Possible cost-reduction possibilities for the NASA portions of MSR

[Phil Stooke]

Not a cost reduction but another strategy: shift part of the cost to another budget line.

Haul those samples up into Mars orbit.  Maneuver to the vicinity of Phobos, matching its orbit fairly closely.  Drop the sample container on Phobos (protected by a small thruster burn plus airbags).  Leave them there for the first human crew to collect during a future mission.

Advantages:  you don't have to loiter in Mars orbit for several years and then be located and met by by another orbiter.  A first human Mars crew can visit Phobos instead of landing on Mars first time - in other words, like an Apollo 8/10 rehearsal mission to buy down risk, but giving the mission a very substantial science return.  The sample collection and return to Earth is not in the PS budget.

Disadvantages: well, OK, the MAV needs to be able to get you to Phobos and I am assuming the landing is not very difficult relative to the orbital rendezvous and robotic return to Earth.

I have also argued in the past that if there is a long delay in pickup of samples, additional sample returns might be accomplished by other players, adding science while not being in the critical path.  For instance, samples from Deimos delivered to Phobos, samples from a Mars Trojan asteroid delivered to Phobos.  Even additional Mars surface samples delivered to Phobos. 

[Robotbeat]

What’s the point of that? May as well just keep the samples on the ground or, if you insist on a crewed mission retrieving the samples in orbit, launch the sample canister to be retrieved by an orbital crewed mission or something.

[mikelepage]

Ingenuity type helicopters could be used to load the Mars samples onto Starship after it lands. I'm not sure how you get them back to Earth though.

At the risk of triggering those who have worked on the program for years, it seems to me like Starship could have a role here, just not with the ISRU conops as proposed by Musk.

One reason I thought the Mars fetch copters were a clever idea is that they could deposit samples into the tip of a MAV, no matter how far off the ground it is.

We do know that that Starship will need to come to some arrangement on planetary protection issues, then will make multiple uncrewed attempts at landing on Mars, presumably at a site far removed from places of scientific interest. Obviously lots of us have have spitballed about combining the SpaceX's test program with MSR, but what would it look like if they do it the SpaceX way, with multiple Starships for redundancy?

Idea: Suppose that NASA contracted/allowed SpaceX to attempt landing 3 times, at a target site some ~50km from the sample repository location, well outside Jezero crater. These landing attempts would happen within days and allow the Starships to double as comms relays for each other. Each Starship would be the same custom build, with its payload bay filled by what is essentially a missile silo, (ie with a nose cone that pops off, and ejectable vent panels at the base of the payload bay.  Under the nose cone of each Starship is housed a single fetch copter, and a landing pad/platform for that copter where it can deposit the sample tubes.

The MAVs (presumably two stage/hypergolic) return directly to Earth, eliminating LMO rendezvous. With the kind of mass margins that Starship has, it ought to be (relatively) easy to implement some kind of double encapsulation system so you're not returning martian regolith/rock to Earth un-contained. 

In this setup it's tempting to give the fetch copters some sample collection capability, such that if the MSR samples successfully liftoff in the first MAV, there's an opportunity to collect more samples - albeit only from where the fetch copters could land. Alternatively can they collect the tubes from Perseverance itself, or even "restock" Perseverance with new sample tubes so it could keep collecting new samples (cue caveats about scope creep, etc etc).

[Zed_Noir]

<snip>
At the risk of triggering those who have worked on the program for years, it seems to me like Starship could have a role here, just not with the ISRU conops as proposed by Musk.

<snip>

Idea: Suppose that NASA contracted/allowed SpaceX to attempt landing 3 times, at a target site some ~50km from the sample repository location, well outside Jezero crater. These landing attempts would happen within days and allow the Starships to double as comms relays for each other. Each Starship would be the same custom build, with its payload bay filled by what is essentially a missile silo, (ie with a nose cone that pops off, and ejectable vent panels at the base of the payload bay.  Under the nose cone of each Starship is housed a single fetch copter, and a landing pad/platform for that copter where it can deposit the sample tubes.

The MAVs (presumably two stage/hypergolic) return directly to Earth, eliminating LMO rendezvous. With the kind of mass margins that Starship has, it ought to be (relatively) easy to implement some kind of double encapsulation system so you're not returning martian regolith/rock to Earth un-contained. 

In this setup it's tempting to give the fetch copters some sample collection capability, such that if the MSR samples successfully liftoff in the first MAV, there's an opportunity to collect more samples - albeit only from where the fetch copters could land. Alternatively can they collect the tubes from Perseverance itself, or even "restock" Perseverance with new sample tubes so it could keep collecting new samples (cue caveats about scope creep, etc etc).

The idea was discuss on page 2 on this thread.

Starship is going to Mars.  The first few will not leave the planet again so one of them could carry a decent sized direct return MAV on a gantry that swings out of the cargo bay.  Just have to get the samples up to be loaded and then launch straight off the gantry.

Doable ??

Simpler just to install the MAV in a pod/cold gas launch tube that is strapped to the Starship exterior. Think either solid or hypergolic propulsion could work with the MAV.

Access to the MAV is through access hatch between the pod and the Starship.

Launch method is the traditional cold gas silo launch common to strategic ballistic missiles.

Think you can have an external launch pod/tube on either side of the main cargo hatch of a relatively lightly modified Starship. Mostly hardware to load the samples to the return capsule in the MAV.

[mikelepage]

<snip>
At the risk of triggering those who have worked on the program for years, it seems to me like Starship could have a role here, just not with the ISRU conops as proposed by Musk.

<snip>

Idea: Suppose that NASA contracted/allowed SpaceX to attempt landing 3 times, at a target site some ~50km from the sample repository location, well outside Jezero crater. These landing attempts would happen within days and allow the Starships to double as comms relays for each other. Each Starship would be the same custom build, with its payload bay filled by what is essentially a missile silo, (ie with a nose cone that pops off, and ejectable vent panels at the base of the payload bay.  Under the nose cone of each Starship is housed a single fetch copter, and a landing pad/platform for that copter where it can deposit the sample tubes.

The MAVs (presumably two stage/hypergolic) return directly to Earth, eliminating LMO rendezvous. With the kind of mass margins that Starship has, it ought to be (relatively) easy to implement some kind of double encapsulation system so you're not returning martian regolith/rock to Earth un-contained. 

In this setup it's tempting to give the fetch copters some sample collection capability, such that if the MSR samples successfully liftoff in the first MAV, there's an opportunity to collect more samples - albeit only from where the fetch copters could land. Alternatively can they collect the tubes from Perseverance itself, or even "restock" Perseverance with new sample tubes so it could keep collecting new samples (cue caveats about scope creep, etc etc).

The idea was discuss on page 2 on this thread.

Starship is going to Mars.  The first few will not leave the planet again so one of them could carry a decent sized direct return MAV on a gantry that swings out of the cargo bay.  Just have to get the samples up to be loaded and then launch straight off the gantry.

Doable ??

Simpler just to install the MAV in a pod/cold gas launch tube that is strapped to the Starship exterior. Think either solid or hypergolic propulsion could work with the MAV.

Access to the MAV is through access hatch between the pod and the Starship.

Launch method is the traditional cold gas silo launch common to strategic ballistic missiles.

Think you can have an external launch pod/tube on either side of the main cargo hatch of a relatively lightly modified Starship. Mostly hardware to load the samples to the return capsule in the MAV.

Apologies, I saw MickQ's post, but not your reply.

I have Delta-V of return to Earth from Mars surface as ~5.7km/s, so this rocket would be at the lower end of the size range for an ICBM. Quick google result says the Trident II (~7km/s dV) is close to 60 metric tons. The lowest mass ICBM with a >5500km range is the RS-26 Rubezh, at 36 tons.

Only going off Kerbal experience here, but putting that much mass outside the main structure of Starship, and behind the center of lift, seems like it would make Starship unstable during EDL. If you're doing direct-to-Earth return with the missile silo approach, it would surely have to be incorporated into Starship's main payload bay.

Or did you mean to retain the ESA orbiter portion of the current MSR plan?

[Zed_Noir]

<snip>
Apologies, I saw MickQ's post, but not your reply.

I have Delta-V of return to Earth from Mars surface as ~5.7km/s, so this rocket would be at the lower end of the size range for an ICBM. Quick google result says the Trident II (~7km/s dV) is close to 60 metric tons. The lowest mass ICBM with a >5500km range is the RS-26 Rubezh, at 36 tons.

Only going off Kerbal experience here, but putting that much mass outside the main structure of Starship, and behind the center of lift, seems like it would make Starship unstable during EDL. If you're doing direct-to-Earth return with the missile silo approach, it would surely have to be incorporated into Starship's main payload bay.

Or did you mean to retain the ESA orbiter portion of the current MSR plan?

Was thinking of something like the 14 meter long MGM-134 Midgetman with a launch mass of about 13.6 tonnes that has a range 11k km delivering one ordnance package.

The potential Mars sample return capsule will likely be much smaller and lighter than said ordnance package.

There will be equivalent ballast mass on the other side of the Starship for center of lift issues.

Having an external pod/launch tube is simpler than try to rework the interior of the Starship. Plus an ejectable hatch cover on side mounted external pod/tube is simpler than a nose hatch that has to pass through Martian reentry.

[vjkane]

NASA's managers announced today that they are "pausing" or "slowing" down the Mars sample return mission development.

Links to two articles below.

Immediate future work will focus on the system to deliver the samples to orbit. The articles don't make it clear what that means. It could mean that NASA would launch the samples into orbit and then collect them some (maybe many years) later. I don't believe that's what is meant. A small canister in orbit could easily be lost to tracking, but alternatively the samples are safe if left on the Martian surface. I think that what this means is that NASA will focus on refining the launch and sample capsule designs. These were elements that the recent review stated were least defined but the definition of which had the greatest impact on the full design of the return architecture.

Space Policy Online: https://spacepolicyonline.com/news/nasa-pauses-mars-sample-return-program-while-assessing-options/

Space News: https://spacenews.com/nasa-slows-down-work-on-mars-sample-return-due-to-budget-uncertainty/

[VSECOTSPE]

Immediate future work will focus on the system to deliver the samples to orbit. The articles don't make it clear what that means. It could mean that NASA would launch the samples into orbit and then collect them some (maybe many years) later. I don't believe that's what is meant. A small canister in orbit could easily be lost to tracking, but alternatively the samples are safe if left on the Martian surface. I think that what this means is that NASA will focus on refining the launch and sample capsule designs. These were elements that the recent review stated were least defined but the definition of which had the greatest impact on the full design of the return architecture.

Figueroa’s report examined several alternate architectures that disaggregated elements and  spread out the resulting launches to fit a lower annual budget level.  See pages 44-46.  That doesn’t necessarily mean that the sample would be left in orbit with a beacon before the ERO arrived.  But it could, and the SpaceNews article may be a tacit admission that the program is headed in this direction.  FWIW…

[vjkane]

Figueroa’s report examined several alternate architectures that disaggregated elements and  spread out the resulting launches to fit a lower annual budget level.  See pages 44-46.  That doesn’t necessarily mean that the sample would be left in orbit with a beacon before the ERO arrived.  But it could, and the SpaceNews article may be a tacit admission that the program is headed in this direction.  FWIW…

VSECOTSPE's reference to the review panel's report led me back to their report. I've copied their figure and text here and added a decoder for acronyms

ERO   Earth return orbiter
SRL   Sample retrieval lander
CCRS   Capture containment and return system
MFR   MSR Fetch Rover
MML   MSR MAV Lander

The IRB developed a programmatic assessment of potential MSR Program cost and schedule scenarios, including the program’s current plan. While the ICE for the current plan shows a range of costs for the ERO/2027 and SRL/2028 LRDs, the JCL indicates a very low probability of achieving the current LRDs. The earliest likely LRDs for both ERO and SRL are in 2030. The subsequent IRB ICE for the delayed (2030/2030) scenario illustrates a rephased cost plan for comparison. This scenario, combined with potential alternate architectures, provides a probable lifecycle cost range of approximately $8-11B.

The alternate architectures considered were not detailed concepts and were intended only to represent a potential solution space for programmatic analysis. The alternate architectures included decoupled projects (where launches are not dependent on one another) with fiscal year budget scenarios that are either constrained (up to $850M, $950M, or $1B/fiscal year) or unconstrained. Scenarios included:

• The “All In” approach, which prioritized SRL to launch as soon as possible with helicopters removed but retaining the arm.
• The “Robust and Resilient (R&R)” approaches, which disaggregated the landed systems into two missions:
1) MSR Fetch Rover (MFR), which includes a 2xMER rover carrying the OS, scaled-down Skycrane EDL, and STHS, and
2) MSR MAV Lander (MML), which includes the MAV and a MAV landing/launch system using a scaled-up Skycrane EDL. The helicopters and the arm were removed from the MAV lander due to the addition of the MFR. One option launched ERO/CCRS first, with a constrained funding level of $950M/year. Another set of options launched the landed systems first and examined the impact of two constrained funding levels – $850M/year and $1B/year.

Overall, the IRB analysis suggests the alternate architecture solution space has an LCC range of $8.4-10.9B with various LRD combinations in the 2030-2035 timeframe.

[Don2]

The 15th slide in this MEPAG presentation from 2021 says that the next good opportunity after 2028 is 2035, and the next slide says that a nuclear lander would be required for 2030.

https://mepag.jpl.nasa.gov/meeting/2021-01/04_MEPAG%201_2021%20V5.pdf

[Don2]

ISRU saves less mass than I hoped. Supplying the power for it is an issue, especially given the variable nature of solar power. Different sources give different estimates for the power system mass, and the highest estimates are greater than the mass of the oxygen produced. The mass of the batteries required is significant. At this point I am unconvinced that it is worth the trouble for MSR. However, it would be an exciting technology demonstration.

I have found an unusually well written system study of MAV options from 1998.
https://dataverse.jpl.nasa.gov/dataset.xhtml?persistentId=hdl:2014/22864

One of the options they look at is a hybrid rocket with HTPB/Al fuel and LOX oxidizer produced via ISRU. They provide some equations for sizing different parts of the system. I have used those to calculate the mass and power needs of the ISRU plant. 

Some people will object that the study is old. However, the laws of physics and chemistry have not changed in the past 25 years. Solar cell technology has improved, and much experience has been gained in operating  solar powered equipment on Mars.

The hybrid LOX design has an Isp of 328, better than the 287 Isp typical of solids. From the rocket equation I estimate this would make the rocket 17% lighter for the same payload capacity. The rocket needs 213 kg of LOX. The ISRU equipment and the power system to run it mass 123kg, giving a saving of 90kg. Overall the rocket and the ISRU plant come in at 284kg, 37% lighter than a solid of equal payload capacity.

Power need = 300 (MOXIE like unit) + 194 (liquefaction and storage) + 100 = 594W
Mass of power system = 58kg
Mass of liquefaction system = 41kg
Mass of MOXIE sized unit based on study = 24kg (actual MOXIE mass = 17kg)
Total mass of system = Power system mass + Liquefaction mass + MOXIE like unit mass = 123kg
Mass of oxygen made in 876 days = 210kg (87 kg saving)

In a real system there would be issues with dust depositing on the panels, the seasonal variability of solar power and battery storage losses. A more modern hybrid would probably use a paraffin based fuel with better low temperature performance and Isp of 330. There has been a surprising amount of work on hybrid rockets using ISRU LOX over the past 20 years at Stanford university and JPL.

A different estimate for the mass of the power system comes from data presented in the following paper. This is a systems engineering paper which looks at the amount of energy needed to keep the MAV warm.
https://dataverse.jpl.nasa.gov/dataset.xhtml?persistentId=hdl:2014/43568    (Table 3, Page 6)
This gives an estimate of 307kg.

Another estimate comes from assuming that half of the power produced by the panels will have to be stored in batteries for night time use. Assuming Li-ion batteries store 265 Whr/kg (Wikipedia), this gives a battery mass of 27kg to store half of the 14.3 kwhr produced from 14.3 sq meters of panel. Those panels mass 43kg. Total mass is 70kg.

I suspect the lighter estimate left out the mass of the batteries required to manage the variability of solar.

[Dalhousie]

ISRU saves less mass than I hoped. Supplying the power for it is an issue, especially given the variable nature of solar power. Different sources give different estimates for the power system mass, and the highest estimates are greater than the mass of the oxygen produced. The mass of the batteries required is significant. At this point I am unconvinced that it is worth the trouble for MSR. However, it would be an exciting technology demonstration.

I have found an unusually well written system study of MAV options from 1998.
https://dataverse.jpl.nasa.gov/dataset.xhtml?persistentId=hdl:2014/22864

One of the options they look at is a hybrid rocket with HTPB/Al fuel and LOX oxidizer produced via ISRU. They provide some equations for sizing different parts of the system. I have used those to calculate the mass and power needs of the ISRU plant. 

Some people will object that the study is old. However, the laws of physics and chemistry have not changed in the past 25 years. Solar cell technology has improved, and much experience has been gained in operating  solar powered equipment on Mars.

The hybrid LOX design has an Isp of 328, better than the 287 Isp typical of solids. From the rocket equation I estimate this would make the rocket 17% lighter for the same payload capacity. The rocket needs 213 kg of LOX. The ISRU equipment and the power system to run it mass 123kg, giving a saving of 90kg. Overall the rocket and the ISRU plant come in at 284kg, 37% lighter than a solid of equal payload capacity.

Power need = 300 (MOXIE like unit) + 194 (liquefaction and storage) + 100 = 594W
Mass of power system = 58kg
Mass of liquefaction system = 41kg
Mass of MOXIE sized unit based on study = 24kg (actual MOXIE mass = 17kg)
Total mass of system = Power system mass + Liquefaction mass + MOXIE like unit mass = 123kg
Mass of oxygen made in 876 days = 210kg (87 kg saving)

In a real system there would be issues with dust depositing on the panels, the seasonal variability of solar power and battery storage losses. A more modern hybrid would probably use a paraffin based fuel with better low temperature performance and Isp of 330. There has been a surprising amount of work on hybrid rockets using ISRU LOX over the past 20 years at Stanford university and JPL.

A different estimate for the mass of the power system comes from data presented in the following paper. This is a systems engineering paper which looks at the amount of energy needed to keep the MAV warm.
https://dataverse.jpl.nasa.gov/dataset.xhtml?persistentId=hdl:2014/43568    (Table 3, Page 6)
This gives an estimate of 307kg.

Another estimate comes from assuming that half of the power produced by the panels will have to be stored in batteries for night time use. Assuming Li-ion batteries store 265 Whr/kg (Wikipedia), this gives a battery mass of 27kg to store half of the 14.3 kwhr produced from 14.3 sq meters of panel. Those panels mass 43kg. Total mass is 70kg.

I suspect the lighter estimate left out the mass of the batteries required to manage the variability of solar.

Can you run the ISRU plant just during daylight hours?

[Don2]

Can you run the ISRU plant just during daylight hours?

That's one possibility. The problem is that the electrolysis cell runs at 800C and degrades quickly if it switched on and off, so that would have to be kept hot through the night. Most of the rest of the energy is used for gas compression. That includes compressing from Mars ambient of 6mb to 1000mb for electrolysis, plus another compression in the liquefaction process. That could be done only in daylight. Perhaps it would be possible to store compressed gas in tanks for use at night, but that would add some weight and inefficiencies to the process.

Another thing I have wondered about is if it would be possible to use an RTG plus a Stirling engine to produce a stable electricity supply for the process. At 24% efficiency that would yield about 500W of power. (The canceled ASRG project claimed 26% efficiency) An MMRTG masses 45kg, plus something extra for the Stirling engine. Compared to my estimates of 58/70/307 kg for the solar system, that looks attractive.

An RTG heating a Stirling engine producing power for an ISRU to fuel a hybrid rocket would make for a real cool technology demonstration. However, I think it is just too many untried technologies to be sensible for the current sample return mission.

Another thing I have started to think about is that the mass of the launch support hardware is likely to be a significant fraction of the mass of the rocket. ISRU really doesn't help with that.

[Don2]

Even if the mass of the rocket and fuel is reduced, that might not reduce the landed payload mass as much as I hoped. For MSR you don't just have to land a rocket on Mars, you also have to land a launch site. I've looked at the mass of the Navy's VLS cells and they are much heavier than the rockets they launch. For the man portable Stinger missile, the launch tube and other hardware is about 40% of the mass of the rocket it launches. The mass of the launch hardware for MSR is likely to be a large fraction of the mass of the rockets. It will need to do the following jobs.

1/Thermal control of the rocket
2/Payload loading
3/ Erection
4/ Fueling if ISRU LOX is used
5/ Management of the hot exhaust gas at launch.

In the end the amount of supporting launch hardware an MAV needs might matter more than its performance. Saving a kilogram on the mass of the MAV does no good if it adds one kilogram to the mass of the launch and support hardware.

Here is a reference for the launch hardware needs of the MAV.
https://dataverse.jpl.nasa.gov/file.xhtml?fileId=48646&version=1.1

[Jim]

3/ Erection
4/ Fueling if ISRU LOX is used
5/ Management of the hot exhaust gas at launch.
 

Here is a reference for the launch hardware needs of the MAV.
https://dataverse.jpl.nasa.gov/file.xhtml?fileId=48646&version=1.1

reference is too old.  The MAV is ejected is into the "air" for launch.

[Don2]

I've seen the video before. It doesn't state the mass of the system that throws the MAV upwards, which I believe is called VECTOR. While I have seen the mass of the MAV quoted many times, I haven't seen the mass of VECTOR quoted anywhere. However, the mass of VECTOR and the rest of the launch systems are just as important as the mass of the MAV. It all has to go to Mars. This might be evidence of a problem in NASA's approach to MAV development.

For most launchers, the launch system is something of an afterthought. For terrestrial launchers, the mass of the launch system usually doesn't matter. MAV is different. The launch system mass is just as important as the mass of the rocket. In my opinion, MAV and its launch systems should be developed as a single integrated system. The fact that the mass of the MAV is usually quoted without mentioning the mass of the launch system could be evidence that that isn't happening.

The type of rocket chosen is going to affect the mass of the launch system. Any throttleable design could reduce the exhaust gas management problem by launching on less than full power and then throttling up when clear of the lander. I hope they studied that sort of thing before they chose the solid rocket over the hybrid rocket design.

[ccdengr]

[Not stating the mass of the launch infrastructure] might be evidence of a problem in NASA's approach to MAV development.

Do you really think that JPL is that stupid?

I hope they studied that sort of thing before they chose the solid rocket over the hybrid rocket design.

Hybrids have been proposed many times for various applications and have rarely if ever worked out.  Look at all the problems Virgin Galactic had with SS2.  To me, the hybrid MAV has always seemed more like it was being pushed by some people in academia to give them something to write papers about, not a real solution.

That said, https://ntrs.nasa.gov/api/citations/20205004450/downloads/2020%20AIAA%20JPC%20MAV%20Hybrid%2007%2020%202020%20Draft.docx.pdf has a lot of info about the hybrid and why it wasn't chosen.  Ultimately the paper claims this was simply because of the lower TRL of the hybrid, but some might conclude that the remaining challenges of the hybrid were being glossed over.

[vjkane]

That said, https://ntrs.nasa.gov/api/citations/20205004450/downloads/2020%20AIAA%20JPC%20MAV%20Hybrid%2007%2020%202020%20Draft.docx.pdf has a lot of info about the hybrid and why it wasn't chosen.  Ultimately the paper claims this was simply because of the lower TRL of the hybrid, but some might conclude that the remaining challenges of the hybrid were being glossed over.

I found this quote from the paper abstract informative: "While the hybrid technology development program made many strides, there was still a list of challenges to overcome. "

[Nomadd]

 There seems to be a lot of proposals in here to save money by making things a lot more complicated and requiring a lot more new tech.
 As much as I loathe the habit, right now the best thing might be to screw around for a few more years and see how certain projects work out.

[Blackstar]

There seems to be a lot of proposals in here to save money by making things a lot more complicated and requiring a lot more new tech.

That's very common to these discussions (and to this forum). It is often phrased with the word "simply," followed by a solution that requires entirely new technology development (in other words, NOT simple).

I think something that people here and elsewhere fundamentally misunderstand is that new technologies can reduce costs in the long run and over time, but that they have a high development cost. If all you are doing is a single mission, or a few missions, then that high up-front development cost will make that solution for a single mission a lot more expensive than using a more proven technology.