totm dec 2023 Artemis Discussion Thread

[Exoscientist]

On 8/19/2023 at 12:30 PM, tater said:

So meets zero Artemis standards.

As we worked through way up thread, the minimum TLI throw for an acceptable surface mission (decent duration for crew on surface, and at least 2 astronauts, possibly more) is pushing 70 tonnes.

A single-stack sortie lander is an option where a habitat is landed ahead of time, so consumables and crew volume can be small. Land, shut down vehicle, EVA to habitat, then reverse to leave.

 

 That was for missions that first went through the Gateway. We’re trying to avoid that. The proposal of using existing ESA space assets to create the lunar lander built by the ESA, would save $3 billion from NASA’s Artemis budget by eliminating the SpaceX lander. And removing the Gateway would delete another $4 billion. That’s $7 billion saved by NASA for the Artemis missions.

   Bob Clark

 

Edited August 20, 2023 by Exoscientist

[tater]

8 minutes ago, Exoscientist said:

 That was for missions that first went through the Gateway. We’re trying to avoid that. The proposal of using existing ESA space assets to create the lunar lander built by the ESA, would save $3 billion from the Artemis budget for the SpaceX lander. And removing the Gateway would delete another $4 billion. That’s $7 billion saved by NASA for the Artemis missions.

No, it wasn't. it was to do a single stack mission at all.

600 m/s doesn't cost the difference between 49t and 70t. The min requirement for HLS was a 6.5 day surface stay for 2 (for the first mission). This is because of Gateway, even though mission 1 does not in fact require Gateway. Regardless, we would want the missions to exceed the capabilities of Apollo I would assume. Longest EVA duration total was 22 hours (over more than 1 day). I would assume that wanting something like 6 days on the surface is entirely reasonable. The later LMs were ~16.4t, so we have to assume that NASA would want a lander better than that.

We need 900 m/s for TEI. Orion has ~1356.

Orion CSM semi-dry mass is ~17.2t. ~8.6t props (inc RCS). Semi-dry meaning that includes consumables, it's "wet" minus props). LM is 16.4t. Gonna treat it as a unit, assume it can do the lunar sortie.

So Orion CSM+LM stack mass to start is 42.9t. At 45.1t to TLI, SLS can send our stack and it can just brake into LLO! LM can sortie! Orion is trapped in LLO, it used all the current props, plus an additional ~2t of props. Orion by itself, with extra tankage needs 900 m/s to head home. So this retro-stack of Orion+LM needs an additional 2.2t just to reach LLO.

Assuming just 400kg of extra tank mass for an extended SM, Orion CSM needs ~6.7t of props for the TLI burn (with 100 m/s excess for contingency). So we need to burn 11t to brake with 6.7t of residuals. Course even adding those 2 together, we now need more to brake. So we're at ~9t of added props min, on top of our 42.9t stack. Call it 10.1, try 53t stack even. 36.6t of CSM. That's 19.2t of CSM props. Except I am not counting added tank mass (~1t?).

It can brake into LLO, leaving 6t of residuals with massless tanks. But we need 6.7t, so no go.

Looks like if we can get SLS to throw the ~55t stack to TLI we can do it! (inc tank mass, and some margin).

So first step is to change lander requirements to sorta suck.

Second step is to make SLS throw 10t more to TLI than it will ever be able to do.

Third step is for the impossible second step to somehow happen before the heat death of the universe, and at zero cost.

 

Just as a reality check, that's an Apollo LM as the lander. Some mass savings in computers, etc will likely make it better, with more capability, but it's not a "sustainable" lander and meets no Artemis goals (have to look at sample return requirements as well). That lander? Yeah, also needs to be built from nothing. A decade? How many billion $?

 

 

With a surface stay requirement to more like a week, and the crew compartment less austere as the LM, the min mass easily increases by ~10-15t in LEO (65-70t to TLI).

[Exoscientist]

SLS at $2 billion per flight  shouldn’t be used for cargo.  It should only be used for carrying astronauts. For cargo and habitats use the Falcon Heavy. For in-space hydrolox stages on the Falcon Heavy, I estimate 15 tons one-way to the lunar surface if you have low-boiloff tech. If only the stage for TLI is hydrolox, so you don’t need low boiloff, and with a storable propellant lander stage I estimate 10 tons one-way. For Falcon 9 launches it would be 1/3rd of those numbers.

 Since the crew lander would be only to transfer astronauts to a lunar habitat already stationed and provisioned there the lander could be LEM sized at ca. 15 tons.

 By the way the total cost for SLS and Orion of $4 billion per launch is unsustainable. For an sustainable presence on the Moon we need cheaper lunar access. The Starship could do it if it gets operational, but I don’t like the multiple refuelings for a single mission. Robert Zubrin noted if you gave the SH/ST a smaller 3rd stage, then it could do single-launch missions to the Moon and to Mars.

Edit: I looked it up and the Falcon 9 with cargo Dragon can carry 3,300 kg to the ISS. So with all-hydrolox low boiloff in-space stages, it could get, at a payload of 5,000 kg to the lunar surface, more payload than the Falcon 9/Dragon gets to the ISS. Or by using a hydrolox stage only for TLI, no low boiloff required, with a storable prop lunar lander stage, it could about the same as the F9/Dragon to the ISS at ca. 3,000 kg.

 The reason why the F9 going all the way to the Moon can get at or more payload than its payload to the ISS is firstly it wouldn’t use the cargo Dragon at 4 tons for the lunar lander and secondly hydrolox is more efficient for upper stages than kerolox.

  Bob Clark

Edited August 21, 2023 by Exoscientist
Added info on Falcon 9 payload to ISS

[tater]

At $2B it would be cheap. It’s more like $4B with crew. It’s too expensive for any use at all.

[tater]

I would say that at ~$50M/seat (even that is too expensive), the best solution would be to stage crew in LEO. If someone wants an alternate cislunar architecture built around existing vehicles, or evolved versions of existing vehicles, design vehicles that are assembled in LEO.

Ie: like Centaur (who doesn't?), actually do ACES, and create cislunar tugs, etc.

[RCgothic]

We absolutely could have built an LEO rendezvous lunar architecture around F9/FH/Delta IVH/FH/Vulcan/NG.

And it turns out that's actually what they're doing for the 2 HLS Landers, so SLS is completely redundant. Just attach Orion or preferably an uprated dragon or starliner to one of the HLS transfer stages and be done with it.

[Exoscientist]

I’ve been thinking about the possibilities of what the upcoming lunar rovers to the Moon’s south pole might find.  Too bad about Luna-25, but at least one of the rovers has to succeed.  Multiple lines of evidence suggest there may be precious metals there. In that case there would be an economic motive for going to the Moon. In such a scenario more commercial approaches to lunar access would be tried.

  Robert Clark

[sevenperforce]

15 hours ago, Exoscientist said:

For in-space hydrolox stages on the Falcon Heavy, I estimate 15 tons one-way to the lunar surface if you have low-boiloff tech. If only the stage for TLI is hydrolox, so you don’t need low boiloff, and with a storable propellant lander stage I estimate 10 tons one-way.

With a hydrolox upper stage, a fully expendable Falcon Heavy can only send 9.3 tonnes of payload to TLI, let alone to the lunar surface. Hydrolox is fluffy; a hydrolox Falcon upper stage would only be able to carry around 30 tonnes of propellant.

[magnemoe]

On 8/12/2023 at 10:32 PM, RCgothic said:

Some combinations are easier than others.

Stick an existing spacecraft on top of  an expendable Starship? They could knock out a new adaptor from stainless steel and GSE to support it in 6-12 months.

Different stage on top of SLS? It'd be over 5 years to get through the design studies contracting and infrastructure mods.

Building a new stage from parts? 5+1d6 years probably.

This, but the issue here is if starship is seriously delayed. The BO lander is an fallback and might be more efficient for exploration landings while SS is the base builder or then you want to land heavy stuff. 

[magnemoe]

1 hour ago, sevenperforce said:

With a hydrolox upper stage, a fully expendable Falcon Heavy can only send 9.3 tonnes of payload to TLI, let alone to the lunar surface. Hydrolox is fluffy; a hydrolox Falcon upper stage would only be able to carry around 30 tonnes of propellant.

I say modifying FH would be much easier than making an new lunar lander from scratch. Length of upper stage is based on RP1 an hydrolox state could be far longer, yes it would need another engine  and stuff like modified pad. 
You have new glen but it has not even done an wet dress test. I say outside of core business Amazon has not done well, video games and streaming  as been an serious loss. 
And that is not rocket science  Now obviously 10^10000 streaming services is an bad idea unless your better than Netflix, people will sub 2 months a year of go pirate for your one good show. 
 

[Exoscientist]

14 hours ago, sevenperforce said:

With a hydrolox upper stage, a fully expendable Falcon Heavy can only send 9.3 tonnes of payload to TLI, let alone to the lunar surface. Hydrolox is fluffy; a hydrolox Falcon upper stage would only be able to carry around 30 tonnes of propellant.

 I’ve seen numbers for Falcon Heavy to TLI in the range of  ~20 tons

NASA chief explains why agency won’t buy a bunch of Falcon Heavy rockets.
“It’s going to be large-volume, monolithic pieces that are going to require an SLS.”
ERIC BERGER - 3/26/2018, 3:23 PM
SpaceX has not publicly stated the TLI capacity of the Falcon Heavy rocket, but for the fully expendable version of the booster it is probably somewhere in the range of 18 and 22 tons. This is a value roughly between the vehicle's published capacity for geostationary orbit, 26.7 tons, and Mars, 16.8 tons.
https://arstechnica.com/science/2018/03/nasa-chief-explains-why-agency-wont-buy-a-bunch-of-falcon-heavy-rockets/

 But to maximize payload don’t use the kerolox FH upper stage to do the TLI burn. Use the 63.8 ton FH capacity to LEO to carry hydrolox stages for the TLI  burn and for the lunar lander stage. For example a 30 ton Centaur-like stage with a 10 ton Centaur-like stage could get about 15 tons one-way to the lunar surface. This assuming low boiloff tech for the lander stage. By “Centaur-like” I mean getting high vacuum Isp and high, for hydrolox, mass ratio also. 

  Robert Clark

Edited August 22, 2023 by Exoscientist

[StrandedonEarth]

9 hours ago, Exoscientist said:

to maximize payload don’t use the kerolox FH upper stage to do the TLI burn. Use the 63.8 ton FH capacity to LEO to carry hydrolox stages

IIRC correctly the payload adapter and other parts can’t handle the full 64t rated payload; the difference is usually residual props for additional burns. 

Im not sure if the max actual payload has been published..

[RCgothic]

Falcon heavy should be able to send about 20-22t to TLI direct, which is a few tonnes more than F9 has demonstrated. The difference is presumably not insurmountable. But the full 63.5t is probably going to need to include propellant residuals, so no sticking 60t additional stages on top of FH.

20t to TLI is enough to brake 17t into NRHO, which is more than the 13t Orion can ever comanifest to NRHO. SLS Block 2 Crew could technically comanifest about 16.5t to TLI, but it would result in a rather radical flyby manoeuvre.

SLS Cargo is never going to be a thing. SLS is overpriced for crew, nevermind cargo, and the crew program can't spare any. So the biggest payloads it's ever going to loft are going to be comanifested alongside crew. Whilst these could technically be wider than a falcon payload, all the illustrated comanifest payloads I'm aware of would fit on a Falcon.

In fact I think FH is sending two of these at once.

So the only remaining niche of SLS is "doesn't need a propulsion module on the payload" as if such modules weren't common and robotic docking wasn't a solved problem that doesn't need an extra $3.5B and an astronaut crew mission.

And all of this presupposes no LEO rendezvous (FH 32t to TLI) or Superheavy (150+t to TLI).

Edited August 22, 2023 by RCgothic

[tater]

39 minutes ago, StrandedonEarth said:

Im not sure if the max actual payload has been published..

There are plans for an extended fairing to meet the Space Force requirements (along with vertical payload integration). This might change the actual payload mass. But yeah, the bulk of payload is residuals.

[sevenperforce]

18 hours ago, magnemoe said:

I say modifying FH would be much easier than making an new lunar lander from scratch. Length of upper stage is based on RP1 an hydrolox state could be far longer

F9 is already close to the maximum fineness ratio for a rocket. Wind conditions at altitude are a problem with high fineness because the bending moment increases dramatically. A significant tank stretch isn't really on the table.

11 hours ago, Exoscientist said:

On 8/21/2023 at 11:46 AM, sevenperforce said:

With a hydrolox upper stage, a fully expendable Falcon Heavy can only send 9.3 tonnes of payload to TLI, let alone to the lunar surface. Hydrolox is fluffy; a hydrolox Falcon upper stage would only be able to carry around 30 tonnes of propellant.

 I’ve seen numbers for Falcon Heavy to TLI in the range of  ~20 tons

Oh, absolutely. FH can absolutely send around that much to TLI. Which is why replacing the upper stage with hydrolox for only ~9-10 tonnes to TLI wouldn't be a good idea.

11 hours ago, Exoscientist said:

But to maximize payload don’t use the kerolox FH upper stage to do the TLI burn. Use the 63.8 ton FH capacity to LEO to carry hydrolox stages for the TLI  burn and for the lunar lander stage. For example a 30 ton Centaur-like stage with a 10 ton Centaur-like stage could get about 15 tons one-way to the lunar surface.

I'm confused -- are you now talking about a four stage vehicle? Or 4.5 stage, counting the FH side boosters?

Look at pages 83 and 88 of the Falcon 9 User Guide. The extended fairing has a cylindrical payload area 478" high and 181" in diameter, and there's an additional 175" of conical payload volume above that cylindrical plane. Let's imagine that your notional TLI hydrolox stage used a cluster of BE-7 engines to maximize utilization of volume and needed about 4" for tank walls, external fittings, and clearance from the fairing. The BE-7 engines are 80" tall. Let's imagine a lunar lander slightly squattier than the Apollo LM, at 5 meters height. That cuts 23" into the cylindrical region. Finally, let's assume a 1" insulated common bulkhead and shave off another 18" equivalent of vertical volume to account for the ellipsoidal caps.

So that leaves a total available tank volume of 137 cubic meters, which gives us space for 38 tonnes of hydrolox. That's best-case-scenario assumptions with a single-stage architecture. If you try to stack two hydrolox stages on top of each other, your total combined propellant load drops to 28 tonnes of hydrolox.

I'm still not quite sure what exactly you're proposing in terms of the mission profile, but from the overall context I think you are envisioning a lander+stage combo being launched on Falcon Heavy, meeting up with separately-launched crew somewhere in cislunar space, and then taking the crew down to the lunar surface and back up. For safety and simplicity, I'll consider a crasher stage architecture, where a zero-boiloff hydrolox stage performs both the braking burn into lunar orbit as well as the descent burn, and drops off to allow the lander to perform the final hovering landing as well as the ultimate ascent. Let's use the 9-12 tonne ascent vehicle concept from the Artemis initial studies; I'll go with 10 tonnes to make sure we have enough extra mass for landing legs and hovering propellant.

The crasher stage needs 2.77 km/s of Δv to brake into cislunar space and then take the lander down to the surface. Assuming 450 seconds of specific impulse, the stack propellant fraction needs to be on the order of 47%. Assuming a relatively decent stage mass ratio of 10:1 including engine(s) and insulation system, the stage needs to be 52% of the total stack mass, giving us a total stack mass of 20.8 tonnes.

But here we see there's no need for stacking up multiple hydrolox stages. Falcon Heavy can already deliver ~20 tonnes to TLI, so adding an additional stage underneath is completely unnecessary.

[Exoscientist]

1 hour ago, sevenperforce said:

'm still not quite sure what exactly you're proposing in terms of the mission profile, but from the overall context I think you are envisioning a lander+stage combo being launched on Falcon Heavy, meeting up with separately-launched crew somewhere in cislunar space, and then taking the crew down to the lunar surface and back up. For safety and simplicity, I'll consider a crasher stage architecture, where a zero-boiloff hydrolox stage performs both the braking burn into lunar orbit as well as the descent burn, and drops off to allow the lander to perform the final hovering landing as well as the ultimate ascent. Let's use the 9-12 tonne ascent vehicle concept from the Artemis initial studies; I'll go with 10 tonnes to make sure we have enough extra mass for landing legs and hovering propellant.

 No. The discussion was about using the SLS for cargo. I think it is too expensive for that purpose. Use the Falcon Heavy for that purpose, or other low cost commercial launchers. No crew module would be included here. I was talking about using two hydrolox stages to get the max cargo on the Falcon Heavy but here’s another way using a single stage:

Suppose we use a 45 ton prop load “Centaur-like” stage carried to LEO by the Falcon Heavy for cargo only transport to the lunar surface one-way.  As it’s Centaur-like, take the ISP as 465.5s(the max Centaur RL-10 Isp with extended nozzle was 465.5s);  and give it a ca. 10 to 1 mass ratio, so a dry mass of 4.5 tons with the 45 ton prop load. Then with 12 tons payload you could get:

465.5*9.81LN(1 + 45/(4.5 + 12)) =  6,000 m/s, which is sufficient for the stage to do both TLI and land on the lunar surface one-way with the 12 tons of payload. But the need for low boiloff for the 3-day flight to the Moon would reduce this payload somewhat.

 Alternatively, you could let the FH do the TLI burn, and say it could get ca. 20 tons to TLI. Use a 10-ton “Centaur-like” stage, at 10 ton prop load, 1 ton dry mass, and 465.5s ISP. Then it could get 9 tons payload one-way to the lunar surface:

465.5*9.81Ln(1 + 10/(1 + 9 )) = 3,160 m/s, sufficient for the lunar landing once already put on the trans-lunar trajectory to the Moon by the Falcon Heavy.

  Bob Clark

[sevenperforce]

17 hours ago, Exoscientist said:

The discussion was about using the SLS for cargo. I think it is too expensive for that purpose. Use the Falcon Heavy for that purpose, or other low cost commercial launchers.

Oh, I certainly agree with that. SLS is far too expensive for cargo. Far too expensive for anything, really, but definitely for cargo.

17 hours ago, Exoscientist said:

I was talking about using two hydrolox stages to get the max cargo on the Falcon Heavy but here’s another way using a single stage:

Suppose we use a 45 ton prop load “Centaur-like” stage carried to LEO by the Falcon Heavy for cargo only transport to the lunar surface one-way.

If it's "Centaur-like" then you're not getting 45 tonnes of hydrolox inside even the Falcon Heavy extended fairing, as I explained above (and re-explain in more detail below).

17 hours ago, Exoscientist said:

 As it’s Centaur-like, take the ISP as 465.5s(the max Centaur RL-10 Isp with extended nozzle was 465.5s)

The RL10-C-2-1 has never been paired with a Centaur, but is still in production and achieves your desired specific impulse. It's a whopping 163.5" long but fortunately 77.2" of that is a deployable nozzle extension, leaving the 64.1" engine and the 22.2" fixed nozzle extension to give a total stowed height of 86.3". 

Here are the internal dimensions of that extended fairing, lifted straight from the F9 user guide. The numbers on the left are height above the payload adapter; the numbers on the right are diameter:

Spoiler

I'm not sure what cargo you're thinking of, but for the sake of simplicity let's cut the hydrolox stage off at ST = 477.976", right where the ogive starts. That means your payload, whatever it is, needs to fit inside a right truncated cone with a height of 175", a base with diameter 180", and a top with diameter 49". Let's allow 4" of clearance all the way around the stage for downcomers, vents, and the like (remember that this thing has to perform some significant maneuvering), giving us a stage diameter of 172". Tank wall thickness is of course negligible. Rule of thumb is that the most efficient dome shape is an ellipsoid of height R/2^(1/2), so this gives an ellipsoidal cap height of 60.8". The double-walled intermediate bulkhead has a thickness of 0.3". Neglecting any volume occupied by stringers and the like, the total available propellant volume is represented as the sum of the volumes of an oblate spheroid of height 121.6" and a cylinder of height 269.8", both with a diameter of 175":

The ellipsoid has a volume of 31.95 cubic meters and the cylinder has a volume of 106.34 cubic meters, giving this "Centaur Plus" frankenstage a total available internal propellant volume of 138.29 cubic meters. At hydrolox's bulk density of 0.28 g/cc, this gives 38.7 tonnes of propellant.

Now, the RL10-C-2 has a propellant mass flow rate of 24.07 kg/s, giving it a total burn time of 1,608 seconds. That's far, far too long -- a full third of an LEO orbit (although obviously it wouldn't happen all at once; I'm just illustrating) -- so the Oberth losses would be immense. You're going to need two engines. It will be a tight squeeze under the payload fairing but it can probably be done.

17 hours ago, Exoscientist said:

...and give it a ca. 10 to 1 mass ratio...

If we're going "Centaur-like" then let's apply the Centaur mass ratio closely. Centaur has an empty mass of 2,247 kg, which drops to 2,032 kg when you subtract the mass of the RL10-C-1. It carries a total of 20,830 kg of hydrolox, giving it an engineless mass ratio of 10.25:1. Thus to hold 38.7 tonnes of propellant, our frankenstage needs a dry mass of 3,775 kg. Add the increased weight of two RL10-C-2s and the stage mass comes up to 4,377 kg.

17 hours ago, Exoscientist said:

Then with 12 tons payload...

With 12 tonnes payload, the combined stack develops 5,540 m/s of Δv.

Falcon Heavy can deliver 63.8 tonnes of payload to LEO, but that necessarily includes its own residuals. The PAF can't handle a ~55 tonne payload so it would need a special stage adapter to brace against the bulkhead or something. Estimating 1.8 tonnes for that structure, Falcon Heavy's "payload" is 56.9 tonnes and it reaches LEO with 6.9 tonnes of its own propellant to spare. Estimating FHUS at 4.6 tonnes, this means FH can burn the last of its props to give the upper stage stack thingy a boost of around 363 m/s past LEO, which is at least something helpful.

Thus the total Δv available to the payload from LEO is 5.9 km/s. TLI is 3.2 km/s, LOI is 0.9 km/s, and descent to the lunar surface from orbit is 1.87 km/s, for a total requirement of 5.97 km/s. Just shy.

However, you do have to actually LAND your payload, and our Centaur Plus frankenstage isn't going to be useful for that. If you go with a crasher stage architecture, then we can easily give the actual payload some small pressure-fed hypergolic thrusters to perform the last 70 m/s of the landing burn, plus whatever is necessary for the actual landing maneuvers.

Of course, a 12-tonne payload is enough for a lunar ascent vehicle. On the other hand, this would require building a completely new intermediate Centaur frankenstage. IIRC, Centaur V was announced ca. 2012 and they still haven't ironed out the kinks, so we can assume a similar timeframe for this sort of retrofit.

If you want to just play rocket legos with the existing Centaur SEC, that's much more doable. With the same intended 12-tonne payload, the existing Centaur SEC develops 3,982 m/s of dV and has a stack mass of 35.1 tonnes. It fits easily in the extended fairing with significantly more room for payload. Launched on Falcon Heavy (using a notional 1.3-tonne payload adapter frame), the Falcon upper stage would reach LEO with an impressive 27.4 tonnes of propellant residuals, allowing it to deliver a heft 1,747 m/s to the Centaur and its payload. This means the whole stack boasts 5.7 km/s of Δv, just 200 m/s less than the first design despite using only about half as much hydrolox. AND that's with the lower specific impulse of the RL10-C-1.

Just goes to show that specific impulse ain't everything. Once you are going beyond LEO, mass ratio becomes more important than specific impulse, which is why the Merlin 1DV and the single RL10-C-1 can beat a pair or RL10-C-2s. The high specific impulse of hydrolox is more useful for lifting large monolithic payloads into LEO in the first place (think: Saturn V second stage).

[tater]

Metzger is the world's leading authority on rocket plume-regolith interactions.

This will be a very interesting paper. If this claim is true, a little dust near Boca Chica was a very small price to pay for this information.

[RCgothic]

One plan for lunar landing pads I'd seen was to coat the surface with adhesive. This would be like a thin concrete skin over zero foundations. Yeah, good catch Dr Metzger!

[darthgently]

2 hours ago, RCgothic said:

One plan for lunar landing pads I'd seen was to coat the surface with adhesive. This would be like a thin concrete skin over zero foundations. Yeah, good catch Dr Metzger!

Adhesives and high temperatures are a tricky combination.   The image of some in situ process to glassify the top surface to an effective depth springs to mind, or turn the top layer into a ceramic of some sort, perhaps requiring some additives and alot of heat

[RCgothic]

The point being that without foundations such a skin could break under weight of thrust resulting in the pad being excavated and throwing debris at the lunar craft resulting in LOCV.

[darthgently]

12 minutes ago, RCgothic said:

The point being that without foundations such a skin could break under weight of thrust resulting in the pad being excavated and throwing debris at the lunar craft resulting in LOCV.

Yes, but given lower gravity and so lower thrust, combined with engines mounted higher up as planned in a few designs, a thick (8 to 12 inches?) ceramic-like surface could be its own foundation for practical purposes.  We will never be firing 33 raptors at 100% on the Moon.  Just throwing the idea out there.  I don't think that pilings and such are necessarily going to be warranted

[magnemoe]

3 hours ago, darthgently said:

Yes, but given lower gravity and so lower thrust, combined with engines mounted higher up as planned in a few designs, a thick (8 to 12 inches?) ceramic-like surface could be its own foundation for practical purposes.  We will never be firing 33 raptors at 100% on the Moon.  Just throwing the idea out there.  I don't think that pilings and such are necessarily going to be warranted

As I understand they will cut the vacum engines some distance above the surface, 100-200 meters, I assume after coming to an stop in practice, then the landing engines who is mounted above the tanks and tilted outward will take over, this these small engines mounted so high I don't think it will be a problem as exhaust expands rapidly in vacuum. 
Not sure if its good enough to not be an danger for astronauts nearby or can damage sensible stuff but the ship would be safe. 

[darthgently]

50 minutes ago, magnemoe said:

As I understand they will cut the vacum engines some distance above the surface, 100-200 meters, I assume after coming to an stop in practice, then the landing engines who is mounted above the tanks and tilted outward will take over, this these small engines mounted so high I don't think it will be a problem as exhaust expands rapidly in vacuum. 
Not sure if its good enough to not be an danger for astronauts nearby or can damage sensible stuff but the ship would be safe. 

Maybe when 100s of tons of harvested resources need to be lifted from the Moon a more robust pad will be needed,  but I'm not seeing a problem for Artemis.  But who knows?

[magnemoe]

1 hour ago, darthgently said:

Maybe when 100s of tons of harvested resources need to be lifted from the Moon a more robust pad will be needed,  but I'm not seeing a problem for Artemis.  But who knows?

Yes here we obviously agree, but at this time we will have launch pad level systems some distance from the industrial zone. Might even something looking like the Starbase launch tower. 
That is before we make the magnetic rail launcher.
http://freefall.purrsia.com/ff900/fv00877.htm
Yes here its used in atmosphere and to get the spacecraft supersonic.
For an cargo pod on the moon I guess one part of the track who is a couple of g to settle stuff then some hundreds, engine on pod circulates and adjust trajectory for getting picked up by a tug for docking.