Slashdot Mirror


San Diego's Fireworks Show Over In 15 Seconds

First time accepted submitter fotoguzzi writes "Garden State Fireworks is investigating how the entire Fourth of July show was launched after a signal was sent to the barges that would set the timing for the rest of the show after the introduction. Can anyone suggest how such a trivial step could go so disastrously wrong?" It's not the first time such a thing has happened, either.

7 of 241 comments (clear)

  1. Re:You shouldn't have made her pay for parking (n/ by Anonymous Coward · · Score: 5, Informative
    Ugh.

    Don't do that. No one reads the comment subjects. When you stick the whole message there, you just look like an idiot.

  2. Lots of WTF in that story by slashmydots · · Score: 5, Informative

    San Diego fireworks show exploded in 15 seconds, ruining show

    Well.....

    Best part about #bigbayboom fail is that EVERYONE has always wondered what would happen if all fireworks went off at once," tweeted @richandcreamy.

    There we go. That's more accurate, lol. I think "ruining the show" is a bit harsh :-P

    By the way...

    "I waited 3hrs in the cold and payd $12 for parking & got one little explosion?" tweeted @aj521z.

    What ****ing planet is this person from?! It is NOT COLD in San Diego at the moment at any time of day.

  3. Re:Wasn't there... by Adult+film+producer · · Score: 5, Informative
  4. Re:Testing Circuit Failure? by PPH · · Score: 5, Informative

    Electric matches on the circuits take 5 milliamps to ignite them.

    That's a bit low (see: http://www.pyromate.com/Basics-of-Electrical-Firing.htm). 5 mA sounds like the test current.

    It's going to be an interesting investigation. Most modern pyrotechnic controllers incorporate a shorting system to keep the squibs from being fired inadvertently by static electricity or single point control failures. To fire each circuit, the safety shunt must be removed and then the firing voltage applied. That's two failures at the lower level of the controller. And on every circuit simultaneously. I doubt it.

    From the video, it appears that the fault was common to three separate sites. They almost certainly used (at least) one controller at each location, tied together through somee communications network to a central control unit handling the timing. My money is on a software failure at that central point.

    --
    Have gnu, will travel.
  5. Re:Wasn't there... by milbournosphere · · Score: 5, Informative

    I was, up on the hill at USD. The sound-wave alone was AWESOME. Probably far more memorable than the actual show would have been.

  6. Fire One by smurd · · Score: 5, Informative

    A am a pryotechnician that works exclusively with computer fired shows. From what I'm hearing on the mailing lists so far, they were using the Fire One controller. We also use them (we have over 100 modules at $795 each). I haven't been involved in the "Loading" of the show into the embedded controller for the past few years, but I was called into action about 3 years ago when we had the same problem with our "semi automatic" shows (press a button for each event). I found there was an additional step when downloading the show from the PC to the firing controller called "Assign Delays" that had to be manually entered when loading. Without that step, all shells for each event fired immediately. I don't know if Fire One ever fixed it because it's now part of our written checklist for loading and we haven't had a problem since, and Fire One is notorious for fixing a problem with one customer, updating the firrmware but not telling the rest of their customer base that there is an update.

    If you are using Fire One, you can thank me for the new Line receivers in the new modules, I had to go to the plant and show them the problem.

    1. Re:Fire One by smurd · · Score: 5, Informative

      Yeah, Don't get my above rant wrong, Fire One is the best system, really the only system out there for firing large shows. I use all of the others, but they are the only ones out there that can handle large pyromusicals (I.E. > 2000 cues).

      It has more then enough juice to fire an entire module (32 cues) at once, and that is a good thing for mine fronts, set peices etc... We need and want that feature.

      I'll be back at the magazines tomorrow and want to run a test - I'm thinking if you either forget to assign delays, or assign them twice for a fully scripted show, you will have the same result.

      As far as I know, there are no commercially available products out there that will let you test ematches with firepower on (the Capacitive Discharge circuit) powered, so I'm pretty sure it's not a testing issue.

      Even though Garden State is a competitor, I feel for those guys, We've had our share of learning curves too.