It's really hard to land on the moon....

[Trafalgar]

So the recent failed attempt by to land a probe on the moon got me thinking, how hard is it? Off of Wikipedia, this is a list of landing attempts since 2020:

• Chang'e 5 - PRC - Dec 16 2020 - Success
• Omotenashi - Japan - Nov 21, 2022 - Fail
• Hakuto-R - Japan - Apr 2023 - Fail
• Chandrayaan-3 - India - Jul 14 2023 - Success
• Luna 25 - Russia - Aug 10, 2023 - Fail
• Peregrine - USA - Jan 8, 2024 - Fail
• SLIM - Japan - Jan 19, 2024 - Success
• IM-1 Odysseus - USA - Feb 22, 2024 - Fail

A few thoughts on this:
-Only 3 of 8 recent attempts to land on the moon have been successful. This is much lower than I would have thought.
-PRC, Japan, and India have been successful while Russia and USA have not. So your historical space exploration heavy weights are being outdone by relative newcomers.
-It's interesting to me that IM-1 has been treated as a partial success by the US media. Wikipedia even says "landed" instead of failure despite the craft toppling over onto its side. If this was a crewed mission it would be a disaster with the crew stranded.
-In January, NASA released a schedule for Artemis in January that planned a crewed moon landing for September 2026. They have contracted both SpaceX and Blue Origin to develop spacecraft for this. There is no way a reliable crewed lander can be developed and tested by then.

[Gnoman]

So the recent failed attempt by to land a probe on the moon got me thinking, how hard is it? Off of Wikipedia, this is a list of landing attempts since 2020:

• Chang'e 5 - PRC - Dec 16 2020 - Success
• Omotenashi - Japan - Nov 21, 2022 - Fail
• Hakuto-R - Japan - Apr 2023 - Fail
• Chandrayaan-3 - India - Jul 14 2023 - Success
• Luna 25 - Russia - Aug 10, 2023 - Fail
• Peregrine - USA - Jan 8, 2024 - Fail
• SLIM - Japan - Jan 19, 2024 - Success
• IM-1 Odysseus - USA - Feb 22, 2024 - Fail

A few thoughts on this:
-Only 3 of 8 recent attempts to land on the moon have been successful. This is much lower than I would have thought.
-PRC, Japan, and India have been successful while Russia and USA have not. So your historical space exploration heavy weights are being outdone by relative newcomers.
-It's interesting to me that IM-1 has been treated as a partial success by the US media. Wikipedia even says "landed" instead of failure despite the craft toppling over onto its side. If this was a crewed mission it would be a disaster with the crew stranded.
-In January, NASA released a schedule for Artemis in January that planned a crewed moon landing for September 2026. They have contracted both SpaceX and Blue Origin to develop spacecraft for this. There is no way a reliable crewed lander can be developed and tested by then.

Peregrine was an abort, not a fail. A mechanical defect was detected and the decision was made to bring it back to Earth for analysis instead of continuing with the landing attempt. It is not certain that it would have failed to land, the Powers That Be merely decided that analyzing the fault was more important than making the attempt.

Odysseus was a perfect landing - until a mechanical defect in one of the landing struts caused it to not be fully upright, making it only partially operational.

The two failed Japanese missions failed due to apparent software bugs.

This means with the possible exception of Luna 25 (where there's no clear idea what happened - official statement is that there was an incorrect burn time, but the nature of the crash is preventing analysis as to why), none of them failed because the people who launched them don't know how to land on the Moon. Note that at least a couple of the failures were in "we're deliberately trying to see what corners we can cut to keep weight down, because every single gram we can save makes moon launches cheaper" situations.

[Chronos]

Landing on the Moon is really, incredibly easy, by space standards. It's so easy that 3 out of 8 recent programs managed to do it on their first try. How many space systems have an initial success rate that high? How many of any major program has an initial success rate that high?

[Trafalgar]

Peregrine was an abort, not a fail. A mechanical defect was detected and the decision was made to bring it back to Earth for analysis instead of continuing with the landing attempt. It is not certain that it would have failed to land, the Powers That Be merely decided that analyzing the fault was more important than making the attempt.

Odysseus was a perfect landing - until a mechanical defect in one of the landing struts caused it to not be fully upright, making it only partially operational.

The two failed Japanese missions failed due to apparent software bugs.

This means with the possible exception of Luna 25 (where there's no clear idea what happened - official statement is that there was an incorrect burn time, but the nature of the crash is preventing analysis as to why), none of them failed because the people who launched them don't know how to land on the Moon. Note that at least a couple of the failures were in "we're deliberately trying to see what corners we can cut to keep weight down, because every single gram we can save makes moon launches cheaper" situations.

A propellant fuel tank on Peregrine ruptured after being over pressurized making a soft landing impossible. They didn't "bring it back to earth for analysis". They caused the probe to brake up during reentry over the Pacific so it wouldn't contribute to the space debris problem.

Odysseus was not a perfect landing. A laser rangefinder failed to operate. This caused the probe to land too hard on too steep a slope resulting in a leg crumpling.

As far as Luna 25, Roscosmos is very tightlipped about their failures. I imagine they have a pretty good idea what happened even if they haven't told the rest of the world...

Landing on the Moon is really, incredibly easy, by space standards. It's so easy that 3 out of 8 recent programs managed to do it on their first try. How many space systems have an initial success rate that high? How many of any major program has an initial success rate that high?

Russia, USA, Japan, and PRC have all landed probes on the moon before. This was India's second attempt to soft land a probe. So I am not sure what you mean by "first try".

Compare these results to NASA's "Surveyor" program which landed 5 out of 7 probes on the moon in the 1960s. And that was using an onboard 1960s computer.

[Maat Mons]

The more complex a device is, the more ways there are for it to malfunction. The simple computers of the 1960's weren't the handicap to reliability you seem to be implying they would have been.

[Sermil]

-It's interesting to me that IM-1 has been treated as a partial success by the US media. Wikipedia even says "landed" instead of failure despite the craft toppling over onto its side. If this was a crewed mission it would be a disaster with the crew stranded.

But it's not a crewed mission. Its mission -- with success metrics decided in advance -- was to send back data about the performance of several new technologies NASA wanted to test. It did so. You can't move the goalposts after the fact and decide that it was a complete failure because it completed its primary objective but not well enough.

It was a success. End of story.

-In January, NASA released a schedule for Artemis in January that planned a crewed moon landing for September 2026. They have contracted both SpaceX and Blue Origin to develop spacecraft for this. There is no way a reliable crewed lander can be developed and tested by then.

Oh, yes, the whole Artemis timeline is complete BS. They have to say certain things that Congress wants to hear.

But back to the main question:

-Only 3 of 8 recent attempts to land on the moon have been successful. This is much lower than I would have thought.

Space is hard. Space is really hard.

There's a lot of reasons for this, but I can list a few:

1. Very hard to test for. There's no real way of doing a small-scale test of, say, Moon landing. There's nothing on Earth that is a good analogue for landing on the Moon. Anything on Earth will have far too much gravity, far too much air, etc. It's like trying to train to be an Olympic figure skater if you didn't have access to any sort of ice. So we usually have to send mostly-untested prototypes (at huge expense) to the Moon to practice Moon landings. To put it another way, for other technologies, all those failures would have happened in someone's private testing range and engineers would have iterated on them before you saw them.
2. There are no humans around to help. Well-trained humans are often great at improvising around the unexpected and come with a well-tested set of sensors -- think Miracle on the Hudson or the Apollo 13 crew doing manually-targeted burns. They're also good at fixing things -- how many car accidents are avoided because someone can just pull off to the side of the road and change a tire? Moon landers are entirely automated and have no way of fixing issues -- when things go wrong, that's usually the end of the mission.
3. For the most part, things happen fast and you can't try again. Even an airplane can often do a go-around if a landing is coming in too high. But once a Moon lander does its retro landing burn, it's committed, the ground is coming at them fast. They have to hit every mark perfectly. The computer can't say "Oops, that engine didn't light properly, let's scrub and try again".
4. Weight. The easiest ways to control for engineering risk are redundancy and margin. Put on one more sensor than you need, add a little fuel, make the strut / tank / spaceship body a little stronger than it needs to be. No airplane would fly with exactly, down to the ounce, the fuel it needs to get to the destination. But the "tyranny of the rocket equation" means that every extra pound of weight on a lander is adding huge amounts to the launcher. You really have to optimize weight (especially for value-conscious private companies like IM). That means making everything as weak as you can get away with.

-PRC, Japan, and India have been successful while Russia and USA have not. So your historical space exploration heavy weights are being outdone by relative newcomers.

*cough* Russia has never put anything on the Moon. The Soviet Union put landers on the Moon. Many of the people in the Soviet space program were not Russian.

And NASA did not launch any of those Moon landers. Private companies did.

[Rockphed]

The inoperative rangefinder was, per a NASA engineer I spoke with, caused by someone forgetting to remove the hardware interlock that prevented the laser from emitting. This is almost as bad as crashing a probe into mars because someone ignored the documentation and used the wrong units. Fortunately, they had an experimental backup, so they could still land.

[Peelee]

The Mod on the Silver Mountain: Let's shy away from discussing dealings with Congress, just to keep things safe.

Anything on Earth will have far too much gravity, far too much air, etc.

Also, this has the interesting analog of how taking off and landing are sort of inversely proportional. Landing on Earth is much easier than landing on the moon, thanks to the atmosphere - you just get your angle right, make a few atmospheric dips if your velocity is too high, and let chutes take care of the rest. The Moon, with no atmosphere, means you're directly controlling the entire landing process by firing your rockets. You orient by rocket. Control descent speed by rocket. Steer by rocket. All things that chutes and math can handle for you on Earth. Conversely, to take off on Earth, you have to break through all that soupy gas before you can really start burning to orbital velocity, which wastes just a ton of energy. On the Moon? Easy peasy, your only hurdle is gravity and that's always the case regardless.

[Radar]

The inoperative rangefinder was, per a NASA engineer I spoke with, caused by someone forgetting to remove the hardware interlock that prevented the laser from emitting. This is almost as bad as crashing a probe into mars because someone ignored the documentation and used the wrong units. Fortunately, they had an experimental backup, so they could still land.

There were many other failures like this. For example, Genesis probe that gathered solar wind particles very close to the Sun crashed full-speed into Earth because the parachutes did not open. The reason? All four, independent accelerometers were mounted upside down.

[tomandtish]

Speaking of hard, Japanese firm Space One tried to be the first Japanese company to put a satellite into orbit. Their rocket, Kairos, exploded 5 seconds after liftoff (Wednesday for them, 10:01pm Eastern Time for U.S.).

Apparently still a few bugs in the system.

[Sean Mirrsen]

As far as Luna 25, Roscosmos is very tightlipped about their failures. I imagine they have a pretty good idea what happened even if they haven't told the rest of the world...

Closest thing I heard to an official explanation was accelerometer failure, leading to the burn time being determined by a cutoff timer instead of actual acceleration data. Which, clearly, was too much burn.

[Trafalgar]

In other news, SpaceX Starship had a test launch that was mostly successful though the reentry test didn't happen because the space craft was in a spin. The video of plasma forming around the spacecraft is pretty spectacular.

[Quizatzhaderac]

If this was a crewed mission it would be a disaster with the crew stranded.

There's a tradeoff that needs to be made between reliability versus cost and more mission objectives.

The main benefit of the unmanned missions is that it allows you to skimp on the reliability.

Two unmanned missions with a 50% success chance are much cheaper than one mission with with a 99% success chance. For manned missions you have to do the cost analysis differently.

[Ionathus]

In other news, SpaceX Starship had a test launch that was mostly successful though the reentry test didn't happen because the space craft was in a spin. The video of plasma forming around the spacecraft is pretty spectacular.

Can you share a link?

[Trafalgar]

Can you share a link?

Here you go.

[Tyndmyr]

In other news, SpaceX Starship had a test launch that was mostly successful though the reentry test didn't happen because the space craft was in a spin. The video of plasma forming around the spacecraft is pretty spectacular.

It's a shame that SpaceX isn't public. I love what they're doing, and I'd love to buy a giant pile of their stock. Nobody else is where they are in the commercial space race.

I guess it's either buy Google for the mild exposure offered or embrace Virgin Galactic.

Still, the launch videos are pretty awesome, even without the having a direct stake in it.

[Chronos]

Quoth tomandtish:

Their rocket, Kairos, exploded 5 seconds after liftoff

But how long, subjectively, was that five seconds? Was it a good five seconds?

[tomandtish]

But how long, subjectively, was that five seconds? Was it a good five seconds?

I can't find anything specific regarding whether they were aware of a problem prior to explosion, although in fairness it would probably be a "Hey, what's that light fo..." Boom.

hard to judge distance in the video but given the rocket is roughly 60 feet long I don't think it even hit the 500 foot mark. So probably not a good 5 seconds.

[tomandtish]

But how long, subjectively, was that five seconds? Was it a good five seconds?

I can't find anything specific regarding whether they were aware of a problem prior to explosion, although in fairness it would probably be a "Hey, what's that light fo..." Boom.

Hard to judge distance in the video but given the rocket is roughly 60 feet long I don't think it even hit the 500 foot mark. So probably not a good 5 seconds.

[Chronos]

(The word "Kairos" means "time as measured by its subjective quality, not by its duration".)