Mars 9 tons at a time.

[gunsandrockets]

<I would need over a dozen of the shuttle orbiter's main RCS thrusters. What do those weigh and what's their ISP?><br /><br />I can ballpark that without looking it up. Shuttle bi-propellant RCS thrusters I bet have an ISP around 300 seconds. Thrust to weight ratio is probably considerably better than an RL-10, since an RL-10 is a high ISP engine. Interesting rule of thumb is -- all else being equal, engine ISP is inversly proportional to T/W ratio, so doubling ISP equals half the thrust to weight ratio.

 

[keermalec]

TIW:<br /><br /><blockquote><font class="small">In reply to:</font><hr /><p>Where do you get that a ballute only works below Mach 3.5? The one I referenced just above, the "HyperCone" had that limit, but there are other designs around that are intended for primary deceleration from orbital velocity. <p><hr /></p></p></blockquote><br /><br />OK good point. I am not very knowledgeable in this domain and did not think ballutes could slow you down from such high speeds. <br /><br />However:<br /><br /><blockquote><font class="small">In reply to:</font><hr /><p>Here is a link to a graph of the entire descent profile. The horizontal scale is velocity and the vertical scale is altitude. Both are logarithmic, so you can see the appropriate amount of detail at each stage. The upper right corner represents 3550 m/s at 150 km altitude. The vertical pink line over to the right represents Mach One. <p><hr /></p></p></blockquote><br /><br />3.55 km/s represents re-entry speed from martian orbit.<br /><br />If you're coming from Earth, your delta-v will be minimum 4.5 km/s according to Borowski (Optimal Mars transfer in 2014). Average delta-v at Mars atmosphere for a Hohmann transfer is 5.47 km/s plus or minus 1 km/s depending on conjuction.<br /><br />That means you must provide for an atmospheric entry velocity of at least 4.5 km/s.<br /><br />Considering our vehicle is inhabited, it may be decided that the trip should be shorter in terms of time spent in space (radiation exposure, muscular atrophy due to lack of gravity ...etc). Borowski suggests cutting the trip time down to 6 months instead of the 9 a Hohmann transfer would give. This increases the Mars atmosphere entry delta-v to ... 8.7 km/s (page 20).<br /><br />So the aerobraking ballute should provide for entry at 8.7 km/s for an inhabited vehicle. <div class="Discussion_UserSignature"> <p><em>“An error does not become a mistake until you refuse to correct it.” John F. Kennedy</em></p> </div>

 

[thereiwas]

The paper by Braun and Manning linked to back a few pages, "Mars Exploration Entry, Descent and Landing Challenges", says that doing a two-part entry has several advantages. So I have been assuming an aerocapture first, using the ballute to slow from Earth/Mars speed to orbital, then doing the final entry later.<br /><br />This also allows for more precise landing because it is not dependent on Earth departure time or transfer orbit. It gives time to locate the landing beacon from the first package down, make adjustments, and carefully time the descent.<br /><br />I am also figuring this for unmanned cargo, since several of those need to get there first and in the spirit of this topic, "what can we do now?" Would you want to make that trip in an 8t vehicle? If a Bigelow habitation module is included we would want to leave that in orbit as well and perhaps descend to the surface in a descent stage modelled on this proposal. I expect the manned missions to be more complicated and larger, thugh not so large as NASA thinks they have to be.

 

[keermalec]

Good answer. A multipass aerobrake is certainly the way to go as it reduces both the size of the ballute and the stresses associated with a high delta-v. <br /><br />I remember now that one of our premises was to use a Delta IV Heavy launcher which necessarily sends our payload on a Hohmann transfer, and not on a faster 6-month trip. The arrival delta-v is therefore 5.47 km/s on average.<br /><br />If we do a first aerobrake pass to slow us down to a 150 x 33793 km orbit (1 martian day orbital period), the necessary delta-v is 0.88 km/s, easilly supported by our ballute aerobrake.<br /><br />A second aerobrake of 1.2 km/s should bring us to a 150 x 250 km orbit and form there to atmospheric entry with a 3.43 km/s delta-v.<br /><br /> <div class="Discussion_UserSignature"> <p><em>“An error does not become a mistake until you refuse to correct it.” John F. Kennedy</em></p> </div>

 

[solarspot]

I can't find any information about the shuttle orbiter's R-40A thrusters on Aerojets website. The only information I can find on these things actually, is via astronautix.com... According to This page, the R-40B engine (slightly better for a lander like this) has an Isp of 293s, thrust of 4kN, and a T/W of about 56. Our lander would need 16 of these at least, with a mass of 116kg or so. As an aside, a 400N hydrazine thruster like This one would have an Isp of around 220s, T/W of 17. The lander would need at least 150 or so of these things, for a total mass of just under 400kg. I guess bigger is better sometimes <img src="/images/icons/laugh.gif" /><br /><br />[note]: There is the issue with the R-40A or R-40B engines, that I am not actually aware whether those engines are still in production. On the other hand, 400N thrusters like the one on EADS's website are quite common, and EADS (a European company) actually (relatively speaking) mass produces them.

 

[thereiwas]

It seems that for a 400kg penalty I can eliminate the solid engines entirely (a considerable simplification). Assuming engines with Isp=320 and T/W of 70kN/180kg, the sweet spot in maximum non-engine mass delivered to the surface seems to require 100 kN peak thrust starting at 450m altitude. This is considerably higher than my previous solutions but allows more time for horizontal corrections to land at the right spot. I am going to keep playing with the thrust modulation equation to see if I can lower the peak to a more constant level. It is not behaving the way I want yet.<br /><br />Starting higher requires less peak thrust because there is more time to slow down. Less peak thrust means smaller engines, but they have to burn longer. I think the 400 kg penalty is because the solid engines had a better T/W. Given how crowded this thing was getting with engine nozzles, perhaps this is for the best. <img src="/images/icons/smile.gif" /> <br /><br />Found the mistake. I got the max thrust down to 50kN and the non-engine landed mass to 7100 kg. Engine start is higher, close to 1km. Lots of time to adjust landing spot.<br /><br />Now, I wonder what happens if I can leave the ballute on until much lower....

 

[gunsandrockets]

<I got the max thrust down to 50kN><br /><br />Huh. That's about 11,000 lbs thrust, right? That seems higher than I would have expected for an 8,000 kg lander. I'll have to go check and see what the T/W ratio was for the old Viking lander and see how it compares.

 

[gunsandrockets]

Viking 1 Lander<br /><br />link... http://nssdc.gsfc.nasa.gov/database/MasterCatalog?sc=1975-075C<br /><br />"At the time of separation, the lander was orbiting at about 4 km/s. After separation rockets fired to begin lander deorbit. After a few hours at about 300 km altitude, the lander was reoriented for entry. The aeroshell with its ablatable heat shield slowed the craft as it plunged through the atmosphere. During this time, entry science experiments were performed. At 6 km altitude at about 250 m/s the 16 m diameter lander parachutes were deployed. Seven seconds later the aeroshell was jettisoned, and 8 seconds after that the three lander legs were extended. In 45 seconds the parachute had slowed the lander to 60 m/s. At 1.5 km altitude, retro-rockets were ignited and fired until landing 40 seconds later at about 2.4 m/s."<br /><br />"Propulsion was provided for deorbit by a monopropellant hydrazine (N2H4) rocket with 12 nozzles arranged in four clusters of three that provided 32 N thrust, giving a delta-V of 180 m/s. These nozzles also acted as the control thrusters for translation and rotation of the lander. Terminal descent and landing was achieved by three (one affixed on each long side of the base, separated by 120 degress) monopropellant hydrazine engines. The engines had 18 nozzles to disperse the exhaust and minimize effects on the ground and were throttleable from 276 N to 2667 N. The hydrazine was purified to prevent contamination of the martian surface. The lander carried 85 kg of propellant at launch, contained in two spherical titanium tanks mounted on opposite sides of the lander beneath the RTG windscreens, giving a total launch mass of 657 kg....Approximately 22 kg of propellants were left at landing...After...landing, the lander had a mass of about 600 kg"<br /><br /><br />Analysis<br /><br />So, Viking has a touchdown mass of 600 kg and descent thrusters totalling

 

[thereiwas]

Ok then. Considering the rocket ignition height and terminal velocity were about the same, but we are planning for something that masses nearly 12 times more, I'd say we're in the right ballpark!<br /><br />But where do you get 8328 N? I read it as saying 2667 N.<br /><br />I don't think we need to worry so much about dispersing the exhaust, since we are not trying to do scientific experiments on the spot.<br /><br />

 

[gunsandrockets]

<I'd say we're in the right ballpark!><br /><br />Yep.<br /><br />< But where do you get 8328 N? I read it as saying 2667 N. /><br /><br />I confess to a little confusion over the passage. But to the best of my reading comprehension, I took the description to mean 2667 N per EACH of the three descent engines. But maybe I got it wrong.<br /><br />As I think about it more carefully, I think I am right. If the total maximum thrust only equalled a T/W of 1.2, then the thrusters would be decelerating the lander with tremendous inefficiency since 83% of the thrust would only be counteracting the gravity of Mars. With the higher T/W of 3.7, only 27% of the thrust is counteracting the gravity of Mars and 73% of the thrust would slow Viking down.<br /><br /><I don't think we need to worry so much about dispersing the exhaust, since we are not trying to do scientific experiments on the spot. /><br /><br />Yep. <br /><br /><br />

 

[thereiwas]

<p>Here are some numbers showing sensitivity to engine size. The first table shows the major initial assumptions. (Sorry about the blank space before each table - I don't know why it is doing that.)<br /><table vspace="10" border="1" align="left" hspace="12" cellpadding="4" bgcolor="white"><br /><tr><th bgcolor="orange" align="center"><b><font color="black">Parameter</font></b></th><br /><th bgcolor="orange" align="center"><b><font color="black">Value</font></b></th><br /></tr><br /><tr><td bgcolor="yellow"><b><font color="black">Ballute radius</font></b>

 

[j05h]

TIW - What kind of safety margins does your descent rockets include? While it may only take 10-12 seconds to optimally decelerate, the lander should be able to hover/translate some during decent. <br /><br />Perhaps once the process is worked out, the margins can be lowered, especially for cargo landers. <br /><br />With this kind of ballute-rocket, do you think the heatshield still needs a "crumple zone"? That might be the safest, final backup for the payload's safety.<br /><br />Josh <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[thereiwas]

First I am trying to see what the minimums are and how sensitive the solution is to different variables. Then to add a safety margin on top of that we make the propellant tank bigger, and start the engines higher. Another consideration is that the Martian atmosphere is quite variable in its density, according to season and weather. A completely fixed choice of ignition altitude is probably a bad idea, because the achievable terminal velocity is dependent on the density <i>that day</i>. It will have to take continuous radar measurements on the way down to calculate what the density is (working backwards from the velocity and altitude) and then decide where the best place to start engines will be.<br /><br />A crumple zone built into the spherical nose of the cone shape is probably a good idea anyway. I have it touching down at around 30 cm/sec, but 7t hitting a brick wall at that speed would be quite a jolt without some sort of cushioning.

 

[j05h]

So realistically the engines start at 30-60s before impact, trottling up around 20s or 15s. <br /><br />For atmospheric density reading and approach data, I'd put the radar in the beacon precursor craft instead of each lander. Put 1-3 of them down to define your LZ and be local weather/seismo stations as well as navigation.<br /><br />Good news on the crumple zone, then. <br /><br />What materials are you assuming the ballute is built from? Any thoughts on the heatshield material?<br /> <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[thereiwas]

Sorry, I don't know much about the materials involved at all. I have just been working on the flight dynamics. I need some more formulas to plug in before I can have any idea of the max Watts per square meter the ballute will experience. I have some educational material on that I have not looked into yet. Peak G is around 2.1G and area is 706.8 square meters if that helps. One materials problem is the ballute doesn't have any mass to absorb heat, so it all has to be done by dispersal methods.

 

[j05h]

alright, I'll look into the materials a little more, but Nextel, Kapton and metal are my current choices because I can find info on them. <br /><br />Does your model produce any numbers for peak heating? That is a critical bit of info for choosing materials. <br /><br />Part of my interest in water-transpiration cooling is that it saves the ballute and main heatshield from having to absorb or disperse as much heat. Instead, the steam carries some of it away. <br /><br />Josh <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[thereiwas]

No, the peak heating is the next part I need to add. It is easy enough to calculate the total energy generated (from loss of kinetic and potential energy each second) but most of that stays outside the boundary layer. How much makes it to the vehicle depends on the details of laminar and turbulent flow just inches away from the leading surface and I don't have the equations for that yet.

 

[thereiwas]

I found the heating equation for an Earth reentry:<br /><br /><font color="yellow">Qdot = 1.83e-4 * v^3 * sqrt( rho/radius )</font><br /><br />Where Qdot=Watts per square meter and rho is the atmospheric density in kg/m^3. I already compute Martian air density all the way down for the drag computations. Using this Earth formula the total heat load is <font color="yellow">14 kJ</font>per square meter (9.9 MJ overall) and the max Q is <font color="yellow">175 Watts</font>per square meter (which seems low). The equation is empirically derived and is not actually correct for Mars. I don't know how close this is.

 

[thereiwas]

This article in <i>Science</i> points out the extreme variability in atmospheric density on Mars, both seasonally and daily. They saw up to 50% variability. There are even standing "waves" in parts of the atmosphere.<br /><br />Density (rho) plays a direct part in the amount of drag generated during entry:<br /> <font color="yellow">drag = 0.5 * rho * v^2 * Area * Cd</font><br /><br />I would not be surprised if this is a factor in the loss of so many Mars missions, due to arriving on an unexpectedly "thin" day.

 

[keermalec]

Maybe this will help:<br /><br />In the June 1998 Addendum to NASA's mars Design reference Mission 3, the peak temperature of the aeroshell is 2800 F for a Mars atmospheric entry velocity of 8.5 km/s and a diameter of 7.6m.<br /> <div class="Discussion_UserSignature"> <p><em>“An error does not become a mistake until you refuse to correct it.” John F. Kennedy</em></p> </div>

 

[thereiwas]

Here are some results varying the atmospheric density from nominal. I just apply a uniform factor to the <font color="yellow">rho</font> returned form the usual formula.<br /><br />In summary, to deal with a 50% reduction in air density, the vehicle would need to carry an extra 63% of propellant to deal with the three times higher engine ignition altitude and the almost doubled velocity at the time of ignition. There is somewhat less aerodynamic heating on the way down, and the engines use closer to full power (45kN in this experiment). My engine control equations attempt to maintain a constant rate of deceleration and so choose the ignition point based on where max thrust would be required on the way down.<br /><br />It comes out a bit better in practice because of the decrease in mass as fuel is used; the engines do throttle down slightly as that happens so there is room for pushing it a bit closer to the limit by taking that into account when choosing ignition point. For those interested in the details, the engines fire when the altitude drops below <br /><center><font color="yellow"><tt>0.5 * v^2 / (maxAcc - g)</tt></font></center><br /> where maxAcc is the maximum acceleration the engines could provide, and g is the Mars gravitational constant. The actual deceleration called for instantaneously on the way down is <br /><center><font color="yellow"><tt>(0.5 * v^2 / altitude) + g</tt></font></center><br /><br />Again, sorry for the blank space.<br /><table border="1" align="center" cellpadding="4" bgcolor="white"><br /><tr><th bgcolor="orange" align="center"><b><font color="black">Air<br />density</font></b></th><br /><th bgcolor="orange" align="center"><b><font color="black">Ignite<br />altitude</font></b><font color="black"><br />m</font></th><br /><th bgcolor="orange" align="center"><b><font color="black">Heat<br />load</font></b><font color="black"><br />kJ/m^2</font></th><br /><br /><th bgcolor="orange" align="center"><b><font color="black">Max Q</font></b><font color="black"><br />W/m^2</font></th><br /></tr></table>

 

[j05h]

Those differences mean either carrying extra propellant down or basing in low Mars orbit, waiting for better weather.<br /><br />Thanks for the modeling, TIW.<br /><br />Josh <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[thereiwas]

The trick is in detecting 'better weather'. MGS did it very indirectly, by sensitive accelerometers. Get a weather station down early to report ground truth for subsequent landers. The Martian atmosphere is very wierd; we need to learn more about it, especially for entry methods sensitive to it.<br /><br />On the other hand, a bit extra rocket fuel has its uses on the surface too.

 

[j05h]

The first payload to Mars would be a beacon and weather station, serving as an early control tower for the base. It would use and refine the reentry technique, probably mass around 5t payload and be self-powered. <br /><br />Potentially, the first lander would have bandwidth and wifi, general weather and seismic sensors, cameras, LIDAR and radar. The beacon would be part of the lidar and radar instruments or it's own transmitter. It would have the ability to measure atmospheric density and dust. It would provide constant weather updates for incoming craft. <br /><br />Ideally the first lander would also have several deployable penetrators that also impact the landing zone. These would provide direct local sampling early on (with appropriate sensors) or at least secondary radio signals. Until there is reliable Mars GPS, the base would rely on local "Loran" style positioning, starting with the precursor weather station.<br /><br />With Mars' thin atmosphere (minus duststorms), it would be interesting to see how optical data transmission would perform. Canon makes a series that could serve as a high-bandwidth "uplink". <img src="/images/icons/wink.gif" /><br /><br />What would you add or remove from the precursor lander?<br /><br />Fuel on the surface is very useful. I'd save it for powering things like saws, rovers and drills/TBMs. Are we still talking about LoX and Methane or hypergolics? Hypergolics might bring up issues with the crumple zones in the heat shield.<br /><br />josh <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[thereiwas]

Propellants on the way out have to be stable long-term storage for at least 9 months. I think that rules out cryogenics because of the boil-off. Unfortunately the kind of stuff that can be made in situ, for long term use on the ground and to power an ascent vehicle, is LOX and Methane. I doubt a single engine could run on both but that is an interesting problem.<br /><br />In "Red Mars" the heavy construction equipment was powered by hydrazine.<br /><br />But maybe the stuff used outbound and left over after landing can be chemically converted into something more usable? I would like the descent engines to be usable for something as well - they mass over 100kg and I hate to waste anything.<br /><br />We can do better than Loran (which requires multiple transmitters). The old VOR/Tacan technique (updated to modern technology) of rotating phasors and transponders can give distance and bearing information for subsequent landers to home in on. Calibrating the transmitter for "north" would have to be done by astronomy, due to no magnetic field.<br /><br />A scattering of very small stationary weather stations around the surface would be very helpful in seeing when bad weather is approaching. Getting them in place is tricky. I wonder if a completely passive device that can be made very robust to take a hard landing could be scanned from orbit and things like air pressure inferred from how multi-frequency laser beams bounce off a retroreflector? But then one dust storm and it's out of action.<br />