing pad to launch pad - that will add complexity, and for no good reason. 



Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 12:08 am
Risk of landing at the launch pad vs not?    It is very much a significant and large increase of risk to the pad
There is no risk when not landing at it.   Landing accident is a higher probability than launch accident.

Moving the rocket from landing pad to launch pad is less complex and risky than landing at the pad.  And there are very good reason not to land at the pad.   

Landing at the launch mount is much like cross feed.  Sounds great but implementation is ......

Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 01:05 am
This logic only works if the scenario of "only destroying the landing pad" is significantly better.  But for a highly reusable rocket, the two pads are equally important.

If you launch from pad A and land at pad B, then an accident at either pad completely takes out the system.  Plus. you need the ground transport system.

Instead, use the same real estate (pad, exclusion zone, etc) and build two launch/land pads - and you have redundancy.

Also, is landing really more dangerous than launch?   The energies involved are certainly only a small fraction of launch, and you have a functioning well-behaved vehicle that can divert up until the last few seconds...   These are not the experimental barge landings.  I am reasonably confident that if the vehicle is not healthy, they won't crash it into the launch pad - they will bring it down into a crash pit.  You don't have such options in a launch.  If it's not going well, it's going to fry the launch pad - there's no way to "get it away from it".

----

Nobody has experience trying to do this.  It is a mistake to project expendable, or even F9 experience directly onto the ITS pad design.


Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 02:07 am
T I am reasonably confident that if the vehicle is not healthy, they won't crash it into the launch pad - they will bring it down into a crash pit.  You don't have such options in a launch. 


What says they have that option?   And if it is not healthy, what says it can divert.  Also, things can go bad after the point of divert (like running out of fuel).
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 02:09 am
Never said landing is more dangerous than than launch.  Just that it is probably of success is less and hence put the pad at more risk.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 02:19 am
There is no risk-free option, and of course a landing rocket may malfunction in the absolute last minute.  But we know all about "dead-man's" divert, so if the launch pad is at risk, the "window of doom" is really short.  It can literally be a few seconds, if the pad is built from scratch with that in mind.

But to state that "landings have a higher risk of failure than launches" - neither you nor anyone else has enough information to make that determination, and SpaceX has more information than anyone - especially since they also know what they intend to design.

What I did above was simply show that in many respects, landings are less risky than launches.  One thing is for sure - landings will get a lot less risky than they are today - by more than launches will.

And again - if a landing failed and took out the landing pad, then what use is the launch pad anymore?  Each mission is comprised of many launches, using a highly reusable booster.  If the landing pad got destroyed, you're equally screwed.

---

As for options, both a launch pad and a landing pad require the actual hardware, and a safety buffer zone that needs to be evacuated during operations.   Of course the launch pad is larger, but as I pointed out, beyond year 1, you need multiple launch pads anyway because you need contingencies - people are expecting supplies on Mars.


Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/23/2017 03:39 am
And again - if a landing failed and took out the landing pad, then what use is the launch pad anymore?  Each mission is comprised of many launches, using a highly reusable booster.  If the landing pad got destroyed, you're equally screwed.

O, rly? That line of argument makes sense if you have either no backup boosters or no secondary landing site for the campaign.

The landing pad is a concrete slab: easily replaced, repaired, or substituted. The launch pad and its associated tower, crane, tanks, densifiers, and other GSE? Not so much.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 03:53 am
And again - if a landing failed and took out the landing pad, then what use is the launch pad anymore?  Each mission is comprised of many launches, using a highly reusable booster.  If the landing pad got destroyed, you're equally screwed.

O, rly? That line of argument makes sense iff you have no backup boosters for the campaign.

The landing pad is a concrete slab: easily replaced, repaired, or substituted. The launch pad and its associated tower, crane, tanks, densifiers, and other GSE? Not so much.

The landing pad in the ITS concept is more complex - it captures the rocket.  You're arguing that it's simpler to replace - which means you need several of them (you certainly don't want to stand down for 2 months while the Mars window slips away)...

So if you've built several of them, the "simple solution" is no longer simple - you now have one launch pad, two landing pads, a transport system...  Booster transfer operations...

Meanwhile, the launch pad is still exposed to the launches, so you didn't cut the risk by much - if a rocket fails on take-off, it'll take out the launch pad.

---

The "separate pads" strategy only makes sense if you consider the landing to be very high risk, but if you do, then you don't have a Mars program.

That's what I meant by projecting the current state of affairs (only 8 landings ever achieved, with a first-iteration model) into the design of a system that's designed to launch and land multiple times per day.

My best guess is:
- More than one launch site (after the first synod),
- Booster lands as depicted, into the launch pad
- A crash pit built into the design of the pad so that even a very late failure does not destroy the pad
- A strict landing criteria that if it fails, the rocket is scuttled into the crash pit.  (No Hail-Mary landing attempts like were done with F9.

Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/23/2017 05:35 am
I was reacting to your statement of equivalence between launch and landing complexes.

SpaceX assumes greater risk by landing on the launch mount, period. It isn't solely about the odds of disaster during launch vs. landing, but also the consequences of each type of disaster. The launch pad is a more valuable asset, and especially so for a campaign with many launches and (likely) redundant vehicles.

So yes, in the separate pad scenario "only destroying the landing pad" is significantly better.

I'm not arguing they won't land on the launch pad; let's take them at their word. To your point, no wanting to schlep the booster around must be the crux of it. But don't argue that losing either launch or landing are equivalent.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 06:03 am
I was reacting to your statement of equivalence between launch and landing complexes.

SpaceX assumes greater risk by landing on the launch mount, period. It isn't solely about the odds of disaster during launch vs. landing, but also the consequences of each type of disaster. The launch pad is a more valuable asset, and especially so for a campaign with many launches and (likely) redundant vehicles.

So yes, in the separate pad scenario "only destroying the landing pad" is significantly better.

I'm not arguing they won't land on the launch pad; let's take them at their word. To your point, no wanting to schlep the booster around must be the crux of it. But don't argue that losing either launch or landing are equivalent.


"Period" doesn't cut it.

If you look at it holistically, at the system level, for what SpaceX is trying to achieve, then landing at the pad makes a lot more sense, even from a risk perspective.

The key here is that you need to compare risk, to the whole system, between the two scenarios - and not only concentrate on "OMG what if a landing booster crashes into the launch pad" - because there are many other "OMG" risks and they play out differently on the two alternatives.

Do you remember the deal with Toray fiber, that first gave us a clue about how many BFSs SpaceX intends to launch?  And now that we know that there will be 5x tanker launches for each BFS going out - all within a launch window - the sheer scale this system is intended to operate at?

It's hard to think at those scales when all we have right now is an F9 once per month, but when you're talking about hundreds and hundreds of launches per synod, things look different, and this system is designed for that.

So first, if you land on a separate pad, you've added more steps to the cycling of the booster, which is the hardest working and largest component of the system. There's risk right there.

And the time to crawl from the landing pad to the launch pad - this reduced pad launch rate.  So now for a given target launch rate, you need more pads - and if you're back to more pads, what exactly did you gain?

You've also doubled your real-estate requirement, but didn't gain redundancy.  If instead you'd have built another launch pad where that landing pad is, you'd have double the capacity.

And for all of that - you've hardly cut risk.  A failure during launch has a high probability of destroying the pad.  A failure during landing, in most cases, will not hit the pad.  So your launch pad is seeing a very similar amount of risk, even if you have a separate landing pad.

--------------

Just like F9 landings, the first few trials will be risky, until the bugs are worked out.  I'd say that during these development flights, it'll make sense to land on a test cradle off-pad.  But that's about it.  Once it starts to auto-land reliably, the odds of a failure that damages the launch pad are really small.

Title: Re: BFR landing cradle discussion and updates
Post by: Semmel on 02/23/2017 06:47 am
I don't agree meekGee. The launch pad has all the ground support equipment that a landing mount does not need. I would argue that a landing mount would be only one tenth of the cost and complexity of a launch mount, even if you factor in transport. So it could be beneficial to have one launch mount and many landing mounts.

Also, Jim and others were talking about the absolute risk of damaging the launch pad. If you have two independent events, launch and landing, the risk to the launch pad is higher compared to only launch. Adding the difference in complexity, separate launch and landing mounts can reduce risk and the expected damage (i.e. Bad day probably times damage over a given time period)
Title: Re: BFR landing cradle discussion and updates
Post by: MikeAtkinson on 02/23/2017 10:53 am
Even if they do eventually land on the launch pad, initially it makes sense to have a separate landing pad. The first landings are likely to be risky, it may take several attempts to land in the cradle successfully, and you would not want to have to rebuild the launch pad several times.
Title: Re: BFR landing cradle discussion and updates
Post by: francesco nicoli on 02/23/2017 11:44 am
This logic only works if the scenario of "only destroying the landing pad" is significantly better.  But for a highly reusable rocket, the two pads are equally important.

If you launch from pad A and land at pad B, then an accident at either pad completely takes out the system.  Plus. you need the ground transport system.

Instead, use the same real estate (pad, exclusion zone, etc) and build two launch/land pads - and you have redundancy.

Also, is landing really more dangerous than launch?   The energies involved are certainly only a small fraction of launch, and you have a functioning well-behaved vehicle that can divert up until the last few seconds...   These are not the experimental barge landings.  I am reasonably confident that if the vehicle is not healthy, they won't crash it into the launch pad - they will bring it down into a crash pit.  You don't have such options in a launch.  If it's not going well, it's going to fry the launch pad - there's no way to "get it away from it".

----

Nobody has experience trying to do this.  It is a mistake to project expendable, or even F9 experience directly onto the ITS pad design.

My impression is that landing pads are extremely simpler to build and refurbish than launching pads. Therefore, they are not equal in value, even if they are equal in importance for the mission.
Transportation is an issue, but I think it's a secondary issue in respect to doubling the risk of losing your launching pad.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 01:38 pm

But to state that "landings have a higher risk of failure than launches" - neither you nor anyone else has enough information to make that determination, and SpaceX has more information than anyone - especially since they also know what they intend to design.


Quite wrong.  It is very easy to see and intuitive to see that "landings have a higher risk of failure than launches".  And it is true. The vehicle is going through a worse environment during return vs ascent.

And you are wrong about the information.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/23/2017 02:36 pm

"Period" doesn't cut it.

If you look at it holistically, at the system level, for what SpaceX is trying to achieve, then landing at the pad makes a lot more sense, even from a risk perspective.

The key here is that you need to compare risk, to the whole system, between the two scenarios - and not only concentrate on "OMG what if a landing booster crashes into the launch pad" - because there are many other "OMG" risks and they play out differently on the two alternatives.

Do you remember the deal with Toray fiber, that first gave us a clue about how many BFSs SpaceX intends to launch?  And now that we know that there will be 5x tanker launches for each BFS going out - all within a launch window - the sheer scale this system is intended to operate at?

It's hard to think at those scales when all we have right now is an F9 once per month, but when you're talking about hundreds and hundreds of launches per synod, things look different, and this system is designed for that.

So first, if you land on a separate pad, you've added more steps to the cycling of the booster, which is the hardest working and largest component of the system. There's risk right there.

And the time to crawl from the landing pad to the launch pad - this reduced pad launch rate.  So now for a given target launch rate, you need more pads - and if you're back to more pads, what exactly did you gain?

You've also doubled your real-estate requirement, but didn't gain redundancy.  If instead you'd have built another launch pad where that landing pad is, you'd have double the capacity.

And for all of that - you've hardly cut risk.  A failure during launch has a high probability of destroying the pad.  A failure during landing, in most cases, will not hit the pad.  So your launch pad is seeing a very similar amount of risk, even if you have a separate landing pad.

--------------

Just like F9 landings, the first few trials will be risky, until the bugs are worked out.  I'd say that during these development flights, it'll make sense to land on a test cradle off-pad.  But that's about it.  Once it starts to auto-land reliably, the odds of a failure that damages the launch pad are really small.

There is an efficiency argument and a risk argument. Then there's a decision based on evaluating the two. We know the apparent outcome of that decision because we've seen the ITS video.

What more needs to be said? Clearly efficiency won the day. Nobody is saying "OMG risk!", so please don't try to pin that on anyone. If anything, things got a bit hyperbolic upthread with "If the landing pad got destroyed, you're equally screwed" [as the launch pad].

Tell me, are you really arguing the risk introduced by moving a booster between pads rises to the level of launch or landing? C'mon. Some efficiency lost? Sure. Significant risk added? No.

Landing on the launch mount is an efficiency win when rapid booster reflight is the objective.

What happens when the booster to launch pad ratio is no longer 1:1? With more than one booster the launch pad becomes the bottleneck for a flight campaign. Rotating boosters between launches becomes an option.

We don't know, of course, how this will trade. If booster turnaround time > 'sliding new booster up to the pad' time, you may want to be landing at a separate site. Best to keep our minds open.

I know we all want this to work like a plane. But when a 777 is undergoing significant maintenance they take it to the hangar and don't leave it occupying a gate.



edit: typo
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 02:40 pm

But to state that "landings have a higher risk of failure than launches" - neither you nor anyone else has enough information to make that determination, and SpaceX has more information than anyone - especially since they also know what they intend to design.


Quite wrong.  It is very easy to see and intuitive to see that "landings have a higher risk of failure than launches".  And it is true. The vehicle is going through a worse environment during return vs ascent.

And you are wrong about the information.
With all due respect, nobody knows how reliable BFR landings will be, and how risky they'll be compared to launches.

As for intuition, two years ago I suggested that for truly rapid reuse, the first stage has to stay vertical throughout its life cycle, and that the second stage has to be integrated to it vertically.

I didn't know of course about the scale of the rocket, but that just made it more obvious.

According to your combined information and intuition, that was a non starter. It was against their "mantra" of horizontal integration.

Ironically, what you're arguing for now is the exact same thing you ruled out then...

But landing back at the pad is just the logical extension of the concept.

Landings can be made incredibly safe to the pad if you have a robust scuttle criteria for the rocket.  The landing may fail, but the pad will be safe.

You have to accept that things change.  Even from what they changed to just a year ago.

Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 02:53 pm


Landings can be made incredibly safe to the pad if you have a robust scuttle criteria for the rocket. 


There is no such thing now
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 03:00 pm

Ironically, what you're arguing for now is the exact same thing you ruled out then...

No, I am not.  And your institution was wrong.  What I stated for F9 two years ago is still applicable.  And your intuition is not applicable to the ITS.  There is no separate payload and second stage.

This is a different vehicle and one that has a drastically different fineness ratio and shape.

And landing at separate pad, doesn't preclude breaking it over.

Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 03:05 pm
Landing for a propulsive rocket returning from ascent is always going to be more risky than the launch it self.
Title: Re: BFR landing cradle discussion and updates
Post by: Semmel on 02/23/2017 03:57 pm
Landing for a propulsive rocket returning from ascent is always going to be more risky than the launch it self.

Why? I don't find that obvious at all.
Title: Re: BFR landing cradle discussion and updates
Post by: GabrielP on 02/23/2017 04:25 pm
For example, if your landing engine(s) don't start or suffer some kind of anomaly that impairs control before the landing burn, you are pretty much guaranteed to get undesirable fireworks.
If the same thing happens before launch, you have the option of being able to stop, fix the problem, then launch.
Title: Re: BFR landing cradle discussion and updates
Post by: R7 on 02/23/2017 05:06 pm
Landing for a propulsive rocket returning from ascent is always going to be more risky than the launch it self.

Why? I don't find that obvious at all.

For starters it is obvious that at T-0 rocket's velocity is guaranteed to be zero. T-landing velocity depends on how well you master a tricky control problem. If it's not zero then you might need a new rocket.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 05:30 pm


Landings can be made incredibly safe to the pad if you have a robust scuttle criteria for the rocket. 


There is no such thing now

Exactly.  Now, there isn't.  If you're gong to land at the launch pad, then of course there will be.

Seriously - do you think SpaceX has missed the risk implications of landing at the pad?

It's just that they have a much better handle on what they plan to design, (or have designed already), what the alternatives are, and how the risks will play out.

Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 05:35 pm
Landing for a propulsive rocket returning from ascent is always going to be more risky than the launch it self.

Why? I don't find that obvious at all.

Yup. Once you commit to launch, you have a fuel-laden rocket that cannot simply "move away" from the pad, and generally a highly energetic process, so if the launch fails, the pad is destroyed.

During landing, you have an empty rocket, that by default will miss the pad and hit the crash pit.  While the landing itself may fail, if it doesn't directly hit the pad, it'll just hit concrete.  You just need to build the pad with a good bunker wall.

With F9 and the Barge, they tried to land no-matter-what, since they wanted the pieces...  So the landing that actually caused damage could have been diverted to the water - but wasn't.

The landing that actually landed gently and collapsed - cause minor damage.

So even with F9, you can already see that landing will be safer.
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 02/23/2017 05:53 pm
Even if the odds of an accident during launch and landing were the same, using the pad for both doubles the risk of an accident disabling the pad.

Using the crawlers at KSC, there can be at least two movable cradles for launch and landing. Land near, but not at the pad and let a crawler move the booster to the pad for launch.

An accident during landing would not disable the pad. An accident at launch would still disable the pad, but recovery would be quicker with a backup cradle ready to go.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 06:09 pm
that by default will miss the pad and hit the crash pit.

Unqualified assumption.  There is no such thing as a crash pit.  Falcon doesn't have one.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 06:12 pm

Seriously - do you think SpaceX has missed the risk implications of landing at the pad?


Yes,  See AMOS-6
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 06:21 pm

Seriously - do you think SpaceX has missed the risk implications of landing at the pad?


Yes,  See AMOS-6

Proving that the risk is not necessarily in the landing...   And demonstrating that the intensity of the damage from a fueled rocket is huge compared to that of a landing rocket (see the tip-over barge landing, whichever that one was).



Title: Re: BFR landing cradle discussion and updates
Post by: matthewkantar on 02/23/2017 06:26 pm
There was the one crash landing that holed the deck of the ASDS. I would think a hit like that would put a pad out of commission.

Matthew
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 06:31 pm
There was the one crash landing that holed the deck of the ASDS. I would think a hit like that would put a pad out of commission.

Matthew

Exactly right - and this crash landing was "known bad" early in the landing process, and could have been splashed safely.  They chose a Hail-Mary attempt because those are experimental landings on a barge which doesn't affect launch capability.

That's exactly how you make the landings pad-safe.  You drag the IIP into the pad, and so the only crash that will actually hit the pad is a slow moving, basically fuel-less rocket.  And we know that this didn't damage the barge.

Everything else crashes off-pad, behind a barrier you put there.

I mean - they clearly know that the launch pad is important... Why wouldn't they do that?
Title: Re: BFR landing cradle discussion and updates
Post by: JasonAW3 on 02/23/2017 06:56 pm
Possibly a dumb thought here, but why not use a sort of reverse Russian launch gantry system?

      When the Russians launch their Soyuz, the towers supporting the rocket fall back as the rocket rises, allowing it to clear them without incident.  (Yes, I know most if not all of you know this already, but some few people might not have seen a Soyuz launch, unlikely though that may be).

      In this case, five or more towers would be equidistantly spaced around the landing/launch cradle, with straps going between the contact surfaces of each of the towers.  As the booster passes a certain altitude, the towers would be released from their down position, using counter weights and physical braking, to begin rising towards the booster.  While this is happening, the straps going between the towers would be tightening in pace with the towers rising.  This system would help nudge a slightly off center (or maybe a bit more than slightly off center) into the center of the cradle ring.

      The towers should actually stop the booster a few feet above the cradle, so it could be finely aligned and lowered onto the cradle properly.

      The towers could then be used as a support for the whole BFR prior to launch, (after removing the strapping between towers, and adjusting the counterweights so that the towers fall back during launch) in the same fashion as the Russian gantry towers.

      It's a crude mechanical system that, with a bit of refining, could prove useful.
Title: Re: BFR landing cradle discussion and updates
Post by: meberbs on 02/23/2017 06:58 pm
that by default will miss the pad and hit the crash pit.

Unqualified assumption.  There is no such thing as a crash pit.  Falcon doesn't have one.
Um.... What?

There have been slides showing that the pre-landing burn trajectory for RTLS would ditch in the ocean in case of an engine restart failure or similar.

There seems to be a fair bit of talking past each other going on:

Is there more risk to the pad with Launch + landing vs just launch? of course.
Is there more risk to the rocket during landing than launch? Almost certain.
Is there more risk to the pad from landing than launch? Unclear: Landing failures would generally be less energetic (less fuel) and it is reasonable to think that most landing failures (with design to account for the probable ones) would be such that a ditch would occur rather than hitting the pad.

What needs to be compared is "chance of landing failure that impacts pad" times "severity of damage" compared to benefits of quick turn around from landing at the launch pad.

The comparison of launch and landing while useful for thinking about isn't actually relevant to the equation, so the answer to the third rhetorical question doesn't actually matter.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 07:00 pm
Possibly a dumb thought here, but why not use a sort of reverse Russian launch gantry system?

      When the Russians launch their Soyuz, the towers supporting the rocket fall back as the rocket rises, allowing it to clear them without incident.  (Yes, I know most if not all of you know this already, but some few people might not have seen a Soyuz launch, unlikely though that may be).

      In this case, five or more towers would be equidistantly spaced around the landing/launch cradle, with straps going between the contact surfaces of each of the towers.  As the booster passes a certain altitude, the towers would be released from their down position, using counter weights and physical braking, to begin rising towards the booster.  While this is happening, the straps going between the towers would be tightening in pace with the towers rising.  This system would help nudge a slightly off center (or maybe a bit more than slightly off center) into the center of the cradle ring.

      The towers should actually stop the booster a few feet above the cradle, so it could be finely aligned and lowered onto the cradle properly.

      The towers could then be used as a support for the whole BFR prior to launch, (after removing the strapping between towers, and adjusting the counterweights so that the towers fall back during launch) in the same fashion as the Russian gantry towers.

      It's a crude mechanical system that, with a bit of refining, could prove useful.

Conceptually, that's not too far from what will be happening....

Except the rocket is designed to be held by its thrust structure, and is kinda fragile along its body, so the mechanism will work down there...

And instead (IMO) of actively coming up to meet the rocket, the slant in the bottom of the fin will create this "closing of the cage" simply due to the downward motion of the rocket...  Like any docking cone...

Everything else remains the same!   :)
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/23/2017 07:06 pm
After the entry burn, the vehicle is heading for the landing site.  The next burn is the landing one.
Title: Re: BFR landing cradle discussion and updates
Post by: meberbs on 02/23/2017 07:09 pm
After the entry burn, the vehicle is heading for the landing site.  The next burn is the landing one.
Have you really never seen this infographic (https://www.reddit.com/r/spacex/comments/3xieex/falcon_9_launch_and_landing_infographic/)?
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 07:23 pm
After the entry burn, the vehicle is heading for the landing site.  The next burn is the landing one.

You don't know where the IIP is during all this time.  The grid fins can give the rocket an AoA, and walk it from off shore to the pad.

That, plus this pad is just a concrete surface.  You can't argue that BFR will put the launch pad at risk because of something F9 does.

---

And again - you're ironically arguing for the very same thing (landing vertically nearby, translating over) that you vehemently opposed a year ago on account of it being dumb and needless and not viable etc.

So once more, what was heresy last year became gospel today, and the next logical step in the process is deemed heresy.

Should we just wait a year or two and save all this back-and-forth?

--

I started this thread to discuss the cradle.  Didn't mean for it to become a thread about whether the cradle is a dumb idea.  Not completely off-topic, but it'd be great if we drift back to talking about cradle implementation ideas.
Title: Re: BFR landing cradle discussion and updates
Post by: JasonAW3 on 02/23/2017 09:00 pm
Possibly a dumb thought here, but why not use a sort of reverse Russian launch gantry system?

      When the Russians launch their Soyuz, the towers supporting the rocket fall back as the rocket rises, allowing it to clear them without incident.  (Yes, I know most if not all of you know this already, but some few people might not have seen a Soyuz launch, unlikely though that may be).

      In this case, five or more towers would be equidistantly spaced around the landing/launch cradle, with straps going between the contact surfaces of each of the towers.  As the booster passes a certain altitude, the towers would be released from their down position, using counter weights and physical braking, to begin rising towards the booster.  While this is happening, the straps going between the towers would be tightening in pace with the towers rising.  This system would help nudge a slightly off center (or maybe a bit more than slightly off center) into the center of the cradle ring.

      The towers should actually stop the booster a few feet above the cradle, so it could be finely aligned and lowered onto the cradle properly.

      The towers could then be used as a support for the whole BFR prior to launch, (after removing the strapping between towers, and adjusting the counterweights so that the towers fall back during launch) in the same fashion as the Russian gantry towers.

      It's a crude mechanical system that, with a bit of refining, could prove useful.

Conceptually, that's not too far from what will be happening....

Except the rocket is designed to be held by its thrust structure, and is kinda fragile along its body, so the mechanism will work down there...

And instead (IMO) of actively coming up to meet the rocket, the slant in the bottom of the fin will create this "closing of the cage" simply due to the downward motion of the rocket...  Like any docking cone...

Everything else remains the same!   :)

In a strong crosswind, which the Cape does get in a fairly short time frame, the booster could be blown off course by several meters.

The current concept might need a bit of tweaking in this case.

As to structure load, I didn't take that into consideration.  As the entire main structure is supposed to be carbon fiber, I was under the impression that it should have a fairly good structure all round.  I wasn't expecting a 'beer can' type of structure.  (Good in verticals compression, but buckles inward from the sides like a empty can with lateral compression).

Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/23/2017 09:11 pm
Possibly a dumb thought here, but why not use a sort of reverse Russian launch gantry system?

      When the Russians launch their Soyuz, the towers supporting the rocket fall back as the rocket rises, allowing it to clear them without incident.  (Yes, I know most if not all of you know this already, but some few people might not have seen a Soyuz launch, unlikely though that may be).

      In this case, five or more towers would be equidistantly spaced around the landing/launch cradle, with straps going between the contact surfaces of each of the towers.  As the booster passes a certain altitude, the towers would be released from their down position, using counter weights and physical braking, to begin rising towards the booster.  While this is happening, the straps going between the towers would be tightening in pace with the towers rising.  This system would help nudge a slightly off center (or maybe a bit more than slightly off center) into the center of the cradle ring.

      The towers should actually stop the booster a few feet above the cradle, so it could be finely aligned and lowered onto the cradle properly.

      The towers could then be used as a support for the whole BFR prior to launch, (after removing the strapping between towers, and adjusting the counterweights so that the towers fall back during launch) in the same fashion as the Russian gantry towers.

      It's a crude mechanical system that, with a bit of refining, could prove useful.

Conceptually, that's not too far from what will be happening....

Except the rocket is designed to be held by its thrust structure, and is kinda fragile along its body, so the mechanism will work down there...

And instead (IMO) of actively coming up to meet the rocket, the slant in the bottom of the fin will create this "closing of the cage" simply due to the downward motion of the rocket...  Like any docking cone...

Everything else remains the same!   :)

In a strong crosswind, which the Cape does get in a fairly short time frame, the booster could be blown off course by several meters.

The current concept might need a bit of tweaking in this case.

As to structure load, I didn't take that into consideration.  As the entire main structure is supposed to be carbon fiber, I was under the impression that it should have a fairly good structure all round.  I wasn't expecting a 'beer can' type of structure.  (Good in verticals compression, but buckles inward from the sides like a empty can with lateral compression).
Did you calculate the "few meters"?

Because it depends on factors we don't know. For example:

How much slam in the hoverslam ?

How much force can the side thrusters generate?

Will there be communication of real-time wind data back up to the booster?

---

That said, 1-2 meters, on a 12 m cylinder, should be possible with pure mechanical centering.  Maybe even more.
Title: Re: BFR landing cradle discussion and updates
Post by: gospacex on 02/24/2017 01:30 am
After the entry burn, the vehicle is heading for the landing site.  The next burn is the landing one.

You don't know where the IIP is during all this time.  The grid fins can give the rocket an AoA, and walk it from off shore to the pad.

That, plus this pad is just a concrete surface.  You can't argue that BFR will put the launch pad at risk because of something F9 does.

In any case, landing on a launch pad has a nonzero chance of damaging it. There are ways to lessen it, such as IIP management, but it can't be made zero.

Landing on a separate landing pad eliminates this risk, at a cost of needing additional hardware and time to move landed stage.

Unless you aim at a situation where you have many redundant pads and need to support launch rate of ~1 launch/day or even more, the second approach looks better to me.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/24/2017 01:34 am
After the entry burn, the vehicle is heading for the landing site.  The next burn is the landing one.

You don't know where the IIP is during all this time.  The grid fins can give the rocket an AoA, and walk it from off shore to the pad.

That, plus this pad is just a concrete surface.  You can't argue that BFR will put the launch pad at risk because of something F9 does.

In any case, landing on a launch pad has a nonzero chance of damaging it. There are ways to lessen it, such as IIP management, but it can't be made zero.

Landing on a separate landing pad eliminates this risk, at a cost of needing additional hardware and time to move landed stage.

Unless you aim at a situation where you have many redundant pads and need to support launch rate of ~1 launch/day or even more, the second approach looks better to me.
That's a fair assessment.  I'll just add that IMO IIP management solves practically all the problems, since it also guarantees that if there is a last second male function, then both speed and fuel levels are very low.

We know the targeted launch rate is > 1/day, and I can't imagine they'll run the Mars effort out of 1 pad.
Title: Re: BFR landing cradle discussion and updates
Post by: gospacex on 02/24/2017 02:41 am
No market for that in a foreseeable future.
With launch rate > 1/day, Musk can launch entire Earth's yearly launch manifest in some 3 months.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 02/24/2017 04:12 am
A booster will cost roughly the same as a launch/landing pad (+/- 50%). If you can already tolerate the booster failure, then I don't see why you can't just build more pads as backup to start with. You have to get the booster failure rate low enough anyway, or all this calculus is moot.

And nominally, you also have more throughput if you build an extra launchpad beforehand.
Title: Re: BFR landing cradle discussion and updates
Post by: meberbs on 02/24/2017 04:38 am
No market for that in a foreseeable future.
With launch rate > 1/day, Musk can launch entire Earth's yearly launch manifest in some 3 months.
This entire discussion only applies to a future where ITS is built. In such a future, the "market" is the 6-7 booster flights per ship sent to Mars, and the ability to scale to sending dozens of ships to support a growing colony. While some cargo ships could be pre-placed, I am guessing most flights would occur in a 3-9 month period. 30 ships is 180 flights minimum, and they want to be able to go well past that. At least 1 launch per day is pretty much a minimum requirement for them. Aiming for less than that would mean they have already failed.

A booster will cost roughly the same as a launch/landing pad (+/- 50%). If you can already tolerate the booster failure, then I don't see why you can't just build more pads as backup to start with. You have to get the booster failure rate low enough anyway, or all this calculus is moot.

And nominally, you also have more throughput if you build an extra launchpad beforehand.

Yeah, if they don't have a very low failure rate <1%, there is no way they can sustain what they are planning. I think they will accept the risk of only 1 pad at first, but once they are supporting people on mars, they will need additional pads as redundancy. They might only have 1, maybe 2 boosters total for the first few synods (This is really strange now that I realize it). This kind of changes the implication of your logic.
Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 02/24/2017 08:19 am
Yeah, if they don't have a very low failure rate <1%, there is no way they can sustain what they are planning.

To reuse a booster 1000 times you need <<0.1%.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 02/24/2017 11:40 am
Yeah, if they don't have a very low failure rate <1%, there is no way they can sustain what they are planning.

To reuse a booster 1000 times you need <<0.1%.
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).
Title: Re: BFR landing cradle discussion and updates
Post by: mnelson on 02/24/2017 12:19 pm
To reuse a booster 1000 times you need <<0.1%.
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).

If the failure rate is 0.1% then you will not average 1000 launches per booster. The probability of making 1000 *consecutive* successful launches with a single booster would be 37% (0.999^1000).

I order to have a 50/50 chance of making 1000 launches with a booster you need a 0.07% failure rate.

I'm struck by just how far the industry is from achieving anywhere near 0.07% failure rate. Seems like a pretty big technological leap independent of the other technological leaps being assumed. So far no one has gotten within a couple orders of magnitude.
Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 02/24/2017 01:13 pm

To reuse a booster 1000 times you need <<0.1%.
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).

I read it as a boster can fly 1000 times. If I want to achieve this for most of the boosters, it must be much better than 0.1%. I stick with <<.

Though I expect boosters to have failures occasionally but would be able to recover and land anyway. The upper stage would separate and land independently after burning off its propellant.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 02/24/2017 01:28 pm
To reuse a booster 1000 times you need <<0.1%.
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).

If the failure rate is 0.1% then you will not average 1000 launches per booster. The probability of making 1000 *consecutive* successful launches with a single booster would be 37% (0.999^1000).

I order to have a 50/50 chance of making 1000 launches with a booster you need a 0.07% failure rate.

I'm struck by just how far the industry is from achieving anywhere near 0.07% failure rate. Seems like a pretty big technological leap independent of the other technological leaps being assumed. So far no one has gotten within a couple orders of magnitude.
Nah, it still works. The reason why is because you'll get some boosters that last a lot longer than 1000. It averages out.
Title: Re: BFR landing cradle discussion and updates
Post by: gospacex on 02/24/2017 04:02 pm
A booster will cost roughly the same as a launch/landing pad (+/- 50%). If you can already tolerate the booster failure, then I don't see why you can't just build more pads as backup to start with.

In a mishap, you are losing the booster AND the launch pad, so it still costs significantly more.

Many failure modes which would result in a mishap when you are trying to land on the launch pad are completely survivable with the landing pad approach: wind gust? No big deal, we landed 2m off the mark.
Title: Re: BFR landing cradle discussion and updates
Post by: mnelson on 02/24/2017 05:31 pm
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).

If the failure rate is 0.1% then you will not average 1000 launches per booster. The probability of making 1000 *consecutive* successful launches with a single booster would be 37% (0.999^1000).

I order to have a 50/50 chance of making 1000 launches with a booster you need a 0.07% failure rate.

I'm struck by just how far the industry is from achieving anywhere near 0.07% failure rate. Seems like a pretty big technological leap independent of the other technological leaps being assumed. So far no one has gotten within a couple orders of magnitude.
Nah, it still works. The reason why is because you'll get some boosters that last a lot longer than 1000. It averages out.

Nah, your handwaving. At a 0.1% failure rate only 37% of boosters making it to 1000 cycles. Another way to look at it is that a 0.1% failure rate "averages out" to an expect life of 692 cycles - half will fail sooner and half will last longer.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/24/2017 06:35 pm
A booster will cost roughly the same as a launch/landing pad (+/- 50%). If you can already tolerate the booster failure, then I don't see why you can't just build more pads as backup to start with.

In a mishap, you are losing the booster AND the launch pad, so it still costs significantly more.

Many failure modes which would result in a mishap when you are trying to land on the launch pad are completely survivable with the landing pad approach: wind gust? No big deal, we landed 2m off the mark.

Except you may not be landing at all, since the ITS booster has no legs.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 02/25/2017 06:38 pm
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).

If the failure rate is 0.1% then you will not average 1000 launches per booster. The probability of making 1000 *consecutive* successful launches with a single booster would be 37% (0.999^1000).

I order to have a 50/50 chance of making 1000 launches with a booster you need a 0.07% failure rate.

I'm struck by just how far the industry is from achieving anywhere near 0.07% failure rate. Seems like a pretty big technological leap independent of the other technological leaps being assumed. So far no one has gotten within a couple orders of magnitude.
Nah, it still works. The reason why is because you'll get some boosters that last a lot longer than 1000. It averages out.

Nah, your handwaving. At a 0.1% failure rate only 37% of boosters making it to 1000 cycles. Another way to look at it is that a 0.1% failure rate "averages out" to an expect life of 692 cycles - half will fail sooner and half will last longer.
I'm not hand waving, you just don't understand the situation. If you keep operating all the boosters that don't fail, and you have a reliability of 99.9%, you'll get 1000 reuses out of them on average. Some of those 37% may keep going, with 37% of /them/ getting 2000 reuses.

Heck, if you stop using the boosters after 1000 uses, you'll need perfect reliability to get 1000 uses on average. The way you're setting up the problem is to make it impossible for SpaceX no matter how good they do.
Title: Re: BFR landing cradle discussion and updates
Post by: gospacex on 02/25/2017 07:13 pm
A booster will cost roughly the same as a launch/landing pad (+/- 50%). If you can already tolerate the booster failure, then I don't see why you can't just build more pads as backup to start with.

In a mishap, you are losing the booster AND the launch pad, so it still costs significantly more.

Many failure modes which would result in a mishap when you are trying to land on the launch pad are completely survivable with the landing pad approach: wind gust? No big deal, we landed 2m off the mark.

Except you may not be landing at all, since the ITS booster has no legs.

I think the setup with separate landing pads assumes that booster will have legs.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/25/2017 07:37 pm
Not <<, just <, to reuse a booster an average of 1000 times (sometimes more, sometimes less).

If the failure rate is 0.1% then you will not average 1000 launches per booster. The probability of making 1000 *consecutive* successful launches with a single booster would be 37% (0.999^1000).

I order to have a 50/50 chance of making 1000 launches with a booster you need a 0.07% failure rate.

I'm struck by just how far the industry is from achieving anywhere near 0.07% failure rate. Seems like a pretty big technological leap independent of the other technological leaps being assumed. So far no one has gotten within a couple orders of magnitude.
Nah, it still works. The reason why is because you'll get some boosters that last a lot longer than 1000. It averages out.

Nah, your handwaving. At a 0.1% failure rate only 37% of boosters making it to 1000 cycles. Another way to look at it is that a 0.1% failure rate "averages out" to an expect life of 692 cycles - half will fail sooner and half will last longer.
I'm not hand waving, you just don't understand the situation. If you keep operating all the boosters that don't fail, and you have a reliability of 99.9%, you'll get 1000 reuses out of them on average. Some of those 37% may keep going, with 37% of /them/ getting 2000 reuses.

Heck, if you stop using the boosters after 1000 uses, you'll need perfect reliability to get 1000 uses on average. The way you're setting up the problem is to make it impossible for SpaceX no matter how good they do.

Hmm... It's almost as if y'all are referring to "a value which is to be expected", statistically speaking, in a "distribution which is one of randomness"?
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/25/2017 07:40 pm
In a mishap, you are losing the booster AND the launch pad, so it still costs significantly more.

Many failure modes which would result in a mishap when you are trying to land on the launch pad are completely survivable with the landing pad approach: wind gust? No big deal, we landed 2m off the mark.

Except you may not be landing at all, since the ITS booster has no legs.

I think the setup with separate landing pads assumes that booster will have legs.

Respectfully disagree. All the upthread discussion of separate landing pads assumes a landing cradle.

This begs the question: what kind of weight penalty has SpaceX avoided with a leg-free booster design? We've been obsessing over the operational efficiency of the landing cradle located on the launch pad. Is it possible the motivating factor for the design is the presence of a landing cradle at all, thereby enabling a leg-free and lighter-weight booster?
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/25/2017 07:43 pm
A booster will cost roughly the same as a launch/landing pad (+/- 50%). If you can already tolerate the booster failure, then I don't see why you can't just build more pads as backup to start with.

In a mishap, you are losing the booster AND the launch pad, so it still costs significantly more.

Many failure modes which would result in a mishap when you are trying to land on the launch pad are completely survivable with the landing pad approach: wind gust? No big deal, we landed 2m off the mark.

Except you may not be landing at all, since the ITS booster has no legs.

I think the setup with separate landing pads assumes that booster will have legs.

These are the observations that make cradle landing on the pad a good idea:

- A rocket is very unlikely to "miss by a little"
- In the unlikely case that the rocket misses by a little (that is, hits the physical cradle, but still fails) it is guaranteed that it will be practically empty of fuel, and have almost no kinetic energy.
- A pad that is hardened enough to survive a BFR launch, and that can be made to be "clean" by the time landing happens (all service masts etc. pull away) can survive either of the above failure scenarios.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/25/2017 07:47 pm
Last sentence is wrong.  There is a difference between surviving blast vs impact
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/25/2017 10:20 pm
Last sentence is wrong.  There is a difference between surviving blast vs impact

And there can't be a significant impact, since if the rocket hits the cradle, it means it is traveling slowly, because of IIP management as described above.

The nice thing about impacts is that if the rocket misses even by a little bit, the momentum is transferred to the ground. (unlike explosions)  Like a bullet that misses you by a hair - a miss is a miss.

I'm also assuming that all the "delicate" stuff, from TSMs to BFS umbilicals will be retracted and shielded in the >5 minutes that elapse between launch and landing.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/25/2017 11:29 pm

And there can't be a significant impact, since if the rocket hits the cradle, it means it is traveling slowly, because of IIP management as described above


Wrong again.  This is a massive vehicile.  And IIP management isn't going to help if it happens during engine burn.

Also, the ' IIP management as described above" is wrong.  The infographic is wrong.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 12:06 am

And there can't be a significant impact, since if the rocket hits the cradle, it means it is traveling slowly, because of IIP management as described above


Wrong again.  This is a massive vehicile.  And IIP management isn't going to help if it happens during engine burn.

Also, the ' IIP management as described above" is wrong.  The infographic is wrong.

It's a massive vehicle, and also a massive pad and cradle.

IIP management can absolutely occur during engine burn.  The engine burn helps drift the IIP onto the cradle.

If the IIP gets to the cradle, it means that the vehicle has slowed down to a very low velocity, and it also means there's almost no fuel left in the vehicle.

The infographic is irrelevant.  Do you have some inside information about how BFR is going to land?

What we're talking about is how it CAN land safely while minimizing risk to the pad/cradle.  You're saying it can't, so you need to show what you're basing that on.
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 02/26/2017 01:30 am


IIP management can absolutely occur during engine burn.  The engine burn helps drift the IIP onto the cradle.


The vehicle and IIP is over the landing pad before the final burn.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/26/2017 01:55 am


IIP management can absolutely occur during engine burn.  The engine burn helps drift the IIP onto the cradle.


The vehicle and IIP is over the landing pad before the final burn.

You're saying all horizontal velocity must be cancelled before the landing burn? That isn't the case with F9 today.
Title: Re: BFR landing cradle discussion and updates
Post by: Patchouli on 02/26/2017 03:28 am
I expect the concept will change a lot as it's developed to accept the fact there will be errors in alignment and they may even forgo having it land at the launch site and the cradle concept all together.

Looks at how F9 went from splash down recovery to them deciding to land it.
Title: Re: BFR landing cradle discussion and updates
Post by: Lar on 02/26/2017 04:54 am
Topic of the thread is landing cradle engineering/ideas/implications, etc....

"Landing cradles are a bad idea" is off topic. They might be. At this stage in what we know I think they are. But this isn't the thread to argue that.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 02/26/2017 05:57 am
I think the setup with separate landing pads assumes that booster will have legs.

Respectfully disagree. All the upthread discussion of separate landing pads assumes a landing cradle.

This begs the question: what kind of weight penalty has SpaceX avoided with a leg-free booster design? We've been obsessing over the operational efficiency of the landing cradle located on the launch pad. Is it possible the motivating factor for the design is the presence of a landing cradle at all, thereby enabling a leg-free and lighter-weight booster?

Yes. That is indeed the primary reason for a leg-less design. As a vehicle scales up, the leg need to scale up in mass and strength much faster. A booster scaled 2x in all dimensions will mass ~ 8X, forcing the legs to become a much more significant part of the vehicle. Or to put it another way: Landing legs scale poorly with size increases.

Using a landing cradle is a way of putting *most* of that leg mass on the ground instead... The landing cradle handles the dampening of the final motion.

Combining the landing cradle with the launch mount is simply intended to increase flight frequency, nothing more.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 06:03 am


IIP management can absolutely occur during engine burn.  The engine burn helps drift the IIP onto the cradle.


The vehicle and IIP is over the landing pad before the final burn.

That might be true today, though I suspect you don't have the details of their trajectory.

The important thing is, we're talking about BFR and a pad-cradle.  If the landing cradle is on the pad, and the pad is valuable, then clearly they'll walk the IIP onto the pad until the very last second.

They do that anyway, effectively, when they're fighting wind during the landing.

Your whole argument that "landing at the launch pad is foolish" because "they won't walk the IIP since they don't walk it today" is circular.  You're proposing a foolish way of landing, and then saying they can't do it safely because it's foolish...

Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 06:06 am
I think the setup with separate landing pads assumes that booster will have legs.

Respectfully disagree. All the upthread discussion of separate landing pads assumes a landing cradle.

This begs the question: what kind of weight penalty has SpaceX avoided with a leg-free booster design? We've been obsessing over the operational efficiency of the landing cradle located on the launch pad. Is it possible the motivating factor for the design is the presence of a landing cradle at all, thereby enabling a leg-free and lighter-weight booster?

Yes. That is indeed the primary reason for a leg-less design. As a vehicle scales up, the leg need to scale up in mass and strength much faster. A booster scaled 2x in all dimensions will mass ~ 8X, forcing the legs to become a much more significant part of the vehicle. Or to put it another way: Landing legs scale poorly with size increases.

Using a landing cradle is a way of putting *most* of that leg mass on the ground instead... The landing cradle handles the dampening of the final motion.

Combining the landing cradle with the launch mount is simply intended to increase flight frequency, nothing more.

Exactly.  And even more:

To increase X-Y precision (the statistical deviation from "nailing it", you need to increase the vertical velocity at capture.  The more you "hover", the more difficult it is to get X-Y exactly right.  (Think about the ultimate example - a guided air-to-ground missile that's not even trying to stop...)

If you want to increase the vertical velocity, you need longer travel in the shock absorbers of the legs.  That would make them even heavier.

By leaving this mechanism on the ground, you allow higher Vz at beginning of capture, therefore allowing better X-Y precision...

Everything really works together here..
Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 02/26/2017 07:32 am
No legs probably require more fuel for a precision landing. So I doubt no legs transfers to less weight. IMO the cradle approach serves the purpose of simplified pad operations and fast turnaround. They do plan for launching at least once a day. So for redundancy they need a full second pad ASAP.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 07:37 am
No legs probably require more fuel for a precision landing. So I doubt no legs transfers to less weight. IMO the cradle approach serves the purpose of simplified pad operations and fast turnaround. They do plan for launching at least once a day. So for redundancy they need a full second pad ASAP.
Why more fuel?
Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 02/26/2017 07:39 am
No legs probably require more fuel for a precision landing. So I doubt no legs transfers to less weight. IMO the cradle approach serves the purpose of simplified pad operations and fast turnaround. They do plan for launching at least once a day. So for redundancy they need a full second pad ASAP.
Why more fuel?

For a slower, more precise final landing approach. Elon Musk said as much, I believe in his IAC presentation.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 07:42 am
No legs probably require more fuel for a precision landing. So I doubt no legs transfers to less weight. IMO the cradle approach serves the purpose of simplified pad operations and fast turnaround. They do plan for launching at least once a day. So for redundancy they need a full second pad ASAP.
Why more fuel?

For a slower, more precise final landing approach. Elon Musk said as much, I believe in his IAC presentation.
Even if it will be slower, a few seconds worth of gravity losses can equal the weight of the legs...

And since the rocket is lighter, you gain even more fuel savings on all three burns, irrespective of the final landing speed.

Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 02/26/2017 08:59 am
OK accepting that no legs equals less fuel even with slower precision landing.

My point that they will not want a separate pure landing cradle to facilitate fast turn around stands.
Title: Re: BFR landing cradle discussion and updates
Post by: OneSpeed on 02/26/2017 10:21 am
... As a vehicle scales up, the leg need to scale up in mass and strength much faster. A booster scaled 2x in all dimensions will mass ~ 8X, forcing the legs to become a much more significant part of the vehicle. Or to put it another way: Landing legs scale poorly with size increases.

The nominal strength of a structure is independent of the structure size when geometrically similar structures are considered. Any deviation from this property is called the size effect, and is only encountered when the relative sizes differ by orders of magnitude. For a booster scaled 2x, all of its components will mass 8x, and their strengths will scale equally. Landing legs scale no differently than any other part of the rocket.
Title: Re: BFR landing cradle discussion and updates
Post by: cartman on 02/26/2017 11:07 am
My opinion is that what we saw at IAC was Elon's idea of how things would look like when they have achieved their goals for ITS. So this is their vision, not how ITS will look like in its first years.
My guess is that the first generation may be with legs, until they manage to never miss their target by more than a meter or so. Then they remove the legs and land onto a pure landing cradle. Once they manage to do that reliably and in general fly without incidents for hundreds of flights then they will begin launching from the landing cradle. Experience will help refine all the designs. At the beginning they will have large fuel reserves and will only fly in very low wind conditions and as they get more comfortable they will lower the reserves and loosen the weather constraints. A tall wind barrier could help here.
Title: Re: BFR landing cradle discussion and updates
Post by: livingjw on 02/26/2017 01:14 pm

Yes. That is indeed the primary reason for a leg-less design. As a vehicle scales up, the leg need to scale up in mass and strength much faster. A booster scaled 2x in all dimensions will mass ~ 8X, forcing the legs to become a much more significant part of the vehicle. Or to put it another way: Landing legs scale poorly with size increases.

Using a landing cradle is a way of putting *most* of that leg mass on the ground instead... The landing cradle handles the dampening of the final motion.

Combining the landing cradle with the launch mount is simply intended to increase flight frequency, nothing more.

Landing legs are probably in the neighborhood of 1% of the landing weight, independent of size.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 03:09 pm
OK accepting that no legs equals less fuel even with slower precision landing.

My point that they will not want a separate pure landing cradle to facilitate fast turn around stands.
Indeed.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/26/2017 04:44 pm
OK accepting that no legs equals less fuel even with slower precision landing.

My point that they will not want a separate pure landing cradle to facilitate fast turn around stands.

Unless:

i.  There is more than one booster available for a given launch pad
ii.  Flight rate improves by rotating the boosters
Title: Re: BFR landing cradle discussion and updates
Post by: Patchouli on 02/26/2017 05:13 pm
One option could be they make the cradle transportable similar to the shuttle mobile launch platform and have several of them.

That way landing occurs away from the service tower etc and if a mishap does occur only the cradle and booster are damaged.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 09:10 pm
One option could be they make the cradle transportable similar to the shuttle mobile launch platform and have several of them.

That way landing occurs away from the service tower etc and if a mishap does occur only the cradle and booster are damaged.

True.  But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.

It's possible, just like landing on a dedicated pad is possible, but it's not a no-brainer - it's a trade-off.

I can't imagine SpaceX didn't cover all three options before deciding on "back-to-the-pad".

I'm not even going to say they'll 100% stick with their choice. They're nothing if not agile.  But to say that "back-to-the-pad" is clearly dumb because they obviously didn't factor in risk to the pad is just nonsense - that was the whole point of the argument upthread.

And now, as Lar hinted - back to the landing cradle itself.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/26/2017 09:15 pm
But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.

They'll need a transporter regardless of where it lands. Extra processing time, yes, that's the tradeoff.
Title: Re: BFR landing cradle discussion and updates
Post by: RobLynn on 02/26/2017 10:12 pm
Landing legs are probably in the neighborhood of 1% of the landing weight, independent of size.

Current Falcon 9 landing legs are "less than a model S" ie less than 2000kg.  assuming clsoe to the upper limit then that is closer to 5-10% of landed weight than 1%.  Of course this is pretty close to a worst case design given wide base for very tall thin stage and aero drag incorporated.

I believe helicopters are about 2% and airliners around 5-8% of MTOW. 

Landing leg (and aircraft undercarriage) mass is dependant on landed mass and length to be bridged.  Bigger longer nozzles mean longer legs, and larger diameter stages also add greater length to be bridged by extra thrust structure or longer legs or adding bending stiffness to the tank to make it more square-bottomed.

1% sounds very optimistic for a Ř12m rocket like ITS.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/26/2017 10:30 pm
But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.

They'll need a transporter regardless of where it lands. Extra processing time, yes, that's the tradeoff.

For the BFR?  As part of the launch cycle?  Why would they? 
Or are you referring to transport from the assembly site?  This may be true, unless they assemble it on a different pad and fly it in. (Such a staging pad also solves the whole "rocket rotation" issue.)
Title: Re: BFR landing cradle discussion and updates
Post by: RobLynn on 02/28/2017 09:31 am
A high speed cable driven motion simulator:
https://www.youtube.com/watch?v=W7nGL-_5SDw
Perhaps this has potential for adaption and scaling up into an actively positioned landing cradle?  It is nice in that it keeps the cradle light with most of the positioning control hardware far from the rocket exhaust, and has potential to accommodate a lot of inaccuracy in speed and height to reduce the landing guidance demands.

With arrestor hooks on the BFR or ITS vehicle it could potentially even absorb some of the falling velocity (or taken to extremes with 3 huge towers all of the landing fuel could be eliminated.
Title: Re: BFR landing cradle discussion and updates
Post by: Hobbes-22 on 02/28/2017 12:03 pm
One option could be they make the cradle transportable similar to the shuttle mobile launch platform and have several of them.

That way landing occurs away from the service tower etc and if a mishap does occur only the cradle and booster are damaged.

True.  But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.


Transporters are far less rare than they used to be. You can rent (or buy) self-propelled modular transporters that can easily scale (by combining modules) to the size needed to transport a BFR.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/28/2017 02:20 pm
One option could be they make the cradle transportable similar to the shuttle mobile launch platform and have several of them.

That way landing occurs away from the service tower etc and if a mishap does occur only the cradle and booster are damaged.

True.  But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.


Transporters are far less rare than they used to be. You can rent (or buy) self-propelled modular transporters that can easily scale (by combining modules) to the size needed to transport a BFR.
He was talking about transporters into which the rocket lands.
Title: Re: BFR landing cradle discussion and updates
Post by: Hobbes-22 on 02/28/2017 02:25 pm
One option could be they make the cradle transportable similar to the shuttle mobile launch platform and have several of them.

That way landing occurs away from the service tower etc and if a mishap does occur only the cradle and booster are damaged.

True.  But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.


Transporters are far less rare than they used to be. You can rent (or buy) self-propelled modular transporters that can easily scale (by combining modules) to the size needed to transport a BFR.
He was talking about transporters into which the rocket lands.

Like the MLP used for the Shuttle, you'd split the fireproof ironmongery from the transporter.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/28/2017 02:31 pm
One option could be they make the cradle transportable similar to the shuttle mobile launch platform and have several of them.

That way landing occurs away from the service tower etc and if a mishap does occur only the cradle and booster are damaged.

True.  But now you've added a crawler transporter, which is a rare beast and much more complex then "just a concrete pad", a system for moving BFR from the transporter to the pad, and extra processing time.


Transporters are far less rare than they used to be. You can rent (or buy) self-propelled modular transporters that can easily scale (by combining modules) to the size needed to transport a BFR.
He was talking about transporters into which the rocket lands.

Like the MLP used for the Shuttle, you'd split the fireproof ironmongery from the transporter.
OK, so the rare thing becomes the movable cradle/flame-duct.

The transporter meanwhile has to handle not just static carrying, but the forces and moments of capture.

And still you have a transfer operation to the pad itself.

Like I said above - all these variants are feasible, but none of them is "simply simpler", and they are all pretty obvious, and so were not selected for the current plan.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 02/28/2017 07:47 pm
Simply simpler than what? The ITS video showed us but a snapshot.

meekGee, you contend upthread that a secondary (non-launch pad) landing cradle would necessitate the creation of a booster transporter. 

I assert they will need a transporter regardless. How could they get by without one? SpaceX would need to:

1. Assemble the booster vertically
2. On a site from which the booster can launch
3. Fly it to a site for test-firing
4. Fly it to the primary/departure launch pad
5. Never have more than one booster per primary launch pad

Now we entertain the thought exercise whereby the locations of 1 and 2 can be consolidated with 3, or 3 can be consolidated with 4, or 1 through 4 are consolidated. But it starts to seem a bit unlikely, doesn't it? And we haven't even gotten to 5.

Unless the launch tower unleashes a factory of nano-assemblers and ITS springs into existence on the launch pad itself, the booster will need to be moved and transported outside of its own power.

I love the idea of a rocket remaining vertical throughout its lifetime, but don't see it happening for ITS.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 02/28/2017 08:15 pm
Simply simpler than what? The ITS video showed us but a snapshot.

meekGee, you contend upthread that a secondary (non-launch pad) landing cradle would necessitate the creation of a booster transporter. 

I assert they will need a transporter regardless. How could they get by without one? SpaceX would need to:

1. Assemble the booster vertically
2. On a site from which the booster can launch
3. Fly it to a site for test-firing
4. Fly it to the primary/departure launch pad
5. Never have more than one booster per primary launch pad

Now we entertain the thought exercise whereby the locations of 1 and 2 can be consolidated with 3, or 3 can be consolidated with 4, or 1 through 4 are consolidated. But it starts to seem a bit unlikely, doesn't it? And we haven't even gotten to 5.

Unless the launch tower unleashes a factory of nano-assemblers and ITS springs into existence on the launch pad itself, the booster will need to be moved and transported outside of its own power.

I love the idea of a rocket remaining vertical throughout its lifetime, but don't see it happening for ITS.

Context, please.

What the vid showed was a booster that lands in a cradle co-located with the launch pad.

A bunch of people jumped up crying "that's insane, the risk, why don't they land somewhere else", which boils down to two alternative:

(1) Land on stationary cradles, and have a "simple" transporter take the rocket (vertical?  horizontal?) back to the pad - thus requiring two transfer operations.

(2) Land on a movable cradle-transporter, drive back to the pad (vertical?  horizontal?), and have one transfer operation.

I was showing that both alternate systems add complexity and time, but don't save a whole lot of risk.

You're arguing they need a transporter anyway, but:

- First, even if they do, it's not a cradle-transporter anymore (so strike option 2)
- Second, for MRO and such, time is not a problem like it is within an hours-turn-around operation, so having a transporter doesn't mean they want to use it every launch cycle.
Title: Re: BFR landing cradle discussion and updates
Post by: RobLynn on 03/22/2017 05:23 am
Automatic bullseye dartboard (moving board around using servo controlled winches), includes visual processing of dart trajectory.
https://www.youtube.com/watch?v=MHTizZ_XcUM
Just need to scale it up a little :)
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 03/22/2017 07:01 am
Automatic bullseye dartboard (moving board around using servo controlled winches), includes visual processing of dart trajectory.
https://www.youtube.com/watch?v=MHTizZ_XcUM
Just need to scale it up a little :)
And clearly it can catch three cores in quick succession too.

Just as long as they don't mislabel the cores and cause it to miss on purpose.
Title: Re: BFR landing cradle discussion and updates
Post by: Norm38 on 03/23/2017 02:52 am
The shuttle transporter had to carry the entire STS.  Orbiter, main tank, payload, and full solid rocket boosters.

This transporter would have to carry an empty booster stage.  Quite a bit less mass to move.
Title: Re: BFR landing cradle discussion and updates
Post by: RoboGoofers on 03/23/2017 03:56 pm
I've attached some images from the ITS reveal video to show the clearest public example of what SpaceX currently invisions.

there are two (out of focus) arms in the first picture, so presumably there are 6 arms, with 2 on each side of the fins, or there are 3 arms with 2 fingers each.

The mound or volcano (roost?) it sit on says to me that however they roll the booster up to the pad, they'll use the crane to put it in place.

the underside of the pad shows 3 'supports' 120 deg. opposed, but they they might not be static, maybe being part of the landing cradle.

also if the fins slide into keyways, they don't slide in much since they stick out of the ground >90%.
Title: Re: BFR landing cradle discussion and updates
Post by: dglow on 03/23/2017 04:02 pm
Thanks, RoboGoofers. Not a lot of cradle-ness in that video. What we're shown is about a very dramatic reveal of the 42-engine cluster, and not much more.
Title: Re: BFR landing cradle discussion and updates
Post by: RoboGoofers on 03/23/2017 04:29 pm
Thanks, RoboGoofers. Not a lot of cradle-ness in that video. What we're shown is about a very dramatic reveal of the 42-engine cluster, and not much more.

Yeah the images don't reveal much but they seemingly eliminate a lot.
Title: Re: BFR landing cradle discussion and updates
Post by: Lars-J on 03/24/2017 04:05 am
Thanks, RoboGoofers. Not a lot of cradle-ness in that video. What we're shown is about a very dramatic reveal of the 42-engine cluster, and not much more.

Yeah the images don't reveal much but they seemingly eliminate a lot.

Do not take anything about the ground infrastructure in that video as anything but an artist impression.
Title: Re: BFR landing cradle discussion and updates
Post by: RoboGoofers on 03/24/2017 02:38 pm
Thanks, RoboGoofers. Not a lot of cradle-ness in that video. What we're shown is about a very dramatic reveal of the 42-engine cluster, and not much more.

Yeah the images don't reveal much but they seemingly eliminate a lot.

Do not take anything about the ground infrastructure in that video as anything but an artist impression.

I don't know... The F9R animation with the Muse soundtrack got a lot of details right.

It's certainly better to assume it's close to SpaceX's plan. Otherwise we get ideas about grabbing the stage out of the air with the crane, or giant robot arms.
Title: Re: BFR landing cradle discussion and updates
Post by: Toast on 03/24/2017 03:13 pm
I don't know... The F9R animation with the Muse soundtrack got a lot of details right.

It's certainly better to assume it's close to SpaceX's plan. Otherwise we get ideas about grabbing the stage out of the air with the crane, or giant robot arms.

I agree that it's probably a lot closer to reality than any of the Rube Goldberg-esque contraptions people have dreamed up. That said, the difference between the ITS unveiling video and the F9R video is that the F9R video came out after Falcon 9 was already flying (late 2011, if I'm remembering right). In contrast, the Interplanetary Transportation System isn't even a finalized design, much less in production.
Title: Re: BFR landing cradle discussion and updates
Post by: BeamRider on 04/10/2017 04:48 pm
Just a fine point: the blast effect on landing will be a fraction of that on liftoff. It might be possible to laterally retract the landing capture hardware for launch, when blast effect will be worst, and then deploy it for landing. Think of the lean-back operation of the new transporter erector as a rough analog.
Title: Re: BFR landing cradle discussion and updates
Post by: rakaydos on 04/10/2017 06:36 pm
So basically, we're arguing pros and cons of 2-3 different landing architectures.

1)Colocated landing/liftoff cradles, payload integration on the pad, no transporter use during a multi-launch campaign.
2)Landing cradle, transfer to crawler, crawler movement, transfer to pad, payload integration on pad.

I think we can all agree there is more risk in option 1 than option 2. Option 2 is also far slower than option 1, and has more moving parts and more staff involved.

IF the risk of option 1 can be made managable, it clearly offers reduced operational costs, at least for nominal operations.

CAN option 1's risk be made managable? It will always be riskier than the same effort put into option 2. But can the risk be made low enough to recoup any disaster costs from the lower operational costs?

Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 04/10/2017 07:56 pm
So basically, we're arguing pros and cons of 2-3 different landing architectures.

1)Colocated landing/liftoff cradles, payload integration on the pad, no transporter use during a multi-launch campaign.
2)Landing cradle, transfer to crawler, crawler movement, transfer to pad, payload integration on pad.

I think we can all agree there is more risk in option 1 than option 2. Option 2 is also far slower than option 1, and has more moving parts and more staff involved.

IF the risk of option 1 can be made managable, it clearly offers reduced operational costs, at least for nominal operations.

CAN option 1's risk be made managable? It will always be riskier than the same effort put into option 2. But can the risk be made low enough to recoup any disaster costs from the lower operational costs?

A combination of the two options would help mitigate risk. Operate as in option 1, but the integrated launch and landing cradle is built to be moved by a crawler. If a crash damages the cradle, once the debris is removed, use a crawler to bring in the spare cradle assemble. That should reduce the pad downtime.
Title: Re: BFR landing cradle discussion and updates
Post by: GORDAP on 04/12/2017 11:40 am
So basically, we're arguing pros and cons of 2-3 different landing architectures.

1)Colocated landing/liftoff cradles, payload integration on the pad, no transporter use during a multi-launch campaign.
2)Landing cradle, transfer to crawler, crawler movement, transfer to pad, payload integration on pad.

I think we can all agree there is more risk in option 1 than option 2. Option 2 is also far slower than option 1, and has more moving parts and more staff involved.

IF the risk of option 1 can be made managable, it clearly offers reduced operational costs, at least for nominal operations.

CAN option 1's risk be made managable? It will always be riskier than the same effort put into option 2. But can the risk be made low enough to recoup any disaster costs from the lower operational costs?

A combination of the two options would help mitigate risk. Operate as in option 1, but the integrated launch and landing cradle is built to be moved by a crawler. If a crash damages the cradle, once the debris is removed, use a crawler to bring in the spare cradle assemble. That should reduce the pad downtime.

I like this idea.  But take it one step further.  Integrate the crawler and landing cradle and have them travel on a set of rails.  Locate the landing area just 500 to 1000 meters from the launch pad.  Once the BFR lands on the cradle, just track on back to the pad with no lifting or even mechanical coupling involved.

Hmm, might need to make the cradle travel faster than just a 'crawl' (when it's empty).  It needs to motor on out to the landing area in time to be in position to catch the BFR when it's done with the boost phase.  Not a lot of time there.
Title: Re: BFR landing cradle discussion and updates
Post by: RoboGoofers on 04/14/2017 01:43 pm
It makes sense to plan for every eventuality, but an ITS crash is going to be catastrophic for SpaceX. A 6-12 month investigation of the crash and they'll miss the Mars window. In addition, the risk of living on Mars predicates landing on Mars. If they can't convince people they can do that reliably, they'll have less success recruiting the best colonists.
Title: Re: BFR landing cradle discussion and updates
Post by: chalz on 04/16/2017 12:46 pm
Is cradle landing a non-negotiable part of the ITS design? Presumably it is needed to achieve a certain mass fraction that enables a spaceship with the biggest payload possible. If ITS is intended to be super reliable and high flight rate then a lowering of maximum payload would be substituted by more flights. Therefore it seems like this will be a feature that will be dropped on the way to the first flight. Like crossfeed was dropped from Falcon Heavy development.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 04/16/2017 01:03 pm
I wouldn't be too surprised if ITS (in whatever form) debuts with legs.

However, it's not mass fraction that I think is the main advantage but turnaround time. Moving around even F9 stages takes a while with cranes, folding legs, the TEL, etc. But BFR is the size of a high rise building. Moving it around would be a huge challenge. Just about impossible to do in just a few hours, and nearly so in 24 hours. Launch cradle avoids all that by installing the BFR right on its launch mount.

So this isn't analogous to crossfeed. It's mostly about speeding & streamlining the processing to enable rapid reuse on a truly enormous scale. Otherwise, rapid reuse is not feasible with such large launch vehicles just due to the time it takes to move such enormous stages.
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 04/16/2017 03:30 pm
SpaceX will need the landing cradle when colonization begins to maintain the massive flight rate. For the initial exploration flights, a LEO fuel depot could replace the need for a rapid turnaround. SpaceX could take their time filling up the depot, then the BFS can refuel and leave for Mars without waiting around for tanker flights.
Title: Re: BFR landing cradle discussion and updates
Post by: docmordrid on 04/17/2017 04:06 am
>
....a LEO fuel depot could replace the need for a rapid turnaround. SpaceX could take their time filling up the depot, then the BFS can refuel and leave for Mars without waiting around for tanker flights.

I've thought that since the IAC presentation, but in what form factor? Its storage tanks would need to hold at least as much as the Spaceship main tanks plus boiloff. Could that fit in the Tanker and have enough for the tanker to RTLS?
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 04/17/2017 04:25 am
>
....a LEO fuel depot could replace the need for a rapid turnaround. SpaceX could take their time filling up the depot, then the BFS can refuel and leave for Mars without waiting around for tanker flights.

I've thought that since the IAC presentation, but in what form factor? Its storage tanks would need to hold at least as much as the Spaceship main tanks plus boiloff. Could that fit in the Tanker?

If SpaceX put extra tanks in what would have been the crew and cargo areas, the depot could be a variant of the BFS. Launch it, refuel with multiple tanker flights, and then use it to refuel an outgoing BFS. Since it will have engines, it can land for maintenance.
Title: Re: BFR landing cradle discussion and updates
Post by: sevenperforce on 04/18/2017 07:24 pm
>
....a LEO fuel depot could replace the need for a rapid turnaround. SpaceX could take their time filling up the depot, then the BFS can refuel and leave for Mars without waiting around for tanker flights.

I've thought that since the IAC presentation, but in what form factor? Its storage tanks would need to hold at least as much as the Spaceship main tanks plus boiloff. Could that fit in the Tanker?

If SpaceX put extra tanks in what would have been the crew and cargo areas, the depot could be a variant of the BFS. Launch it, refuel with multiple tanker flights, and then use it to refuel an outgoing BFS. Since it will have engines, it can land for maintenance.
The tanker has plenty of capacity to fuel an outgoing ITS Spaceship. The only reasonable way for this to work is that the booster launches a tanker, then sends a series of tankers up to refuel the first tanker, and then finally launches the manned spaceship which refuels from the full tanker before departing.
Title: Re: BFR landing cradle discussion and updates
Post by: CameronD on 04/20/2017 01:56 am
>
....a LEO fuel depot could replace the need for a rapid turnaround. SpaceX could take their time filling up the depot, then the BFS can refuel and leave for Mars without waiting around for tanker flights.

I've thought that since the IAC presentation, but in what form factor? Its storage tanks would need to hold at least as much as the Spaceship main tanks plus boiloff. Could that fit in the Tanker?

If SpaceX put extra tanks in what would have been the crew and cargo areas, the depot could be a variant of the BFS. Launch it, refuel with multiple tanker flights, and then use it to refuel an outgoing BFS. Since it will have engines, it can land for maintenance.
The tanker has plenty of capacity to fuel an outgoing ITS Spaceship. The only reasonable way for this to work is that the booster launches a tanker, then sends a series of tankers up to refuel the first tanker, and then finally launches the manned spaceship which refuels from the full tanker before departing.

One tiny problem with this:  The technology for on-orbit refuelling isn't mature enough (yet!) to work and there is still much research to be completed on the behaviour of liquids in a zero-G environment before it is technically possible to design and build a system to do it.

Maybe in a year or 5.. but not right now.

Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 04/20/2017 02:29 am
>
....a LEO fuel depot could replace the need for a rapid turnaround. SpaceX could take their time filling up the depot, then the BFS can refuel and leave for Mars without waiting around for tanker flights.

I've thought that since the IAC presentation, but in what form factor? Its storage tanks would need to hold at least as much as the Spaceship main tanks plus boiloff. Could that fit in the Tanker?

If SpaceX put extra tanks in what would have been the crew and cargo areas, the depot could be a variant of the BFS. Launch it, refuel with multiple tanker flights, and then use it to refuel an outgoing BFS. Since it will have engines, it can land for maintenance.
The tanker has plenty of capacity to fuel an outgoing ITS Spaceship. The only reasonable way for this to work is that the booster launches a tanker, then sends a series of tankers up to refuel the first tanker, and then finally launches the manned spaceship which refuels from the full tanker before departing.

One tiny problem with this:  The technology for on-orbit refuelling isn't mature enough (yet!) to work and there is still much research to be completed on the behaviour of liquids in a zero-G environment before it is technically possible to design and build a system to do it.

Maybe in a year or 5.. but not right now.
...uh, SpaceX has been studying the behavior of liquids in zero-g for pretty much every single Falcon 9 launch from what I can tell:
https://www.youtube.com/watch?v=PPnCKK1isMI

It's not a huge mystery for how to transfer fluids in microgravity. Just use ullage thrusters to settle the liquid and a pressure differential to transfer it from one tank to another. Start with more thrust than you think you need at first, then optimize later. It's absolutely technically possible to design this now. And they can do test launches to make sure it works, too, it's not like they just get one shot at it with no chance for tweaking.

For some really dumb reason, a lot of people seem to think fluid transfer is some exotic tech. I really don't understand where that comes from.
Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 04/20/2017 06:40 am
For some really dumb reason, a lot of people seem to think fluid transfer is some exotic tech. I really don't understand where that comes from.

But......... but TRL. It has not been done so it can not be done. At least not without a few billions and a decade of research.
Title: Re: BFR landing cradle discussion and updates
Post by: sevenperforce on 04/20/2017 12:36 pm
The tanker has plenty of capacity to fuel an outgoing ITS Spaceship. The only reasonable way for this to work is that the booster launches a tanker, then sends a series of tankers up to refuel the first tanker, and then finally launches the manned spaceship which refuels from the full tanker before departing.
One tiny problem with this:  The technology for on-orbit refuelling isn't mature enough (yet!) to work and there is still much research to be completed on the behaviour of liquids in a zero-G environment before it is technically possible to design and build a system to do it.

Maybe in a year or 5.. but not right now.
That may or may not be the case, but it doesn't really speak to the tanker capacity...
Title: Re: BFR landing cradle discussion and updates
Post by: philw1776 on 04/20/2017 01:20 pm
For some really dumb reason, a lot of people seem to think fluid transfer is some exotic tech. I really don't understand where that comes from.

But......... but TRL. It has not been done so it can not be done. At least not without a few billions and a decade of research.

I think we found the name Tory Bruno posts under here
Title: Re: BFR landing cradle discussion and updates
Post by: Jim on 04/20/2017 02:42 pm

One tiny problem with this:  The technology for on-orbit refuelling isn't mature enough (yet!) to work and there is still much research to be completed on the behaviour of liquids in a zero-G environment before it is technically possible to design and build a system to do it.

Maybe in a year or 5.. but not right now.



What technology?   on-orbit refueling has been happening for more than 30 years.  See Progress spacecraft.  Cryogenic?   Centaur has been doing cryogen fluids management for a half a century.  All that is need is a cryogenic disconnect and those already exist.
Title: Re: BFR landing cradle discussion and updates
Post by: CameronD on 04/21/2017 02:33 am

One tiny problem with this:  The technology for on-orbit refuelling isn't mature enough (yet!) to work and there is still much research to be completed on the behaviour of liquids in a zero-G environment before it is technically possible to design and build a system to do it.

Maybe in a year or 5.. but not right now.

What technology?   on-orbit refueling has been happening for more than 30 years.  See Progress spacecraft.  Cryogenic?   Centaur has been doing cryogen fluids management for a half a century.  All that is need is a cryogenic disconnect and those already exist.

Good point, Jim!  I was thinking only of refuelling as part of 'satellite servicing' tech and had forgotten about Progress/ISS..   :-[

..although both the examples you give are in Lowish Earth Orbit/microgravity and not GEO/zero gravity, if that's what they intend.
 
 
Title: Re: BFR landing cradle discussion and updates
Post by: Oersted on 04/21/2017 08:21 pm
Integrate the crawler and landing cradle and have them travel on a set of rails.  Locate the landing area just 500 to 1000 meters from the launch pad.  Once the BFR lands on the cradle, just track on back to the pad with no lifting or even mechanical coupling involved.

Hmm, might need to make the cradle travel faster than just a 'crawl' (when it's empty).  It needs to motor on out to the landing area in time to be in position to catch the BFR when it's done with the boost phase.  Not a lot of time there.

Have rails going both East and West from the launch pad. Have one crawler/cradle at the launch pad and another crawler/cradle at one of the two landing pads. After a launch & landing shift the crawler/cradles into position for the next launch & landing.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/22/2017 01:14 pm
Moving Landing Funnel

Another iteration of the ITS/BFR landing cradle is the Moving Landing Funnel. Rather than thrusters on the booster to guide the 150t 275t rocket into the cradle, the 5t funnel moves under the rocket during the last three seconds. It has no equipment for moving, just the Raptor flames hitting the sloped sides of the funnel causing it to (quickly) move toward the center of the plume. in the last second the bottom edge of the rocket will contact the sloped sides and finalize the centering. Edit correction: Dry tonnage is for booster rather than for spaceship.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/22/2017 01:15 pm
Moving Landing Funnel (2)

The base ring of the funnel has multiple roller bearings built into the bottom. These rollers allow it to move quickly in any direction in response to a force exerted against a sidewall. It acts like the dolly that an auto mechanic employs to move around underneath a car.

The base ring also has many openings around the sides as blast slots to allow the landing plume to escape in all directions during the last second.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/22/2017 01:16 pm
Moving Landing Funnel (3)

The moving landing funnel requires a smooth hard surface, such as a steel deck. In the sketch, there is a 12m round blast door with a flame trench underneath. In the open position, it accommodates the blast plume of re-launching the rocket. in the closed position, it provides a continuous flat surface to accommodate the moving funnel.

In the sketch, the landing booster is off-center from the blast door. Once landed, however, the funnel base and booster can be push-pulled into position over the launch blast door. It can also be rotated to face the fueling umbilicals. The equipment for moving the vehicle is entirely on the ground, not on the booster. If SpaceX were to adopt this option, the landing guides at the bottom of the booster would be unnecessary.

Thank you who preceded me to create this excellent thread: meekGee, Robotbeat, Jdeshetler, Req, ZachF, rakaydos, Oersted, RoboGoofers, and Jim, among others.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/27/2017 12:23 pm
Since this idea has received no comment so far, one of the following must be true: (a) Superb approach that simply must be adopted, (b) Good approach but some aspects need work, (c) Certain aspects must be  redone to avoid disaster, or (d) Inevitable RUD.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 04/27/2017 03:03 pm
Since this idea has received no comment so far, one of the following must be true: (a) Superb approach that simply must be adopted, (b) Good approach but some aspects need work, (c) Certain aspects must be  redone to avoid disaster, or (d) Inevitable RUD.

or:

(e) cute, but why, if the rocket which is already mobile can land on a stationary fixture?

Also, passive alignment such as you describe is (again) cute, but not likely to work.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/27/2017 07:28 pm
(e) cute, but why, if the rocket which is already mobile can land on a stationary fixture?

Also, passive alignment such as you describe is (again) cute, but not likely to work.
Thanks. You may be right.
To me, however, there is a trade-off in fuel usage and reliability between hovering for a longer period to get the last few meters of accuracy and just letting the rocket down hard within 2m of center. Otherwise, why F9 hoverslam, for example?
Title: Re: BFR landing cradle discussion and updates
Post by: rakaydos on 04/27/2017 08:09 pm
(e) cute, but why, if the rocket which is already mobile can land on a stationary fixture?

Also, passive alignment such as you describe is (again) cute, but not likely to work.
Thanks. You may be right.
To me, however, there is a trade-off in fuel usage and reliability between hovering for a longer period to get the last few meters of accuracy and just letting the rocket down hard within 2m of center. Otherwise, why F9 hoverslam, for example?
Actually, x/y accuracy gets easier the faster you hoverslam, because there's less time for random real life effects to mess with your pure physics.

It's the Z accuracy that's the issue, as the faster you hoverslam, the easier it is to miss your exact targeteed height. THAT's what a cradle should try to help with.
Title: Re: BFR landing cradle discussion and updates
Post by: sevenperforce on 04/27/2017 08:57 pm
Actually, x/y accuracy gets easier the faster you hoverslam, because there's less time for random real life effects to mess with your pure physics.

It's the Z accuracy that's the issue, as the faster you hoverslam, the easier it is to miss your exact targeteed height. THAT's what a cradle should try to help with.
So, basically, some giant-ass shock absorbers.

What about hydraulics? Not of the closed-cycle oil variety, but just water? On contact with the booster, the cradle arms are released and sink into water reservoirs, forcing the fluid out as they absorb any remaining momentum from the booster. Cheap, reusable, and easy to build...plus, that area is already plumbed for the flame trench anyway, so they're used to dealing with large amounts of water.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 04/27/2017 09:09 pm
(e) cute, but why, if the rocket which is already mobile can land on a stationary fixture?

Also, passive alignment such as you describe is (again) cute, but not likely to work.
Thanks. You may be right.
To me, however, there is a trade-off in fuel usage and reliability between hovering for a longer period to get the last few meters of accuracy and just letting the rocket down hard within 2m of center. Otherwise, why F9 hoverslam, for example?
Actually, x/y accuracy gets easier the faster you hoverslam, because there's less time for random real life effects to mess with your pure physics.

It's the Z accuracy that's the issue, as the faster you hoverslam, the easier it is to miss your exact targeteed height. THAT's what a cradle should try to help with.
'sactly.

That's the beauty of the fixed cradle.

Mobile cradles, by virtue of being mobile, will have less Z tolerance, thus requiring less hoverslam, and increasing x-y uncertainty even further.  It's a self-defeating direction.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/27/2017 09:54 pm
Actually, x/y accuracy gets easier the faster you hoverslam, because there's less time for random real life effects to mess with your pure physics.

It's the Z accuracy that's the issue, as the faster you hoverslam, the easier it is to miss your exact targeteed height. THAT's what a cradle should try to help with.
So, basically, some giant-ass shock absorbers.

What about hydraulics? Not of the closed-cycle oil variety, but just water? On contact with the booster, the cradle arms are released and sink into water reservoirs, forcing the fluid out as they absorb any remaining momentum from the booster. Cheap, reusable, and easy to build...plus, that area is already plumbed for the flame trench anyway, so they're used to dealing with large amounts of water.
I like the hydraulic piston. Would suggest placing it to one side to make room under the landing/launch table to make room for a blast trench for re-launch.

Up thread it was determined that +/- 2-meter (edit: x-y) accuracy should be planned for in a nominal landing. Need to accommodate that.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/28/2017 10:36 am
...
...
Actually, x/y accuracy gets easier the faster you hoverslam, because there's less time for random real life effects to mess with your pure physics.

It's the Z accuracy that's the issue, as the faster you hoverslam, the easier it is to miss your exact targeteed height. THAT's what a cradle should try to help with.
'sactly.

That's the beauty of the fixed cradle.

Mobile cradles, by virtue of being mobile, will have less Z tolerance, thus requiring less hoverslam, and increasing x-y uncertainty even further.  It's a self-defeating direction.
Your point about accommodating hoverslam is well received. AIUI the rocket slams into the landing pad or cradle with about 1g force.
 
To accommodate the slam with the landing funnel, I would set springs over each ball bearing to elevate  the 5t/20t funnel base 5 cm above the surface. When the 275t rocket slams into the tunnel it would push the balls to the ground and allow the ground to receive the shock. A second set of heavy wheels would deploy from the base to move  the 300 ton load 2 meters to the landing pad center.

Hydraulic pistons/de-celerators should also be incorporated to accommodate the shock of landing (per Sevenperforce). The funnel base  should be enlarged and hydraulic de-celerators inserted into the spaces between blast slots. Since water is used rather than hydraulic fluid, the water could just squirt into the air and be replenished between landings.   
Title: Re: BFR landing cradle discussion and updates
Post by: sevenperforce on 04/28/2017 01:25 pm
So, basically, some giant-ass shock absorbers.

What about hydraulics? Not of the closed-cycle oil variety, but just water? On contact with the booster, the cradle arms are released and sink into water reservoirs, forcing the fluid out as they absorb any remaining momentum from the booster. Cheap, reusable, and easy to build...plus, that area is already plumbed for the flame trench anyway, so they're used to dealing with large amounts of water.
I like the hydraulic piston. Would suggest placing it to one side to make room under the landing/launch table to make room for a blast trench for re-launch.

Up thread it was determined that +/- 2-meter (edit: x-y) accuracy should be planned for in a nominal landing. Need to accommodate that.

I was thinking of having each of the three capture arms mounted on its own hydraulic piston table; I just didn't picture it in the diagram. It would be more like the attached diagram (though this has bisymmetry when obviously the ITS will use trisymmetry).
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 04/28/2017 01:43 pm
...
...
Actually, x/y accuracy gets easier the faster you hoverslam, because there's less time for random real life effects to mess with your pure physics.

It's the Z accuracy that's the issue, as the faster you hoverslam, the easier it is to miss your exact targeteed height. THAT's what a cradle should try to help with.
'sactly.

That's the beauty of the fixed cradle.

Mobile cradles, by virtue of being mobile, will have less Z tolerance, thus requiring less hoverslam, and increasing x-y uncertainty even further.  It's a self-defeating direction.
Your point about accommodating hoverslam is well received. AIUI the rocket slams into the landing pad or cradle with about 1g force.
 
To accommodate the slam with the landing funnel, I would set springs over each ball bearing to elevate  the 5t/20t funnel base 5 cm above the surface. When the 275t rocket slams into the tunnel it would push the balls to the ground and allow the ground to receive the shock. A second set of heavy wheels would deploy from the base to move  the 300 ton load 2 meters to the landing pad center.

Hydraulic pistons/de-celerators should also be incorporated to accommodate the shock of landing (per Sevenperforce). The funnel base  should be enlarged and hydraulic de-celerators inserted into the spaces between blast slots. Since water is used rather than hydraulic fluid, the water could just squirt into the air and be replenished between landings.
The thing the cradle needs to absorb is residual velocity, not acceleration.

By ”absorb" I mean "protect the rocket by slowing it down over distance", not "survive".

So IF the rocket had 10 m/s residual, and IF you could spread it evenly over 1 meter, you'd get a=5g - and have to make sure the interface mechanics can handle that.
Title: Re: BFR landing cradle discussion and updates
Post by: sevenperforce on 04/28/2017 02:39 pm
The thing the cradle needs to absorb is residual velocity, not acceleration.

By ”absorb" I mean "protect the rocket by slowing it down over distance", not "survive".

So IF the rocket had 10 m/s residual, and IF you could spread it evenly over 1 meter, you'd get a=5g - and have to make sure the interface mechanics can handle that.
Which is why the easiest way to handle this...I think...is to let the control arms sink into a meter or two meters or five meters of water.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 04/28/2017 03:03 pm
The thing the cradle needs to absorb is residual velocity, not acceleration.

By ”absorb" I mean "protect the rocket by slowing it down over distance", not "survive".

So IF the rocket had 10 m/s residual, and IF you could spread it evenly over 1 meter, you'd get a=5g - and have to make sure the interface mechanics can handle that.
Which is why the easiest way to handle this...I think...is to let the control arms sink into a meter or two meters or five meters of water.
They actually have those real nice long tapered "fins" that look to me like they can function as guide rails for final alignment and slow-down before capture.

There is a small chance that they actually thought this through.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/29/2017 01:53 pm
...
...
The thing the cradle needs to absorb is residual velocity, not acceleration.
...

The term de-celleration is to indicate negative acceleration, i.e. from x m/s to zero.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/29/2017 02:05 pm
...
...
Which is why the easiest way to handle this...I think...is to let the control arms sink into a meter or two meters or five meters of water.
Yes, indeed. Your control arms are analogous to the widened funnel base with spaces between blast slots, where hydraulic shock absorbers would be placed. I also like your shock absorbers placed under the 3 landing fins.

Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/30/2017 12:10 pm
I often jump to a solution before laying out the problem. Please allow me to rectify that.

A 275t booster rocket is to land directly onto a launch cradle with an accuracy of +/- 2m in any x-y direction. In the z direction, it must shed 10 m/sec (edited) velocity (per meekGee) by de-celeration using the thrust of the landing engine(s). The vehicle must be rotated so that fuel ports align with umbilical fuel lines.  To perform the landing, two approaches are proposed:

(1) Guide the rocket 2 m horizontally by RCS thrusters and the bottom edge of the vehicle bumping into the side of a sloped surface in fixed position. Thrusters and the slope will push the vehicle sideways to the center of the cradle for final touch-down.

(2) Guide a moveable cradle 2 m horizontally by employing the landing plume pushing against the sloped side of a cradle funnel to center it under the rocket. After landing, the rocket and cradle will be pushed 2 m to a final position over the launch blast trench.
Edit - References: Replies #265, #266 and #267.


Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 04/30/2017 01:40 pm
Just a note - 10 m/s was just an example...  And I don't know where the 2 m capture radius came from.

Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/30/2017 04:28 pm
Just a note - 10 m/s was just an example...  And I don't know where the 2 m capture radius came from.
Seems like a good example. I'll search for the 2 m.
Title: Re: BFR landing cradle discussion and updates
Post by: Patchouli on 04/30/2017 05:27 pm
The thing the cradle needs to absorb is residual velocity, not acceleration.

By ”absorb" I mean "protect the rocket by slowing it down over distance", not "survive".

So IF the rocket had 10 m/s residual, and IF you could spread it evenly over 1 meter, you'd get a=5g - and have to make sure the interface mechanics can handle that.
Which is why the easiest way to handle this...I think...is to let the control arms sink into a meter or two meters or five meters of water.

Standing water under the pad might be necessary to keep it from destroying the pad and itself during launch as was suggested with the similarly sized ROMBUS and Nexus concepts.

OT but I think a squat shape for the booster like NEXUS or ROMBUS might make reentry and landing easier.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/30/2017 08:54 pm
meekGee, I tried to find a good reference for the 2 m radius of landing error, but its pretty soft. I tried and failed to find hard data on the degree of F9 boosters' off-center landings. I found that most accuracy data on USA rockets are now classified and unavailable to the public. Your comment in Reply #73 of this thread that the F9 boosters have demonstrated accuracy of "a few meters" is about as good as we can expect.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/30/2017 10:06 pm
It occurred to me that the moveable landing cradle  (Replies 265, 266, 268, and 282) could work even if the plume from the landing Raptors was directly over the upper edge of the funnel. The pressure differential inside and outside the funnel would quickly accelerate it toward plume center.

The upper opening of the funnel has a radius of seven meters, so the theoretical error radius would be 7 m. The real radius would depend on how quickly the funnel could translate horizontally while the booster descends vertically during the last 3 seconds.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 04/30/2017 10:15 pm
...
...
Standing water under the pad might be necessary to keep it from destroying the pad and itself during launch as was suggested with the similarly sized ROMBUS and Nexus concepts.
...
...
I do not believe the ITS booster will launch without the full array of ground support equipment and a blast trench beneath it. It will not launch from a flat surface any more than any other large rocket. Otherwise we could have saved many millions on launch equipment over the past 50 years.
Title: Re: BFR landing cradle discussion and updates
Post by: robert_d on 04/30/2017 11:27 pm
Six water cannons aimed for contact about 25 meters up could help with earlier alignment. Differential pressure based on radar.
Title: Re: BFR landing cradle discussion and updates
Post by: guckyfan on 05/01/2017 05:53 am
Six water cannons aimed for contact about 25 meters up could help with earlier alignment. Differential pressure based on radar.

There are plenty of powerful RCS thrusters that can align the bottom of the ITS, both lateral translation and lateral speed.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 05/01/2017 09:59 am
Six water cannons aimed for contact about 25 meters up could help with earlier alignment. Differential pressure based on radar.

There are plenty of powerful RCS thrusters that can align the bottom of the ITS, both lateral translation and lateral speed.
Yes, but it may not be necessary. According to the IAC presentation, there will be a crane at the landing site to lift an entire spaceship 77 m into the air and accurately place it onto the booster. This is the same alignment accuracy required to align the booster in the launch cradle. We might as well use it.
Title: Re: BFR landing cradle discussion and updates
Post by: Ionmars on 05/08/2017 11:02 am
This is me talking to me: Forget the moveable landing cradle. The two options are:
(1) Super-accurate repeatable landings on a fixed mount.
(2) Landing accuracy within a few meters -- just land onto a flat hard surface a safe distance away from the blast trench, bur within reach of the crane. Just use the crane to lift and re-allign the booster onto the fixed mount.

Edit: I posted this before coffee this morning.  :-X Option 2 doesn't work without legs (obviously), which is not the case for the booster in this thread.
Title: Re: BFR landing cradle discussion and updates
Post by: ChrML on 05/08/2017 11:52 am
I might be missing something obvious, and I'm not a mechanical engineer, but why would you need a launch/landing cradle at all?

SpaceX will have to find a solution for launching the ITS without launch cradle anyway to be able to launch it from Mars where no such infrastructure is available /hard to create. It should launch standing on its own legs, but perhaps be moved/aligned with a flame trench below. Then land on the same legs.

If you can get it to reliably launch from earth that way, it will also launch reliably from Mars. If you design a system thats only for earth, you have to design another system for mars at som point later anyway and test it again. A cradle able to handle a nearly 300 ton rocket will easily weigh 50-100 tonnes. That is a lot to bring/build on Mars.

Landing it on its own legs will also allow it to be moved quite easily with some specialised car (like airplane towers, but in this case jacks on wheels). Then you can land it anywhere near the launch site and just move it to place. Then strap in place in case of wind until next launch. On mars pressure is low so no strapping should be necessary.

Reasoning for this: The simpler system, less components = Less cost and less risk.

Just my 2 thoughts :)
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 05/08/2017 11:58 am
I might be missing something obvious, and I'm not a mechanical engineer, but why would you need a launch/landing cradle at all?

SpaceX will have to find a solution for launching the ITS without launch cradle anyway to be able to launch it from Mars where no such infrastructure is available /hard to create. It should launch standing on its own legs, but perhaps be moved/aligned with a flame trench below. Then land on the same legs.

If you can get it to reliably launch from earth that way, it will also launch reliably from Mars. If you design a system thats only for earth, you have to design another system for mars at som point later anyway and test it again. A cradle able to handle a nearly 300 ton rocket will easily weigh 50-100 tonnes. That is a lot to bring/build on Mars.

Landing it on its own legs will also allow it to be moved quite easily with some specialised car (like airplane towers, but in this case jacks on wheels). Then you can land it anywhere near the launch site and just move it to place. Then strap in place in case of wind until next launch. On mars pressure is low so no strapping should be necessary.

Reasoning for this: The simpler system, less components = Less cost and less risk.

Just my 2 thoughts :)
Yup - BFR never goes to Mars, only BFS does - and so BFS has legs and lands within reach of the crane, just as you propose.
Title: Re: BFR landing cradle discussion and updates
Post by: ChrML on 05/08/2017 12:04 pm
Is there a reason the booster can't do the same? Of course it will weight a few tons more, but seems easier to sacrifice a few tons of payload now and save millions of $ of development/building. And just have 1 landing solution to worry about for now.

Could use the exact same landing leg system for both crafts. Weight should not differ too much.

Then implement the cradle solution as a "optimization" in 5-10 years later. Priority 1 should be to get it going and not gaping over too much cost :)
Title: Re: BFR landing cradle discussion and updates
Post by: docmordrid on 05/08/2017 05:12 pm
Per Musk at TED there's an ITS update coming in a few weeks, so details may be updated then. It's also possible they backtrack to the ITS Booster landing on legs. No one here knows yet, or can't talk because of a NDA.
Title: Re: BFR landing cradle discussion and updates
Post by: RoboGoofers on 05/09/2017 03:37 pm
Is there a reason the booster can't do the same? Of course it will weight a few tons more, but seems easier to sacrifice a few tons of payload now and save millions of $ of development/building. And just have 1 landing solution to worry about for now.

Could use the exact same landing leg system for both crafts. Weight should not differ too much.

Then implement the cradle solution as a "optimization" in 5-10 years later. Priority 1 should be to get it going and not gaping over too much cost :)

They are already landing the F9 booster with probably ~10 cm accuracy, on the Z axis. That's the hardest axis. XY should be easy.

Here's a small scale test with a similar control system.
https://www.youtube.com/watch?v=KBMU6l6GsdM

Are people questioning this posability because of the scale of the BFR?
Title: Re: BFR landing cradle discussion and updates
Post by: JamesH65 on 05/09/2017 06:03 pm
Are people questioning this possibility because of the scale of the BFR?

I'm not, I think the size makes it easier, relatively speaking. The booster will have a lot more mass and therefor momentum, and will be more difficult to divert off it trajectory. Get that right (and they already get that right with F9), and it shouldn't make much to be accurate to <1m.
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 05/09/2017 06:16 pm
Are people questioning this possibility because of the scale of the BFR?

I'm not, I think the size makes it easier, relatively speaking. The booster will have a lot more mass and therefor momentum, and will be more difficult to divert off it trajectory. Get that right (and they already get that right with F9), and it shouldn't make much to be accurate to <1m.

I'm sure SpaceX will be able to land BFR in a landing cradle. My issue with the concept is if there is a failure not only do they lose the rocket and landing cradle, but they lose the launch pad as well. Kind of a risky way to save time.

I think the launch/landing cradle should be on rails so it can be moved to the launch pad and tower. Land nearby, but not close enough to damage the launch pad if there is an accident, then roll the vehicle to the pad. A spare cradle would be a good idea to speed up a return to operations.
Title: Re: BFR landing cradle discussion and updates
Post by: rsdavis9 on 05/09/2017 08:02 pm

I think the launch/landing cradle should be on rails so it can be moved to the launch pad and tower. Land nearby, but not close enough to damage the launch pad if there is an accident, then roll the vehicle to the pad. A spare cradle would be a good idea to speed up a return to operations.

I think at first it makes sense to keep landing and launch separate. But when you have 3 or 4 launch/landing cradles why not go for the single one? It wont make a difference if you got 3 or more and cream one.
Title: Re: BFR landing cradle discussion and updates
Post by: ChrML on 05/09/2017 09:39 pm
Not concerned about it being hard. Just seems very unnecessary to focus on building a cradle and the development and potential problems it might have now.

When there's missing funds and plenty of other unadressed issues to solve with the mission to Mars.

It's pretty easy to just move the Booster that has landed on legs 200meters from the launch pad back into place. Then share the exact same landing solution on BFR, VFS (and F9 and FH). There is no money to save on implementing cradle landing now before the first ship has flown and the ship is proven feasible (even if payload is reduced 5 tons due to landing equipment in beginning).
Title: Re: BFR landing cradle discussion and updates
Post by: docmordrid on 05/09/2017 11:07 pm
Well, there's an ITS update in a few weeks and for all we know this may change back to legs - among other things. They may be on ITS 1.1 or 2.0 by now.

Look at the original "Muse" stage recovery concept video and there were no grid fins, the legs didn't look anything like the Dan Gurney All American Racers legs do now etc.
Title: Re: BFR landing cradle discussion and updates
Post by: Oersted on 05/10/2017 10:12 am
Two facts that seem to point in a logical direction:

- Landing on a flat surface necessitates substantive legs to avoid impingement and damage to the stage.

- Actively directing a landing stage into a tight, fixed cradle is extremely difficult and dangerous.

What solution avoids both of these problems?

I think a big grid landing pad does. A big grid allowing the rocket blast to blow through, while still supporting the weight of the stage as well as allowing a certain lateral freedom for the touch-down point.

As for the actual support mechanism when the stage touches down, I think it will consist of short, fixed stubby legs on the stage, combined with active latch-and-locking mechanisms traversing rapidly on the surface of the big grid. Imagine several Roomba-like horizontal landing leg extensions homing in on the landing spot and hooking on to the vertical fixed stubby legs of the stage. Upon touchdown the extensions will be pushed onto the stubby legs, click in place, and dampen any residual movement.

Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 05/10/2017 01:39 pm
If you have multiple launch pads, you don't need to separate landing cradle from launch pad. And on the other side, if you aren't launching enough to need at least 2 pads, you probably don't need the improvement in turnaround time that the launch/landing cradle gives you.

If there's only one pad, then just use legs. If there's at least two pads and you have enough flight rate that legs slow you down, then use the combination launch/landing cradle. I don't see the use of separating the landing cradle from the launch cradle.
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 05/10/2017 01:54 pm
If you have multiple launch pads, you don't need to separate landing cradle from launch pad. And on the other side, if you aren't launching enough to need at least 2 pads, you probably don't need the improvement in turnaround time that the launch/landing cradle gives you.

If there's only one pad, then just use legs. If there's at least two pads and you have enough flight rate that legs slow you down, then use the combination launch/landing cradle. I don't see the use of separating the landing cradle from the launch cradle.

Why not just use legs and skip the whole cradle idea?

A LEO fuel depot makes more sense than a rapid launch cadence to refuel the ITS spaceship. The crew/passengers spend less time waiting around in LEO. Launch, refuel at the depot, head to Mars.
Title: Re: BFR landing cradle discussion and updates
Post by: docmordrid on 05/10/2017 02:34 pm
In the IAC presentation my main issue, besides legs, was launch order. I'd launch the tanker first, assure it's operational, and only then launch the crewed Spaceship.
Title: Re: BFR landing cradle discussion and updates
Post by: DanielW on 05/10/2017 02:54 pm
In the IAC presentation my main issue, besides legs, was launch order. I'd launch the tanker first, assure it's operational, and only then launch the crewed Spaceship.

I assume the video was edited for clarity of mission profile rather than exact chronology. The spaceship launches then get refueled.
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 05/10/2017 03:08 pm
In the IAC presentation my main issue, besides legs, was launch order. I'd launch the tanker first, assure it's operational, and only then launch the crewed Spaceship.

Refueling the spaceship requires multiple tanker launches. With only one tanker the spaceship has to go first or there has to be a depot. Since the plan doesn't use depots, a very rapid turnaround for the booster and tanker is required. The launch/landing cradle would enable a rapid reflight for the booster.

It seems to me a depot would take the pressure off of the launch schedule, but then again a depot might be too expensive since it's something else to design, build, and put in orbit.
Title: Re: BFR landing cradle discussion and updates
Post by: DanielW on 05/10/2017 03:17 pm
In the IAC presentation my main issue, besides legs, was launch order. I'd launch the tanker first, assure it's operational, and only then launch the crewed Spaceship.

Refueling the spaceship requires multiple tanker launches. With only one tanker the spaceship has to go first or there has to be a depot. Since the plan doesn't use depots, a very rapid turnaround for the booster and tanker is required. The launch/landing cradle would enable a rapid reflight for the booster.

It seems to me a depot would take the pressure off of the launch schedule, but then again a depot might be too expensive since it's something else to design, build, and put in orbit.

A depot does not need to be anything other than the first tanker you launch.
Title: Re: BFR landing cradle discussion and updates
Post by: docmordrid on 05/10/2017 03:39 pm
In the IAC presentation my main issue, besides legs, was launch order. I'd launch the tanker first, assure it's operational, and only then launch the crewed Spaceship.

Refueling the spaceship requires multiple tanker launches. With only one tanker the spaceship has to go first or there has to be a depot.
>

Perhaps, or the pre-launched tanker shortens the time the Spaceship is in LEO before TMI. Or it is the depot.
Title: Re: BFR landing cradle discussion and updates
Post by: ChrML on 05/10/2017 05:27 pm
A LEO fuel depot makes more sense than a rapid launch cadence to refuel the ITS spaceship. The crew/passengers spend less time waiting around in LEO. Launch, refuel at the depot, head to Mars.
Yes, it makes more sense for comfort. But complicates the mission significantly.

How do you bring a large fuel tank to LEO?

In multiple pieces, and assemble them in space? With the launch costs and complicated assembly that means. Even sealing the parts to each other with no leakage becomes complicated in space.

Launch it in one big payload? Perhaps possible, but requires massive design/fairing change.

How to keep the fuel cold?

Mission complication: Instead of one fuel transfer, you get two.


Perhaps easiest way to keep a fuel depot is launching another ITS to LEO, then use it as a fuel depot. Fuel cooling equipment, transfer equipment, and tank already in place.

But if you can barely afford one ITS for first mission, how is creating two more feasible at this point?

I think rapid reuse is the solution here of the tanker. No requirement for very much rapid first launches either as they will not be carrying humen. The first pioneers can probably also survive a few weeks on orbit before Mars trip, after all people have been in the space station for a full year.

Again: Save money where you can, learn to crawl, prove the mission and get 1 crewed ITS over. Then make it better, walk, run etc..
Title: Re: BFR landing cradle discussion and updates
Post by: meberbs on 05/10/2017 05:38 pm
2 comments:

-People seem to have forgotten or missed that the original presentation explicitly mentioned that if the whole set of refueling trips takes too long, or for other unknowns, they have the option of launching crew last. (I'd have to double check whether this referred to an in-orbit crew transfer, or using the first launched tanker to store the propellant from the others. This might not have been clarified.)

-What does any of this have to do with the landing cradle? Can we please take this conversation to a relevant thread?
Title: Re: BFR landing cradle discussion and updates
Post by: RonM on 05/10/2017 08:52 pm
2 comments:

-People seem to have forgotten or missed that the original presentation explicitly mentioned that if the whole set of refueling trips takes too long, or for other unknowns, they have the option of launching crew last. (I'd have to double check whether this referred to an in-orbit crew transfer, or using the first launched tanker to store the propellant from the others. This might not have been clarified.)

If they have two tankers, then the first can be used to store the propellent from the other tanker flights.

The entire mission can be conducted with one booster, one spaceship, and one tanker flown four or five times per mission. SpaceX isn't going to build five tankers for every spaceship.

-What does any of this have to do with the landing cradle? Can we please take this conversation to a relevant thread?

It questions the need for even having a landing cradle at the launch pad. However, I guess that goes against the spirit of the thread.

Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 05/11/2017 04:30 am
You can launch the first BFS empty, then refuel it, then launch a BFS with people, have them transfer to the first BFS, and you have an empty BFS you can now refuel.  Repeat.

This only makes sense if refueling takes a long time, or if you consider the risk during refueling to be more than other mission risks (e.g. launch, interplanetary travel, Mars EDL, etc).  If you don't, you just have to realize that when colonizing Mars with BFSs, you run the risk of losing a fully populated BFS.
Title: Re: BFR landing cradle discussion and updates
Post by: ChrML on 05/11/2017 04:44 am
Most commercial aircrafts have the capability to reverse thrust out of the gate before taxi, yet they choose to push it for safety reasons due to risk of debris into engine.

I think this will be a similar case. Specialized ground vehicles are probably a lot easier to move the booster back than cradles. Also keeps the solution flexible. Ie if one lander fails and needs repair (or rapid disassembly...), neither the launch site is bound or the rocket. Plus a lot safer.

Btw looking forward to ITS update. Been a while with silence since the idea was launched.
Title: Re: BFR landing cradle discussion and updates
Post by: Robotbeat on 05/12/2017 01:03 am
The launch cradle makes integration faster and cheaper. It takes like a week just to get a landed Falcon 9 moved into a building, let alone back to a launch pad. A bigger rocket would be much harder and would take longer.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 05/12/2017 02:21 am
Most commercial aircrafts have the capability to reverse thrust out of the gate before taxi, yet they choose to push it for safety reasons due to risk of debris into engine.

I think this will be a similar case. Specialized ground vehicles are probably a lot easier to move the booster back than cradles. Also keeps the solution flexible. Ie if one lander fails and needs repair (or rapid disassembly...), neither the launch site is bound or the rocket. Plus a lot safer.

Btw looking forward to ITS update. Been a while with silence since the idea was launched.

As far as airplane analogies go, you chose a specific aspect in which airplanes are very much unlike rockets.

The whole trade-off between legs, cradles, landing on pad, landing off-pad - simply does not have a parallel in airplane world.

I could equally argue that airplanes land on their "launch pad".  But this is meaningless, since the way the landing-launch cycle interacts with the ground infrastructure is completely different.
Title: Re: BFR landing cradle discussion and updates
Post by: ChrML on 05/14/2017 09:45 am
The launch cradle makes integration faster and cheaper. It takes like a week just to get a landed Falcon 9 moved into a building, let alone back to a launch pad. A bigger rocket would be much harder and would take longer.
I'm not sure why moving the rocket should take a week. In the beginning, maybe, but not with the ground crew/equipment in place.

If you land it say 100 from the launch pad, they should be able to move it back within a few hours.

Refurbishment, checks and refilling will take longer.
Title: Re: BFR landing cradle discussion and updates
Post by: meekGee on 05/15/2017 08:06 am
The launch cradle makes integration faster and cheaper. It takes like a week just to get a landed Falcon 9 moved into a building, let alone back to a launch pad. A bigger rocket would be much harder and would take longer.
I'm not sure why moving the rocket should take a week. In the beginning, maybe, but not with the ground crew/equipment in place.

If you land it say 100 from the launch pad, they should be able to move it back within a few hours.

Refurbishment, checks and refilling will take longer.
The recent interview with Tom Mueller explains it very well
Title: Re: BFR landing cradle discussion and updates
Post by: Jet Black on 05/15/2017 08:44 pm
The launch cradle makes integration faster and cheaper. It takes like a week just to get a landed Falcon 9 moved into a building, let alone back to a launch pad. A bigger rocket would be much harder and would take longer.

If the Roomba works well enough, it could just be a matter of picking the rocket up and moving it a few dozen meters into exactly the right position. It doesn't have to be spun over on an axis, loaded on to a tray and then put into a building.