Landing legs for the BFR

louis · 2018-05-08 14:22:01

Just thought it might be worth having a separate thread on the landing leg design for the BFR spaceship (Mars lander) since it is so crucial.

Musk appears to have upped the design from 3 legs to 4. Arguments have been put forward for even 5 or 6 legs.

How big should the leg pads be? Should they be flat? I was wondering whether on Mars rounded (convex or maybe even concave) pads would be more effective in dealing with boulders.

Presumably the legs should be hydraulic but to what extent?

Should the legs have some lateral give - an arc through which they can move? (Or should they be fixed?)

Should the legs have a "pump" function so that they can be extended or indeed reduced in length where necessary after landing, so as to increase stability after landing.

Are there any other designs that might help? I am thinking possibly of a "spear" attachment to each foot to drive into the surface. Or maybe subsidiary "mini" legs that extend after landing.

Another design I was just thinking about would be a tripod but each of the three pods would be a straight bar with three hydraulic pads below. So there would be nine legs in total. If one landed on a boulder it would rise up leaving the work of support to the other two.

Last edited by louis (2018-05-08 14:28:32)

SpaceNut · 2018-05-08 16:54:53

Think office desk rolling chair with regards to safety and then you look at stability of leg count to spread and you will come to know that 4 legs are quite unstable hence office chairs now have 5 to make them more stable. Legs need to control a center of balance along the mid line of the rock from its base to the top. If all the mass once landed is in the top we have for a very unstable lander if the legs do not spread out. Also they need to compensate for the uneven landing area that may be sloped to still allow for the center line to stay vertically plum...

Oldfart1939 · 2018-05-08 17:22:07

More is involved than simply the number of legs; the points of contact with the ground determines some polygonal surface area inside the lines drawn between them. The stable configuration is when the center of gravity falls well within that area; when it's outside, the whole works tips over. That's for static stability, but in landing we're dealing with a momentum vector which could be outside that polygon. That's what GW has references by having the span of the legs greater than the height of the vehicle as an "insurance policy" against tip-over. The real advantage of having more legs is spreading the ground pressure over a larger contact area. The sheer massive nature of a partially or fully fueled BFR needs solid soil in order to bear this mass without sinking into the regolith.

My take on things is for the 9 meter diameter BFR the legs should be at least 5 in number, preferably 6; the spread should be at least 3X that value, if not more, such as 4X to 5X. A lot depends on the location of the center of mass of the landing and landed vehicle.

GW Johnson · 2018-05-09 08:57:58

Statics: spacecraft weight (not mass, weight!) divided by landing leg pad contact area (summed for all legs) is an effective bearing pressure exerted upon the soil. This applied bearing pressure must be less than the soil's inherent bearing strength. I put those numbers for the BFS at 10 sq.m of pads and two approximations to common Martian soils over at "exrocketman".

The weight vector from the center of gravity must fall within the polygon defined by the landing pads. For vehicles resembling BFS with the 4 legs Spacex shows, this is max 18 degree slope if the slope is directly toward 1 pad, and 13 degrees if the slope is directly toward the center between 2 pads. That is one place where the short height/span ratio really excels, while BFS is at risk. Many photos from Mars show localized terrain slopes exceeding 13 degrees.

Dynamics: describing the transient analysis I did over at "exrocketman" takes too many words to spell out here. Suffice it to say that a pad coming down on a 1 meter boulder, and suddenly slipping off that boulder after the leg strokes a bit, is a very serious risk of toppling over. It becomes almost a certainty if the local terrain slope exceeds about 5 degrees.

There is also the transient load effect on the statics as the vehicle touches down. If it falls a bit, it whacks the ground. Most of us engineers double the static load to estimate small erratic transients like this. What it means is that your applied bearing pressure upon the soil needs to be less than half its inherent strength.

There is no surviving a topple-over accident with a rocket of any kind. The tanks burst, the propellants mix, and "kaboom!", violent explosion. We have seen this vividly in films as long ago as the testing of the German V-2/A-4 during 1943. A toppled manned BFS is a dead crew, period.

And nothing is as expensive as a dead crew, especially one killed by bad management decisions. THAT is why NASA is so fatally risk-averse now. Spacex is still focused on designing and building this BFR/BFS vehicle. Ultimately, they will have to address this rough field landing issue, which so far they so very obviously have not.

It is this rough-field risk that impels me to insist that BFS's to Mars (or the moon) should be landed on hard-paved, large, very level landing pads, as early in the process as possible. The first ones must make rough-field landings, but that risk can be reduced by a proper pathfinder probe to the site. I have already described such in another thread.

GW

Last edited by GW Johnson (2018-05-09 09:04:33)

Quaoar · 2018-05-09 13:53:12

GW Johnson wrote:

And nothing is as expensive as a dead crew, especially one killed by bad management decisions. THAT is why NASA is so fatally risk-averse now. Spacex is still focused on designing and building this BFR/BFS vehicle. Ultimately, they will have to address this rough field landing issue, which so far they so very obviously have not.
It is this rough-field risk that impels me to insist that BFS's to Mars (or the moon) should be landed on hard-paved, large, very level landing pads, as early in the process as possible. The first ones must make rough-field landings, but that risk can be reduced by a proper pathfinder probe to the site. I have already described such in another thread.
GW

It may be possible to put the sea-level engines on the belly to perform a much safer landing on the belly, given the unknown terrain, then take off from Mars from the belly and at some altitude flip the BFR and use the vacuum engines on the tail?

Last edited by Quaoar (2018-05-09 13:56:35)

Terraformer · 2018-05-09 14:08:05

I'm sure I've seen concept art for Lunar landers that used such a system. I think it was probably the ULA ACES idea? The thrusters were mounted on the side, whilst the engine was at the rear, since it was based on the Centaur upper stage.

louis · 2018-05-09 16:35:59

GW Johnson wrote:

Statics: spacecraft weight (not mass, weight!) divided by landing leg pad contact area (summed for all legs) is an effective bearing pressure exerted upon the soil. This applied bearing pressure must be less than the soil's inherent bearing strength. I put those numbers for the BFS at 10 sq.m of pads and two approximations to common Martian soils over at "exrocketman".

Do flat pads provide best stability in rock fields? Might not something like the equivalent of a spaghetti server be better - here's one of those:

https://www.google.co.uk/shopping/produ … GwodsiYLSw

GW Johnson wrote:

The weight vector from the center of gravity must fall within the polygon defined by the landing pads. For vehicles resembling BFS with the 4 legs Spacex shows, this is max 18 degree slope if the slope is directly toward 1 pad, and 13 degrees if the slope is directly toward the center between 2 pads. That is one place where the short height/span ratio really excels, while BFS is at risk. Many photos from Mars show localized terrain slopes exceeding 13 degrees.

Presumably computer-controlled hydraulics on the legs can greatly aid stablisation ie the leg(s) on the higher end of the slope are hydrauically reduced in length and vice versa, to obtain a stable set up...

GW Johnson wrote:

Dynamics: describing the transient analysis I did over at "exrocketman" takes too many words to spell out here. Suffice it to say that a pad coming down on a 1 meter boulder, and suddenly slipping off that boulder after the leg strokes a bit, is a very serious risk of toppling over. It becomes almost a certainty if the local terrain slope exceeds about 5 degrees.
There is also the transient load effect on the statics as the vehicle touches down. If it falls a bit, it whacks the ground. Most of us engineers double the static load to estimate small erratic transients like this. What it means is that your applied bearing pressure upon the soil needs to be less than half its inherent strength.

I think my spaghetti server suggestion would give you better odds on not slipping off boulders. Another possibility is that there could be a system of inflatable supports that can be deployed within seconds of landing to bolster stability. Should pads be fixed or have some pivoting action? My instincts say some pivot would be useful as more likely to kick a boulder out of the way.

GW Johnson wrote:

There is no surviving a topple-over accident with a rocket of any kind. The tanks burst, the propellants mix, and "kaboom!", violent explosion. We have seen this vividly in films as long ago as the testing of the German V-2/A-4 during 1943. A toppled manned BFS is a dead crew, period.
And nothing is as expensive as a dead crew, especially one killed by bad management decisions. THAT is why NASA is so fatally risk-averse now. Spacex is still focused on designing and building this BFR/BFS vehicle. Ultimately, they will have to address this rough field landing issue, which so far they so very obviously have not.
It is this rough-field risk that impels me to insist that BFS's to Mars (or the moon) should be landed on hard-paved, large, very level landing pads, as early in the process as possible. The first ones must make rough-field landings, but that risk can be reduced by a proper pathfinder probe to the site. I have already described such in another thread.

Message received and understood re keeling over.

Might there be a possibility that the Mission One Cargo BFS model might carry a landing pad which it could fire on to the rock field? This 13 metre by 13 metre (say) landing pad could be self-deploying and would have some self-correcting internal system of gas or liquid cells that would allow for a reasonably level surface on top. So it would be fired like a big flare from the BFS cargo ship and then automatically deploy on the ground say 50 metres or so away from the BFS cargo ship (following a visual survey of the surrounding area so the best position was aimed for).

Anyone have any ideas how a self-activate landing pad with internal gas/liquid cells might make itself flat on top in relation to the uneven ground below? How robust would it have to be to allow a safe landing?

SpaceNut · 2018-05-09 16:53:14

Tried the link but it came up blank....

So not flat but with a pattern of raised diamonds plate if I get what you are looking too...sort of an antislip foot.

The final portion of the leg is most likely a piston obsorber to the pad with some sort of retractable or extending portion of the leg being on a pivot point hinge of sorts to allow for it to compensate...

Oldfart1939 · 2018-05-09 16:58:58

All of this brouhaha about fancy landing legs has one thing that drives me to drink: it violates my KISS principle! Instead of building (overly)complex landing legs, use the weight to have more of them and a large enough span to overcome the potential instability of one pad coming down on a boulder. In fact, the entire BFS program has the odor of too much optimism, and too little initial exploration and small scale before we go big, about it.
I'm not being overly critical and pessimistic; simply realistic.

SpaceNut · 2018-05-09 19:26:14

As far as small landers go pop a hole through the heatshield with the extending landing pad and be done with it as the seal will be just fine with retropropulsion landings. Remember temperatures will be lower than going in with nothing...

louis · 2018-05-09 19:46:27

Just tried the link myself and it was OK.

Yes basically a rounded (concave) pad with numerous sharp extensions beneath.

I favour (on the basis of my instinct, rather than any engineering skill!) the pad being pivoted to a few degrees and yes hydraulic extension/retraction of the leg would be vital in my view. I would hope with modern technology you could have the required extension/retraction of the legs within a few milliseconds of touchdown.

SpaceNut wrote:

Tried the link but it came up blank....
So not flat but with a pattern of raised diamonds plate if I get what you are looking too...sort of an antislip foot.
The final portion of the leg is most likely a piston obsorber to the pad with some sort of retractable or extending portion of the leg being on a pivot point hinge of sorts to allow for it to compensate...

Terraformer · 2018-05-10 03:23:11

Oldfart,

I don't see any good reason for the proposed architecture, myself. Why land the entire ship on Mars, then need a launching pad that can take it's fully fuelled weight, then launch the entire ship back to Terra? Keep orbital ships in orbit, and use smaller reusable landing craft to move supplies and fuel between it and the surface.

elderflower · 2018-05-10 03:23:31

You need only have six (preferably more) legs so that failure of one leg doesn't result in toppling. That failure could be structural, failure to deploy or foot-lands-on-unstable-ground or boulder. More redundancy, less chance of multiple failure.
Use of two different designs and two different actuation mechanisms would be good, so that common mode failures can be avoided.

louis · 2018-05-10 04:11:50

I tend to agree, but then I was coming from a few years back when it seemed only a small lander was feasible.

The great advantage of Space X's architecture is that you can land hundreds of tonnes of supplies and equipment each cycle and so, in essence, provide everything you need to create an industrial and agricultural infrastructure from the get-go.

The main drawback with the big tonnage drop is definitely the landing of such large rockets and it's interesting, if nothing else, to speculate about how Space X are going to address the challenges of landing a 47 metre tall, 100 tonne rocket on what is likely to be a rock field.

Terraformer wrote:

Oldfart,
I don't see any good reason for the proposed architecture, myself. Why land the entire ship on Mars, then need a launching pad that can take it's fully fuelled weight, then launch the entire ship back to Terra? Keep orbital ships in orbit, and use smaller reusable landing craft to move supplies and fuel between it and the surface.

Quaoar · 2018-05-10 15:25:03

Terraformer wrote:

I'm sure I've seen concept art for Lunar landers that used such a system. I think it was probably the ULA ACES idea? The thrusters were mounted on the side, whilst the engine was at the rear, since it was based on the Centaur upper stage.

The Centaur Based Lunar Lander:

http://selenianboondocks.com/2008/11/lu … d-landers/

The descend stage was supposed to land on the belly, then the head detach and take off vertically. It was nice, because the empty tank with solar panels remains on the Moon and can be used to build a modular base.

SpaceNut · 2018-05-10 17:46:03

I recall the Lunar Depot concept...which was all the rage under constellation before it got cancelled.

The landing legs on the block 5 have now been given the retraction ability after a landing has occured to which with some practice we can now launch then retract for a lunar or mars mission in the same rocket.

louis · 2018-05-11 03:36:45

That's interesting - another design step in the BFR concept seems in play. Excellent!

SpaceNut wrote:

I recall the Lunar Depot concept...which was all the rage under constellation before it got cancelled.
The landing legs on the block 5 have now been given the retraction ability after a landing has occured to which with some practice we can now launch then retract for a lunar or mars mission in the same rocket.

GW Johnson · 2018-05-11 10:27:40

The BFR/BFS vehicles are too far along in their design process to make a major structural change like moving propellant tanks to emplace engines for a belly landing. The propellant tanks literally ARE the structural airframe of the vehicle. Screwy shapes are NOT pressure vessels, even low-pressure ones. And propellant tanks must be low-pressure pressure vessels.

So, I think you're looking at a tail-sitter from Spacex, just like their presentations have indicated since the larger version was revealed at Guadalajara in 2016. It'll be not far from the reduced-size version presented in 2017. The numbers are starting to work out rather well for that, as I showed reverse-engineering it over at my "exrocketman" site.

As I have identified, there are rough-field safety issues landing that design on Mars or the moon. Spacex will eventually have to come to terms with that, and I think they will. They're rather heavily occupied designing the thing well enough to actually build and test a prototype vehicle, plus get Falcon-Heavy flying more routinely, plus improve the reusability of Falcon-9 as Block 5, plus get man-rated Dragon-2 off the ground while hamstrung by NASA's "dry empty skull" problems.

That's a lot on anyone's plate. I'm not surprised that some of the issues with BFR/BFS are unresolved. But eventually, they will have to be. That takes time, and you always run into the unexpected. And THAT is why Musk's timelines have been factor-2 or more optimistic.

GW

Last edited by GW Johnson (2018-05-11 10:29:03)

kbd512 · 2018-05-11 15:10:15

Why can't they just design a two-piece strut?

The shock absorbers would be housed in the "fin" fairings and use the upper propellant tank and the lower part of the cabin pressure hull as the attachment points. The leg would extend, lock into position and then three shocks attached to two different strongpoints would permit a very wide set of landing gear. The first shock would be attached to the cabin pressure hull and the second two would be attached to the propellant tank. It should be workable without a complete redesign of the rest of the ship. It wouldn't quite be as good as what GW said would be ideal, but the width of the landing gear would be much closer to the length of the ship and there would be three versus one shock absorbers per landing gear leg to absorb the load.

The landing strut would slide through the attachment points for the shock absorbers, just like the telescoping crane arms on mobile cranes.

Anyway, just a thought.

Edit:

Hasn't anyone here ever seen how a friction fit bipod or tripod leg extension works to increase the bearing surface over which the load is applied to support the weight of a heavy gun or missile launcher?

That's what I had in mind for the landing gear. It'd be a triangular composite rod that uses larger bearing surfaces attached to the shock absorber attachment points to distribute the load over a larger section of the rod. The rod would slides through the attachment points using jacking gear and then the shock absorbers achieve a friction fit near the end of the rod when it's fully deployed or extended. The upper shock then actuate to push on the end of the leg to deploy the leg outwards, away from the body of the ship.

Last edited by kbd512 (2018-05-11 15:35:30)

louis · 2018-05-11 17:11:02

Do you agree they will do multiple tests on rock fields on Earth before setting off for Mars?

GW Johnson wrote:

The BFR/BFS vehicles are too far along in their design process to make a major structural change like moving propellant tanks to emplace engines for a belly landing. The propellant tanks literally ARE the structural airframe of the vehicle. Screwy shapes are NOT pressure vessels, even low-pressure ones. And propellant tanks must be low-pressure pressure vessels.
So, I think you're looking at a tail-sitter from Spacex, just like their presentations have indicated since the larger version was revealed at Guadalajara in 2016. It'll be not far from the reduced-size version presented in 2017. The numbers are starting to work out rather well for that, as I showed reverse-engineering it over at my "exrocketman" site.
As I have identified, there are rough-field safety issues landing that design on Mars or the moon. Spacex will eventually have to come to terms with that, and I think they will. They're rather heavily occupied designing the thing well enough to actually build and test a prototype vehicle, plus get Falcon-Heavy flying more routinely, plus improve the reusability of Falcon-9 as Block 5, plus get man-rated Dragon-2 off the ground while hamstrung by NASA's "dry empty skull" problems.
That's a lot on anyone's plate. I'm not surprised that some of the issues with BFR/BFS are unresolved. But eventually, they will have to be. That takes time, and you always run into the unexpected. And THAT is why Musk's timelines have been factor-2 or more optimistic.
GW

louis · 2018-05-11 17:14:28

I take your point...that must surely be a possibility. I am presuming Space X are looking to keep it simple and avoid compromising internal cargo space.

I think the landing is the most crucial issue for a successful Mars mission so Space X really need to get this right. Let's hope they are focussed on the problem but maintaining an open mind to solutions.

kbd512 wrote:

Why can't they just design a two-piece strut?
The shock absorbers would be housed in the "fin" fairings and use the upper propellant tank and the lower part of the cabin pressure hull as the attachment points. The leg would extend, lock into position and then three shocks attached to two different strongpoints would permit a very wide set of landing gear. The first shock would be attached to the cabin pressure hull and the second two would be attached to the propellant tank. It should be workable without a complete redesign of the rest of the ship. It wouldn't quite be as good as what GW said would be ideal, but the width of the landing gear would be much closer to the length of the ship and there would be three versus one shock absorbers per landing gear leg to absorb the load.
The landing strut would slide through the attachment points for the shock absorbers, just like the telescoping crane arms on mobile cranes.
Anyway, just a thought.
Edit:
Hasn't anyone here ever seen how a friction fit bipod or tripod leg extension works to increase the bearing surface over which the load is applied to support the weight of a heavy gun or missile launcher?
That's what I had in mind for the landing gear. It'd be a triangular composite rod that uses larger bearing surfaces attached to the shock absorber attachment points to distribute the load over a larger section of the rod. The rod would slides through the attachment points using jacking gear and then the shock absorbers achieve a friction fit near the end of the rod when it's fully deployed or extended. The upper shock then actuate to push on the end of the leg to deploy the leg outwards, away from the body of the ship.

SpaceNut · 2018-05-11 19:20:05

The next high on the pole after a long stay and fueling is the relaunch as we will not have a factory at our disposal or engineers or parts to make it home.

louis · 2018-05-11 20:46:06

True, true...and is launching from a rockfield the same as launching off a smooth pad? My view is that Space X must and do recognise these problems and are urgently looking how to address them but it will be v. interesting to see how they do meet the challenges.

SpaceNut wrote:

The next high on the pole after a long stay and fueling is the relaunch as we will not have a factory at our disposal or engineers or parts to make it home.

Quaoar · 2018-05-12 08:05:32

GW Johnson wrote:

That's a lot on anyone's plate. I'm not surprised that some of the issues with BFR/BFS are unresolved. But eventually, they will have to be. That takes time, and you always run into the unexpected. And THAT is why Musk's timelines have been factor-2 or more optimistic.
GW

We have not to forget that build a perfectly working rocket is only half of the issue: they also have to project, develop and test a perfectly working device able to produce a thousand of tons of LOX-CH4 propellant from martian atmosphere CO2 and water from a buried glacier that is supposed to be in equatorial latitudes...

Last edited by Quaoar (2018-05-12 08:06:43)

Oldfart1939 · 2018-05-12 10:48:42

All launches made so far have been from prepared launch pads with a strongback utilized for the rocket. Maybe they should attempt a launch from Cape Canaveral using just the landing legs as a starter? Use one of the recycled Block 4 rockets and see how THAT goes. Sooner or later, SpaceX needs to demonstrate that technology. Maybe some landings in Texas out in the desert? That would be a great stability demonstrator. No amount of enthusiasm and "trusting Elon to have something up his sleeve" can ever substitute as actual bad case scenarios experiments.

New Mars Forums

Announcement

#1 2018-05-08 14:22:01

Landing legs for the BFR

#2 2018-05-08 16:54:53

Re: Landing legs for the BFR

#3 2018-05-08 17:22:07

Re: Landing legs for the BFR

#4 2018-05-09 08:57:58

Re: Landing legs for the BFR

#5 2018-05-09 13:53:12

Re: Landing legs for the BFR

#6 2018-05-09 14:08:05

Re: Landing legs for the BFR

#7 2018-05-09 16:35:59

Re: Landing legs for the BFR

#8 2018-05-09 16:53:14

Re: Landing legs for the BFR

#9 2018-05-09 16:58:58

Re: Landing legs for the BFR

#10 2018-05-09 19:26:14

Re: Landing legs for the BFR

#11 2018-05-09 19:46:27

Re: Landing legs for the BFR

#12 2018-05-10 03:23:11

Re: Landing legs for the BFR

#13 2018-05-10 03:23:31

Re: Landing legs for the BFR

#14 2018-05-10 04:11:50

Re: Landing legs for the BFR

#15 2018-05-10 15:25:03

Re: Landing legs for the BFR

#16 2018-05-10 17:46:03

Re: Landing legs for the BFR

#17 2018-05-11 03:36:45

Re: Landing legs for the BFR

#18 2018-05-11 10:27:40

Re: Landing legs for the BFR

#19 2018-05-11 15:10:15

Re: Landing legs for the BFR

#20 2018-05-11 17:11:02

Re: Landing legs for the BFR

#21 2018-05-11 17:14:28

Re: Landing legs for the BFR

#22 2018-05-11 19:20:05

Re: Landing legs for the BFR

#23 2018-05-11 20:46:06

Re: Landing legs for the BFR

#24 2018-05-12 08:05:32

Re: Landing legs for the BFR

#25 2018-05-12 10:48:42

Re: Landing legs for the BFR

Board footer