Great shoutout to Arizona State University for the images. I like to see when someone or some group gets some recognition for work they have performed in relative obscurity for a very long time. Years of craft, getting better and better at something most don't bother with. Then suddenly your expertise gets a spotlight in a meaningful way.
(Yes, I suppose there are many other sources that could provide images. These happen to be from ASU.)
Welp, I worked on this one. Specifically, I worked on the laser rangefinders which are under so much scrutiny. I no longer work at Intuitive Machines, but I'm certainly interested in finding out what happened to the lasers this time.
Ooo. Not sure if you can tell us anything. If you can’t I totally understand.
But in case you can: Was a radar based altimeter considered?
How do you guys deal with kicked up regolith? (I have seen first hand how hard heavy snow is on lidars, and would imagine that regolith “shower” is similar, but what do I know.)
some of the comments here are suggesting the lander chose that spot, as opposed to crashing and skidding across the surface before settling in the spot purely due to inertia, what's the merit to that?
I'm not sure what you find unclear. Navigation was fine - "Athena knew where it was relative to the surface of the Moon" - but without a working altimeter it was kinda fucked for actually touching down.
> As a result, the privately built spacecraft struck the lunar surface on a plateau, toppled over, and began to skid across the surface. As it did so, the lander rotated at least once or twice before coming to a stop in a small, shadowed crater.
Oh yeah, we've all Kerbaled it in like that at one point or another.
If you take the time to study the documentation from the 1950s & 1960s, the engineering culture of that era appears to be markedly different from the engineering culture prevalent today. And I think it's deeply rooted in the symbiotic relationship between computing, Baumol's cost disease and our obsession with precision, results-oriented, MBA-style-min-maxing, "good enough for government work" engineering.
Robert Truax, the designer of the Sea Dragon, loved to promote the design paradigm of Big Dumb Boosters. Instead of many small, sophisticated rocket engines, what if we made one big robust one that can take a lickin' and keep on kickin'.
The idea was to relax the mass margins and to create big. dumb. boosters. It's the approach TRW explicitly followed for the Lunar Module engine,
> "There was an amusing but instructive side to this program. TRW farmed-out the fabrication of the engine and its supporting structure, less the injector that they fabricated themselves, to a "job-shop" commercial steel fabricator located near their facility . The contract price was $ 8000. Two TRW executives visited the facility to observe the fabrication process. They found only one individual working on the hardware, and when queried, he did not know nor care that he was building an aerospace rocket engine."
> " I had arrived late to witness the test, and only saw the firing. I was told by others who witnessed the entire test procedure that the engine was pulled out of outdoor storage where it lay unprotected against the elements. Before it was placed on the launch stand, the test crew dusted off the desert sand that had clung to it. This unplanned inlcusion [sic] of a bit of an environmental test also demonstrated hardware ruggedness of the kind no other liquid rocket eingine [sic] could approach."
The Surveyor program managed to make it "just work" 5 out of 7 times by adopting this approach. It had robust landing legs and RADAR. They would decelerate and then shut off the engine 11' above the surface. The wide, sturdy legs would then absorb that final impact of coming stand still from free fall.
These programs had a lot of capital behind them. Some components required precision engineering, but there's a very clear through line and embrace of the "we gotta make stuff that can take a lickin' & keeps kickin'" philosophy.
Modern engineering approaches seem to be the opposite of that. I think we've become so accustomed to living in a silicon driven world where our personal devices are engineered at microscopic level that we've forgotten how to do things the Apollo-era way.
For example, to the best of my knowledge, IM-2 doesn't use RADAR — they're using LIDAR and optical navigation instead. Perhaps it is to save on mass and power so that more payload reaches the surface. Perhaps optical navigation was declared to be "good enough." Perhaps it doesn't make sense from a minmaxing of capital perspective. But this philosophy may not be suited to an untamed frontier.
China adopted the Surveyor / Apollo-era philosophy. Their first successful lander, Chang'e 3, used the same hover & fall technique as Surveyor.
> The vehicle will hover at this altitude, moving horizontally under its own guidance to avoid obstacles, and then slowly descend to 4 m above the ground, at which point its engine will shut down for a free-fall onto the lunar surface. The landing site will be at Sinus Iridum, at a latitude of 44º.
It chose the terminal landing sites with the help of LIDAR and its cameras, but it relied on RADAR and a suite of sensors to have robust navigation.
The follow up missions up-ed the ante every time, but they seem to have consistently focused on the robustness of their craft over precision, MBA-spreadsheet-oriented minmax-ing.
> "I think we've become so accustomed to living in a silicon driven world where our personal devices are engineered at microscopic level that we've forgotten how to do things the Apollo-era way."
This is a really interesting point. I think a practical issue in modern times as well is that companies are being inspired by SpaceX while forgetting that it took SpaceX alot of work to get to the point of being able to do things like casually land a 20 story tower in the middle of the ocean on a barge, let alone the even more ridiculous 'stunts' they're doing with Starship.
Apollo was starting from the perspective of trying to do something where it was even debatable about whether it was possible. And so I think there was a lot more 'humility' in design, for lack of a better word.
You're criticizing the prioritization of cost, not the concept of trying to solve for constraints. Engineering is about constrained optimization to meet customer needs.[1] Learning this is a core part of the curriculum at my accredited engineering school.
> Engineering design is a process of making informed decisions to creatively devise products, systems, components, or processes to meet specified goals
based on engineering analysis and judgement. The process is often
characterized as complex, open-ended, iterative, and multidisciplinary.
Solutions incorporate natural sciences, mathematics, and engineering
science, using systematic and current best practices to satisfy defined
objectives within identified requirements, criteria and constraints.
> Constraints to be considered may include (but are not limited to): health and
safety, sustainability, environmental, ethical, security, economic, aesthetics
and human factors, feasibility and compliance with regulatory aspects, along
with universal design issues such as societal, cultural and diversification
facets.
It's not an MBA philosophy but is intrinsic to the profession. Apollo didn't go up because of vibes, it went up because engineers knew the goals going in and to figured out how much fuel was needed to go to the moon. It also went up because the United States was willing to spend over a quarter of a trillion dollars (adjusted for inflation) on getting there,[2] and ignored the arguments that it was a giant waste of money while there were social problems at home.[3]
> If you take the time to study the documentation from the 1950s & 1960s, the engineering culture of that era appears to be markedly different from the engineering culture prevalent today. And I think it's deeply rooted in the symbiotic relationship between computing, Baumol's cost disease and our obsession with precision, results-oriented, MBA-style-min-maxing, "good enough for government work" engineering.
I wonder how much of that is because of public attitudes to government spend. Like if a SpaceX rocket blows up, they're taking innovative, risk-taking approaches to rocket development. If a NASA rocket blows up they're wasting tax payer funding.
Similarly the pressure on NASA to have fewer programs for cost saving is similar. If NASA has two rocket programs, one of which is at a "good enough" level for launching satellites economically into space and one of them is a "safety conscious" rocket for manned launches at a higher per-mission cost, then people look at this and think why is NASA duplicating work and spending. So now they get only one program, so then even launching a GPS satellite is the expensive, human-safe rocket.
> China adopted the Surveyor / Apollo-era philosophy. Their first successful lander, Chang'e 3, used the same hover & fall technique as Surveyor.
Dropping the last 4 metres isn't a sign of having a ruggedized, over-speced "takes a lickin' and keeps on kicking' approach". In lunar gravity, you could drop a raw egg from that height and not perturb the chick inside.
Instead the aim is to avoid throwing up too much moon dust with retro rockets.
Luna 9 (1966) really did need to withstand a bit of a bump, but it was 22km/h, comparable with a fast running pace or a car in first gear, not a high speed impact.
It is all downstream of the loss of the manufacturing industry in America. In the 50s you could entrust a random guy to build a liquid rocket engine in a dusty garage because he spent every day of his career building various pipes and combustion chambers. All of these guys are now dead or retired so when you try to build hardware today you get new grads who settle on LIDAR and computer vision not because it is the best choice but because it is literally all that they are familiar with; the old solutions have all ceased to exist within the minds of employees and classrooms.
This reads like a “comment” version of Destin’s speech to a NASA group a few years ago [0]. The loss of institutional knowledge and fundamentals philosophical differences seem like they’ll need to be overcome.
> Perhaps it is to save on mass and power so that more payload reaches the surface.
It doesn't matter how much mass was saved and how much more payload that allowed to reach the surface if the landing isn't successful. Successful landing is mandatory for anything else to matter. The obviousness of this baffles me that it is taken so haphazardly.
If you look at Change 5 and 6, they seem to do the same image processing based landing control. This doesn't seem to be a cost cutting measure, since the image processing is computationally much more complex than using a radar altimeter.
Definitely not a dumb question. The first lander to land on the Moon (after many failures) is pretty amusing. [1] The Soviets a designed a lander that'd be launched right into the Moon but, just before impact would jettison the lander which itself was a highly reinforced ball that was then designed to simply pound into the Moon at 54kph, but survive the crash. The egg then unfurled and finally humanity had achieved a 'soft' landing on the Moon. Somehow it kind of makes one think of a really elaborate egg drop contest paired with a 'what happens if you jump right before the elevator crashes.'
Like another comment mentioned, complexity and size are big issues. Some more are power/mechanics (fluids, such as for hydraulics, and -280F aren't gonna play well together) and then there's the fact that there's not even a guarantee it'd work. Your legs could get damaged, you might end up in an orientation where none of the legs are appropriate, and so on. So you may be adding a whole bunch of complexity for stuff that might not even save you in the situation it was designed for!
Mass. Each kilogram costs what, millions? Hundreds of millions?
There's a small chance that navigation or landing fails in a way that would make those legs useful, and an even smaller chance that they'll save the mission.
Given tight budgets, this is almost certainly not a gamble worth taking
Because:
1. It cannot fail in this mode.
2. Testing is done by the user, test results are sent by telemetry and the fix will be done, when the bug can be reproduced on developer's computers.
This company's PR team is doing incredible work getting all this type of coverage out of a 0/2 track record with a design everyone else seems to think is obviously flawed.
Kinda explain why Neil Armstrong burned up all their fuel except for a few seconds scoping out the landing site in paranoia.
Instead of building all these expensive to launch big landers, why not get some pizza-box sized probes into earth orbit AND THEN do like a slo-mo golf shot arcing to where the moon will be for a super slow/soft landing?
Some will fail but if you launch 100 and get 20-30 working, there you go.
As technology progresses, get it down to a shoe-box sized probe and then in 10 years smartphone sized (in 100 years tic-tac sized).
It's definitely possible to target a certain surface location on the moon from low Earth orbit and set off on a trajectory to get there with a single burn. However, as the craft(s) approach the moon and enter its sphere of influence, gravity will kick in and increase their relative velocity to the surface. Another burn (suicide burn if you're feeling lucky) would be needed for the soft touchdown.
The moon is also gravitationally very "lumpy", so some small corrections might be needed along the way as well.
Space applications of all sorts are screaming out for mass production approaches. With so much design work and verification the actual manufacturing cost tends to be trivial by comparison, the work readily adapted to concurrent manufacturing processes.
Combine that with leaving the long-range comms (and higher-powered equipment) in lunar orbit as the "master" for all the probes scattered on the surface, and maybe the problem becomes simpler by breaking it in two.
TL;DW: It had far too much sideways velocity immediately before touchdown, likely due to some guidance-system failure. It would have crashed even if it was crab-shaped instead of tower-shaped.
How can you have temperature without a medium? The lander would have steadily lost temperature through heat radiation, but it's not like the vacuum had a temperature in the crater.
Well, there's a natural barrier to entry, i.e. it costs millions on dollars and years of time to be able to "crash" your "junk" onto the surface. So it's not like a bunch of hobby rocketeers are throwing away their McDonald's garbage.
By the way, in comparison to the cost and complexity of said "junk", everything you own is cluttering up the Earth. You should really do something about that.
At what point do you just fire your entire "Land on Moon" software team, and hire a couple young Neil Armstrong wanna-be's, who can hand-land your spacecraft remotely? (In spite of the moon-earth-moon signal lag.)
I mean, because it's in the dark I'd expect it to reach equilibrium with space background thermal radiation which is around 3K. Yet its 100K. Where does that heat comes from? It radiates from earth? Conduct through the floor coming from the inner of the moon itself? (Is there some kind of geothermal gradient on the moon BTW?)
There are multiple factors. The biggest ones are reflected sunlight; infrared and thermal conduction from surrounding rocks; and the Moon's internal heat (the region between the core and the mantle has a temperature of over 1,300 C).
Exactly. I look forward to humanity genuinely progressing towards successfully landing stuff on the moon. One day, we may even be able to land humans there! Intuitive systems is really pushing the envelope here, and I think that's worthy of admiration.
This is horribly disappointing but I'm surprised the altimeter failed as I'd have thought this would have been one of the more reliable aspects of the mission.
How does its altimeter work, exactly what tech does it use? It's worth remembering that radar-type altimeters have been around for a long time and are well developed. For example, Little Boy that was dropped on Hiroshima 80 years ago used radar altimeters in a redundancy arrangement (four devices) and that worked on first attempt.
So what went wrong? Second question, was redundancy employed in the altimeter's design? Third, if the altimeter employed redundancy then why weren't its multiple sections of different designs to allow for the possibility that the reflected signal may be weak and noisy?
(The strength of a returned wave from a radar transmission depends on various factors including its wavelength and the properties of the surface it's being reflected from. If there's any doubt the returned signal's S/N would be such that noise could be a problem it'd make sense for a redundant system to employ multiple wavelengths whose frequencies are far enough apart to take advantage of the fact that the moon's surface would reflect different wavelengths in different ways and at different signal strengths.)
> For the second mission in a row, the lander's altimeter failed
That's a bummer. Altimeters are relatively simple and defined hardware as far as I know. Send a ping, receive a ping, calculate. Too bad they didn't incorporate a backup solution.
[+] [-] 1970-01-01|1 year ago|reply
https://www.lroc.asu.edu/images/1408
[+] [-] freeopinion|1 year ago|reply
(Yes, I suppose there are many other sources that could provide images. These happen to be from ASU.)
[+] [-] CarRamrod|1 year ago|reply
[+] [-] dang|1 year ago|reply
Athena spacecraft declared dead after toppling over on moon - https://news.ycombinator.com/item?id=43292471 - March 2025 (340 comments)
The Moon Lander Athena's Fate on the Lunar Surface Is Uncertain - https://news.ycombinator.com/item?id=43283136 - March 2025 (1 comment)
[+] [-] inamberclad|1 year ago|reply
[+] [-] krisoft|1 year ago|reply
But in case you can: Was a radar based altimeter considered?
How do you guys deal with kicked up regolith? (I have seen first hand how hard heavy snow is on lidars, and would imagine that regolith “shower” is similar, but what do I know.)
[+] [-] yieldcrv|1 year ago|reply
[+] [-] 1970-01-01|1 year ago|reply
[+] [-] martin_drapeau|1 year ago|reply
[+] [-] ceejayoz|1 year ago|reply
Hard landing, skid, tip.
[+] [-] russdill|1 year ago|reply
[+] [-] shadowgovt|1 year ago|reply
Oh yeah, we've all Kerbaled it in like that at one point or another.
[+] [-] areoform|1 year ago|reply
Robert Truax, the designer of the Sea Dragon, loved to promote the design paradigm of Big Dumb Boosters. Instead of many small, sophisticated rocket engines, what if we made one big robust one that can take a lickin' and keep on kickin'.
The idea was to relax the mass margins and to create big. dumb. boosters. It's the approach TRW explicitly followed for the Lunar Module engine,
The Surveyor program managed to make it "just work" 5 out of 7 times by adopting this approach. It had robust landing legs and RADAR. They would decelerate and then shut off the engine 11' above the surface. The wide, sturdy legs would then absorb that final impact of coming stand still from free fall.These programs had a lot of capital behind them. Some components required precision engineering, but there's a very clear through line and embrace of the "we gotta make stuff that can take a lickin' & keeps kickin'" philosophy.
Modern engineering approaches seem to be the opposite of that. I think we've become so accustomed to living in a silicon driven world where our personal devices are engineered at microscopic level that we've forgotten how to do things the Apollo-era way.
For example, to the best of my knowledge, IM-2 doesn't use RADAR — they're using LIDAR and optical navigation instead. Perhaps it is to save on mass and power so that more payload reaches the surface. Perhaps optical navigation was declared to be "good enough." Perhaps it doesn't make sense from a minmaxing of capital perspective. But this philosophy may not be suited to an untamed frontier.
China adopted the Surveyor / Apollo-era philosophy. Their first successful lander, Chang'e 3, used the same hover & fall technique as Surveyor.
It chose the terminal landing sites with the help of LIDAR and its cameras, but it relied on RADAR and a suite of sensors to have robust navigation.The follow up missions up-ed the ante every time, but they seem to have consistently focused on the robustness of their craft over precision, MBA-spreadsheet-oriented minmax-ing.
[+] [-] somenameforme|1 year ago|reply
This is a really interesting point. I think a practical issue in modern times as well is that companies are being inspired by SpaceX while forgetting that it took SpaceX alot of work to get to the point of being able to do things like casually land a 20 story tower in the middle of the ocean on a barge, let alone the even more ridiculous 'stunts' they're doing with Starship.
Apollo was starting from the perspective of trying to do something where it was even debatable about whether it was possible. And so I think there was a lot more 'humility' in design, for lack of a better word.
[+] [-] jjmarr|1 year ago|reply
> Engineering design is a process of making informed decisions to creatively devise products, systems, components, or processes to meet specified goals based on engineering analysis and judgement. The process is often characterized as complex, open-ended, iterative, and multidisciplinary. Solutions incorporate natural sciences, mathematics, and engineering science, using systematic and current best practices to satisfy defined objectives within identified requirements, criteria and constraints.
> Constraints to be considered may include (but are not limited to): health and safety, sustainability, environmental, ethical, security, economic, aesthetics and human factors, feasibility and compliance with regulatory aspects, along with universal design issues such as societal, cultural and diversification facets.
It's not an MBA philosophy but is intrinsic to the profession. Apollo didn't go up because of vibes, it went up because engineers knew the goals going in and to figured out how much fuel was needed to go to the moon. It also went up because the United States was willing to spend over a quarter of a trillion dollars (adjusted for inflation) on getting there,[2] and ignored the arguments that it was a giant waste of money while there were social problems at home.[3]
[1]https://egad.engineering.queensu.ca/wp-content/uploads/2023/...
[2] https://www.planetary.org/space-policy/cost-of-apollo
[3] https://en.wikipedia.org/wiki/Whitey_on_the_Moon
[+] [-] Macha|1 year ago|reply
I wonder how much of that is because of public attitudes to government spend. Like if a SpaceX rocket blows up, they're taking innovative, risk-taking approaches to rocket development. If a NASA rocket blows up they're wasting tax payer funding.
Similarly the pressure on NASA to have fewer programs for cost saving is similar. If NASA has two rocket programs, one of which is at a "good enough" level for launching satellites economically into space and one of them is a "safety conscious" rocket for manned launches at a higher per-mission cost, then people look at this and think why is NASA duplicating work and spending. So now they get only one program, so then even launching a GPS satellite is the expensive, human-safe rocket.
[+] [-] dmurray|1 year ago|reply
Dropping the last 4 metres isn't a sign of having a ruggedized, over-speced "takes a lickin' and keeps on kicking' approach". In lunar gravity, you could drop a raw egg from that height and not perturb the chick inside.
Instead the aim is to avoid throwing up too much moon dust with retro rockets.
Luna 9 (1966) really did need to withstand a bit of a bump, but it was 22km/h, comparable with a fast running pace or a car in first gear, not a high speed impact.
[+] [-] nexuist|1 year ago|reply
[+] [-] j_bum|1 year ago|reply
[0] https://youtu.be/OoJsPvmFixU?si=EUxpp6C9vRAYD3kA
[+] [-] dylan604|1 year ago|reply
It doesn't matter how much mass was saved and how much more payload that allowed to reach the surface if the landing isn't successful. Successful landing is mandatory for anything else to matter. The obviousness of this baffles me that it is taken so haphazardly.
[+] [-] 1970-01-01|1 year ago|reply
I think this is the smoking gun. RADAR is usually successful, while LIDAR has a poor record.
[+] [-] throwaind29k|1 year ago|reply
[+] [-] IX-103|1 year ago|reply
[+] [-] dist-epoch|1 year ago|reply
Seems it's the second time they fail in this mode.
[+] [-] somenameforme|1 year ago|reply
Like another comment mentioned, complexity and size are big issues. Some more are power/mechanics (fluids, such as for hydraulics, and -280F aren't gonna play well together) and then there's the fact that there's not even a guarantee it'd work. Your legs could get damaged, you might end up in an orientation where none of the legs are appropriate, and so on. So you may be adding a whole bunch of complexity for stuff that might not even save you in the situation it was designed for!
[1] - https://en.wikipedia.org/wiki/Luna_9
[+] [-] moffkalast|1 year ago|reply
[+] [-] ragebol|1 year ago|reply
[+] [-] mystified5016|1 year ago|reply
There's a small chance that navigation or landing fails in a way that would make those legs useful, and an even smaller chance that they'll save the mission.
Given tight budgets, this is almost certainly not a gamble worth taking
[+] [-] notTooFarGone|1 year ago|reply
[+] [-] hulitu|1 year ago|reply
/s
[+] [-] lionkor|1 year ago|reply
[+] [-] Aerroon|1 year ago|reply
[+] [-] ghostly_s|1 year ago|reply
[+] [-] ck2|1 year ago|reply
Instead of building all these expensive to launch big landers, why not get some pizza-box sized probes into earth orbit AND THEN do like a slo-mo golf shot arcing to where the moon will be for a super slow/soft landing?
Some will fail but if you launch 100 and get 20-30 working, there you go.
As technology progresses, get it down to a shoe-box sized probe and then in 10 years smartphone sized (in 100 years tic-tac sized).
[+] [-] accrual|1 year ago|reply
The moon is also gravitationally very "lumpy", so some small corrections might be needed along the way as well.
[+] [-] mapt|1 year ago|reply
[+] [-] btbuildem|1 year ago|reply
[+] [-] TimByte|1 year ago|reply
[+] [-] jiggawatts|1 year ago|reply
TL;DW: It had far too much sideways velocity immediately before touchdown, likely due to some guidance-system failure. It would have crashed even if it was crab-shaped instead of tower-shaped.
[+] [-] villmann|1 year ago|reply
[+] [-] driggs|1 year ago|reply
This startup has already crashed two pieces of space junk onto the lunar surface. Can any startup able to get there do whatever they want?
[+] [-] kQq9oHeAz6wLLS|1 year ago|reply
By the way, in comparison to the cost and complexity of said "junk", everything you own is cluttering up the Earth. You should really do something about that.
[+] [-] bell-cot|1 year ago|reply
[+] [-] littlestymaar|1 year ago|reply
I mean, because it's in the dark I'd expect it to reach equilibrium with space background thermal radiation which is around 3K. Yet its 100K. Where does that heat comes from? It radiates from earth? Conduct through the floor coming from the inner of the moon itself? (Is there some kind of geothermal gradient on the moon BTW?)
[+] [-] antonvs|1 year ago|reply
[+] [-] Koshkin|1 year ago|reply
[+] [-] TimByte|1 year ago|reply
[+] [-] wesselbindt|1 year ago|reply
[+] [-] hilbert42|1 year ago|reply
How does its altimeter work, exactly what tech does it use? It's worth remembering that radar-type altimeters have been around for a long time and are well developed. For example, Little Boy that was dropped on Hiroshima 80 years ago used radar altimeters in a redundancy arrangement (four devices) and that worked on first attempt.
So what went wrong? Second question, was redundancy employed in the altimeter's design? Third, if the altimeter employed redundancy then why weren't its multiple sections of different designs to allow for the possibility that the reflected signal may be weak and noisy?
(The strength of a returned wave from a radar transmission depends on various factors including its wavelength and the properties of the surface it's being reflected from. If there's any doubt the returned signal's S/N would be such that noise could be a problem it'd make sense for a redundant system to employ multiple wavelengths whose frequencies are far enough apart to take advantage of the fact that the moon's surface would reflect different wavelengths in different ways and at different signal strengths.)
[+] [-] accrual|1 year ago|reply
That's a bummer. Altimeters are relatively simple and defined hardware as far as I know. Send a ping, receive a ping, calculate. Too bad they didn't incorporate a backup solution.