New Mars Forums

Yes, I'm still alive. And here to run some thoughts past you guys.

I've been thinking about the problem of getting to Mars and back. (I'm less concerned with what we do there, or what stuff we need to land there permanently)

Key to that is the landing/ascent problem and I think critical is the ascent problem. Hence my interest some months ago in reusable landing/ascent vehicles.

Then I've considered how the overall architecture works. How to achieve simplicity, synergy and robustness. Which means integrating hardware in different ways.

First, I'm going to suggest something radical. And I'll come at it this way. Earlier I went into some detail as to how a fully (many years) reusable landing/ascent vehicle might look. It turned out that such a beast is technically possible, but with CO/O2 fuel is marginal - so marginal and also heavy as to be excluded. The next phase was again an attempt at a fully reusable landing/ascent vehicle using methane as the ascent fuel. This looked more promising. I found it also made sense that if you were to import (land) hydrogen then the descent phase was more economical run on hydrogen. But even so, given the low density of liquid hydrogen the tankage had to be scaled around the task of landing that volume of hydrogen.

As a side note here. For all the talk about mass leverage in importing hydrogen and then converting it to methane, for me the mass leverage is at best 4:1 (methane is 25% by mass hydrogen). And the reason I say this is that the process of making oxygen on Mars can happen without any imported precursor. Oxygen in a sense, is essentially free. (Yes, it does take energy). For me at least, it makes most sense to land methane on Mars, rather than hydrogen. At least until there is a good indigenous source of hydrogen.

What this points to is an ascent powered by methane/oxygen where the methane is imported (landed) and the oxygen is produced locally. I felt I had to spell that out in case it wasn't obvious what I"m going to say next.

The other problem that keeps arising with an ascent vehicle is sheer mass. Neglecting the crew, the mass at takeoff basically scales around the mass of the structure itself. Now, if that structure is also a lander then the mass increases for two reasons. One is heat shield and the other is the forces incurred by a conventional heat shielded landing.

As a consequence you inevitably end up with a lander/ascent vehicle massing well over 30 tonnes fully fueled. And that's optimistic.

And this raises the obvious question: what if you could reduce the mass of the ascent vehicle. Now stepping aside from any form of reusability the answer is a purpose built, single use vehicle. The problem is that in terms of overall mission design you've still got to land the ascent vehicle in the first place, and then you need a separate lander for the crew. Those mass multiplier effects have to be kept in mind before ruling out a reusable design. Now, within the context of reusable lander/ascent vehicle how do you minimise the mass at takeoff? Well the answer is to take away hypersonic reentry. That eliminates the need for a conventional heat shield. It also considerably reduces the stresses on the vehicle and allows it to be pared down to a few tonnes.

So to cut a long story short, this is a lander/ascent vehicle that masses 18 tonnes fully fueled. The essential vehicle is a few tonnes. With crew, and basic life support its closer to 5 tonnes. Those numbers give it a delta-V capability of 4.5Km/s with methane/oxygen fuel. Which is sufficient for an ascent to any orbit, with margin.

Which then takes me to my next controversial decision. That is to use mostly propulsive landing. The key assumption here is to start from a low Mars orbit and then deorbit leaving your velocity at just over 3.5Km/s that's a velocity reference to a inertial frame. Remember, the planet itself spins within this frame.

At a first approximation, were you to do an burn that zeroed your velocity you'd have used 3.5Km/s of delta-V from your engines. Do this at a sensible altitude (say 100Km) before you've the density is of any consequence. Now if there were no atmosphere (and the planet wasn't spinning) you'd free fall vertically and hit the surface 229 seconds later at 871m/s. In this theoretical case you might apply another 0.9Km/s of delta-V. Now you've consumed 4.6Km/s. That's your worst case. And as it happens, the vehicle, fully fueled is capable of almost that.

But physics is on your side. Even in the stall and drop scenario above, you've still the assistance of air drag. It turns out you hit the surface at typically 300-400m/s (depending on assumptions about Beta - mass to effective drag area). You never go past Mach 3.

And that's according to a simulator I wrote earlier this year. What I didn't factor into the simulator was that the planet (and its atmosphere) are rotating - typically around 400m/s in zones near the equator. What this means is that relative to the air and to the surface you're already going that much slower.

So in the worst case (non optimised) you need just over 4Km/s of effective delta-V from your lander.

Now, further optimisation means taking advantage of the modest thermal loads at supersonic or near hypersonic velocities, again relative to the air. Which means slowing down to a relative velocity or more like Mach 2 - about 500m/s. Taking that into account, and tweaking the simulator (bending the frame of reference) what happens is your vehicle again accelerates to over Mach 3 before again meeting the thickest part of the atmosphere (around 25Km) and then slowing to a terminal velocity closer to the above 300-400m/s mark.

Now, allowing for terminal maneuvering and other losses and my best guesstimate is you can land mostly propulsively with the equivalent of 3.8Km/s of delta-V. Now, this figure may be improved upon, but there is a diminishing return from doing so. Likewise there is a chance I'm out, but that doesn't make much difference to the masses of fuel we're dealing with below.

That's a mass ratio of just on 3, or an all up mass of 15 tonnes. 10 tonnes fuel.

What this amounts to is a light weight vehicle that relies upon propulsion to avoid hypersonic air travel. It does rely upon some modest and light weight thermal protection in critical areas. But the temperatures involved are in the hundreds of degrees C. It also enables other design refinements like not having to have deploy-able landing gear or engines hidden away behind a shield. There are basically very few surfaces not made of metal or protected by cryogenic fuel. There are key systems that do need protection but this involves only a small amount of thermal protection mass. Think high temp textiles and ceramic fibres. Even the landing legs are not going to get too hot for a decent high temp alloy.

Supersonic retro propulsion is unavoidable here. But SRS is unavoidable in any large payload delivered to Mars. Its a key enabler. We need to research it, and then we'll discover where the envelope is. You'll notice that I'm doing hypersonic to supersonic deceleration high enough that the density of the air poses no stability problems (by definition basically).

Now that I've argued for the feasibility of a most propulsive landing, at least for small payloads such as a crew, I'd like to show how that principle can then inspire and integrate into an overall architecture.

Aside from a reusable lander/ascent vehicle I need only two more kinds of vehicles. One is a big dumb booster (LH2/LOX). The other is what I've usually called a space hab, but lots of others refer to this as a transit vehicle. My transit vehicle is unsurprisingly a crew compartment, tankage, and propulsion. One element of design is to put a docking portal axially on both ends. An arrangement that allows chaining. One end of the vehicle the crew compartment narrows (think sleeping quarters, radiation shelter). Around that is the tankage and a minimal propulsion unit (I'll explain why later). So the pressurised crew compartment has the full diameter on one end but at the other is basically a tube that connects to the portal.

I'm going to propose that minus fuel but including life support, consumables and so on, this vehicle is going to mass 20 tonnes. Now I know what you're thinking. Just hang on a sec.

In transit to Mars, and return, there will always be (barring emergencies) two of these vehicles, either flying in formation or docked together. Hence an all up mass of 40 tonnes (without fuel) and a target crew of four. In practice during flight you will probably want to dock these two units and then optimise their internal space, allowing more room in one, and more mass and thus radiation protection in the other.

Its a concept I originally hit upon because it allows mass to be transferred out of one vehicle and into the other, so that one vehicle can do something particularly costly in terms of energy, like moving from high Mars orbit to low Mars orbit and then back. As it turns out I don't think that's needed, but there's a dozen good reasons for having two essentially identical vehicles, including recovery from damage, systems failures and so on. I should add that these have the capability to transfer fuel between them.

Now, as far as the overall architecture goes we're going to use L2 as a staging point. (Look I'm not that fond of integrating moon based activities but what appeals to me about L2 is basically avoiding a high orbit that dips in and out of the radiation belts).

So conceptually this mission begins and ends at L2, but we'll discuss how to get there an back too.

Each transit vehicle has fuel capability for around 2.2Km/s or about 18 tonnes of fuel each. That's rated as sufficient to return from high Mars orbit to L2 under nearly all circumstances and that also includes propulsive capture. Now I did toy with the idea of aerocapture but in the end, I'd rather have the reserve of fuel even if its not fully used. The on board fuel is there for the return journey from high Mars orbit to L2. A typical mission would use more like 15 tonnes of fuel or less (per hab unit). Minimal energy trajectories might use as little as 12 tonnes. Understandably the systems on the vehicle include the ability to keep the fuel at near zero boil off.

The forward journey from L2 to high Mars orbit is simply a boost with a conventional H2/LOX booster. There are of course other ways to do this but I'm going to keep the exposition simple.

One thing I would like to have (not essential) is a top side docking port for the lander/ascent vehicle. Doing so means it can dock with the transit vehicles (at either end).

Now, the stack as it travels to Mars is a booster docked with a transit vehicle, which in turn is docked with a transit vehicle, which in turn is docked with an (upside down) lander. What we have here is bump into place assembly. No need to get out with a spacesuit and a spanner.

The lander at the top of the stack is pre-fueled with enough fuel to reliably transfer to low Mars orbit. Which is about 2.5 tonnes of fuel.

Now, lets step back in time. Prior to any of this happening, a number of missions have taken place. Leaving infrastructure in place, including the oxygen production plant on the surface and a store of methane. You could go so far as to have a fully fueled ascent vehicle waiting as a backup. I'm also going to gloss over all the other stuff you need including the surface hab, rovers, etc. All of which is out of scope here.

What is important is that for every manned flight to Mars, there is another unmanned flight. Now this happens on a slower, lower energy trajectory, again using a big dumb booster. This unmanned flight places key hardware in low Mars orbit. This is time to arrive in low Mars orbit prior to the crew arriving in high Mars orbit. Arguably this could or should happen one cycle earlier.

The unmanned flight delivers to low Mars orbit two items. One is another identical transit vehicle / space hab. Identical in form and design, but limited in consumables. However its fuel tanks are completely filled. And one little quirk. The other item is a lander which is also fully fueled. One design modification places a liquid methane tank (3 tonnes) in place of the crew compartment. The basic mass of this vehicle is 3 tonnes. Again, the booster, space hab and lander stack end to end through their docking ports.

Upon arrival on Mars, the lander that forms part of the unmanned flight is free to land. Thereupon it delivers its cargo (3 tonnes of methane) to the surface. And an oxygen production plant in the form of a rover, provides the necessary cooling system. The lander is also used as an oxygen storage vessel. The space hab (it doesn't quite make sense to call it a transit vehicle) now remains permanently in low Mars orbit. Its function is to provide a safe haven - because the lander has limited life support.

Now back to the manned mission. On arrival in a high Mars orbit, the crew transfer to the lander and use its fuel to transfer to the space hab in low Mars orbit. They wait whilst the lander takes on fuel. The crew transfer again to the lander and it departs and lands on the surface. A full surface mission takes place. Sometime during this period the lander is refueled. The crew again use the lander to ascent to orbit, docking with the space hab. Again they wait for the lander to take on sufficient fuel (about 2.5 tonnes) to return and dock with the transit vehicle which has remained for this time in high Mars orbit.

The stack in high Mars orbit comprises the two separate space habs docked together (as they were when the crew arrived) and the lander which is docked to one end. Returning from high Mars orbit to L2 involves using the engines of the lander. Note that the two space habs have their own minimal propulsion which could achieve the same goal, albeit with reduce efficiency (a much longer burn). Note also that the lander draws down fuel from the space habs.

In a nominal mission this will leave roughly a third of the fuel on board. This fuel will be consumed to propulsively capture back to L2. During transit the fuel is shifted towards one end to provide a lower radiation environment on that end - up until its used.

From L2 it is possible to either use the lander to approach low Earth orbit, or to supply a lander capable of directly landing on Earth.

Now taking a longer view, what happens is that the space habs are cycled through several phases over multiple missions. A typical life cycle involves going to L2, becoming part of a manned mission, returning to L2, then returning to low Earth orbit to be used as a vessel for resupply. Then returning to L2 and delivering consumables. Then the same unit is sent as part of an unmanned mission being delivered to low Mars orbit. So a typical unit would see service over 2 cycles or close to 5 years.

You know personally, I'd prefer to see things being recycled more than this but in the end something has to deliver fuel to low Mars orbit. The practical consequence of all of this is that over several missions there may be 2 or 3 (or even 4) functional space habs in low Mars orbit. Which is probably to the good. In the end though, something that is designed to have multiple copies made is probably going to benefit from economies of scale and will improve over time.

Notice also that I've basically designed this so that one singular methane/lox engine design accounts for both the multiple engines (probably 8) on the lander and the 2 or so on each space hab. And as time goes by there are going to be spares everywhere.

And its exactly the same life support system and ancillary equipment everywhere.

The really nice thing about this architecture is that its very resilient to failures. You can for instance limp home in one space hab. You can transfer fuel where needed. Which means in emergencies you can borrow fuel from one vehicle at the expense of crashing the other. In a similar fashion its also resilient to irreparable damage to fuel tanks, life support systems etc.

Another feature is that for just a little more fuel you can use the lander as a taxi to the Martian moons. I think everyone would jump at the opportunity to tie together both a surface mission and sample returns from the moons.

Now, seen from L2 the mass of the manned mission is about 80 tonnes. And the unmanned part about 60 tonnes. Given that the assumed delta-V from L2 to high Mars orbit is 2Km/s then the mass ratio for a H2/LOX booster is about 1.56. So a booster of about 45 tonnes for the manned mission. Which places it strategically below 50 tonnes. Likewise the two fueled space habs and the lander are flown separately. Similarly for the booster for the unmanned mission.

Whilst this takes multiple separate parts, none are over 50 tonnes. The only point where mass does add up is considering the LEO to L2 part of the system.

Oh, before I forget. The lander that flew back with the crew. It switches roles back at L2 and goes on to the next unmanned flight and ends up as the vehicle that transfers fuel to the surface.

Now between LEO and L2 things get more complex because not everything has to be brought up to L2 every time. So I'm not going to bore with detail, but in the worst case, using conventional technology, you're looking at very roughly 250 tonnes all up in LEO terms. Plus some basic infrastructure at L2.

This is where I think ultimately we're going to end up with a solar tug and where it makes the biggest difference. And that's from LEO to L2 and back. Sooner or later may wish to refurbish things in LEO and also consumables and fuel make up the majority of the mass. If so the combined mass per mission is around, or a bit under 200 tonnes.

At this point I have to say that even the fully conventional boost to L2 situation is a considerable improvement in terms of overall mass than anything I've seen from NASA.

Now step back and consider this. Initial mass in low Earth orbit is not everything. I can think of ways to halve it. Cutting corners, using aeorcapture etc. But as the cost of launching fuel into low Earth orbit comes down, it will become irrelevant. Indeed, a solar tug only becomes really viable over multiple missions.

The thing I keep coming back to is simplicity, redundancy, interoperability of parts. I've basically proposed two key vehicles. And anything else is just a minor design change on those two vehicles. This is why I've proposed to land fuel (for the next ascent) propulsively with another lander. Its not like I couldn't save a few tonnes there, but it means design effort can be focused on just one thing.

Now, there will be other things we'd need to land on Mars. I'd argue there's no real need to land anything heavier than about 15 tonnes. And the reason is a lot of mass is just consumables and fit out. Its not like the crew aren't being paid well. There are design challenges for such large loads but I'd submit two things.

One is it isn't as bad as NASA thinks. You don't need 100 tonne class entry mass vehicles to land 30 something tonne payloads. You'd never need more than half that. Second, whilst its a challenge, its not beyond our current knowledge plus (and again the key enabler) a better understanding of SRP.

My key point in all of this (and its why I came here in the first place) is to demonstrate why its a good idea to stop thinking in terms of landing people inside very heavy payloads and instead land them separately in purpose built landers with higher margins.

And the flip side of that is, it also makes it easier to design the landing system for the non manned large payloads, because we're not worrying about the crew there.

There are of course other features to this architecture I like. For instance you can extend it to provide (a small amount of) gravity en-route.

Anyhow, enough verbage. What do you guys think?