Forgot your password?
typodupeerror
Space Hardware

Carmack's Armadillo Aerospace Rocket Crashes and Burns 353

Posted by ScuttleMonkey
from the doomed-rockets dept.
mcgrew (sm62704) writes "New Scientist is reporting that John Carmack's 'Armadillo Aerospace' has suffered a large setback in the Northrop Grumman Lunar Lander Challenge after one of its two main rockets crashed and burned. 'During the test, Texel lifted off and hovered without incident, then descended again and touched the ground. But it then rose again unexpectedly and began accelerating upward. "Crap, it's going to fly into the crane, I need to kill it," Carmack recalls thinking. He hit the manual shutdown switch, turning off the vehicle's engine in mid-flight. Texel was about 6 metres above the ground and fell like a stone. One of its fuel tanks broke open when it hit the ground, spewing fuel that ignited and engulfed the vehicle in flames. "It made a fireball that would make any Hollywood movie proud," Carmack says.' No one was hurt in the crash, but the vehicle was destroyed."
This discussion has been archived. No new comments can be posted.

Carmack's Armadillo Aerospace Rocket Crashes and Burns

Comments Filter:
  • by faloi (738831) on Wednesday August 22, 2007 @03:06PM (#20322127)
    Gravity will make you it's bitch!
  • by 0racle (667029) on Wednesday August 22, 2007 @03:07PM (#20322131)
    On one hand what they were working on was completely destroyed, on the other the explosion was AWESOME!
  • by zsouthboy (1136757) on Wednesday August 22, 2007 @03:07PM (#20322135)
    ...anisotropic filtering.
    • Re: (Score:3, Funny)

      by Dr. Eggman (932300)
      Nah, vertical sync. just was broken, better wait till Nvidia releases new drivers before trying again.
  • by ircmaxell (1117387) on Wednesday August 22, 2007 @03:10PM (#20322157) Homepage
    This is a good thing. It proves (again and again) that new technology is never perfect. Just think, no computer program is ever completely bug free the first time it's compiled. The first car is never perfect... There are always bugs in any system. The point is that the safety mechanisms in the system worked well (after all, acording to the inputs of the lander, it was falling). As with any "accident", there are many failures that lead up to those incidents. That's the price of achievement. Nobody was hurt, so learn, build bigger and build better. If you learn from it, it wasn't a "mistake"...
  • by Odin_Tiger (585113) on Wednesday August 22, 2007 @03:10PM (#20322159) Journal
    Always ensure you have enough HP to survive the landing or an invulnerability artifact when performing a rocket jump.
  • it had to have been caught on video. who's got the youtube link? or perhaps an HD torrent?
    • Re: (Score:3, Informative)

      by XaXXon (202882)
      blog posting said the guy who does the videos is out for a week and a half, but they will be posting it - presumably to their website.
  • It was a bad weekend for Armadillo. We set out to put some flights on Texel, the backup Quad vehicle, and it didn't go so well. We have video that we will be releasing, but Matt had to leave for Germany the next day, so it won't be digitized for a week and a half.

    We started out with a normal 90 second elevated / tethered hover test, but we ran into a problem with the actuator power. We initially thought it was a bad main power switch, but it turned out to be the lithium-polymer battery pack cutoff circuit incorrectly shutting down at 16 amps of load instead of 40. This was a new battery pack ( www.batteryspace.com HPL-8059156-4S-WR), and it had passed all the individual actuator checks, but when the igniter started firing with both high amp NOS solenoids, the battery shut down (went to 0.3 volts indicated) after one second and stayed there until it was physically disconnected. Russ made a fairly heroic field repair, cutting open the battery pack and wiring around the protection circuit while sitting on top of the rocket. The total time spent on this after three attempts was 90 minutes, and enough lox had boiled off that the vehicle hit lox depletion at 60 seconds of flight. We got a few good data points from this: the batteries need to be checked at full current load, with vents open we boil off about two pounds of lox a minute, and lox-depletion runs are benign, if a little flamey.

    For the second flight we were going to do a ground liftoff (still tethered for runaway protection) to test the automatic ground contact engine shutoff code. We have had several reasons to want to automate this: We get a fair bit of bounce on touchdown, because the engine is essentially keeping the vehicle weightless during the terminal descent. A computer controlled shutdown would be at least a half second faster than my manual punching of the shutdown when I visually see ground contact. The quads will just safely bounce around on the ground a bit if the engine just goes to idle and doesn't shut down, but the module, with the gimbal below the CG, will try to tip itself over when a landing leg becomes a pivot point, so there is extra incentive to get it shut off fast. You can see that in our XPC '05 vehicle flight. We also need to handle the case of the vehicle landing in a situation where I can't shut the engine off promptly, either because there was a telemetry problem, or when we are doing high altitude flights, it lands out of direct sight. There is a separate shutdownTime parameter that will keep it from sitting there at idle for ten minutes, but a telemetry abort could still have it on the ground and cooking for the better part of 220 seconds. We could still shut the flight safety fuel valve, which would result in just idle level lox pouring out of the engine, but that has its own problems.

    I have been very hesitant to put in ground contact shutoff code, because shutting the engine down for some incorrect reason would be catastrophic, and I would feel awful if that ever happened. We had some switch based ground contact sensors on the old VDR, but they never got tested. We have concluded that the landing jolt, as seen by the IMU accelerometers, is a good enough ground contact signal. There is always the worry that combustion instability, or a nozzle ejection event, might trigger the signal level, so there are additional guards about it only functioning when you are within three meters of the ground (we must leave some slop for uneven terrain or GPS innacuracy) and trying to descend.

    We loaded up again, being very thankful that we now pack three six-packs of helium for each test trip after we were forced to cancel the second flight on a previous test session due to insufficient helium after troubleshooting a problem forced a repressurization on the first flight. Liftoff and hover was fine, and at the 45 second mark (no sense pushing it on a ground liftoff), I had it come in for a landing. It hit the ground, and I saw it bounce back up. My first thought was "That didn't seem to help at all".
    • by Linker3000 (626634) on Wednesday August 22, 2007 @03:57PM (#20322631) Journal
      Executive summary:

      "hotwired the battery...we don't need no stinkin' ground shutoff code...Sensors - never got around to testing them...we left some slop...ya think something rated at 4G would work up to 6G?...we know the GPS receivers are vibration sensitive so we stuck some bubble wrap round them and hoped...we checked earlier telemetry and yup - they're darn vibration sensitive...hold on lads; I've got an idea...The rocket has gotta return to the ground at some point; if only we'd done some testing on this...John's doing some fancy flying - oh, sh*t, he's not...now the tanks are scrap we're probably going to do some useful tests on them that we wouldn't have done with usable ones - heck those things cost money, baby...some of the wiring harness is wrapped in leather so we're going to alienate the vegan customer base...flammable foam catches fire."

      I think I'll walk.

      PS: The captcha I had to type in to submit this was "Piloting" - BWAHAHAHAHAHA
      • Re: (Score:3, Insightful)

        by Bryan Ischo (893) *
        Modding this as flamebait was really unfair, whoever did that. It's actually a pretty good short summary of all of the problems that led up to the crash. I read some of them and wondered how they could have ever made those mistakes (like not drop-testing the ground shutoff code, something that would be very cheap and easy to do and give much greater confidence in that critical part of the system), but I am going to have to assume that there is much more to the development and testing of these things than
    • by XenoPhage (242134) on Wednesday August 22, 2007 @04:02PM (#20322683) Homepage
      For those that are not aware, this was John's post on a PUBLIC forum. John has continually posted information regarding his team's experiences and any important information they have learned. He's taken the open source mentality into the rocketry arena and many teams are all the better for it. This is the type of information that NASA would happily write a few hundred page reports on and they encase in cement and bury.

      I've been lurking on the rocketry group for a while now and it's great to see the open discussions about everything from rocket design to safety. I've learned more in a few months that I ever did watching all those NASA shuttle launches over the years.
    • by Radon360 (951529) on Wednesday August 22, 2007 @04:14PM (#20322791)

      John goes on about the use of GPS in the control for acceleration for a bit. Understanding that where you have no nearby reference points, such as in space, this may be a good solution. At the same time, you usually don't have anything nearby to worry about crashing into (such as the ground). Although GPS can be very accurate, it often takes more datapoints that can be obtained in a very short timeframe to get that accuracy.

      I wonder if there's a reason why they aren't using some means of LASER or RADAR rangefinding when in close proximity to landing for obtaining positioning (altitude) and velocity/acceleration information. The update rate could easily be several orders of magnitude faster than GPS could ever provide...especially since you need two position reports from GPS to find velocity and three to determine acceleration.

      • Re: (Score:3, Informative)

        by Lank (19922)
        The use of GPS for the craft's altitude is indeed a problem. GPS units have the potential to give horrible altitude readings! Quite often, they do. 95% of the time their readings are within 15 meters of the actual altitude. That means two readings up to 30 meters apart would be considered normal. In actuality 95% of the time they are within 23 meters (source: http://gpsinformation.net/main/altitude.htm [gpsinformation.net]) And then for the other 5% of the time they can literally be any value whatsoever. It's mind-boggli
    • by LWATCDR (28044)
      Sounds like they should also add redundant GPS systems. Three with a hot back up would be a good start.
  • well (Score:4, Funny)

    by flynt (248848) on Wednesday August 22, 2007 @03:12PM (#20322189)
    It took me a while to get the hang of rocket jumping, too. Keep at it!
  • So (Score:3, Funny)

    by CaffeineAddict2001 (518485) on Wednesday August 22, 2007 @03:22PM (#20322297)
    When can we buy parts of the wreckage on ebay?
    • Well I say that and it appears they no longer sell armadillo droppings. At least I cant find it on their page anywhere. Back in the beginnings, they sold scraps/broken stuff as "Armadillo Droppings" for a donation to their cause, small donations got smaller droppings, like nuts/bolts/etc, larger donations got larger parts. I guess they discontinued so they could concentrate more on the actual research/dev stuff (if not for just legal reasons, you know, selling rocket parts to lybian nationalists might not l
  • by 8127972 (73495) on Wednesday August 22, 2007 @03:24PM (#20322317)
    He should fly the rocket from a first person perspective.
  • X-Prize Cup (Score:4, Insightful)

    by pavon (30274) on Wednesday August 22, 2007 @03:42PM (#20322485)
    Ahh, I'm bummed out now. I was really looking forward to seeing them at the X-Prize Cup in October. They were expected to claim the prize (for the level 1 lander challenge), as they had already completed flights matching the profile on their own, and just had to repeat it a the cup for it to be official. I don't know if they'll have enough time to rebuild the craft in time for the event.
    • First, Armadillo is not out of the race. Second, there are 8 other in this. One is new shepard who keeps VERY quiet. I believe that this year, there will be a winner.
    • Yeah, after RTFA and John's forum post, it appears that they only crashed their smaller unit and their larger craft is still fine. I also hadn't heard that they were shooting for the level 2 with pixel. That should be interesting.
    • by peacefinder (469349) <[alan.dewitt] [at] [gmail.com]> on Wednesday August 22, 2007 @04:15PM (#20322813) Journal
      Texel was one of two essentially identical vehicles that Armadillo put together last year for the Lunar Lander Challenge. The other is Pixel, which is the one they actually flew last year (and that had a good shot at winning) at the LLC level 1 event. Pixel is still flightworthy. This crash of Texel doesn't take them out of the LLC race, although it will lower their chances of success; it is going to make them much more cautious about banging Pixel up ahead of the next LLC competition and therefore they'll get flight less testing in.

      They're also working on a set of new vehicles they call Modules, of which I gather they have one essentially complete and five in production.
    • by FleaPlus (6935)
      Ahh, I'm bummed out now. I was really looking forward to seeing them at the X-Prize Cup in October. They were expected to claim the prize (for the level 1 lander challenge), as they had already completed flights matching the profile on their own, and just had to repeat it a the cup for it to be official. I don't know if they'll have enough time to rebuild the craft in time for the event.

      That was just one of their vehicles. They still have Pixel. From Carmack's post:

      We still have Pixel and Module 1 in flyable shape at the shop, so this doesn't have a critical impact on us, but it does change our testing plans for the next two months before the X-Prize Cup. We are cancelling the untethered 180 second flights for Pixel at OKSP. We will plan on doing two sets of back-to-back 180 flights under tether, but if we are going to risk a crash, it might as well be for the money at XPC now that we don't have a backup. We are going to finish up Module 2 in the next couple weeks so we have a backup for level 1. Modules 3 through 5 should also be at least frame constructed by XPC, but whether we get them wired and tested will depend on how our flight testing goes. If we manage to destroy a module in the next two months, we can crunch hard and get an extra one put together if necessary.

  • Harsh (Score:5, Insightful)

    by Hawthorne01 (575586) on Wednesday August 22, 2007 @03:57PM (#20322627)

    I have no opinion on Carmack one way or another, but tagging this story with 'haha' and 'hesnorocketscientist' seems a tad mean.

    So he's a game designer dabbling in space exploration. It's not like he ran a bicycle shop [wikipedia.org] or something. Now *there's* a logical starting point for a career in aeronautics!

    • Re:Harsh (Score:4, Informative)

      by SCHecklerX (229973) <thecaptain@captaincodo.net> on Wednesday August 22, 2007 @04:28PM (#20322957) Homepage

      So he's a game designer dabbling in space exploration. It's not like he ran a bicycle shop or something. Now *there's* a logical starting point for a career in aeronautics!


      Actually, it is. Bikes and airframes are VERY similar. You are trying to get a very strong structure with as little weight as possible. With a bike, as with an airplane, you can't just slap a factor of safety of 9 on the thing. You have to really design it, and pay attention to materials science. (Hint: Bikes, like planes, take advantage of lightweight aluminum alloys, carbon fiber, high torsional rigidity, etc).

      Then there is knowing how to use the right materials in the right places for minimum cost/weight, or for rigidity / flex.

      Today's bikes are what they are mostly from Aerospace research.
    • Re:Harsh (Score:5, Insightful)

      by Reality Master 101 (179095) <RealityMaster101 ... NBSDom minus bsd> on Wednesday August 22, 2007 @05:20PM (#20323407) Homepage Journal

      I have no opinion on Carmack one way or another, but tagging this story with 'haha' and 'hesnorocketscientist' seems a tad mean.

      I've noticed that Carmack gets a lot of flack whenever Armadillo stumbles, and it's an interesting psychological phenomena. You'd think that especially on Slashdot, there would be a lot of people who like seeing smart people succeed, but in Carmack's case, there seems to be a lot of resentment about a "mere" video game programmer daring to learn something like rocket science. Not only learn about, but actually be *serious* about it! And doing it without any sort of engineering degree! The gall!

      This seems to be especially true of amny "real" engineers, who seem jealous that an outsider with money is trying to do what they can't seem to do, which is produce very low cost access to space. "Yeah, if I had Carmack's money, I could do what he's doing better than he could do it..."

      Never mind that Armadillo is one of only a few VTVL ships to actually fly.

      Carmack is an incredibly smart guy, and he's not given near enough credit for raw intelligence, rather than just being a good game hacker.

      • Re: (Score:3, Interesting)

        by DerekLyons (302214)

        I've noticed that Carmack gets a lot of flack whenever Armadillo stumbles, and it's an interesting psychological phenomena. You'd think that especially on Slashdot, there would be a lot of people who like seeing smart people succeed, but in Carmack's case, there seems to be a lot of resentment about a "mere" video game programmer daring to learn something like rocket science.

        No, the resentment comes because he's largely the Keystone Kops of the alt.Space community - constantly blundering about and making b

        • Re: (Score:3, Interesting)

          by bentcd (690786)

          He suffered a lot of failures early on because he didn't use aerospace grade wiring harnesses ("They are expensive and probably overengineered" was his reason as best I can recall) for just one example.

          For this to be something worthy of criticism, we need to know how many other calls of the type "X is too expensive and probably overengineered" he has made and which turned out to be true. If Carmack had successfully debunked 10 different high-cost items as unnecessarily expensive and found that aerospace harness is the one exception, then that is an excellent result and probably worth a few failures to figure out. It is not reasonable to expect that he should have been able to reason his way into such kno

  • by klossner (733867) on Wednesday August 22, 2007 @04:07PM (#20322727)
    But the touchdown did have a big enough effect to jostle the onboard GPS unit that Texel relied on to track its motion.

    Why would a candidate for a mock lunar lander be designed to depend on GPS? There won't be GPS service on the moon in the foreseeable future.

  • Minor nitpick -- this was NOT an X-prize competition (Rutan already won X-prize last year). This was a NASA sponsored competition for design of a lander.
  • UAC (Score:5, Funny)

    by samwh (921444) on Wednesday August 22, 2007 @04:17PM (#20322843)
    Shame, I was already to invest in his new company, dubbed the "Union Aerospace Corporation"
  • The tagging system on Slashdot is getting really pathetic. What kind of jerkoff tags this posting "haha". You think you can do better?
  • Cover the basics (Score:5, Insightful)

    by Dan East (318230) on Wednesday August 22, 2007 @04:24PM (#20322903) Homepage Journal
    I'm hesitant to criticize a group that is breaking so much new ground so quickly, but this sounds like some really amateurish mistakes when it comes to electrical engineering. Basically they added new sensors to detect when the craft impacts the ground. The computer monitoring the sensors was expecting a signal of a certain strength to indicate it had touched the ground, however the value the computer was expecting was higher than what the sensors could physically produce. So it sounds like they either engineered the electronics wrong making it impossible for the sensor to produce a meaningful response, or they misread the sensor datasheet which resulted in flawed software.

    Now it's one thing to make an engineering mistake, but it couldn't have taken them an hour to rig up a simple test rig they that they could drop onto the ground, or tap with a mallet, or something similarly simple, to see if the computer could register a landing.

    I just can't imagine strapping something new onto an entire rocket assembly, going to all the risk and expense to actually launch the thing and fly it around, hoping that all the new circuitry and software will work perfectly the first time.

    It makes me wonder about the whole process NASA has in place with these contests. Even if a craft can meet various flight goals, does it result in anything of worth to NASA? For example, take a piece of software. Say there is this program that really does something impressive (game engines come to mind). So you take a look at the source, and find it is a total and complete mess. Maybe it is full of memory leaks and other bugs, so it just can perform a specific task right, but given other scenarios it crashes. Maybe the code is insecure, or is not scalable, or cannot be extended, or is not maintainable, or is not portable to other platforms. Any of those things could practically render the sources useless. But yet the program does a specific task and does it really well. For some reason I feel that NASA is going to end up with crafts with similar engineering caveats.

    Dan East
    • by ak_hepcat (468765) <leif AT denali DOT net> on Wednesday August 22, 2007 @04:40PM (#20323087) Homepage Journal
      Yes, because NASA never installed any sensors backwards, thus never indicating when to pop open a drag 'chute.

      Not that I'm not a fan of NASA. I am. I own the Space Shuttle Operators Manual, and when I was 11 (when I got it) I probably
      could have flown the shuttle, or at least co-piloted that darn thing.

      Point is, mistakes happen. That's fine. What's great about Carmak and co. is that they tend to not only admit, but they also
      learn from them. Because only half the fun in building rockets is watching 'em blow up.
    • Re:Cover the basics (Score:4, Informative)

      by hxnwix (652290) on Wednesday August 22, 2007 @06:20PM (#20323865) Journal
      NASA makes mistakes that fall under your "amateurish" rubric. EG, burning up a Mars probe due to unit conversion errors.

      Now it's one thing to make an engineering mistake, but it couldn't have taken them an hour to rig up a simple test rig they that they could drop onto the ground, or tap with a mallet, or something similarly simple, to see if the computer could register a landing.
      In hindsight, yes, I'm sure Armadillo wishes that out of the nearly infinite variety of conceivable tests, they had performed this particular one. Nonetheless, they are operating on a shoestring budget and producing impressive results.

      Even if a craft can meet various flight goals, does it result in anything of worth to NASA?
      According to Wikipedia, "the Challenge offers a series of prizes for the teams that launch a VTVL rocket that achieves the total delta-v that would be equivalent to those needed for a vehicle to move between lunar orbit and the lunar surface." This is something that NASA has not achieved and that would be of immense scientific and commercial value.
  • FYP (Score:3, Funny)

    by zero1101 (444838) on Thursday August 23, 2007 @08:12AM (#20329097) Homepage
    "Crap, it's going to fly into the crane, I need to kill it," Carmack recalls thinking. He fired his railgun into the vehicle several times before grabbing a nearby quad damage and finishing it off with a rocket. "It made a fireball that would make any Hollywood movie proud," Carmack says.

    YES!

CCI Power 6/40: one board, a megabyte of cache, and an attitude...

Working...