Beresheet, SpaceIL Israeli Moon mission

[ChrisWilson68]

This makes SpaceIL look pretty amateurish.

The software wasn't properly designed if re-starting the broken IMU caused things to lock up.

That same software wasn't properly tested since the problem wasn't found before launch.

The software wasn't properly designed if a reboot deleted the updated software that they had loaded.

The person or people who decided it was OK to manually restart the second IMU in the middle of the critical landing maneuver made a horribly bad decision.

Having only a small team and a small budget isn't an excuse for any of this.  A very small team with better judgement wouldn't have made any of these mistakes.

[Steven Pietrobon]

Lesson 1: Don't fix it if it isn't broken during critical maneuvers! IMU1 was working so no need to send a reset command. Its far more likely the reset could cause a problem than IMU1 failing in the next couple of minutes to landing.

Lesson 2: Don't store your critical software changes in SRAM or DRAM! Use nonvolatile memory like MRAM.

Lesson 3: Check what happens if your computer reboots during landing! If you crash, rewrite the code so that it can quickly recover and continue from where it left off, just like Apollo 11 did during its landing.

[Eerie]

This makes SpaceIL look pretty amateurish.

Gee, really? A first time private lunar lander, amateurish? Impossible!

[QuantumG]

It was a good attempt, I hope they get the funds to try again... but (to me) it doesn't look like they will.

[Comga]

So they programmed the ability to remotely reset the IMU but never tested what would happen if they actually tried to run that command.

Unfortunately I'm not surprised.

Edit to add: during the landing attempt if I remember correctly at least one engineer can be heard saying not to reset, apparently they didn't listen to him.

To your addition, to me it sounded like the lead gave the command to not do a reset.
It appeared that it was done anyways, perhaps by someone intent on a screen and not listening sufficiently.

There is only so much that practice can prepare a team.   Even Apollo didn't land until the third mission to the Moon.  When it's critical, but you have the world's largest economy behind you, you go just slow enough.
That's not the case for Beresheet.
One guy out of hundreds remembered the "SCE to AUX" switch that saved Apollo 12, and he had to repeat himself before it was sent up.  The inverse of this didn't occur for Beresheet.  These missions exist at the margins of what's possible.

Everyone saying how much smarter they are having all the answers in hindsight are being foolish themselves.
Why not a TRIPLY redundant IMU? Cost, mass, complexity.  That's why
Why not use different form of RAM? There were reasons.  These people know about SRAM, etc.  They made choices for how to get from one mode to the other quickly and cost effectively. 
Don't do this! Do that! How could they be so stupid?
Please stop.

[Comga]

I'll translate for ya: SpaceIL needed more money after the GLXP bombed out, and we needed credibility after years of achieving nothing, so we applied for the CLPS program as a shell company to avoid the US-national-requirements - hey, it worked for RocketLab! So here we are, trying to somehow shoehorn a lunar lander into our product offering. It makes sense!

No really. A lunar lander has all the basic technology you need to solve to do modern autonomous rockets ala SpaceX, Masten (who I'm actually quoting here) and Armadillo (now Exos). This is the technology I've watched get born over the last 10 years. It's just all so blatantly obvious to a computer geek like me - stop using embedded systems that are emulating shit from the 1960s and apply the full power of this ready and eager industry to your teams - stop pretending that anything you don't understand is bad and trust the experts in other fields. This technology is so readily available now - you're welcome.

Pro Tip: if you are getting all your knowledge about a field from the people in that field, you better be good at sorting the truth from the Incorrect, because there's a lot of confidence guys and pretend spooks (no really, you work for the NSA?! Me too!) hiding amongst this lot. It really burns when you figure out one of your heroes is everything they said he was. 

Of course there's so many stories you can't tell - because you weren't on the inside, so you don't already have a narrative - but they all come together eventually. Little hints here and there. The side-show to keep the tourists interested while the real action gets reported in the media. Ho-boy, was that a bad move? Oh, you're actually doing a pivot with an Israeli partner? Well - that's gotta win some votes.

Chinese partner: might as well write your bankruptcy up now.

Who has LinkSpace BTW? You can't just hold a card like that and think we're not gunna remember you've got it! Is it you Bezos? Was that the plan? Turn dollars into invention in Shenzhen to take away the competitive advantage of everyone who currently competes with you while broadcasting out as loud as you possibly can that you're not doing anything?

Hey Mom!

Mom: Yes, son.

I'm just going to go into the kitchen to get an apple.

Mom: Really...?

Yep, just an apple.

Nothin' untoward going on here.

<Thank you, tip your server>

OK Can someone translate QG for me?
Seriously, you have a wealth of information and insights, but some of us have little to no idea what you are talking about. 
"Mom!  Make him stop!"
Translated: A reply in plain English, with fewer obscure allegories, would be of interest to many of us.   It's not obvious to those of us not in computer science or engineering or whatever you feel would be the major advance.  We are really interested in what you meant.

PS  Did you also reverse some of your sentences, like meaning "when you figure out one of your heroes is NOT everything they said he was"?

[FutureSpaceTourist]

One year after the launch of Beresheet we've not had much information from official sources about how exactly the spacecraft failed leading to its crash into the lunar surface. However a new article on an Israeli news site by an individual close to the project offers up a few bits of information which weren't previously in the public sphere. The article is in Hebrew, but the google translation is pretty good and has been confirmed by individuals who understand the language.
https://www.ynet.co.il/articles/0,734...

However we still don't have the level of detail on the operation of the computer, its software extensions and the process for starting the main engine, so there may be more to find out in the future.

[mn]

One year after the launch of Beresheet we've not had much information from official sources about how exactly the spacecraft failed leading to its crash into the lunar surface. However a new article on an Israeli news site by an individual close to the project offers up a few bits of information which weren't previously in the public sphere. The article is in Hebrew, but the google translation is pretty good and has been confirmed by individuals who understand the language.
https://www.ynet.co.il/articles/0,734...

However we still don't have the level of detail on the operation of the computer, its software extensions and the process for starting the main engine, so there may be more to find out in the future.

Here is the full link to the ynet article
https://www.ynet.co.il/articles/0,7340,L-5681547,00.html

I read the article quickly and watched the video. One point to correct on the video: when the IMU failed he says it also blocked the signal from the other IMU which than triggered the infamous reboot. What the article actually says is that they could theoretically have continued with the one remaining IMU (and take the risk of the remaining IMU failing later in the landing sequence). They decided to restart the failed IMU, but they failed to realize (or did realize, but too late, as apparently someone did say not to restart) that the restart will cause the computer to very briefly lose data from BOTH IMU's (for less than a second), this brief loss of IMU data triggered a reboot of the computer (as the computer was programmed to try to reboot if it loses the IMU signal) which led to the failure as he goes on to describe.

[zubenelgenubi]

New thread for Beresheet 2: https://forum.nasaspaceflight.com/index.php?topic=52496.0