Stories
Slash Boxes
Comments

SoylentNews is people

SoylentNews is powered by your submissions, so send in your scoop. Only 16 submissions in the queue.
posted by hubie on Monday December 01, @09:26PM   Printer-friendly
from the details-matter-even-really-little-details dept.

The root cause of the collapse of Baltimore's Francis Scott Key Bridge when hit by container ship Dali has been identified. It was the wrong placement, by a few millimeters, of the label on one wire. As usual, the National Transportation Safety Board has taken their time and done a detailed investigation--summarized in this short video
      https://www.youtube.com/watch?app=desktop&v=bu7PJoxaMZg

tl;dr - the wire was not completely inserted into a terminal block, due to the wire label wrapped over the ferrule. Over time the connection became intermittent and eventually shut off power on the ship...after which it drifted into the bridge. Of course there were additional contributing problems as well.

The YT video comments include some more interesting details.

[Ed. note: For those not inclined to watch the YouTube video, the narrative summary of the video is listed in the spoiler below.]

1. The Dali electrical system distributes power and control signals throughout the vessel.
2. The control circuits contain hundreds of terminal blocks that organize thousands of wires.
3. The wires on the Dali were terminated with metal sleeves called ferrules that allowed for easier assembly into the terminal blocks.
4. Each wire was identified with a labeling band.
5. This image shows several terminal blocks on the Dali with wires connected.
6. To assemble a wire into a terminal block, a tool inserted into a side port opens a spring clamp, which allows the wire's ferrule to slide into place.
7. Removing the tool closes the spring clamp, securing the ferrule firmly against the terminal block's internal conductor bar.
8. Labeling bands identify wires and are typically positioned on the wire insulation.
9. However, many labeling bands on the Dali wires were placed partially on the ferrules, which increased the ferrules' overall circumference.
10. As a result, during vessel construction, some of the ferrules could not be fully inserted in the terminal blocks, including the ferrule on wire 1 from Terminal Block 381.
11. On that wire, the labeling band prevented full insertion of the ferrule, so the spring clamp gripped only the ferrule's tip, resulting in an inadequate connection.
12. Due to this unstable connection, over time the ferrule on wire 1 slipped out of the spring clamp to rest atop the spring clamp face, resulting in a precarious electrical connection.
13. When a gap occurred between the ferrule and the spring clamp face, the electrical circuit was interrupted, leading to a blackout on the Dali.

Original Submission

 
This discussion was created by hubie (1068) for logged-in users only, but now has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by krishnoid on Tuesday December 02, @03:03AM (6 children)

    by krishnoid (1156) on Tuesday December 02, @03:03AM (#1425565)

    The video's only two minutes and they included a *very* clear animation of the fault, error, and failure [wiley.com]. I ran into something comparable when someone was trying to crimp an RJ-45 end onto solid-core [wikipedia.org] category 5 plenum cable rather than stranded patch cable, and it produced intermittent failures until I asked him to punch down the solid wire into a RJ-45 socket so I could plug a patch cable into it.

    I suspect a lot of electrical issues on newer installations come down to incompletely stripped/too-short ends that go into the terminations. People will screw/crimp down a connection properly and wind a wire around the post, because they involve physical force and clear indications of something wrapping around something else. But electricity flowing through enough length of a wire that's contacting the metal of a terminal post? That's magic, from a mammalian evolutionary standpoint.

    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 3, Interesting) by RS3 on Tuesday December 02, @03:23AM (5 children)

    by RS3 (6367) on Tuesday December 02, @03:23AM (#1425567)

    I'm not 100% authoritative so I'm not sure, but I've heard, and been with people who were authoritative, who asserted there are RJ-45 connectors for stranded, and a different type of connector for solid wire. In fact I have a coworker who is absolutely knowledgeable and I'll ask him. Your solution is (obviously) the correct one.
     

    • (Score: 3, Informative) by VLM on Tuesday December 02, @02:50PM (4 children)

      by VLM (445) Subscriber Badge on Tuesday December 02, @02:50PM (#1425596)

      Like this for stranded about $36 for 50:

      https://www.startech.com/en-us/cables/crj45c6str50 [startech.com]

      vs this for solid about $43 for 50:

      https://www.startech.com/en-us/cables/crj45sol50pk [startech.com]

      The little pokie things that pierce the cable insulation are different.

      I used to work in the telecom biz at multiple places "a long time ago" and its a matter of statistics; the failure rate couldn't be more than 2 times higher with the "wrong" type. If you test all your cables with a scanner and they pass and the crimp on the jacket is good it'll never work free because the jacket crimp is holding it in place. However I wouldn't use/reuse solid core as a jumper on a regular basis that'll eventually work loose.

      I'm struggling to think of a _good_ reason to crimp solid core. You're always better off terminating solid core on a female connector and stranded on a male connector as is industry standards. You'll always regret doing something else, even if it was "emergency" or "temporary"

      • (Score: 2) by krishnoid on Tuesday December 02, @06:10PM (2 children)

        by krishnoid (1156) on Tuesday December 02, @06:10PM (#1425611)

        A _common_ enough reason among people unfamiliar with it is that they have one or more boxes of solid core plenum that spool out conveniently sitting around during construction, and the crimp ends and tool are easy to get at Home Depot and cheaper (and faster) than the female ends and punch tool.

        • (Score: 2) by VLM on Tuesday December 02, @08:30PM

          by VLM (445) Subscriber Badge on Tuesday December 02, @08:30PM (#1425627)

          Yeah I could see that if someone needed a long jumper and had a spool of cable laying around... It probably makes sense if its a temporary install, or "its only temporary.... unless it works..." situation. The only other justification I can come up with is extreme physical space limitations like an IP webcam on top of a flag pole type of thing, just barely enough space to shove the cable in there and call it good. There are more professional ways to install stuff like that but there are special situations sometimes where is just has to be forced to fit...

          Labor costs and Lost time costs during the inevitable intermittent outages make it a hard sell vs the cheaper in the long run "just buy the correct parts and do it the correct way".

        • (Score: 3, Funny) by RS3 on Wednesday December 03, @04:26AM

          by RS3 (6367) on Wednesday December 03, @04:26AM (#1425670)

          Guilty.

      • (Score: 2) by RS3 on Wednesday December 03, @04:35AM

        by RS3 (6367) on Wednesday December 03, @04:35AM (#1425671)

        The little pokie things that pierce the cable insulation are different.

        Yeah, I just don't know the difference, but I'll do some research. I remember one guy I was with somewhere- he looked at some RJ-45 connectors and he commented that they were the wrong ones for whichever the wire was, despite the bag's label. IE, they were new but somehow incorrectly labeled.

        If you test all your cables with a scanner and they pass and the crimp on the jacket is good it'll never work free because the jacket crimp is holding it in place. However I wouldn't use/reuse solid core as a jumper on a regular basis that'll eventually work loose.

        I'm struggling to think of a _good_ reason to crimp solid core. You're always better off terminating solid core on a female connector and stranded on a male connector as is industry standards. You'll always regret doing something else, even if it was "emergency" or "temporary"

        Extremely good advice, thank you. I guess I don't feel too bad using solid wire in correct connectors, but only if it's not getting moved, vibration, etc. But I'm going to make a point of doing them correctly- solid only in "V" IDC female.

        Hmmm, I'm glutton for punishment enough to solder a foot or two of stranded onto a long solid...