Red Dragon Discussion Thread (1)

[go4mars]

How is everything installed?

  Do you mean in terms of timing?  If bigger than any available openings, the dragon would need to be built around the tanks I guess.  Do you mean piping through the pressure hull to tie into the current hypergol system? I don't know enough about it to suggest anything in that regard. 


Robotbeat's post may have made my "extra tanks in dragon" suggestion irrelevant anyways.

SpaceX's "Red Dragon" concept skips several expensive/risky steps that the older EDL concepts (for MER, MSL, etc) have: no drogues, no parachutes, no shedding the heatshield or backshell. Also, precision guidance on the entry part combined with propulsive descent and landing can also allow landing very precisely, since you aren't being blown around while hanging from parachutes...Really, SpaceX needs to know how to do this anyway for their propulsively landed crewed Dragon. The only difference are the different Martian entry characteristics (something we know FAR more about now than we did when Viking landed) and that the abort thrusters need to start firing when Dragon is still supersonic ...I don't see why more than 500-700m/s delta-v would be needed...I get a figure more like 15-25% fuel payload, and for a relatively light Dragon, that's actually a pretty typical fuel load.

 

15-25% fuel payload.  Does that include the "several tonnes" of scientific equipment? 

[Robotbeat]

Getting to Mars and especially landing is ridiculously hard. Of the 4 Mars landers that Russia successfully got into space, 3 failed completely and the other one lasted only for 20 seconds on the surface. And countless flybies and orbiters failed, as well. Even veteran space agencies (other than JPL recently) have at best a 50% failure rate for Mars missions.

Here's a good overview:
http://www.space.com/13558-historic-mars-missions.html

[oldAtlas_Eguy]

Getting to Mars and especially landing is ridiculously hard. Of the 4 Mars landers that Russia successfully got into space, 3 failed completely and the other one lasted only for 20 seconds on the surface. And countless flybies and orbiters failed, as well. Even veteran space agencies (other than JPL recently) have at best a 50% failure rate for Mars missions.

Here's a good overview:
http://www.space.com/13558-historic-mars-missions.html

I love the one about the lander missing the planet. How do you miss a planet?

[ugordan]

I love the one about the lander missing the planet. How do you miss a planet?

Mars 7:
"Due to a problem in the operation of one of the on-board systems (attitude control or retro-rockets) the landing probe separated prematurely (4 hours before encounter) and missed the planet by 1300 km. The early separation was probably due to a computer chip error which resulted from degradation of the systems during the trip to Mars."

[Zed_Noir]

Got a query. Can the Dragon with the trunk can have supplemental  hypergolic propellants feed from tanks inside the trunk?

Requires substantial redesign.

Why would it be substantial? There is already lots of umbilical connections between the trunk and the capsule for power and radiator flow plus other things. My guess is just some plumbing fixtures and lines is needed to be added to feed the propellants from the trunk tanks to either the capsule tanks or directly to the SuperDracos. If SpaceX can make the F9's propellants plumbing work, they should be able to move propellants from the trunk to the capsule.

Jim is right. The trunk nomenclature is awkward.
 

[Jason1701]

Got a query. Can the Dragon with the trunk can have supplemental  hypergolic propellants feed from tanks inside the trunk?

Requires substantial redesign.

Why would it be substantial? There is already lots of umbilical connections between the trunk and the capsule for power and radiator flow plus other things. My guess is just some plumbing fixtures and lines is needed to be added to feed the propellants from the trunk tanks to either the capsule tanks or directly to the SuperDracos. If SpaceX can make the F9's propellants plumbing work, they should be able to move propellants from the trunk to the capsule.

Jim is right. The trunk nomenclature is awkward.

Transferring hypergolics is complicated, especially in the space and Martian environments. It's not like fixing your toilet. F9's propellants are different. Jim, do you agree?

[go4mars]

Is it unusual for hypergolic engines like super dracos to be able to throttle so much?  The superdraco system has to be able to go from multi-g full proppelant-load aborts on earth to low-prop load touchdowns on Mars (in ~1/3 g).  That's seems pretty significant.  Would the efficiency be a lot less at vastly lower thrust with hypergols? 

[Jim]

1`.  If bigger than any available openings, the dragon would need to be built around the tanks I guess. 
15-25% fuel payload.

2.  Does that include the "several tonnes" of scientific equipment? 

1. not feasible. 

2. Anymore than a couple 100's of pounds is unlikely.

[kevin-rf]

Why would it be substantial? There is already lots of umbilical connections between the trunk and the capsule for power and radiator flow plus other things. My guess is just some plumbing fixtures and lines is needed to be added to feed the propellants from the trunk tanks to either the capsule tanks or directly to the SuperDracos. If SpaceX can make the F9's propellants plumbing work, they should be able to move propellants from the trunk to the capsule.

Jim is right. The trunk nomenclature is awkward.
 

You do realize that at the same time for Boeing's new 737-MAX they are doing all sorts of study's and limiting the fan diameter of the upcoming model because they do not want to touch the wiring that routes around the front landing gear wheel well...

...And that is for something as simple as an airline that they will sell thousands of.

[Patchouli]

Maybe by 2018 a big dollop of Pu-238 will be easier to come by.  It would certainly appear to solve a lot of the issues here.  Is its current scarcity the primary reason that "RTGs would cost too much"?  As to allowing it on FH, by 2018, FH might not be as new of a LV. 

If they do use an RTG heat source most likely it would be Am-241 since large quantities of it would be easier and cheaper to obtain.
It's also a much safer isotope to handle then other candidates.
The short coming is it has only 1/4 the power for a given mass as Pu-238.
Though a core 4x larger does not necessarily mean the rest of the RTG is 4x larger.
Besides it's primary purpose would be to just keep the vehicle warm in the Martian night as the vehicle would likely be solar powered.

[Jim]

If they do use an RTG heat source most likely it would be Am-241 since large quantities of it would be easier and cheaper to obtain.

It is not "easier".  It is not usable for spaceflight.
How many times do you have to be told that isn't going to happen.  They aren't going to qualify another source.  Repeating it isn't going to make it happen.

[oldAtlas_Eguy]

Is it unusual for hypergolic engines like super dracos to be able to throttle so much?  The superdraco system has to be able to go from multi-g full proppelant-load aborts on earth to low-prop load touchdowns on Mars (in ~1/3 g).  That's seems pretty significant.  Would the efficiency be a lot less at vastly lower thrust with hypergols? 

Since the 80’s a technology for use with hypergolic engines has been available to do effective throttling from 0% to 100%. EE’s know this a Pulse Width Modulation. A high speed actuator valve that is a simple on/off that can go from full off to full on in 10milli second or less that is then operated on a cycle of 10-100hz so that the time off to time on in 1 cycle can be accurately controlled giving an effective throttle position for the engine of 0% to 100%. This method works so well that most modern throttling hypergolic thruster designs use this methodology. An acoustic sensor can be used to monitor the engine/actuator to check for potential failure modes during operation because the acoustic signal will be off nominal in an engine/actuator that is about to fail.

[Kaputnik]

Skycrane is just refining an existing concept? Hmmm...

Absolutely. It uses the same aeroshell and DGB chute. Just the prop module for final approach and landing is on top of the payload, rather than underneath it, for egress reasons.

SpaceX are apparently relying on hypersonic retro-propulsion

Supersonic, maybe, but where is your proof that it is hypersonic retropropulsion? A Dragon capsule, according to my calculations, has a terminal velocity of approximately the speed of sound at Mars.

And I calculate that Dragon's terminal velocity would be around Mach 2 on Mars. In addition, you need a good bit of altitude to transition to final approach and landing (on the order of a kilometre or more) which pushes the terminal velocity figure up somewhat.

Importantly, is it deccelerating rapidly enough to even reach that terminal velocity? I doubt it, although I lack the ability to prove it.

It's not really that different, unless you have only a single engine pointing directly upstream. I'm not going to trivialize what needs to be done, but you seem to have only a very superficial understanding of hypersonic retropropulsion. Dragon's landing thrusters would be on the side, like crewed Dragon, and thus wouldn't be directly in the airstream... And what simulation/experiment that has been done on supersonic retropropulsion (it's not likely to be hypersonic for an unmanned spacecraft) has shown that thrusters to the side like that work quite well, keeping the vast majority of the drag. And actually, hyper- and/or supersonic retropropulsion HAS been done on Earth (remember reading about it), just not operationally, since there's exactly no reason to do it operationally at Earth since the atmosphere is far, far denser than at Mars.

I admit I am not an expert on super/hypersonic retropropulsion. But I know that is something that is considered pretty hard to pull off, and virtually impossible to test in a lab. I would suggest that SpaceX are not experts in this field either, nor is this operational environment the principle design goal for the Super-Draco system, which is already fulfilling two radically different purposes.

Does it pass the sniff test that it would somehow be cheaper to develop this new capability than it would be to clone something that is known to work?

Something "known" to work (part of the time...), yet costing hundreds of millions of dollars for even a comparable payload and thus eating up almost all of your budget. We actually HAVEN'T proven the capability to land something weighing more than a few hundred kilograms on the surface of Mars, and that capability is rather spendy with usually a small actual payload (the MERs are only 180kg). The Skycrane concept has not been demonstrated, yet, and is pretty expensive.

Why concentrate on the failures? The current state of the art may have failed once, but it is also the only method that has actually succeeded, ever.

SpaceX is developing much of the stuff needed for this mission, anyway. They need abort thrusters, which are already funded partly, and they will be building a version of them that will be landing with those thrusters. Why not clone that for Mars to allow a greater payload instead of copying a typically expensive EDL concept that's ultimately quite limited in its payload mass?

Because the Mars application is so fundamentally different to the Earth one. What is your proof that a propulsive landing lands a greater payload than an aero-decel one?

Also, I'm not sure you're aware that Dragon can already take quite a high propellant load.  When it's empty, it's not that massive, either. Also, Dragon uses bipropellant landing thrusters, not monopropellant like both Viking and MSL, thus Dragon will be capable of more impulse from the same amount of fuel. And I don't see why more than 500-700m/s delta-v would be needed, thus I don't see where you get the figure of 1/3 of spacecraft mass being fuel comes from. I get a figure more like 15-25% fuel payload, and for a relatively light Dragon, that's actually a pretty typical fuel load. Remember, since parachutes wouldn't be required, you can use that mass for fuel instead.

Even Viking had more than 10% pmf, so 15% sounds a bit optimistic to say the least.
Braun et al (2005) consider entry vehicles of up to 450kg/m2 ballistic coefficient; Dragon would be more than double that 'density' so goes right off the top of their graphs. But anyway, they consider up to 30% pmf to be about right, and obviously the Dragon will need more than that. But hey, perhaps they are wrong,

One thing that hasn't been mentioned yet is the cosine losses from the oblique angle of the Super-Dracos. All we have to go on at the moment are artists' impressions, but they seem to show an angle of about 40 degrees, which is a big loss on effective isp.

[Robotbeat]

Primarily aerodynamic decelleration is quite limited in the mass it can land on Mars because of scaling laws (think of the difference in terminal velocity between a grain of sand and a boulder...). This is one reason more propulsive techniques are better in some ways since they can scale better, even if the TMI mass might be a little greater (which is debatable as well). It's already established that you need propulsive help in landing on Mars no matter what (except for incredibly small payloads), so by eliminating several spacecraft configuration changes, you're probably making a lot fewer opportunities to make a mistake and far fewer spacecraft configurations to qualify for flight.

And regarding to firing a rocket engine partially into a supersonic airstream... Shuttle did this a bit with its RCS thrusters, at least for yaw control. And yeah, it produces some counter-intuitive effects, but that's no longer an unknown-unknown.

And this can actually be tested to a certain extent (though not all the aerothermal aspects of entry will be the same, at least you can validate your models of supersonic descent control) by SpaceX if they wanted to in the upper atmosphere of Earth. Might not be a bad idea, actually, though it could be expensive if they don't find a way to integrate it into some other mission.

But as the many failed flyby and orbiter missions to Mars show, EDL isn't the only difficult thing about going to Mars.

[Kaputnik]

I think some very optimistic assumptions are being made about the delta-v required for an a combined aero/prop landing.
Terminal velocity is, I think, a bit of a red herring. e.g. I would calculate that MPF would have had a terminal velocity under parachute of under 50m/s. Yet it only reached 270m/s before entering final powered approach.
So whilst a Dragon might, in theory, only have to shed 5-700m/s under power, in practise that number is going to be much higher because it will have to switch to propulsive mode long before it ever reaches a stable terminal velocity. Just how much of the 5km/s minimum entry speed is is going to have left? I would say that we are talking at least 1km/s delta-v, likely more.

Cosine losses are going to have the effect of dropping effective isp to around 250, possibly less. Discounting gravity losses, that means at least a third of the entry mass needs to be propellant. The actual delta-v and effective isp could easily shrink the remaining mass considerably.


I do see the benefits of propulsive descent for larger vehicles, and it could make sense for SpaceX to pioneer this 'brute force' approach on a smaller scale first. However there is no free lunch, and the price looks like significantly greater IMLEO.

[mmeijeri]

I do see the benefits of propulsive descent for larger vehicles, and it could make sense for SpaceX to pioneer this 'brute force' approach on a smaller scale first. However there is no free lunch, and the price looks like significantly greater IMLEO.

Brute force is likely to be attractive only after we've set up the infrastructure for prepositioning propellant to a Mars Lagrange point and LMO by SEP, long after the timeframe imagined for Red Dragon.

[Seer]

At least 40% or more likely 50%. The delta v is around 1200 m/s. The reason is that the velocity to be killed is quite high, a lot higher than the terminal velocity as Kaputnik noted.

Here are some rough calculations.  Suppose the dragon retropulsion is 4 g of deceleration. In 20 secs it reduces speed by 800 m/s. This is roughly the deceleraton burn required.  Then add maneuvering and hover delta v. Then note that there are large cosine losses.

[Patchouli]

If they do use an RTG heat source most likely it would be Am-241 since large quantities of it would be easier and cheaper to obtain.

It is not "easier".  It is not usable for spaceflight.
How many times do you have to be told that isn't going to happen.  They aren't going to qualify another source.  Repeating it isn't going to make it happen.

If Spacex does this as a private mission it might be easier simply due to the huge cost and red tape associated with Pu-238.
Am-241 would be more readily available.
Still considering the supply issues even NASA probably should look into qualifying other RTG heat sources.
I suspect Spacex will likely try to pull this off completely solar powered without even RHUs.

[Jim]

If Spacex does this as a private mission it might be easier simply due to the huge cost and red tape associated with Pu-238.
Am-241 would be more readily available.
Still considering the supply issues even NASA should look into qualifying other RTG heat sources.
I suspect Spacex will likely try to pull this off completely solar powered without even RHUs.

Again, don't make statements that have no basis in fact.  You don't know what you are talking about.

Any launch large quantity of nuclear material is going to have huge cost and red tape associated with it regardless of whether it is commercial or gov't managed.

Am-241 is not "more readily available", since it hasn't been launched as an RTG.   The red tape would take just as long as obtaining Pu.

[Kaputnik]

At least 40% or more likely 50%. The delta v is around 1200 m/s. The reason is that the velocity to be killed is quite high, a lot higher than the terminal velocity as Kaputnik noted.

Here are some rough calculations.  Suppose the dragon retropulsion is 4 g of deceleration. In 20 secs it reduces speed by 800 m/s. This is roughly the deceleraton burn required.  Then add maneuvering and hover delta v. Then note that there are large cosine losses.

Well that sounds more likely, to me. Dragon has a dry mass of 45% so accomodating that propellant could get tricky.
I'm not aware of an isp number for the super-draco yet. The best, pump-fed, MMH/N204 systems achieve around 335s. I would hesitate to presume that SpaceX are able to meet this number, but maybe give them the benefit of the doubt for now.
The artists' impressions of the super-dracos firing appear to show an angle of about 45 degree... that presents a cosine loss of about 0.7, so reducing effective isp to 237s. Maybe the real thing will not be quite so bad.

If the required delta-v really is as high as 1.2km/s, then the propellant must account for at least 40% of the entry vehicle. For Dragon, that leaves only 15% left for landing gear and other systems, and of course payload.


The real issue I have with this is that I think people are taking Elon at face value when a grain of salt woudl be more applicable. He said that with the LAS, Dragon would be 'capable of landing on any solid body in the solar system'. That doesn't really stand up to scrutiny. Mercury? Venus? Io?
Of course it wouldn't be the first time that projects got a lot of people excited when the fundamental idea was flawed. Venturestar springs to mind; I'm sure there are other examples.