Author Topic: MCT Speculation and Discussion Thread 4  (Read 504467 times)

Offline envy887

  • Senior Member
  • *****
  • Posts: 2946
  • Liked: 1375
  • Likes Given: 915
Re: MCT Speculation and Discussion Thread 4
« Reply #2700 on: 08/29/2016 02:00 PM »
Legs need to be very strong, don't they. What's the weight on Mars of the fuel needed to launch direct to Earth?

In a ideal scenario (empty 65t mass vehicle, slow 6 km/s return) about 120t of weight.

More realistic scenario (75t mass vehicle, 25t mass payload, 7 km/s return) about 250t of weight.

Even in a worst case scenario (100t mass vehicle, 25t mass payload, 8 km/s fast return) about 400t of weight.

However, I'm not sure this is the primary design constraint on the legs, since they have to decelerate some 160 to 200 tonnes of mass from some small velocity (~2  m/s?) during landing. It's entirely possible that the landing "impact", if at dynamic acceleration of 1 to 2 gees, will generate more leg loading than the static weight of the return fuel under Mars gravitational acceleration of .38 gees.

Offline Paul451

  • Full Member
  • ****
  • Posts: 1305
  • Australia
  • Liked: 643
  • Likes Given: 544
Re: MCT Speculation and Discussion Thread 4
« Reply #2701 on: 08/29/2016 02:09 PM »
Would deployable thrust deflectors make sense? Rather than gimballing the engines, just redirect the exhaust away from the vehicle at 30 or 45 degrees. I'm thinking an ablative-coated panel set just below the engines (which I think will be mounded on the sidewalls), that pops out into the exhaust stream.

If you mean the deflectors would be attached to the vehicle itself, then at 45, the vertical thrust would push down on the panels (and hence the ship) exactly as much as it pushes up; resulting in zero vertical net force. The sideways thrust would presumably be balanced in each horizontal direction by ensuring the panels point in different direction, so you wouldn't get any horizontal movement either. Essentially is would be a very elaborate scheme to stop the rockets from doing anything at all. It would just sit there, burning away, until something gave out.

If you mean to mount the deflectors on the surface, you are essentially talking about a flame-trench. Which means a large concrete construction, along with some way of suspending the vehicle over it, so a launch platform/gantry of some kind. Which wouldn't be a good place to land (too much stuff in the way), so you'd need a vehicle capable of hoisting and transporting the BFR.... basically you've added an awful lot of hardware.

[edit: typo]
« Last Edit: 08/30/2016 03:24 AM by Paul451 »

Offline envy887

  • Senior Member
  • *****
  • Posts: 2946
  • Liked: 1375
  • Likes Given: 915
Re: MCT Speculation and Discussion Thread 4
« Reply #2702 on: 08/29/2016 03:36 PM »
Would deployable thrust deflectors make sense? Rather than gimballing the engines, just redirect the exhaust away from the vehicle at 30 or 45 degrees. I'm thinking an ablative-coated panel set just below the engines (which I think will be mounded on the sidewalls), that pops out into the exhaust stream.

If you mean the deflectors would be attached to the vehicle itself, then at 45, the vertical thrust would push down of the panels (and hence the ship) exactly as much as it pushes up; resulting in zero vertical net force. The sideways thrust would presumably be balanced in each horizontal direction by ensuring the panels point in different direction, so you wouldn't get any horizontal movement either. Essentially is would be a very elaborate scheme to stop the rockets from doing anything at all. It would just sit there, burning away, until something gave out.

If you mean to mount the deflectors on the surface, you are essentially talking about a flame-trench. Which means a large concrete construction, along with some way of suspending the vehicle over it, so a launch platform/gantry of some kind. Which wouldn't be a good place to land (too much stuff in the way), so you'd need a vehicle capable of hoisting and transporting the BFR.... basically you've added an awful lot of hardware.

The deflectors would be attached to the vehicle.

And Newton says that the only way to get zero net vertical thrust is to redirect the net flux EXACTLY horizontally, which is neither the object, or realistically feasible. If the panels deployed to 45 degrees, the exhaust would be travelling at nearly 45 degrees after deflection, because exhaust particles hitting an impenetrable surface is not an elastic collision, and the pressure from the rest of the exhaust flux will keep the flow moving as much as possible in the direction it started.

If you do the FBD on this system, the down-force on the panels is equal to the net loss of exhaust massflow*velocity in that direction, which is any friction forces, plus the cosine loss of the redirection. The up-force however doesn't have cosine losses, so the net effect on thrust is no different than canting the engines. Picture the variable thrust vectoring nozzles on modern fighter jets for a similar concept: high-temp control surfaces redirecting hot exhaust gas to give more control authority. The V-2 rocket did something similar with graphite vanes in the exhaust nozzle, instead of gimballing the whole engine.

« Last Edit: 08/29/2016 03:37 PM by envy887 »

Offline Impaler

  • Full Member
  • ****
  • Posts: 1283
  • South Hill, Virgina
  • Liked: 363
  • Likes Given: 0
Re: MCT Speculation and Discussion Thread 4
« Reply #2703 on: 08/30/2016 05:20 AM »
Ok lets make that comparison, touch down engines would be as simple as a dozen super draco engines around the nose of the vehicle. [...]

None of which helps for the re-launch.

Actually you could fire them at take off for a small additional thrust but that's not the point, you take off with your main propulsion system which is pointed strait down for maximum efficiency.  The comparison here is between having your main engines be able to gimbal out by huge amounts to avoid having the thrust crater the ground under you on landing, or just having second set of small canted engines at the top of the vehicle for landing.

Taking off again is going to be a whole other problem and will require a surface under the engine that won't be destroyed at liftoff because angling the engines isn't an option because you can't afford cosine loss at takeoff.


I assume you mean weight, the mass would be at least 250mT, and much higher if the ship was to carry enough fuel to return directly to LMO or Earth. MCT is a fuel rich architecture, allowing for many mission profiles.

First the vehicle is not going to be full of propellant at landing, that is nonsense.

The mass I wrote was wrong because I just wrote the cargo mass without including dry vehicle mass which I estimate at around 75 mt.  So a total of 175, add another 10 for residual propellants, at Mars that's a weight of 703 kN.  Super Draco produces 73 kN so 10 would hover the vehicle if pointed strait down, with cosine losses from 40 degree canting that brings the total to 12, for deeper angles like 60 degrees the number of engines would go up to 20 which is still quite reasonable if that angle proves necessary, by placing the engines at the nose of the vehicle they are far from the ground, farther then the MSL decent engines were from the ground, though admittedly their thrust was just 3 kN each.


The landing velocity would be close to zero in either scenario.

It would be targeted at zero, yes but the range of actual touchdown velocity that the landing gear must handle has to be higher when you have only huge raptor engines to slow down with.  The deceleration has to be high all the way to the ground and large turbo-pump engines are far less responsive then small pressure fed engines.  All that put together means a different touch down speed and a different landing gear design driven by thouse speeds.


Who said anything about gimbaling more than a few degrees? The idea is to allow the canting of the entire engine mount in a single degree of freedom. Canting is required anyway by SRP, so there is no extra engineering required for landing.

All of the issues I raised about propellant lines, thrust structures and the like are going to be problems even if your only trying to move in one axis.  I'd already assumed that would be the mode of action as it is obviously simpler then moving in two axis, to gimbal in one axis you need 1 gimbal, to gimbal in 2 axes you need 2 gimbals, canting simply means something being at an angle in this case an angle pointing off from vertical.

To launch you can't be canting, you need vertical thrust or your engine efficiency is ruinous.  So if you want the same engine to be used for landing without the plume cratering the ground and throw rock ejecta back at the engine and underside of the vehicle it would need to be pointed nearly sideways at the moment of touchdown, I've been saying 45 degrees as a conservative estimate, maybe rocks will all fly away from the vehicle at that kind of angle but in reality it would likely need to be even more, that's where I get a 45 degree rotation range from.  SRP dose not require anywhere near this level of rotation as we know it will work with just a 20 degrees canting which is at the upper limit of what has been done before.

Online Oersted

  • Member
  • Full Member
  • ****
  • Posts: 795
  • Liked: 394
  • Likes Given: 239
Re: MCT Speculation and Discussion Thread 4
« Reply #2704 on: 08/30/2016 10:01 AM »
Making an engine gimbal a huge amount will basically require a completely second thrust structure to keep the inward thrust vectors from crushing the vehicle, thrust structures are usually more massive then the actual engines they hold.  Propellant lines need to use bellows to allow them to flex but they are like springs, the bend radius is large so to get a greater deflection the engine needs to be at the end of a longer and wider thrust structure, the mass of all this would quickly become prohibitive.

Your alternative, canted Super-Dracos, would presumably also require a second thrust structure. So, in any case, gimballing the main engines or using separate thusters, a truss structure would presumably be needed. You could even say, perhaps, that a main engine truss structure for gimballing engines would probably perform dual use to some extent.

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 549
  • Liked: 582
  • Likes Given: 480
Re: MCT Speculation and Discussion Thread 4
« Reply #2705 on: 08/30/2016 01:24 PM »
To launch you can't be canting, you need vertical thrust or your engine efficiency is ruinous.

You can't be canting for the entire LMO and TEI burns, but what if it was just for the first few seconds, enough to avoid the risk of ejecta impacting the ship?

The heaviest estimate I've seen for BFS dry mass + payload from Mars to Earth is about 160mT. Assuming 7.5 km/s of ΔV is required, that is a takeoff mass of about 1280mT. If there are 5 Raptors, each having around 270mT of thrust, that is 1360mT of thrust at takeoff. For Mars gravity (0.38g), that is a thrust to weight ratio of 1360 / 1280 * 0.38 = 2.8, far more than necessary to achieve liftoff. Allowing a worst case T/W of 1.15, cos^-1(1.15/2.8) = 65.7, probably more than enough to mitigate the ejecta risk. After a few seconds of flight, the cant could be reduced to zero, and the flight continue with no further cosine losses to TEI.

First the vehicle is not going to be full of propellant at landing, that is nonsense.

As shown above, the ship wouldn't need to be completely full of propellant to relaunch, and it certainly wouldn't be necessary once ISRU is established, but why is it a nonsense? Do you have any supporting calculations?

Offline envy887

  • Senior Member
  • *****
  • Posts: 2946
  • Liked: 1375
  • Likes Given: 915
Re: MCT Speculation and Discussion Thread 4
« Reply #2706 on: 08/30/2016 02:29 PM »
Why would a lander want to land with enough propellant to relaunch? Landing prop for earth return is prohibitive, so where would it go with landed prop, and what would it do when it got there?

Offline JamesH65

  • Full Member
  • ****
  • Posts: 676
  • Liked: 406
  • Likes Given: 8
Re: MCT Speculation and Discussion Thread 4
« Reply #2707 on: 08/30/2016 06:08 PM »
Why would a lander want to land with enough propellant to relaunch? Landing prop for earth return is prohibitive, so where would it go with landed prop, and what would it do when it got there?

To reduce the time taken to generate enough for relaunch? Use as a raw material?

Offline envy887

  • Senior Member
  • *****
  • Posts: 2946
  • Liked: 1375
  • Likes Given: 915
Re: MCT Speculation and Discussion Thread 4
« Reply #2708 on: 08/30/2016 06:52 PM »
Why would a lander want to land with enough propellant to relaunch? Landing prop for earth return is prohibitive, so where would it go with landed prop, and what would it do when it got there?

To reduce the time taken to generate enough for relaunch? Use as a raw material?

It doesn't make sense to land some propellant unless it's coming out of your long-term payload baseline, which as we understand it is 100 tonnes. Eventually ISRU will be fully established and then payload can rise to the baseline, but until then there's no reason to spec margin for prop. Just count the propellant mass as a fraction of the 100t of generic payload delivered to the surface, or more likely as 100 tonne batches of fuel delivered as payload.

I don't think landing propellant (CH4 or more likely jsut LH2) is necessarily a bad idea until IRSU is fully established, but designing an architecture around what would be a short-term crutch is definitely a bad idea.


Offline Burninate

  • Full Member
  • ****
  • Posts: 1129
  • Liked: 345
  • Likes Given: 72
Re: MCT Speculation and Discussion Thread 4
« Reply #2709 on: 08/30/2016 07:09 PM »
I've tried to fit landing some methalox in, but even for a small portion of the vehicle (an 'escape pod'), the math doesn't work for return to orbit (much less return to Earth), it swamps other payload.  If you could wring 4.5km/s out of an escape pod, you could return to MLO and a waiting lifeboat there for return to Earth.  But that's too much.  Landing mostly dry just works a lot better.  Every kilogram of propellant landed reduces the ISRU gear landed by 1kg, which reduces the propellant generated by many kilograms per synod.

Now...  that's not to say landing *completely* dry is the preferred option.  Liquid propellants are useful for a number of things other than full-on Mars ascent.  Solar storm shielding, EDL mass redistribution maneuvers, and high-thrust RCS landing maneuvers are all made considerably easier if you pack a sizable quantity of, say, monomethyl hydrazine on board.
« Last Edit: 08/30/2016 07:15 PM by Burninate »

Offline Impaler

  • Full Member
  • ****
  • Posts: 1283
  • South Hill, Virgina
  • Liked: 363
  • Likes Given: 0
Re: MCT Speculation and Discussion Thread 4
« Reply #2710 on: 08/30/2016 07:34 PM »
To launch you can't be canting, you need vertical thrust or your engine efficiency is ruinous.

You can't be canting for the entire LMO and TEI burns, but what if it was just for the first few seconds, enough to avoid the risk of ejecta impacting the ship?

The heaviest estimate I've seen for BFS dry mass + payload from Mars to Earth is about 160mT. Assuming 7.5 km/s of ΔV is required, that is a takeoff mass of about 1280mT. If there are 5 Raptors, each having around 270mT of thrust, that is 1360mT of thrust at takeoff. For Mars gravity (0.38g), that is a thrust to weight ratio of 1360 / 1280 * 0.38 = 2.8, far more than necessary to achieve liftoff. Allowing a worst case T/W of 1.15, cos^-1(1.15/2.8) = 65.7, probably more than enough to mitigate the ejecta risk. After a few seconds of flight, the cant could be reduced to zero, and the flight continue with no further cosine losses to TEI.

First the vehicle is not going to be full of propellant at landing, that is nonsense.

As shown above, the ship wouldn't need to be completely full of propellant to relaunch, and it certainly wouldn't be necessary once ISRU is established, but why is it a nonsense? Do you have any supporting calculations?

Your not seriously trying to equate a thrust structure that hold 4 or more Raptor engines which produce 2,300 kN EACH with a structure that is going to handle around 900 kN TOTAL.  Even throttled to 40 percent and canted 45 degrees out 4 Raptor engines will be putting 2600 kN of inward crushing force on the vehicle, all that force has to be resisted with higher dry mass.

As for not landing with propellant, landed cargo mass on mars is one of the most expensive things in the whole architecture, it will take something like 10 kg IMLEO for each kg of cargo delivered to surface.  Taking propellant with you for return is very very expensive and avoiding that has always been the point of ISRU.  But even if for some reason this were to be done the propellants would come out of the 100 mt cargo capacity because the vehicles EDL profile would be completely ruined if it were over it's normal mass.

So the whole point of that tangent which was to claim that touchdown mass might be crazy high is moot.  Touch down mass will just be 100 mt cargo be it in the cargo hold or in tanks plus vehicle dry mass and a small propellant residuals for safe landing.  This is the only thing that makes any sense.

Online GORDAP

  • Full Member
  • *
  • Posts: 162
  • St. Petersburg, FL
  • Liked: 57
  • Likes Given: 20
Re: MCT Speculation and Discussion Thread 4
« Reply #2711 on: 08/31/2016 10:55 AM »
Impaler, I'm drawn to your suggestion re smaller, more responsive engines for Mars landing.  But I think Super Dracos aren't the answer for a couple of reasons. 

First, you still need the Raptors for Mars ascent (unless you are arguing that the Dracos will be able to lift a fully fueled BFS from Mars surface).  And the Raptors will have to be canted, so as to avoid backsplash during ascent.  This is exactly the same problem as on descent.  So if you have to solve it for ascent, why include a separate method for descent?

Second, the hypergolic fuel used on the Super Dracos is not easily made via ISRU.  That means that the fuel used for Earth landing must be carried from Earth to Mars, then back to Earth again.  This would really cut into the available payload landed on Mars (given the dictatorship of the rocket equation).  Right?

I think ideally a 1/2 to 1/3 scale Raptor would be ideal for the BFS, but I think SpaceX has indicated they are going to only develop one size of this guy.

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 549
  • Liked: 582
  • Likes Given: 480
Re: MCT Speculation and Discussion Thread 4
« Reply #2712 on: 08/31/2016 12:13 PM »
I think ideally a 1/2 to 1/3 scale Raptor would be ideal for the BFS, but I think SpaceX has indicated they are going to only develop one size of this guy.

Interesting that you should say that. In Nol Bakhtian's thesis 'Drag Augmentation via Supersonic Retropropulsion for Atmospheric Deceleration' she finds that the optimum number of engines for Mars SRP is actually 15 engines in a peripheral array, canted outboard. 15 1/3 scale Raptors would give the same thrust as 5 full scale.
« Last Edit: 08/31/2016 12:13 PM by OneSpeed »

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 6442
  • Germany
  • Liked: 1620
  • Likes Given: 1485
Re: MCT Speculation and Discussion Thread 4
« Reply #2713 on: 08/31/2016 12:38 PM »
Didn't really read it I admit. Looks interesting and I hope it will be discussed more. But interesting theme for a doctorate in philosphy.

Offline Bynaus

  • Full Member
  • ****
  • Posts: 419
  • Planetary Scientist
  • Switzerland
  • Liked: 280
  • Likes Given: 183
Re: MCT Speculation and Discussion Thread 4
« Reply #2714 on: 08/31/2016 02:09 PM »
"Doctorate in philosophy" = PhD. Doctorates in the natural and engineering sciences are typically all "doctorates in philosophy".

Online AncientU

  • Senior Member
  • *****
  • Posts: 5210
  • Liked: 3120
  • Likes Given: 4442
Re: MCT Speculation and Discussion Thread 4
« Reply #2715 on: 08/31/2016 02:11 PM »
I think ideally a 1/2 to 1/3 scale Raptor would be ideal for the BFS, but I think SpaceX has indicated they are going to only develop one size of this guy.

Interesting that you should say that. In Nol Bakhtian's thesis 'Drag Augmentation via Supersonic Retropropulsion for Atmospheric Deceleration' she finds that the optimum number of engines for Mars SRP is actually 15 engines in a peripheral array, canted outboard. 15 1/3 scale Raptors would give the same thrust as 5 full scale.

Nice find!
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline Impaler

  • Full Member
  • ****
  • Posts: 1283
  • South Hill, Virgina
  • Liked: 363
  • Likes Given: 0
Re: MCT Speculation and Discussion Thread 4
« Reply #2716 on: 08/31/2016 10:13 PM »
Impaler, I'm drawn to your suggestion re smaller, more responsive engines for Mars landing.  But I think Super Dracos aren't the answer for a couple of reasons. 

First, you still need the Raptors for Mars ascent (unless you are arguing that the Dracos will be able to lift a fully fueled BFS from Mars surface).  And the Raptors will have to be canted, so as to avoid backsplash during ascent.  This is exactly the same problem as on descent.  So if you have to solve it for ascent, why include a separate method for descent?

Second, the hypergolic fuel used on the Super Dracos is not easily made via ISRU.  That means that the fuel used for Earth landing must be carried from Earth to Mars, then back to Earth again.  This would really cut into the available payload landed on Mars (given the dictatorship of the rocket equation).  Right?

I think ideally a 1/2 to 1/3 scale Raptor would be ideal for the BFS, but I think SpaceX has indicated they are going to only develop one size of this guy.

I think your misunderstanding what I'm saying, I've never said you don't have Raptor engines.  The only questions is you ALSO have smaller engines.

I'm proposing the vehicle would have a cluster of small canted engines at the nose AND a set of vertical Raptor engines on the bottom.  The small canted engines are used just for the final touch down at landing, any SRP higher in the atmosphere is done via Raptor engines.  And the Raptor engines also do most of the landing burn, they bring the vehicle to a standstill just above the height where they would impinge and the ground, then the smaller engines take over and do a slow decent for the final few hundred meters to touchdown.  Lastly their might be some use small engines in orbital maneuvering or to steer during decent but these are secondary functions.

While I'm siting the super-Draco numbers for thrust and engine count necessary to do the job I do suspect that a modified version running on Metho-Lox might be preferable and I've speculated on such an engine myself in the past, allowing it to use ISRU propellants is a big advantage on Earth return as you pointed out.  The point was just to demonstrate the size of the system with a reasonable approximation engine rather then to down select to a specific engine.  I think a methane-Lox version of super-Draco is more likely to be developed then small version of Raptor because pressure fed engines are so much simpler.

Online GORDAP

  • Full Member
  • *
  • Posts: 162
  • St. Petersburg, FL
  • Liked: 57
  • Likes Given: 20
Re: MCT Speculation and Discussion Thread 4
« Reply #2717 on: 08/31/2016 10:47 PM »
Got it.

But I think you're ignoring the problem of take off from Mars surface.

The Dracos (or their methalox equivalents) are not going to lift you off the surface.  So you're left with using the Raptors.  If they are bottom mounted and can't cant much, you have the same problem with liftoff as you've just solved for touchdown.  So whatever you do to the Raptors to avoid backwash of regolith on ascent, you could have used for descent.

Offline Impaler

  • Full Member
  • ****
  • Posts: 1283
  • South Hill, Virgina
  • Liked: 363
  • Likes Given: 0
Re: MCT Speculation and Discussion Thread 4
« Reply #2718 on: 09/01/2016 02:44 AM »
No we have many more options once we are landed on the surface that we don't have when doing an initial landing, namely to modify the surface directly under the engines to make them resilient enough for take off.

Some kind of wire mesh screen can be put down, or even an inflatable 'kidi pool' which can be filled with water and fiber then allowed to freeze into a pykrete pad, the fiber is usually sawdust and by treating it with chemicals that produce and endothermic reaction so the water can be made to freeze almost immediately. 

Along with some stakes to keep it in place his should be strong enough to take the full force of the Raptor engine without breaking as pykrete has physical properties comparable to concrete and is an excellent insulator so it won't melt under the exhaust and it would take a long time to sublimate to so launch windows are not too tight.  It will obviously take a good deal of water but if we don't have that available the vehicle isn't getting filled with propellant anyway.  With say 1 ton of fiber and 10 tons of water you could create 4 pads each 6 meters across and 10 cm thick.  Package each one as a plastic pouch with fiber already inside and just roll it out and fill with water.

It is even possible that an automated means of deploying simple under engine pads like this can be developed, they only need to work once and for a few initial missions until more permanent pads are created.  This is far better then compromising the vehicle itself with under-powered engines on assent or trying to do crazy degrees of movement on the engines.  And far better then the wand-waving some engage in that say the vehicle will only land on naturally occurring bed rock.

Offline jpo234

  • Full Member
  • ****
  • Posts: 687
  • Liked: 488
  • Likes Given: 73
Re: MCT Speculation and Discussion Thread 4
« Reply #2719 on: 09/01/2016 06:55 AM »
Isn't the simplest solution to foreign object damage an engine that is robust enough to survive the odd pebble? And maybe some redundancy in case one engine breaks?
I have a really hard time to imagine how something could enter an active combustion chamber against the flow of the combusted fuel.
« Last Edit: 09/01/2016 07:01 AM by jpo234 »
You want to be inspired by things. You want to wake up in the morning and think the future is going to be great. That's what being a spacefaring civilization is all about. It's about believing in the future and believing the future will be better than the past. And I can't think of anything more exciting than being out there among the stars.

Tags: