Landing on Mars

Rune · 2012-06-18 08:56:19

GW wrote:

(My slide rule was never this difficult to figure out.)

Are you kidding? All your answers are a google away, if you know the keyword... in this case, it's "BBCode" (Bulletin Board Code). Check the link, and you will find out posting anything is a cheat sheet away. Though I haven't figured out tables myself, that's mostly 'cause I'm lazy.

GW wrote:

PS - does anybody have a decent figure for the altitudes above Mars where hypersonic entry begins and where speed is down to around Mach 2.5-ish? Or does anybody have a good temperature vs altitude "standard" profile for Mars?

I know I have seen a NASA paper with reentry charts of speed vs. altitude and aerodynamic heating or something like that, but I have no idea when or where, so I have been after some decent source on that for a long time. I am also coming to believe that turning on a rocket engine in the middle of hypersonic flight is not that big of a deal (my reasoning: Dragon and Soyuz control attitude with a RCS system all the way through reentry, it may not point directly downwards, but those are rockets turned on and off while hypersonic). So why not just turn the engines on when you hit your natural terminal velocity at reasonable altitude, and forget chutes altogether. The mass ratio won't look that bad for reasonable terminal velocities, like 1,000-2,000 m/s, and acceleration profiles can't be that bad (one of the reasons I would like to see at which altitude MSL hits terminal velocity and opens chutes). T/W requirements can't be that much more than in a terminal landing system, so the only added weight is fuel in a bigger tank: less systems, more reliability, I would like to work out weights.

So I really, really would like to see some numbers run on it, which is why I'll check it out when you post it.

Rune. On an unrelated point, I wouldn't know what to do with a slide rule.

Rune · 2012-06-18 09:02:57

And trying that google thing for myself like I always recommend, "Mars atmospheric model" turns out this from the lips of NASA itself:

http://www.grc.nasa.gov/WWW/k-12/airplane/atmosmre.html In imperial units, like you like it

http://www.grc.nasa.gov/WWW/k-12/airplane/atmosmrm.html ...And for the rest of the world.

Rune. Enjoy! Turns out it wasn't that difficult, like always.

GW Johnson · 2012-06-18 09:50:49

Thanks, Rune. The Martian atmosphere data really will come in handy, I think. In both sets of units.

As for BBcode, I know those were English words, mixed with a bunch of acronyms, but that's about it. We do not share a common dictionary. Still lost as regards images.

Slide rules were sticks you added lengths with, marked logarithmically down the sticks, but labeled normally. It was a mechanical analog hand calculator. We used them for about 300 years before electronic calculators appeared. I designed my first airplane and my first half a dozen supersonic missiles with a slide rule.

GW

TwinBeam · 2012-06-18 10:03:08

louis wrote:

Well, people often end up saying that to me..."well, yes you could..." To which I respond - well why not? You then reduce the problem to simply getting enough fuel into LEO, which is probably a lot cheaper than developing new landing systems - which might take ten years, and involve hundreds possibly thousands of people being employed on the project. My approach in any case would be to have a pretty small descent craft and rely on supplies pre-landed for life support once you get to the surface. By making the descent craft fairly small, you reduce the problem to a manageable size.
So are you talking about entering the Mars atmosphere at a narrower angle so you pass through more atmosphere? Doesn't that create problems in terms of heat shield protection?

Hey, I get it - you're doing what everyone else here is doing - trying to think outside the box. The only thing is, you've started by looking at some (once) out-of-box (but now kind of conventional) solutions and then noticed the virtues of the basic "inside the box" solution - essentially "send a big rocket, slow down out of atmosphere, stay relatively slow all the way down - and you can avoid many of these complex issues and systems".

And that is correct - the problem with a big rocket isn't so much technical as it is political. It requires much larger chunk of money to build and launch. So most alternative solutions focus on spending smaller amounts over longer periods to develop less brute-force approaches, leading to still expensive but hopefully more tolerable build and launch costs. The latter has the problem that you may have to wait decades with low-level R&D going on, before you get tech good enough to convince those holding the purse strings that the more affordable mission is practical. On the other hand, because it is more affordable, you may then be able to sustain a longer series of missions; maybe even indefinitely long - i.e. settlement. You'll notice that we do NOT have a lunar colony - politically it was decided that we could not afford to sustain, let alone scale up, Apollo - or at least that the value of doing so was not worth it.

As to my specific tweak of the aerobraking plus rockets proposal - it would initially come in at about the same angle as any aerobraking entry, but by slowing only descent with rockets, would end up taking a shallower angle that does indeed pass through more atmosphere - but a controlled path that keeps heat shield demands within the limits of what it can handle, but not lower. I.e. get maximum value out of aerobraking.

Rune · 2012-06-18 10:33:47

GW Johnson wrote:

As for BBcode, I know those were English words, mixed with a bunch of acronyms, but that's about it. We do not share a common dictionary. Still lost as regards images.
Slide rules were sticks you added lengths with, marked logarithmically down the sticks, but labeled normally. It was a mechanical analog hand calculator. We used them for about 300 years before electronic calculators appeared. I designed my first airplane and my first half a dozen supersonic missiles with a slide rule.

You figured out links and quotes, right? I mean, you use them, so I assumed you do. Pics go the same way, and bold, and italics... the cheat sheet for what to put in the brackets (as in [magicword] the stuff you want magicked [/magicword] so you see it) is in the link to the wiki I posed earlier (BTW, whatever is highligthed in blue can be a link, there's a trick to make them look like words). In the case of an image, which is what you want, it would be (conveniently fouled up so it shows as text): (img)http://url-of-your-image.it's-got-to-be-online-already(/img). Substitute the "(" and ")" with "[" and "]", and the forum will try to show the url written in the middle as a picture.

And I know how a slide rule works in theory, but I wouldn't know how to work it until someone familiarized with it shows me. Which is what I'm trying to do, in reverse. Hope it helps!

...and hitting the preview button I see forums get smarter all the time and this one figures out links by itself, so you might not have figured it out yourself still. In that case, the basics are very simple, whatever gets between the bracketed command the forum is programmed to accept is affected by those commands. Useful commands without the first bracket (so you can read it and the forum doesn't): url] (links), img] (pictures), b] (bold), i] (italics)...

There are some other tricks, but it is all in the wiki I linked to in the earlier post. It will take you like 10 minutes to figure out reading it, promise. I figured it out from watching the unedited version of posts when I hit the quote button, so...

Rune. I'm pretty sure the slide rule takes more time (and math) to learn

TwinBeam · 2012-06-18 14:06:10

Actually, I've been wondering if we haven't gotten so caught up in landing decent-sized habs for long-duration missions, and multi-person human landers, that we aren't missing another "back in the box" approach. I.e. send down much smaller packages using simpler methods, and rely much more on building and putting things together on the surface. (This fits in with the idea of staying in orbit for a while and using tele-robotics to set things up, before descending.)

Most of the mass of a mission would be dropped as modest quantities of raw materials in simple but useful form, and would descend using only aerobraking and parachutes, and airbags for things you care about damaging. The idea being to provide raw materials in easy to use forms, instead of finished products, as much as possible. Think more like "empty sandbags" than "hab wall plates". Inflatables and patches, not metal sheets. Slabs of plastic to melt and extrude for pipes. Etc.

Humans would land individually in small "landing frames" - focusing all mass on getting them down safely and on target. It'd be unpressurized, with the human explorer in a pressure suit and carrying some survival gear in case the craft lands off course and needs to be retrieved. It might well follow Luis' approach of killing most orbital velocity using rockets, then descending sub-sonic with parachutes and using canted rockets for a soft landing. Or it could rely on aerobraking in a shell, followed by parachute and rockets - riskier but cheaper. Maybe a mix - rockets for humans and test landings before sending down humans, aerobraking otherwise.

As small landing frames would be cheaper per unit, a dozen could be tested by landing other important and somewhat fragile equipment, before the first human goes down. If one does fail with a person on board, you don't lose your entire crew - it's a tragedy but not a disaster for the mission (though if it's the first human to descend, it may mean a mission abort). Once down, the landing frame becomes a resource to be disassembled - support struts become support beams for structures, cryoliquid storage tanks and electronics are put to use, etc. Or take off the rockets and attach wheels to make a rover or tractor.

GW Johnson · 2012-06-18 14:35:40

The real trouble with landing on Mars is that, once you're over a ton or two in sizes around 3+ meters across, there's not enough "air" for your heat shield to slow you into the supersonics before you're too low; it's mostly an a = F/m problem, the lower gee doesn't help that problem to any significant degree. Then, further, once you're only supersonic where a chute can be opened, there's not enough "air" for the chute to slow you into subsonic before you hit ground; again, it's mostly an a = F/M problem and the lower gee doesn't help.

Yes, a really huge vehicle might do rocket braking outside the "air", or all the way down, but it will be quite enormous, and likely 2 stages just to land if chemical. There is some aerobraking to be had, that can very significantly reduce the size of the vehicle, just not nearly as much aerobraking as is available here on Earth. There's a reason we were testing parachutes at over 100,000 feet altitudes for Viking.

The real "out of the box" solution is to do rocket braking at low thrust during the entry hypersonics, and during the chute deceleration to subsonic, so that you are ready for a high-thrust landing from a suitable altitude, instead of from below the surface! (ha ha)

The idea is to use some rocket thrust to make up the aerobraking deceleration deficit. You'd like to be way subsonic and just about fully stopped, as your path inclines to vertical, just hundreds to a thousand meters or so off the surface. At that point you shed the chute as a stability risk, and just ride the thrust down, the last several seconds.

We've never fired retro thrust in hypersonics before, although we have fired attitude control thrusters during entry on every single spacecraft we have ever flown. Dragon will do it with its Super Draco thrusters, canted about 45 degrees out the sides. I think you could fire right through ports in the heat shield, if you sealed the "engine room" for no through-flow, and you canted about 10-15 degrees for plume stability.

No one has used combined rockets and chutes since the Russians used that to land battle tanks from aircraft ca. 1960. But that does not mean we cannot do it today. Of course we can. That's a lot easier than the hypersonics problem, and even the hypersonics don't scare me off.

I'm not yet sure what the aero-deceleration deficit looks like on Mars. I suspect it's a function of vehicle size and mass. But, I see no fundamental reason at all why the "big lander" problem cannot be solved. And in a lot less than 10 years. So what if it hasn't been done before?

After all, we went from nothing to an Apollo LEM in about 5 years, never having done a vacuum landing of any kind but once before (Surveyor 3).

Solve this problem, and you will be capable of landing anything from 2-3 tons on up to any size whatsoever, and in the most economical manner possible.

We should have done it for the new Curiosity rover. Then we'd already be a leg-up on sending men to the surface of Mars.

But, we didn't. Budget constraints once again ruled out doing something really smart.

GW

Last edited by GW Johnson (2012-06-18 14:38:50)

Rune · 2012-06-19 06:11:18

GW Johnson wrote:

No one has used combined rockets and chutes since the Russians used that to land battle tanks from aircraft ca. 1960. But that does not mean we cannot do it today. Of course we can. That's a lot easier than the hypersonics problem, and even the hypersonics don't scare me off.

Next week when Senzhou lands the same way Soyuz does (i.e, heatshield, chutes, AND terminal solid rockets) you might get a surprise: it's almost the standard option outside the west... there is way more operational data for that than for shuttle-style gliding landings!

GW Johnson wrote:

We should have done it for the new Curiosity rover. Then we'd already be a leg-up on sending men to the surface of Mars.

Curiosity aerobrakes with it's heatshield while maintaining attitude control with RCS thrust. Then it deploys supersonic drogues, then supersonic chutes, then it uses a terminal rocket system for final approach. What exactly did you say they haven't used there? Because my point of contention with MSL's EDL is the use of chutes, I would have substituted them for more rocket fuel to prove manned systems, and not built the most complex landing system imaginable.

Rune. Again, why chutes for big payloads.

GW Johnson · 2012-06-19 12:22:27

There's two things that haven't quite been done yet: (1) firing significant retro thrust into the oncoming slipstream at hypersonic speeds, and (2) firing significant retro thrust into the oncoming slipstream while hanging from a chute, at supersonic and transonic speeds. The Soyuz thing is like the battle tank: very subsonic.

Yet neither of these is particularly daunting, unless you are so over-bureaucratized as to attempt nothing that has not been done before. The key to firing engines through ports in a heat shield is no throughflow: a sealed engine compartment. Had shuttle Columbia's wing structure been sealed cell spaces inside, she would have brought her crew home safe in spite of the leading edge hole.

The key to retrofire while on a chute at Mach 2 is plume mixing with slipstream before it hits the chute, so that the chute is not damaged by hot gas. Not too much thrust, and stand the chute off well behind the vehicle on a heat-protected strap or cable. Easy enough.

Curiosity didn't really need a hovering skycrane. That kind of thing will never land men and habitats on Mars. Ridiculously big and complicated and wasteful.

GW

TwinBeam · 2012-06-19 15:25:45

GW Johnson wrote:

The real "out of the box" solution is to do rocket braking at low thrust during the entry hypersonics, and during the chute deceleration to subsonic, so that you are ready for a high-thrust landing from a suitable altitude, instead of from below the surface! (ha ha)

GW - I'm curious why you think it would be better to brake with rockets simultaneous with aerobraking - thereby reducing the time and net benefit of aerobraking - rather than using rockets to slow the vertical component of your descent, to prolong and maximize deceleration from aerobraking. And eventually braking with rockets to kill the remaining horizontal motion and land.

Also - is there any inherent benefit to using a single heavier lander, rather than, say, a dozen smaller landers with higher surface area to mass ratios (i.e. better aerobraking)? All I can think of, is that it is "all or nothing" - if it makes it down safely, the heavy lander delivers all the needed mass to one location. Smaller landers will need more heatshield mass, but less fuel mass. The "or nothing" downside seems like a big gamble, especially compared to making "safe as possible" lander(s) for humans, and "acceptable risks" landers for equipment.

GW Johnson · 2012-06-19 20:45:48

TwinBeam:

As I understand it, anything over a couple of tons mass cannot be slowed by sequential aerobraking enough to do a thrust touchdown, not without hitting dirt first. The "air" there is too thin to help enough for deceleration, but too thick to ignore in terms of entry aeroheating. That couple of tons is the entire vehicle mass at touchdown, not just the payload it contains.

A two or three ton limit is fine for small one-way probes, but they already smacked into the wall with the new Curiosity rover. That's what the complicated "skycrane" rig is supposed to address. Sure is a lot of stuff. Complicated stuff risks failure, and throwing away all that "skycrane" sure put some extra mass into what we had to shoot to Mars. I'd almost bet my idea both saves mass and reduces risk.

But any vehicle that might land 2-6 men is going to mass dozens of tons at touchdown, several to many dozen tons if it's not going to be a one-way trip. The dead-head payload crew cabin going back up might be about like a Spacex Dragon, and that's around 10 tons by itself. The ascent booster will be 10-20 times as big, with anything but nuke propulsion. That's around half a hundred to a hundred tons you have to land in the one vehicle. And we haven't included any sort of habitat or survival equipment, although that can be sent down one-way in separate smaller landers, given a beacon to home on. A 1-ton rover car will likely be close to the 2-ton lander limit, all by itself. It's hard to send down lots of tonnage if you're restricted to 2-3 total tons of landed mass per vehicle.

Anything with a blunt heat shield can be flown tipped slightly off-axis to generate a lift force, which can be used to adjust the trajectory to be whatever you need. No need for rocket thrust to do that lift job. We started using that with Gemini back in 1965. It works just fine. The lift force is comparable to the drag force, but on Mars supports 38% of the weight. You'll probably have to tip off-axis a bit more to compensate for the lower density better.

But, with Mars's too-thin "air", you're down to a small handful of km from the surface before the entry hypersonics are even over. You can pop a chute at about Mach 2.5-ish, but on Mars, if you're over the 2-3 ton limit, you'll strike before you can decelerate subsonic. Steepen the trajectory to hit denser "air" earlier, and you just hit the ground sooner. The trajectory has to be very shallow to work at all on Mars. It's already just about "flat horizontal". Plus, once you pop a chute, there's effectively no lift, and it steepens very quickly to near-vertical. That's just chutes. Ballutes would be no different.

So, to me, it appears there is a deficit in the aero-deceleration available at Mars. We need more, but it just ain't there to be had. No one can figure out how to make chutes and ballutes work during the hypersonics, so the only other option I see is adding a little rocket retro thrust. To my knowledge, no one has run the numbers for how much, but I bet it's low thrust during the hypersonics, and during the chute decel from supersonic to subsonic. Then throttle up and do a rocket touchdown.

GW

Rune · 2012-06-20 05:34:28

And the point I've tried to make is, since chutes won't be nearly as mass efficient to slow payloads with ballistic coefficients about 2-20 times bigger (assuming 10-100 times more mass, and scaling with r^3 (volume, proportional to mass)/r^2 (surface), that's more or less what I get)... why use them at all? You are going to have a rocket system anyway, so just put the parachutes' weight in additional rocket propellant. I'm pretty confident you get lower total system mass, but I won't be sure until I see a plot of descent curves versus ballistic coefficient in mars.

Oh, and it is a limitation of fairings, too. We cannot improve on the ballistic coefficient basically, 'cause we can't make our payloads wider and fit them in our launchers. Deployable heatshields (either inflatable or rigid) should work with that, up to some point.

Also, your ascent vehicle needs to mass so much only if you can't fuel it on the surface, either from prepositioned fuel caches or ISRU. That would put a MAV at about 20mT at touchdown (assuming 10mT payload like you suggest, 10mT structure and engines), with empty tanks good to load ~five times as much fuel (isp ~300) to get back to orbit when refueled. If terminal velocity were, for example, 1000m/s (and mach one on Mars is about 244m/s at sea level, so that's pretty hypersonic), and you trusted your engines to do their thing when they are needed, then mass ratio with 300s isp is ~1.4, so entry mass about 28mT, the eight tonnes being fuel for deceleration. For comparison with the real world, MSL touches mars atmosphere massing 2.8mT to land 1.5mT of payload, out of which 0.77 is useful stuff. The no-chute option doesn't compare badly with that, not at all, at least at this conceptual stage.

Rune. Of course the altitude at which you start decelerating on rocket power is important, but I haven't even considered gravity losses. So this is very rough, concept-stuff.

Rune · 2012-06-20 06:28:04

I was bothered enough by all this to dedicate a 5-min google search to it. This is the best that turned out (Guided Entry Performance of Low Ballistic Coefficient Vehicles at Mars). For the nice simple chart, go to page 2. Problem is, the worst beta they work with is about 150kg/m2. That would be MSL-like, or about 11.8mT if you used a 10m diameter heatshield, which is quite ridiculously low for the roomiest rocket I can think of (SLS). The good part, I found out how THAT works: 540m/s at 20km of altitude, and they pop chutes. Maximum load, 10G. Since I just calculated the drop from 1000m/s, applying the 10G limit I find out the lowest you can turn on the engines is 10km, which is almost twice as fast at half the altitude. Looks like you can make do with a worse beta than 150kg/m2. How much worse, twice that? Three times? I wish I knew.

Rune. Somebody else has sources? 'Cause I can't believe nobody has worked this out already.

RobS · 2012-06-20 07:27:02

I am curious; if you used a solid-core nuclear engine to land on Mars, how "hot" is the engine after it is turned off? Is it safe to descend a ladder right past the reactor? With the tanks empty of fuel and providing no shielding, how safe is it to live in the descent vehicle? This is especially serious if we were to use the engine for electricity as well.

Glandu · 2012-06-20 07:47:31

TwinBeam wrote:

Actually, I've been wondering if we haven't gotten so caught up in landing decent-sized habs for long-duration missions, and multi-person human landers, that we aren't missing another "back in the box" approach. I.e. send down much smaller packages using simpler methods, and rely much more on building and putting things together on the surface. (This fits in with the idea of staying in orbit for a while and using tele-robotics to set things up, before descending.)
(.../...)

Link in french : http://salotti.pagesperso-orange.fr/concept242.htm . That guy belongs to the french branch of the Mars society & did speak about its project to Zubrin. He's still in need of refining, though. But the basics is to divide the travel in empty cargos PLUS 2 vehicles of 2 people/32 tons each.

I won't translate it, it's a tough text, & I need to read it a few more times before understanding it, there are a lot of subtle design choices.

el_slapper. Wish I was better in rocket computations.

Rune · 2012-06-20 08:36:46

Glandu wrote:

Link in french : http://salotti.pagesperso-orange.fr/concept242.htm .

There's a chart on high beta reentries there! Thanks! (Not that I can make out much more than that, if it is in French). And going by it, it seems that with L/D 0.2 and beta of 300, you can make it to about 900m/s at 9kms. Close enough to what I suggested as acceptable, only even less fuel. That works out to about 23 mT behind a 10m heatshield, or close to seven (6,870kgs) if you take the 5.4m diameter of EELV's and Falcon Heavy fairings as upper limit to diameter. You can kind of squint a little and see some MAV fitting in there, some tiny pressurized capsule with some fuel tanks, engines, and heatshield (fit it all in under 5mT, 'cause the rest will be propellant). No Dragon can pull off that beta with it's 3.66m heatshield (it would have to mass under 3,156kgs, including landing fuel, and Dragon is right now 4,200kgs without SuperDracos), so what I am wondering right now how is Musk offering his "Red Dragon" mission. Do they deploy high hypersonic decelerators to lower beta? Before someone cries "chutes", I'm considering the very high hypersonic here, so no chutes allowed. Dragon would be a new crater on the ground before it got anywhere near Mach 2.5 on Mars' atmosphere on it's own, that much I am sure of.

Either he is calling for much more than 10G's at a lower altitude (it's directly proportional, half the room to stop, twice the acceleration needed), or for a heavier use of rockets prior to that terminal velocity, hurting both mass ratio and therefore beta.

Rune. BTW, "beta" stands for ballistic coefficient, and me not wanting to find out how the greek letter is written in here (again).

Last edited by Rune (2012-06-20 08:38:05)

TwinBeam · 2012-06-20 22:27:12

@Glandu - thanks for the link. Yes that goes in the direction I indiated - except I'm thinking to take it even further for the human landers - land ONE human at a time, not two.

@Rune and others - with Internet Explorer 9, right clicking on the link brings up Translate with Bing. (Google has an equivalent.) Very readable translation for the most part.

@GW - I'm a little frustrated - you don't seem to be catching the essence of my proposal, namely: Use a rocket NOT in-line-of-flight to slow down, and NOT to get lower faster - but rather aimed at the surface, to SLOW the rate of descent - to maximize the net benefits of aerobraking. When the value of aerobraking falls below the value of rocket braking at the same thrust level being used to slow your descent, switch over to rocket braking. So long as aerobraking is yielding more deceleration than the rocket thrust needed to stay aloft longer, it's a net win.

Impaler · 2012-06-21 03:06:06

Splitting up the Crew while seemingly an attractive solution to the current mass limitations on EDL it is a huge multiplier for lose of crew and mission. Also the surface habitat will almost certainly need to be monolithic or very near that and it's mass is going to drive the EDL tech, if we can't land a suitable habitat then theirs no mission even if Scotty could 'beam' them down. Once we possess the EDL tech for the habitat, the crew is a rounding error.

As for trying to thrust upwards to slow-down laterally, I get your general drift but it seems to me that kind of flight-path would be best achieved by having a lifting shape and/or angle of entry so aerodynamics are converting that forward speed into gravity counteracting lift. As crazy as it might sound perhaps we should examine something like the delta winged configuration. A delta-wing vehicle can be placed nose-up or side-mount on a launch vehicle allowing a vastly larger heat-shield area vs round shields inside narrow payload fairings (Imagine how hard it would be to launch the shuttle if was put on the end of a rocket in the horizontal position). Obliviously the lack of runways on Mars means a glider landing is out of the question so I'd just go to parachutes and rocket landing for the final descent (possibly detaching the whole delta-wing after it's no longer useful the way the heat-shield is typically dropped now).

I don't have any idea what kind of peak heating such a shield would need to withstand, to be practical it would presumably need to be lower heating and lower weight then the shuttles system (can a launch-once, land-once system could do that). Though when you consider that shuttle was able to bring down nearly as much cargo as it lifted (a capacity foolishly mandated and then virtually never used) and it was permanently stuck with the heavy engines in its rear (Unlike the Buran), plus a modest sized crew cabin. All that together must add up to a considerable amount of potential down mass if it were all being used for payload.

No idea if it would really work but the basic principle of putting the heat-shield in a vertical position during launch is the only way were ever going to get a ridged one of that size to Mars. If your not a fan of Delta-wing shapes a traditional conic shield might be launched in the same way.

Rune: I think the Shuttle orbiter shaped vehicle would be in the Beta range your looking for, I saw a figure of 376 m^2 for it's full profile but couldn't find the Ballistic coefficient it actually achieves on its entry angle-of-attack. Maximum landing mass is 100 mt, well above what's needed so Beta could be brought down considerably by lightening the payload. Also the total mass of thermal protection (belly, nose cone, leading edges, rudder cockpit all of it) is only 8500 kg, considerably less then I'd been lead to believe, naturally the tile system would be replaced with good-old ablative materials thus creating something vastly cheaper to build and probably lighter too. If a heat-shield type separation is done then I imagine the whole wing-belly-nosecone being a kind of 'sled', the cylindrical payload just slides aft-ward being pulled by parachutes and then lands horizontally via rocket.

Last edited by Impaler (2012-06-21 06:20:37)

louis · 2012-06-21 06:18:04

Impaler wrote:

Splitting up the Crew while seemingly an attractive solution to the current mass limitations on EDL it is a huge multiplier for lose of crew and mission. Also the surface habitat will almost certainly need to be monolithic or very near that and it's mass is going to drive the EDL tech, if we can't land a suitable habitat then theirs no mission even if Scotty could 'beam' them down. Once we possess the EDL tech for the habitat, the crew is a rounding error.
As for trying to thrust upwards to slow-down laterally, I get your general drift but it seems to me that kind of flight-path would be best achieved by having a lifting shape and/or angle of entry so aerodynamics are converting that forward speed into gravity counteracting lift. As crazy as it might sound perhaps we should examine something like the delta winged configuration. A delta-wing vehicle can be placed nose-up or side-mount on a launch vehicle allowing a vastly larger heat-shield area vs round shields inside narrow payload fairings (Imagine how hard it would be to launch the shuttle if was put on the end of a rocket in the horizontal position). Obliviously the lack of runways on Mars means a glider landing is out of the question so I'd just go to parachutes and rocket landing for the final descent (possibly detaching the whole delta-wing after it's no longer useful the way the heat-shield is typically dropped now).
I don't have any idea what kind of peak heating such a shield would need to withstand, to be practical it would presumably need to be lower heating and lower weight then the shuttles system (can a launch-once, land-once system could do that). Though when you consider that shuttle was able to bring down nearly as much cargo as it lifted (a capacity foolishly mandated and then virtually never used) and it was permanently stuck with the heavy engines in its rear (Unlike the Buran), plus a modest sized crew cabin. All that together must add up to a considerable amount of potential down mass if it were all being used for payload.
No idea if it would really work but the basic principle of putting the heat-shield in a vertical position during launch is the only way were ever going to get a ridged one of that size to Mars. If your not a fan of Delta-wing shapes a traditional conic shield might be launched in the same way.
Rune: I think the Shuttle orbiter would be in the Beta range your looking for, I saw a figure of 376 m^2 for it's full profile but couldn't find the Ballistic coefficient it actually achieves on its entry angle-of-attack. Maximum landing mass is 100 mt, well above what's needed so Beta could be brought down considerably by lightening the payload. Also the total mass of thermal protection (belly, nose cone, leading edges, rudder cockpit all of it) is only 8500 kg, considerably less then I'd been lead to believe, naturally the tile system would be replaced with good-old ablative materials thus creating something vastly cheaper to build and probably lighter too.

I have always favoured crew splitting - two lots of 3 for me is best. I think there are all sorts of advantages including an easier EDL task.

Rune · 2012-06-21 09:32:45

Impaler wrote:

As for trying to thrust upwards to slow-down laterally, I get your general drift but it seems to me that kind of flight-path would be best achieved by having a lifting shape and/or angle of entry so aerodynamics are converting that forward speed into gravity counteracting lift. As crazy as it might sound perhaps we should examine something like the delta winged configuration. A delta-wing vehicle can be placed nose-up or side-mount on a launch vehicle allowing a vastly larger heat-shield area vs round shields inside narrow payload fairings (Imagine how hard it would be to launch the shuttle if was put on the end of a rocket in the horizontal position). Obliviously the lack of runways on Mars means a glider landing is out of the question so I'd just go to parachutes and rocket landing for the final descent (possibly detaching the whole delta-wing after it's no longer useful the way the heat-shield is typically dropped now).
I don't have any idea what kind of peak heating such a shield would need to withstand, to be practical it would presumably need to be lower heating and lower weight then the shuttles system (can a launch-once, land-once system could do that). Though when you consider that shuttle was able to bring down nearly as much cargo as it lifted (a capacity foolishly mandated and then virtually never used) and it was permanently stuck with the heavy engines in its rear (Unlike the Buran), plus a modest sized crew cabin. All that together must add up to a considerable amount of potential down mass if it were all being used for payload.
No idea if it would really work but the basic principle of putting the heat-shield in a vertical position during launch is the only way were ever going to get a ridged one of that size to Mars. If your not a fan of Delta-wing shapes a traditional conic shield might be launched in the same way.

Ok, on the topic of delta-wings and "thrusting laterally" and all that: Capsules fly, they don't fall down. That L/D 0.2 often quoted for apollo-shaped stuff means if you have 10G of deceleration, by very simple aerodynamics, you have several G's of lift available to play with and guide your reentry (L+D, added as vectors, are what give you those 10G, and L is 0.2 times D). So you know, capsules can fly horizontal, even upwards, as long as they are breaking hard in the direction of travel. No fancy wings or anything required, just know that when you slow down enough, and G's go down, you'll drop like a rock.

Impaler wrote:

Rune: I think the Shuttle orbiter shaped vehicle would be in the Beta range your looking for, I saw a figure of 376 m^2 for it's full profile but couldn't find the Ballistic coefficient it actually achieves on its entry angle-of-attack. Maximum landing mass is 100 mt, well above what's needed so Beta could be brought down considerably by lightening the payload. Also the total mass of thermal protection (belly, nose cone, leading edges, rudder cockpit all of it) is only 8500 kg, considerably less then I'd been lead to believe, naturally the tile system would be replaced with good-old ablative materials thus creating something vastly cheaper to build and probably lighter too. If a heat-shield type separation is done then I imagine the whole wing-belly-nosecone being a kind of 'sled', the cylindrical payload just slides aft-ward being pulled by parachutes and then lands horizontally via rocket.

As to the shuttle orbiter specifically: 75mT (empty orbiter) divided by 376m^2 (best case area) gives ~200 as beta. That is worse than MSL by 50, but that is because it is way, way easier to fly on Earth. Which is kind of obvious. And while we are at it, you might consider a shroud to launch a shuttle-sized payload inside, 'cause the shuttle stack is quite retired. Is it too obvious I dislike wings for rockets a lot?

A more ingenious idea to increase surface area and lower beta for heavy payloads is to reuse the whole shroud for a heatshield, then drop it to expose engines. Have a look here, by the end of the pdf, they analyze a shroud/aerodynamic reentry shell combo for the Ares V. It shallows almost half the payload, but it can drop stuff at mach 2.2-3.1 at 5.1-7.3 kms of altitude in Mars. The shame is that you would have to drop the shroud, which kind of rules out reuse. And this is the way clever aerodynamic tricks go in Mars: since there is no air to speak of, you end up shallowing all the mass in aero surfaces.

Rune. As I said, I'd stop banging my head on the wall and start figuring out hypersonic rocketry.

GW Johnson · 2012-06-21 12:27:27

Answering TwinBeam in post number 117 above:

I think I understand what you were proposing. Using rocket thrust as lift to hold the entry vehicle in a flat trajectory long enough to slow down. Yes, that would work. However, if you have a blunt heatshield facing into the slipstream, and you tip the top edge forward a few degrees, you can generate a lift force comparable in magnitude to your drag force during the real hypersonics. That's lift without rocket thrust at all. It works down to around Mach 4-ish, with most any blunt shapes.

Depending upon whether we are entering at escape-class speeds, or orbital speeds, the Mach number is very definitely hypersonic. Here on Earth at orbital entry speeds, the initial hypersonic Mach number is around 25 as the vehicle grazes into sensible aero effects somewhere close to 90 km altitude. It was 36 coming back from the moon.

These are definitely some sort of free molecule slip flow aerodynamic conditions, not continuum flow, as the mean free path between air molecules up there is pretty close to 3 cm. I think most folks use some sort of modified Newtonian flow model, and an awful lot of correlations and Mollier diagrams to account for heat transfer and ionization effects.

From what I read about Curiosity, it enters at around 6 km/sec, "flies" on a tipped heat shield until the hypersonics are well over, presumably a little under Mach 3, pops a chute, and then sheds the heat shield. The chute takes it sort-of barely subsonic, but in that thin "air", the terminal velocity (drag = weight speed) is about 300 m/sec, which is close to speed of sound, unlike chutes here at home. From there it's rocket braking. They chose a skycrane rig to do the rocket braking.

As for lift during hypersonic braking, the tipped heat shield lift will be low initially, high up. Then in the middle of the deceleration deeper in the "air", there is lots of lift as the wind pressures are large at high speed and higher density. Late in the run, in the Mach 3-to-5 range, there is density, but not much speed, so lift forces are falling into insignificance again, and the trajectory inherently steepens. Rocket thrust lift would be advantageous here. And in the deceleration from below-Mach 4 to around-Mach 2.5, there is no aero lift, so rocket thrust lift could help there, too. Below Mach 2.5 is when you pop the chute.

GW

Impaler · 2012-06-21 17:20:40

Rune wrote:

As to the shuttle orbiter specifically: 75mT (empty orbiter) divided by 376m^2 (best case area) gives ~200 as beta. That is worse than MSL by 50, but that is because it is way, way easier to fly on Earth. Which is kind of obvious. And while we are at it, you might consider a shroud to launch a shuttle-sized payload inside, 'cause the shuttle stack is quite retired. Is it too obvious I dislike wings for rockets a lot?
A more ingenious idea to increase surface area and lower beta for heavy payloads is to reuse the whole shroud for a heatshield, then drop it to expose engines. Have a look here, by the end of the pdf, they analyze a shroud/aerodynamic reentry shell combo for the Ares V. It shallows almost half the payload, but it can drop stuff at mach 2.2-3.1 at 5.1-7.3 kms of altitude in Mars. The shame is that you would have to drop the shroud, which kind of rules out reuse. And this is the way clever aerodynamic tricks go in Mars: since there is no air to speak of, you end up shallowing all the mass in aero surfaces.

Rune. As I said, I'd stop banging my head on the wall and start figuring out hypersonic rocketry.

As I said the payload would need to come down, but it's certainly achievable to get a low Beta with a payload of substantial mass. I know very well the shuttle is retired and I'm only using shuttle as a stake-in-the-ground for the launch-ability of that kind of SHAPE, Energia-Buran also validates the side-mount launch concept with the main engines in the bottom of the external tank. Congress also seems to be determined to create a SHLV virtually identical to SLS and such a vehicle would certainly be able to accommodate a side-mount payload.

To get the same 376 m^2 area from the conventional conic shields in payload fairings you'd need a 22 meter diameter fairing, so again the only way to launch a shield of that size is via side-mount. If you have some blind-prejudice for 'wings' then call it a "large triangular side-mount launched heat-shield" cause that's all it really is, a way to get a large heat-shield into space without needing absurd payload fairings. I went and explicitly stated that it the payload detaches at the normal heat-shield separation point and their is no 'glider' phase as that obviously won't work in thin Martian atmosphere so it dosn't need to have any wing-like aerodynamics. The question is just, how much would a monolithic hollow disposable delta-wing shaped heat-shield mass relative to its payload at acceptable Beta values? I suspect it won't be much more then the conventional conic heat-shield.

Now if we can overcome the rigid heat-shield limitation through either inflatables or folding techniques then fairing diameter is no longer setting a hard upper limit on heat-shields then this all goes out the window and we would just use that new tech inside a conventional fairing which is certainly more reliable.

Last edited by Impaler (2012-06-21 17:23:56)

Rune · 2012-06-21 18:37:04

Sorry if it sound like I'm teasing you but... have you read the link I provided? There's you rational way of getting a big surface for aero-deceleration, and it cuts the useful payload of an Ares V to 60mT apart from the massive, heat-shielded shroud, which masses 50mT in itself. As to shape, that only affects L/D, not the ballistic coefficient (well, it's not the dominant factor, to be more precise, Cd also plays a part), so it makes no difference one way or the other, and a delta wing is more wasteful of structural weight than a conic section close to a circle. Just... trust me, as long as you don't intend to fly, just decelerate in a controlled trajectory, then you are better served with more-or-less oval shapes. There's a reason they are the ones used and suggested.

As to sidemount configurations, while they may make a lot of sense for clustering first stages around second stages (like Angara, Delta IV, Falcon Heavy... they must be in fashion), they are not so nice for payloads. Too close to the business end, and it works very bad from a structural standpoint (all those torques on the primary structure take weight to handle).

Parashields /inflatable heatshields may be an ever better solution than shrouds, with regards to weight fractions. Or so their researchers claim, at any rate. I would put money in certifying both for interplanetary reentry speeds so at least one pans out and we can skip shroud limitations completely, while potentially adding reusability. I think some inflatable test has actually flown, but at much lower suborbital velocities.

Rune. Sadly, no real money goes to the useful stuff being developed... this problem first appeared decades ago, and the solutions suggested then, I'm sure.

TwinBeam · 2012-06-21 19:30:08

GW Johnson wrote:

Answering TwinBeam in post number 117 above:
I think I understand what you were proposing. Using rocket thrust as lift to hold the entry vehicle in a flat trajectory long enough to slow down. Yes, that would work. However, if you have a blunt heatshield facing into the slipstream, and you tip the top edge forward a few degrees, you can generate a lift force comparable in magnitude to your drag force during the real hypersonics. That's lift without rocket thrust at all. It works down to around Mach 4-ish, with most any blunt shapes.
GW

Assuming a L/D of 0.2, a capsule starts falling when drag force goes below about 2g's (4m/s^2 / 0.2 = 20m/s^2) - no point using rocket before that, as you get free lift to stay in the air and get free braking.
Assuming a 30T, 8m diameter vehicle, and getting low enough that air density is around 0.01kg/m^3, at 2g's you'd still be clipping along at about 1430m/s.
If you're going to rocket decelerate at constant 2g's beyond that point, you need another 72 seconds with average rocket deceleration of 1g (+ avg 1g aerobraking) to kill horizontal velocity - about 720m/s deltaV equivalent.
Assuming an average of 2m/ss falling acceleration (4m/ss minus average 2m/ss aerodynamic lift), you'd be heading down at 144m/s if you did no vertical braking. So rocket braking at 2g (net 1.6g after gravity, 16m/ss), you'd need to slow the fall for another 9 seconds - 180m/s more. Total 900m/s effective delta-V.

If on the other hand you use the rocket purely for lift from 2g drag (20m/ss) down to 0.4g (4m/ss), you'll have shed all but 450m/s horizontal velocity, or 980m/s shed before it's cheaper to use the rocket to brake. Average aerodynamic lift of 0.2*(12m/ss avg drag accel) = 2.4m/ss average lift, so average of 1.6m/ss rocket acceleration needed to stay at a fixed altitude. With average of 12m/ss avg deceleration, that's 980/12 = 82sec at 1.6m/ss for 130m/s rocket deltaV. Then rocket brakes at 1.8g avg (2g with 0.2g avg drag) for another 23sec - 400m/s deltaV. And it's falling at 90m/s after that, so another 5.6sec at 2g (net 1.6g) to kill that, for 110m/s more deltaV. Total net of 640m/s rocket effective deltaV.

Or about 30% less rocket acceleration required.

TwinBeam · 2012-06-21 20:40:39

Impaler wrote:

Splitting up the Crew while seemingly an attractive solution to the current mass limitations on EDL it is a huge multiplier for lose of crew and mission. Also the surface habitat will almost certainly need to be monolithic or very near that and it's mass is going to drive the EDL tech, if we can't land a suitable habitat then theirs no mission even if Scotty could 'beam' them down. Once we possess the EDL tech for the habitat, the crew is a rounding error.

Splitting the crew might multiply the chance of losing one crew member - but greatly reduces the chance of losing the full mission.

Assume equal 5% chance of a fatal crash for a 4 person lander versus each of four 1 person landers. Chance of killing all 4 crew is 5% for 4 person, but only 0.000625% for 4 separate landers. Of course, the chance of ANY crew dying is 5% for 4 person, versus about 18.5% for 4 separate landers.

So which would you rather have - a 5% chance of all 4 crew dying, or a ~20% chance that 3 crew will have to carry on after one dies?

I'd be willing to bet that if all 4 died, it'd be at least a decade delay, if ever we went back. If one died? Global mourning and pride in the bravery of the other 3, and "we must carry on to show his sacrifice was not in vain" - while in the background engineers scramble to fix the specific problem revealed, so the next landers each have only a 3% chance of a fatal crash.

And that's before considering that a one person lander MIGHT be safer, if you make the heat shield the same radius as the four person lander, since it should get far more aerobraking value, and so have more fuel reserve for control and landing. You'd end up putting more total mass into human landers, but split 4 ways each lander will have more margin for error.

Regarding the hab: I don't think it does need to be monolithic. If you send it down in parts and teleoperate robots from Phobos to assemble and test it, you've gotten away from that requirement.

Even better, just send down structural supports and long skinny air-tight sandbags for the robots to fill and weave between inner and outer structural supports and cement together before burying the whole thing. Early Martians could live in "log cabins", like earlier pioneers.

New Mars Forums

Announcement

#101 2012-06-18 08:56:19

Re: Landing on Mars

#102 2012-06-18 09:02:57

Re: Landing on Mars

#103 2012-06-18 09:50:49

Re: Landing on Mars

#104 2012-06-18 10:03:08

Re: Landing on Mars

#105 2012-06-18 10:33:47

Re: Landing on Mars

#106 2012-06-18 14:06:10

Re: Landing on Mars

#107 2012-06-18 14:35:40

Re: Landing on Mars

#108 2012-06-19 06:11:18

Re: Landing on Mars

#109 2012-06-19 12:22:27

Re: Landing on Mars

#110 2012-06-19 15:25:45

Re: Landing on Mars

#111 2012-06-19 20:45:48

Re: Landing on Mars

#112 2012-06-20 05:34:28

Re: Landing on Mars

#113 2012-06-20 06:28:04

Re: Landing on Mars

#114 2012-06-20 07:27:02

Re: Landing on Mars

#115 2012-06-20 07:47:31

Re: Landing on Mars

#116 2012-06-20 08:36:46

Re: Landing on Mars

#117 2012-06-20 22:27:12

Re: Landing on Mars

#118 2012-06-21 03:06:06

Re: Landing on Mars

#119 2012-06-21 06:18:04

Re: Landing on Mars

#120 2012-06-21 09:32:45

Re: Landing on Mars

#121 2012-06-21 12:27:27

Re: Landing on Mars

#122 2012-06-21 17:20:40

Re: Landing on Mars

#123 2012-06-21 18:37:04

Re: Landing on Mars

#124 2012-06-21 19:30:08

Re: Landing on Mars

#125 2012-06-21 20:40:39

Re: Landing on Mars

Board footer