June 17, 2017

Mapping the ACX Crystal's collision with the USS Fitzgerald using publicly available info

USS Fitzgerald (MMSI:338839000 Callsign:NFTZ)
(2017.06.17 edit)
Per the USNavy's 7th Fleet public affairs office;
USS Fitzgerald (DDG 62) was involved in a collision with a merchant vessel at approximately 2:30 a.m. local time, June 17, while operating about 56 nautical miles southwest of Yokosuka, Japan.

The merchant vessel was the Filipino-flagged ACX Crystal container-ship (IMO:9360611) and she did have her AIS transponder on at the time of the incident.

From the news footage below you'll notice an area of severe damage which looks to me to be from an impact at a perpendicular angle, and not a grazing strike, since there is no scraping or dragging down the length of the USS Fitzgerald. I'm not suggesting the impact was deliberate, only that the vessels would have deflected if they had hit with a glancing strike, where to me it looks like the bow of the Crystal embedded itself for a short period in the USS Fitzgerald. I originally thought the USS Fitzgerald was stationary before the impact, but I've since changed my mind, since I've been told that there would be no operational reason to be stationary near an area of high traffic, on a moonless night. Valid point. Then if she wasn't stationary, why was she crossing the path of the shipping lane and how didn't they notice the 30,000 Ton ship on a collision course with them?



Here is the MarineTraffic.com animation of the entire collision:


I have provided a Google Fusion map below, and linked to the original source data that might be useful to you if you want to do a deeper investigation into the original story yourself.



The data looks a little confusing, so watch the video to get an idea of how the strike happened. The colour coding is to show the speeds of the ACX Crystal when moving. Red is almost stopped, Green is full speed ahead.

From a discussion with JJ I think the UTC/JST conversion may have been messed up, or the reported time by the US Navy and Japanese Coast Guard was not representative of the time of the collision, only when the collision was reported.  I doubt anyone in the US military would mix up time zones, since they're very accurate regarding time. The US Navy press release reports the incident happened at 2:30am "LOCAL TIME", which is Japan Standard Time or JST, and which is UTC+9.  The AIS data I scraped from MarineTraffic.com shows accurate to-the-second (or less) data, so that is what I can rely on for accuracy. Could the ACX Crystal have hit the USS Fitzgerald at full speed just before 16:30Z, rather than ~17:30Z as the US Navy said in their press release? I think so.  With that in mind, watch the video again.  Did the ACX Crystal strike the USS Fitzgerald while on a 70 degree course before 16:30Z, then while on autopilot, correct itself after the USS Fitzgerald was knocked free?  If so, it took another hour for the crew to figure out what happened, turn the ACX Crystal around, and return to the USS Fitzgerald - it's unclear if they even knew what they struck.  JJ suggested maybe the time the accident was called in was ~2:30am JST, but the strike had happened earlier. This makes significant sense to me, and explains the "U turn" they performed, especially if you realize the impact was one 30 minutes before the u-turn.

AIS Data



Timestamp (UTC)SourceSpeed (kn)LatitudeLongitudeCourse
2017-06-16 16:19 (UTC)Terr-AIS18.134.5037139.00869
2017-06-16 16:21 (UTC)Terr-AIS18.434.50739139.020870
2017-06-16 16:24 (UTC)Terr-AIS18.434.51177139.035870
2017-06-16 16:27 (UTC)Terr-AIS18.534.51718139.05470
2017-06-16 16:30 (UTC)Terr-AIS17.334.52216139.072388
2017-06-16 16:33 (UTC)Terr-AIS11.234.51329139.0761135
2017-06-16 16:36 (UTC)Terr-AIS14.634.50962139.0878118
2017-06-16 16:38 (UTC)Terr-AIS13.134.5119139.094441
2017-06-16 16:40 (UTC)Terr-AIS15.334.51949139.104370
2017-06-16 16:43 (UTC)Terr-AIS15.234.52352139.114556
2017-06-16 16:46 (UTC)Terr-AIS15.434.52807139.122656

Those who know my blog will not be disapointed; of course I want to show you the minute-by-minute account of the ACX Crystal's journey and try and tease out what we can from it. Above you have the AIS data from what I believe covers the entire horrific event, and the ACX Crystal leaving the scene of the collision. Yes. LEAVING the scene, only to return an hour later. I'll get back to that.  I believe this shows that 1) nobody was on the bridge of the Crystal and 2) "Iron Mike" was in controls for a full 15min *after* the collision; see item 1.  There have been no reports that any distress call went out until after the ACX Crystal came back at 17:30Z, an hour after the collision.

Some people hate analogies, but here's one anyway.  If you were in the driver seat of a self-driving car, hit another car, and your self-driving car kept driving along the road... how long would it take you to hit the off button?  Well, it took the crew of the Crystal, who I'm positive were not on the bridge, 15 minutes to find the autopilot off button.  Either they were very disoriented by the impact, or they weren't on the bridge to begin with, and had to get up to the bridge in order to shut down the autopilot.

16:27Z
All is well, the ship is on a 70° course, sailing at a fast 18.5kn.

16:30Z
Course has changed +18°, speed dropped slightly, 1.2kn. The collision has likely already happened

16:33Z
Course is now +65° off original, and speed is down to 11.2kn.  Something is clearly wrong, the collision has likely happened, and the 30,000ton container ship has been spun 65° off course by the impact or thrust of the USS Fitzgerald's engines.

16:36Z
It gets weirder. Speed is back up to 14.6kn, and the ship is coming back on course, now "only" +48° off her original course.  Did the ACX Crystal spear the USS Fitzgerald, and just shake it off? That's right, after smashing into another ship, the 30,000 ton container ship is swinging around and increasing speed.  If anyone was at the helm, this would not be happening.

16:38Z
ACX Crystal has now swung around in 2 minutes to a course of 41°, -22° off the original course, and is seemingly trying to correct its course toward her original destination.  Almost like nobody is at the helm, and the ship is being controlled by Iron Mike (the autopilot)

16:40Z
Course corrected, back on a heading of 70°, speed is up to 15.3kn, and the ship is steaming away from the accident. It's been ten minutes since the collision, and nobody has disengaged the autopilot yet.  How do I know? Because they haven't reduced speed or turned around.

16:43Z, 16:46Z
Course corrections, but still no slowing down, now up to 15.4kn. How far are the crew quarters from the bridge on that ship?

Timestamp (UTC)SourceSpeed (kn)LatitudeLongitudeCourse
2017-06-16 16:49 (UTC)Terr-AIS1434.53545139.14373
2017-06-16 16:52 (UTC)Terr-AIS12.934.53923139.157270
2017-06-16 16:55 (UTC)Terr-AIS12.234.54282139.16862
2017-06-16 16:58 (UTC)Terr-AIS12.334.5469139.177863
2017-06-16 17:00 (UTC)Terr-AIS12.434.55079139.187162
2017-06-16 17:03 (UTC)Terr-AIS11.934.55532139.197664
2017-06-16 17:06 (UTC)Terr-AIS7.634.56187139.199305
2017-06-16 17:09 (UTC)Terr-AIS9.434.56249139.1927263
2017-06-16 17:11 (UTC)Terr-AIS1034.56231139.1911261
2017-06-16 17:14 (UTC)Terr-AIS12.434.5595139.172260
2017-06-16 17:18 (UTC)Terr-AIS12.934.55806139.1623260
2017-06-16 17:20 (UTC)Terr-AIS13.734.55479139.1448256
2017-06-16 17:24 (UTC)Terr-AIS14.134.5519139.1302255
2017-06-16 17:26 (UTC)Terr-AIS14.434.54954139.12254
2017-06-16 17:29 (UTC)Terr-AIS14.634.54608139.1029259
2017-06-16 17:33 (UTC)Terr-AIS1334.54381139.0875260
2017-06-16 17:35 (UTC)Terr-AIS11.734.54243139.0776259
2017-06-16 17:38 (UTC)Terr-AIS9.934.53842139.0665243

17:03Z
While the ship has slowed to just under 13kn, it still hasn't turned around.

17:06Z
The Crystal is turning hard and has slowed somewhat.

17:09Z
The Crystal has turned completely around and is doing ~13kn toward the location where they hit "something".  I don't think anyone aboard the ACX Crystal knows by this point what they hit. There are no reports that any distress calls by this time.  There has been no suggestion they called the Japanese Coast Guard by this time for any information either. What did they think they hit?

17:38Z
Over an hour after the initial collision. The US Navy reported the accident happened around this time. I believe what they meant was the distress call was received by the Japanese Coast Guard around this time, because it would be unthinkable that you, a 30,000 ton container ship, would collide with another ship, and not mention it to the Coast Guard, for an hour?


Timestamp (UTC)SourceSpeed (kn)LatitudeLongitudeCourse
2017-06-16 17:52 (UTC)Terr-AIS2.934.52554139.055590
2017-06-16 17:56 (UTC)Terr-AIS3.234.52657139.058759
2017-06-16 18:00 (UTC)Terr-AIS3.234.52952139.0609354
2017-06-16 18:02 (UTC)Terr-AIS434.53196139.0605352
2017-06-16 18:06 (UTC)Terr-AIS434.53695139.060212
2017-06-16 18:11 (UTC)Terr-AIS3.934.54041139.063849
2017-06-16 18:15 (UTC)Terr-AIS3.934.5432139.06839
2017-06-16 18:18 (UTC)Terr-AIS3.734.54664139.0684330
2017-06-16 18:23 (UTC)Terr-AIS4.834.54985139.0638309
2017-06-16 18:25 (UTC)Terr-AIS5.234.55213139.0606305
2017-06-16 18:28 (UTC)Terr-AIS4.934.55527139.0573339
2017-06-16 18:36 (UTC)Terr-AIS134.56038139.0575
2017-06-16 18:41 (UTC)Terr-AIS0.434.56104139.0567296
2017-06-16 18:43 (UTC)Terr-AIS0.534.56106139.0564266
2017-06-16 18:47 (UTC)Terr-AIS0.934.561139.0554270
2017-06-16 18:57 (UTC)Terr-AIS2.434.5608139.055692
2017-06-16 19:00 (UTC)Terr-AIS3.434.56094139.058384
2017-06-16 19:03 (UTC)Terr-AIS434.5612139.061882

17:52Z onward
There is a 14 minute gap that I can't explain, did they turn off AIS?
From this time forward, regular updates keep coming in at 2-3min intervals.  Speed and course changes seem to suggest they were looking for survivors or trying to otherwise assist.

In conclusion I have absolutely no idea what happened out there; I'm trying to think through the story with the available evidence and string it together.  By my understanding of the events I believe there was nobody on the bridge of he ACX Crystal at the time of the collision, and for ~15 minutes afterwards showed no signs of being manned with anyone on the bridge as the course auto-corrected, and increased speed, moving way from the impact site, trying to come back up to the original speed. The damage on the bow, below the water line, slowed the ship down from it's original 18kn due to drag.  Knowing when the Japanese Coast Guard was called is a detail that has not been revealed yet, and we all know there's a recording of it out there somewhere.

I'll update as I find out more, or more evidence is revealed.

For updates on the situation follow the 7th Fleet on Twitter, or check their webpage for updates: http://www.c7f.navy.mil/Media/News/

190 comments:

  1. One theory: sea jihad. Third Mate Muhamid Pongo on duty. Realizes from AIS they just passed a USN "kafir infidel" warship. Makes a U-turn on his own. Looks at the CPA plot, uses it in reverse to create a collision. Last few minutes, lies on the ship to ship radio. (Taqiyya/Kitman, "Holy Lying.) "Don't worry captain, I'm turning to port, you turn to starboard." (As he turns to starboard." Next on the USN recording, which, following Orlando Pulse night club protocals will be kept secret from the world: "Allahu Akbar!"

    Just a theory. You come up with a better theory for the container ship's U-turn and collision course.

    ReplyDelete
    Replies
    1. Nope, sounds correct to me

      Delete
    2. That theory doesn't hold water. ACX Crystal could not have hit Fitzgerald with her port bow while making a starboard turn.

      I think Steffan has a pretty good working theory. Crystal and Fitzgerald were probably on very similiar east north east courses with Crystal overtaking on Fitzgeralds starboard side. The angle of impact appears shallow. Most of the damage is below the waterline, so invisible, but the damage to Fitzgeralds superstructure suggests that the impact angle was shallow and that Crystal heeled over into Fitz's superstructure. Crystal's rolled over port hurricane bow supports that scenario as well.

      Delete
    3. Considering​ the Fitzgerald was struck on the starboard side wouldn't the Crystal have been the stando on vessel?

      Delete
    4. No. Rule 13 states that the overtaking vessel (ACX Crystal) shall always stay clear, no exceptions to this rule.

      Delete
    5. What if the Fitzgerald was heading East-Northeast, and the Crystal turned around and headed West to deliberately hit them? Picture it. The Fitzgerald turns to port, and then the Crystal turns to starboard. Impact occurs soon after the Crystal's port side begins to face the Fitzgerald's starboard side. Crystal heels over into Fitz, then glances off to starboard while Fitz goes port. I suppose if that had happened, you would've seen more paint traded along the side of the hull, though.

      Delete
    6. Rule 15 -- Crossing Situation

      INTERNATIONAL

      When two power-driven vessels are crossing so as to involve risk of collision, the vessel which has the other on her own starboard side shall keep out of the way and shall, if the circumstances admit, avoid crossing ahead of the other vessel.

      Delete
    7. For reasons of security, US Navy vessels disable the Tx beacon side of their AIS. We have them set to receive only so we can see what all other civilian vessels see, but we don't broadcast our position for public knowledge. If we did, they'd have our course info on the same website as the ACX Crystal, or at least be capable of doing so, and from a tactical and OPSEC standpoint that would be a major concern.
      So the Crystal never would have seen Fitzgerald on AIS. Sorry bud, that part doesn't hold water.

      Delete
    8. Matt you Theory is more realistic than the author of the article. In restricted water the ship may have been on Iron Mike, But there is no way there was no one on the bridge, that claim would be laughable if it was not so stupid. If the Crystal hit the destroyer perpendicular at 14+ knots the destroyer would have almost been cut in half. The bow above the water line would have been a knife and the damage would be cutting not ripping. If you are on a ship and it hits another ship at a perpendicular angle you know it even if you are asleep. To use the authors anoligy you are asleep in the passenger seat of a self drive car. The car going 50 MPH T bones a smaller car and you sleep through it. And the damage to the Crystal would have been much different. The Crystal's bow would have crunched back on both sides not scraped and torn on the one side. Plus the bublous bow would have been chrushed back a large distance. Plus a perpendicular collision would not have pulled on the anchor of the Crystal. The article is poor fiction writing. The damage to the Destroyer and Crystal could have happened if one ship was trying to overtake the second ship and the his with a course difference of maybe as little as 10 degrees to as much as 30 degrees, mor that that and the damage would have been much more.

      Delete
    9. The cowardly anonymous shit poster sure sounds familiar to me. Is that you TR? Perpendicular in this case is relative. A shallow hit would have deflected. A true perpendicular hit at 90 degrees in the middle of the center of gravity might split the ship in two, and would wrap it to either side of the bow of the Crystal; however, a (say) 70 deg angle, the the front 1/3 (as we see from the damage) would not wrap the ship around both sides of the bow, and while not technically perpendicular/90 degrees, is what I thought I'd conveyed.

      Delete
    10. I hold an unlimited Masters ticket, was OOD in the navy and a ship pilot in two ports and find MOST of the posts here to be highly intelligent and well thought out morons notwithstanding. I think we can all agree that absent Fitz's track it's all still speculative. My Navy side tells me no CO will allow a Jr officer to attempt an overtaking situation making himself the burdened vessel in the first place, no less be in his cabin at the time especially since we all know MOST civilians will be on the ironmike with at best just two people on the bridge. SO, my guess is the Chrystal was overtaking Fitz.These are high ships, the running lights are way up high some say they were not lit, but the lookouts if scanning the horizon and not looking up may have never even seen them..it happens. Possibly the close proximity of a large mass of metal threw the autopilot off and she sheared to port and glanced off the Fitz, if so they would have never seen her coming down on them. The damage to the PORT bow alone says it was glancing and not a crossing type situation. Just my 2 cents, I want to see BOTH tracks.

      Delete
    11. I was a Operations Specialist CIC Watch Supervisor on a FFG from 92-96 and have a couple of comments/questions.

      We tracked all the surface contacts with consoles and if a contact had a closest point of approach of less than 2000 yards we would adjust course or speed to keep that range to greater than 2000 yards. We would not only track the contact with the consoles but start running MO Boards to get a more accurate track of the other ship. The OOD would also start running a MO Board and we would constantly compare the two as well as talking to the lookouts pointing out the direction to look and what lights etc they saw. I dont know how a much more advanced ship like the Fitz would track a surface contact but, I would assume that it would be much more accurate and almost instantaneous. Would a ship like the Fitz have a contact alarm in the tracking system that would alert the CIC watch that there is a problem? If the Fitz was struck at an angle, wouldnt there be a secondary collision along the hulls as the ships momentum pushed the aft ends together (Fitz starboard hull and the Crystals port side). If there was a collision I dont know that we would have called the Coast Guard right away, if the radio room was damaged then comms (sat and HF) could have gone out. First thing we would have done is radio our command/other US ships. With the Fitz starting to flood as the OOD I would have been totally focused on immediate damage control/rescue operations instead of calling for help from the Coast Guard.

      Delete
    12. There is an island nearby and may have hidden the running lights of the container ship. I have sailed offshore and you can lose a large ship in the lights of the land on a moonless night. Not sure if it would have any influence the radars. Likely not.

      Delete
  2. In recent reports one of the flooded spaces on the Fitzgerald was the radio room. Thus Fitzgerald may not have been capable of sending out message. Perhaps Crystal was going to ignore the whole thing until someone else reported accident. After all, if there is no report made, then "Nothing happened."

    ReplyDelete
    Replies
    1. Probably the Captain of the Crystal knew he'd be fired if they knew the bridge wasn't manned, so they waited to confirm they *had* to report it... or that's my cynical take.

      Delete
    2. They have Bridge to Bridge transceivers that would not have been affected by radio being flooded.

      Delete
    3. when sailing in a sea-line, u have at least a mate and a rorsman (deckhand, OS or AB) on the bridge at all time. under some circumstances u even have a lookout at the forepeak. anyhow..... used to be like that when I was in the merchant navy 40 years ago.

      Delete
    4. Power was probably out on Fitzgerald for at least a period of time. They do have battery operated VHF, SATCOM, Cell phone etc. that are there, but they had a lot going on, and it was dark, things were probably thrown around.

      Delete
    5. @Ken So you think that the Fitz had a power outtake and motionless in the middle of a busy shipping lane for quite some time while the captain was still sleeping in his bunk? NOOO WAY.

      Delete
    6. Absolutely not. Power outage after the collision, not before.

      Delete
  3. How would the Fitz not see the Crystal on their radar? Would the Navy fire on a container ship?

    RIP to the lost sailors.

    ReplyDelete
    Replies
    1. And thank you, Steffan, for your reporting. Great information on this tragic incident.

      Delete
    2. That's the big question; how did it happen? I can only see it very remotely as dots on a map, but there's a lot that isn't known and will need to come out of a long inquiry. RIP Sailors.

      Delete
    3. As a former Radarman aboard a destroyer in the early 60's we in CIC would be intent on our radars tracking unknown contacts and be aware of collision courses and report this to the bridge to take evasive action with course and speed changes. Why did this never happened????

      Delete
    4. I don't have any answers as to why the Fitz didn't avoid the collision, but it wouldn't take much inattention to cause this collision.

      I notice that the ACX Crystal seems to have slightly increased speed (from about 17.5kn to about 18.5kn) before its turn from 90 deg to 70 deg. Imagine the Fitz heading 90deg at about 17.5kn (which would likely be her most efficient speed with two engines, two screws). ACX Crystal is seen on a constant bearing, which means one of two things: the ships are on a collision course, or the ships are on an exactly parallel course with same speed. After careful observation, the crew on the Fitz establishes that the ships are both heading 90deg at 17.5kn and there is about a mile and half separation between the ships.

      Now, what would it take for the ship's parallel course to change into a collision course? If one ship changed course and simultaneously increased its speed by the exact right amount, the bearing between the ships would stay unchanged while the distance between the ships closed. If one ship turned 20deg toward the other, it would have to increase its speed by about 6% to keep its bearing constant (as observed from the other ship). 18.5kn is about 6% faster than 17.5kn.

      The crew of the Fitz, if they were just checking the bearing and not carefully observing the slowly closing range of the ACX for about 15 or so minutes, would have missed the impending collision.

      Delete
    5. Good answer, I believe CIC AND the bridge weren't paying as close attention to the range as they should have been. Tracking includes range and bearing readings. As far as I'm concerned this collision should have been avoided. On the old surface search radars (SPS-10) of my time we kept the curser on the contact and the range was read on a dial and the range and bearing was called out to the watch supervisor every few seconds which is passed on to the OOD.

      Delete
    6. I'm an ex SPS-10 tech and as such, know that the SPS-10 cant track vessels that close.. Too much sea scatter, thus the lookouts. I'm thinking that for some unknown reason, the lookouts miscalculated the Crystal course and speed, or, possibly never had her within sight.

      Delete
    7. Thank you all for your knowledgeable thoughts.
      I wondered if the ACX Crystal was too close to be picked up by radar. Seems that FITZ would have known ACX locale at one point (close) but w/o radar on a moonless night, maybe ACX had no lights (?), and even with lookouts it seems possible that a ship 4x your size might be missed in the dark. But, that sounds bad for the Navy to admit.

      Delete
    8. A lot really depends on Fitzgerald's course in comparison to Crystal. Remember there was a course change about 10-15 before the presumed collison. It would take 2-3 fixes after that course change to determine a course change happened. Even every few minutes, there was a lot of traffic they had to cycle through.

      If you're the lookout and there is a large container ship that you have seen, everyone is aware of and isn't a danger, at least initially, at night with lights only, how long before you realize it's actually getting closer and has changed direction. Not easy to identify immediately in the dark.

      It could easily take several minutes or more to identify the course change, and plot it in relation to your own to realize there is a problem. Depends on other traffic they were tracking, and how close attention they were paying to a target they had initially identified as not an issue.

      Delete
    9. If it was that hard to pay attention, I would think the busy channels would have more accidents. Also, a Destroyer class navy ship not identifying a container ship crossing path until impact makes it hard to conceive how it fulfills its missions. Hard to know how redundant safeguards failed (watch, radar, AIB, more).

      Delete
    10. Remember it was dark and moonless by accounts. Seeing a ship during the day and picking up profile and bearing changes is a lot different than night time. Night time, pitch dark with nothing but lights to go by it's hard to judge distances, speeds and bearing changes in those conditions. If the contact was already identified, and deemed safe, it would take something to change their mind and that could have easily been 5-10 minutes into the port turn course change. Lookout or radar, something would have to change to get them to pay more attention. Radar would pick it up possibly before a lookout at night, but it would still need 2-3 fixes to determine that it was an actual course change and not just an adjustment.

      Delete
    11. It's been a long time since I worked on any of them, but the best I remember, the SPS-10 was not that great for really close in work. That's why we had the LN-66.

      Delete
  4. A lot of these third world shipping vessels have ragtag crews from all over the world, most not even speaking the same language. The only ones with any real education being the handful of officers and even there, disciple would be severely lacking.

    There may very well have been no one on the bridge, or perhaps one person, barely trained in their duties and not paying any attention.

    Given the assumption that no one was on the bridge, why the long delay before turning back? While no one aboard the merchant was injured, that's not to say that there was not confusion and panic aboard, perhaps the lighting circuit breakers were blown and the ship plunged into darkness even though the auto-helm and engines were still operating.

    Looks to me like the collision took place, the ship attempted to continue on programmed course while confusion was aboard the merchant. Then they returned to the scene of the incident and stood by for a while until released to continue on into port.

    There does not appear to be any evidence of deliberate ramming or a turn back to a collision course as some have suggested in ignorance.

    Just a blind hit while plodding along on course with no one paying attention, then confusion of an undisciplined crew as they tried to figure out what happened, and finally a return to the scene, remaining on station, and finally released to continue.

    ReplyDelete
    Replies
    1. Yup. I think that the actions of the Crystal show something was very wrong in the bridge; either nobody, or whoever was there were not trained, or they were unable to cut autopilot for some technical reason... I'm betting on nobody there at the moment; we'll see!

      Delete
    2. if the merchant ship's crew is undisciplined and untrained, then the destoyer's highly trained crew with their state-of-the-art warship are sleeping on the job.

      Delete
    3. Having been in the vicinity of one of these events and investigated a couple others, I believe the the ACX was on autopilot with crew on the bridge and the collision happened at the first slow down and hard turn to starboard. After that the ACX autopilot tried to bring it back on coarse and speed. The crew meanwhile was first accessing the condition of the ship and crew. They knew they were involved in a collision, all the while attempting to contact the other party involved. turned off the autopilot turned around (when safe to do so) and returned to the scene of the collision, as required by International law. The tracks explain all of this. Turning a ship that size around safely while moving ballast around and getting back in 25 min seems reasonable. Without the USS Fitz track we will not know weather this was a crossing situation or an overtaking one. I do know that last minute maneuvers can change the angles of impact from the original situation. My speculation: A crossing situation and trying to cross the "T" realized they weren't going to make it and turned to port at the last minute (less then 400yds) changed the angle of impact to glancing rather then T bone. If it was overtaking at what point do you take action to avoid collision, based on the damage and loss of life, you don't what to get hit, so you can be right in a court of maritime law. Either way by Navy standards this is the USS Fitz's fault and should have been avoided. Heads will roll.

      Delete
    4. With regard to the Crystal's continuing at 17 knots for 10 minutes... on a merchant ship the engine is often run at an economical speed with heavy fuel. The engineers customarily ask for 10 minutes notice to reduce rpm. Something to do with block temperatures. They can, of course, slow down more quickly in an emergency, which this was; but we don't know if they were asked to initially. Also, if they were using heavy fuel, the engineroom would need to convert over to diesel for maneuvering speeds. Stopping the engine while on heavy fuel can gunk up the engine. I have to say I don't know if either of these happened on the Crystal, but they are questions that should be asked. It doesn't seem unusual to me for the ship to have taken that long to return to the scene. They probably weren't able to contact the Fitz to confirm what had happened. At the pilot station, when we have to ask a large ship to make a round turn and come back to the pilot station it often takes them close to an hour - in daylight, no confusion.

      Delete
  5. Great analysis on the USS Fitzgerald collision, particularly in that the detailed data you present suggest the possible mis-reporting of the "2:30 am local time"

    Regarding that, CNN initially reported 1:30 local time / 1630 UTC, citing Japanese Coast Guard.

    “The two ships collided around 1:30 a.m. local time Saturday in the Pacific Ocean, about 56 nautical miles southwest from the port of Yokosuka and 12 miles off the Izu Peninsula, the Japanese coast guard said.”

    http://www.cnn.com/2017/06/16/politics/us-navy-destroyer-collides-with-merchant-ship-japan/

    Reuters has remarked on the confusion:

    “The U.S. Navy said the collision happened at about 2:30 a.m. local time (1730 GMT), while the Japanese Coast Guard said it took place at 1:30 a.m. local time.”

    ReplyDelete
    Replies
    1. ... sorry, the Reuters link got truncated, here it is:

      http://uk.mobile.reuters.com/article/UKNews1/idUKKBN1972T1



      Delete
    2. Thank you! I think that time difference makes a huge change in the story and narrative of what happened. Not the "why", but what sort of catastrophe it was. It seems like the ship was severely damaged, taking on a lot of water, without comms to shore, and unable to call for help. Hard for the US Navy to say that without having a lot of uncomfortable questions. The kinetic force of a 30,000Ton blunt spear hitting the hull below the water line shouldn't be downplayed; that ship took a torpedo+ sized impact in kinetic, rather than explosive, force.

      Delete
  6. Could this be Chrystal's track to render assistance after the collision? It's going to be very difficult to understand the full situation until the Fitzgerald's track can be over-layed.

    ReplyDelete
    Replies
    1. Having a track for the USS Fitzgerald would be optimal, but I'm sure they won't provide one. The impact almost surely happened at ~16:30Z; after that the Crystal's actions speak for themselves. Still on autopilot for 15 min past the accident, almost a full hour from impact to return, and possibly not calling the JP CG for that time either. The USNavy should correct their latest PR that states the impact happened at 1720Z, which is clearly incorrect, per the AIS data, which can't realistically be wrong/faked.

      Delete
  7. Why did the Crystal break from the standard shipping lanes after the incident?

    ReplyDelete
    Replies
    1. Can you be specific regarding at what time (I provided a whole lot of times and with all the wiggling I'm not sure which path you're referring to)

      Delete
    2. macro charts suggest the tokyo lane travels in a n'easterly direction around 45 inside the island of oshima

      before the incident she appeared to be on the main lane to tokyo which should have put her on the lane west of oshima,

      the time lapse appears to show she was turning north on to that lane just prior to the incident

      but after the incident and after her decision to return, she left the scene and acted as if she was headed for sendai or tomakomai on the outside route south of oshima, before correcting to 360 back to tokyo

      loved your analysis by the way


      Delete
    3. would love to see the speed and headings from 14:00 to 16:00z for the crystal. if crystal had done her best to save fuel in a thin margin industry, she should have turned n'easterly much earlier and avoided a collision course

      Delete
    4. Zoom out on the dynamic Google Fusion Tables Map, and click any of the points for course, location and speed :)

      Delete
    5. I'm not sure that my post went through. Is it possible that the freighter's auto-pilot was hacked?

      Delete
    6. Wow, that was great, but more confounding. acx crystal could/should have turned NE about an hour earlier

      she stayed on the international lane to tomakomai for at least 18k AFTER the normal turn on tokyo lane
      she slowed crossing suruga canyon (understandable) before the turn but sped up on 90 E when she should have made 54 NE
      she waited until she was about to cross a local lane from shimoda to kozushima before turning n'east
      google maps shows at least 0.2 degrees longitude at 34.51 north or 18 k (using http://www.nhc.noaa.gov/gccalc.shtml)
      google earth shows about the same between geographical features and lat/lon (with measuring tool)
      if she was programmed, there was no logical reason to travel an extra 25/30k distance using tomakomai without incident
      after incident she returned to Easterly then turning 360N east of oshima island and staying off lanes until well into the sagami-nada sea, but, what the hell

      why be logical?

      thanks.

      Delete
    7. Malware conceivably could be programmed into autopilots at the factory or through infected software updates. I would expect it to be "zero-day", where all infected autopilots are pre-set to malfunction at the same time. That time could vary according to user-selected set-up options (factory default time zone, GMT, home port time zone, etc.), but I think a lot of vessels with such an infection would have autopilot problems at the same time, if it ever were to happen. Always be ready to switch to hand steering. Hacking to enable remote-takeover of autopilot control seems much less feasible and usually impossible with today's ships and yesterday's technology. But this could be a big danger with unmanned, autonomous ships in the foreseeable future.

      Delete
    8. Great point about malware, the future is ours to save

      Conceivably is a scientific way to explore an inconceivable incident. Just how the hell did this happen. Until we know the truth (if ever) every comment is a concept. It is a bit frustrating that data is so closely held, like someone thinks the common man cannot handle the truth.

      The autopilot concept doesn't make much sense to me, unless and only if the Crystal was asleep at the wheel (iron mike) for the extra hour on the outside lane to tomakomai, burning precious fuel, had some auto warning she was crossing a local lane, woke up and turned north late, straight into a arleigh burke class destroyer.

      Again, if she had programmed the autopilot installed 2008, it should have taken a heading of 54 an hour before she crossed the local lane from shimoda to kozushima, so although a hack is conceivable to me, it doesn't subscribe to occams razor and as you say, malware would have infected more than one vessel

      all the best

      Delete
    9. This comment has been removed by the author.

      Delete
    10. [tinfoil-hat]
      Hacking the autopilot would not be necessary since the GPS signal could be more easily hacked. The GPS signal is a lot weaker than the old LORAN system and as such, it wouldn't take much of a transmitter to give erroneous data.
      [/tinfoil-hat]

      Delete
    11. OK ... that is conceivable ... I still don't get why the Fitz felt she could be a good bow thrust for the crystal ... and opted out of avoidance ... but there i go again, relying on logic in a logic free zone

      Delete
  8. US Surface Navy vet here. I don't see any analysis of the actions of the USN vessel here. With the limited amount of info here combined with what my experience as the OOD of a USN cruiser, this is what I see.
    On the USN ship you have multiple people tracking the paths of all contacts.
    There are standing orders from every CO as to when to contact him should a ship be on track to pass within a couple thousand yards. If it is REALLY close, the CO usually will come up to the bridge.
    Junior Officers have been (ahem) known to prefer to NOT wake the old man if at all possible and will adjust course to stay out of the "notify the Captain" parameters.
    Also, late watches can sometimes slacken and contacts do sometimes sneak up on you - especially if folks are tired after a demanding exercise etc.
    Anyway, the rules of the road state that if a contact is on track to pass in front of you from right to left, they have the "right of way." We would often try to radio the merchant, but often there was no response or a language barrier. We never expected a merchant to change course - unless we had the right of way. For them, time and distance are money, they have a very specific route and don't like to change it.
    My guess, from the location of impact, is that the USN vessel tried to cut IN FRONT of the merchant instead of what would normally happen - pass behind it.
    Major screw-up.
    And if the CO was in his cabin (as appears to be the case) and not on the bridge, he was either notified of the contact and didn't come up (unlikely), or his OOD thought he could handle it with a "hold my beer" judgement call.
    My 2 cents.

    ReplyDelete
    Replies
    1. Same thoughts here. 11 years at sea, 3 as a CO. The fact that the vessels this size got that close without the CO on the bridge says something.

      Delete
    2. Agreed. Retired QMC here, 20 years deployed on ships out of homeports in Far East. Practically impossible for the destroyer to suffer that damage and not have been the give-way vessel. Standing out of Tokyo Wan, they may have been just changing watches from Sea and Anchor/nav detail to normal at-sea watch and this contact somehow got lost in turnover confusion, but several people were not watching what was going on...or several people's watches no longer EXIST. The Navy has been relying more on computers for nav and target solution, combining jobs on the bridge and reducing watch team members. not the FIRST time this has bitten them in the ass. MK 1 MOD 0 eyeballs and common sense have given way to trusting digital readouts. Either this is a simple case of that, or there is a LOT MORE negligence to this story.

      Delete
    3. I am a civilian, and not in the shipping industry. My interest in this case stems from the issue of accountability, and because I am a Filipino. (The Crystal was registered as a Philippine vessel, with Filipino.) This updated NYTimes story reports that the CO of the Fitzgerald was in his stateroom at the time of the collision, and that he will likely be relieved of his command for what may have been a failure in communication aboard his ship. I hope that the actions taken by the civilian ship will be fully investigated as well. Human accountability is important.

      https://www.nytimes.com/2017/06/18/world/asia/navy-uss-fitzgerald-japan.html?hp&action=click&pgtype=Homepage&clickSource=story-heading&module=first-column-region&region=top-news&WT.nav=top-news

      Delete
    4. I concur with QMC Boyles. ACX was likely on autopilot, likely with the bridge unmanned and no 'looking' lookouts. She was also likely an "overtaking vessel" and was seen as that by the Fitz bridge watch standers. Speed difference was small; ~zero bearing drift, sort of an ad hoc convoy. ACX got close, rolled Fitz at about 2 knots faster, got pushed right, disengaged hulls, and the lonely autopilot took several minutes to resume course and turns. After the ACX 'watch stander' awoke, brushed his teeth, and made it to the bridge, he had someone check the bow, turned the ship around, gapped the AIS for 14 minutes in a bid for career retention, and then returned to the near-foundering Fitz. Thucydides was right. (former USN QM1)

      Delete
    5. I see it that way as well, only damage to the boxboat is on the port bow..overtaking situation, dark night. never saw her coming down on them. Sure wish we had Fitz's track though.

      Delete
    6. Long ago an OOD(UW) in an ESSEX Class CVA, in my eyes some/most indications point to an "overhauling" situation by CRYSTAL which places on the shoulders of the CO/Watch Officer (or someone)the obligation to maneuver to avoid. On the other hand, FIRZGERALD's Bridge watch, OOD, JOOD, JOOW, and lookouts, plus the watch (CICWO++) in the OPS Center (CIC to us old guys) had absolute indisputable duty to recognize a potentially dangerous situation and develop alternative for avoiding it. To have failed to call or further insure that the CO was on the Bridge seems a failure likely to curtail the careers of a couple of JOs. As for the CO, to have failed to instill among his watch standers the appropriate reactions, seems pretty damning, one of those "snip off his gold buttons and break his sword" career-ending "What we have here is a failure to communicate!" moments. It doesn't matter whether the collision was entirely due to ACX CRYSTAL's mishandling. The CO and the OOD of FIRZGERALD remain accountable for the damage to the ship and the loss of members of her crew.

      Delete
    7. great analysis... glad I kept reading til someone parsed the USN actions/inactions.

      Delete
    8. Tons of speculation, ounces of information. Just because the auto pilot of the Crystal was on does not mean no one was on the bridge. Merchant ships use the auto pilot to steer about 99% of the time in open water. Modern container ships usually have pretty good crews. The rag-tag crews described by some of the USN commenters are found, if at all, on small trampers. That said, questions to ask involve the bridge layout of the Crystal. Often the conning station is behind a series of connected consoles with lighted displays. The mate might not have had any real night vision. If there were only two persons on watch the lookout would have been on the starboard side (but looking all around, hopefully) and the mate more or less keeping track to port. Radar settings on both ships should be looked at. If everyone was looking at 6 or 12 mile displays, stuff inside a mile would not be so easily seen - and I suspect the warship had some features designed to make them less visible to radar(?). Small target, background lights, night blind mate, no AIS on Fitz... she could have been missed. I can't even speculate what happened on the warship. Is a large bridge team like those found on military vessels a liability in some situations?

      Delete
  9. Resembles the Andrea Dorea/Stockholm collision.

    ReplyDelete
    Replies
    1. Not really. In the 1956 case, the ships were on courses differing by almost 180 degrees, i.e. head-on. Stockholm turned to starboard, as required by colregs, but AD turned to port ... AD got T-boned. Here the course difference was probably about 30 degrees, and the Crystal was probably passing.

      Delete
  10. Hmmmm, how do I say this nicely. In 1994 I was on the USS Puget Sound AD-38 when she ran OVER a large catamaran around 2 am. We blew her whistle (and that damned thing is massive). No response from the catamaran and we went over. We did not turn around. Soooo.... The Puget Sound could not turn on a dime.

    ReplyDelete
  11. This comment has been removed by the author.

    ReplyDelete

  12. Based on the AIS data and plot, I disagree that the timing is off. I think the collision did take place around 1730 UTC, when the reported SOG began dropping after the 1729 AIS hit.

    I can see a container ship hitting some small sailboat and not stopping (as they didn't feel anything). However, not a 6,900 long ton ship.

    Why the 180 degree turn around 1703? Don't know. Perhaps they received a change in destination from corporate? It's definitely not a rare occurrence.

    ReplyDelete
    Replies
    1. So I understand, you're saying the Japanese Coast Guard's statement the collision was at 01:30, and my analysis of the AIS data, is incorrect? That's your perogative, of course, I just think you should look at the data again.

      Delete
    2. With only Iron Mike at the helm, there is no reason for the ship to shop for anything.

      Delete
    3. ...and they DID stop, it just took them a while to turn off Iron Mike, consider their options, and turn around.

      Delete
    4. Sir, the report of the collision was reported by the Crystal "before 2:30" to the JP CG "3rd district". Therefore the Crystal only radioed it in at 2:30, when it got back and found a severely damaged vessel. This supports that the accident happened at 1:30 and Fitz couldn't radio for help itself.
      reference:
      http://www3.nhk.or.jp/news/html/20170617/k10011020981000.html

      Delete



    5. Do you have sourcing for the JCG stating the collision was at 0130? I've only been able to find one article which states

      "The JCG strongly believes that the two ships collided around the time the container ship made an abrupt change of course at around 1:30 a.m." but even that wasn't sourced.

      It's not that I'm not open to all logically possibilities. However, I do place some credibility in the USN statement. I don't think 7th Fleet would go forward with a press conference with prepared remarks without ensuring what they were presenting was accurate.

      The first ~40 degree course change at 1630 could have easily been the Crystal maneuvering for another vessel (any data on other AIS tracks in the area at that time?), then returning back to base course.

      Delete
    6. My google translate says:
      "A message from the Philippine flagged container ship "Collision with a US Navy ship" was reported to the 3rd Regional Coast Guard Headquarters off the coast of Ibaraki, Minamiizu-machi, Shizuoka prefecture, 2:30 am before 2:45 am."

      I would still caution on drawing conclusions based solely on the AIS data. However, given the data thus far, it matches that the collision occurred around 02:30 local time, NOT 01:30. The collision happening an hour earlier (at 01:30) is based solely on the Crystal's course change then. Without the FITZGERALD's track, or knowing if there were any other ships in the area, it's a huge leap to accusing the Crystal of hitting a 8,900 ton ship and just continuing on for a half hour.

      Delete
    7. I wholeheartedly but politely disagree with your assessment.

      I believe you mean it would be *unbelievable* that a captain of a MV would strike *something* knocking the ship off course ~65 degrees, and NOT immediately call it into the CG. I agree. It's wholly and totally negligent, criminally so. I stand by my original assessment in the blog post.

      Impact at 1:30, Iron Mike engaged until 1:45, turn around and get back to Fiz at 2:30, and call it in to the CG.

      "it's a huge leap to accusing the Crystal of hitting a 8,900 ton ship and just continuing on for a half hour"
      No, I think it's grossly negligent and evidence they are going to spend a long time in prison. It's totally outside any reasonable person's actions, but in line with someone who knows they didn't have anyone on the bridge and truly has no clue what they hit, but can't call that into the JP CG.

      Delete
    8. We'll see where the investigation leads.

      However, by your assessment, USN 7th Fleet is mistaken when they've publicly stated that the collision occurred around 02:30 am?

      Delete
    9. Yes sir. It would probably not be popular to openly state the Fitz was totally disabled, right down to emergency comms, and could not call for help itself. It was the Crystal that called in the distress call to the JP CG; no indication the Fitz was able or did send any radio distress call.

      Delete
    10. Conspiracy theories aside, I'm fairly confident the ship was in contact with their Tactical/Operational commander via secure means.

      Delete
    11. I wish I could ask you for more details in that regard, but of course you can't tell me :)
      Let me put this way then. Imagine how bad the damage would it have to be for them to lose all comms.
      It was news to the JP CG at 2:30. If the US Navy knew their ship was dead in the water for an hour, I would have thought they would have engaged the JP CG by then. Does that logic hold water at all?

      Delete
    12. Not as logical as the collision taking place around 1730GMT, as stated by the Navy, and both vessels establishing communications immediately following.

      Delete
    13. Okay, so that throws out my running AIS blip to blip rundown; how do you explain the path the Crystal took from 1:30-2:30 then?

      I know, we'll go in circles until there's an investigation and published info. I hope some info comes out that convinces you the impact was at 1:30. Hopefully the 7th Fleet commander will correct the record publicly soon.

      Delete
    14. Crystal alters course to starboard for another vessel around 0130, then returns to her base course.
      Sometime between then and 0200, they receive orders from their company to turn around and head to a different destination.
      Collision occurs around 0230.

      The above is just as plausible than them getting into a collision at 0130 and keep steaming while the US Navy covers up the real time of the collision.

      Delete
    15. The Japanese media seems to report the correct time of the incident tho, because they're listening to their Japanese sources I presume?
      http://www.asahi.com/ajw/articles/AJ201706180025.html

      Delete
    16. OK, so given our conversation, I went back to the data and looked at it much more closely. A third option exists...

      The last AIS hit at 1627UTC had the Crystal traveling at 18.5kts over ground. The linear distance from that point until AIS has her coming about is little over 8NM. Without knowing the tactical data of the Crystal, it is WELL within the realm of possibility that's how long it took to stop a 39k ton ship going that speed. That much mass traveling at that speed takes a lot of time and energy to slow down.

      So, the third option is that the collision did happen around 1630UTC, and the Crystal began slowing down to come about, and return to the site of the collision (which means the FITZGERALD drifted about 2NM to the north during that time).

      I still contend that the CRYSTAL didn't just ignore a collision, then come back. It wasn't until I actually looked at the distances involved (which the graphics scale can be confusing) that this started to make sense as another explanation.

      Delete
    17. This comment has been removed by the author.

      Delete
    18. Thank you Alex, I hadn't seen this:
      https://www.voanews.com/a/navy-identifies-sailors-killed-in-collision/3906043.html

      Delete
  13. Maybe the pilot of the crystal was asleep (or drunk) on the bridge and knocked around by the impact as to not be coherent enough to realize what had happened. By the time the Captain or other officers got to the bridge, (maybe the door was locked?) grabbed the drunk/wounded pilot it took 15 minutes to control the ship and bring it to a stop to find out what they had hit.

    ReplyDelete
    Replies
    1. I like the colour you added :)
      Yes, something like that, clearly, since they were on autopilot for 15 min after the collision.

      Delete
    2. Considering the size difference, I'm sure there was a "jolt" on Crystal. If it wasn't severe enough on there to cause fire/flooding, no alarms, then they may not have realized what happened. Check the logs, check autopilot, send someone forward to check the bow, and find damage. Listen for distress calls, and don't hear anything. Should we turn back and check or shouldn't we? Ultimately decide to turn around and check finding Fitzgerald.

      If that is the case, what would have happened if they had decided to continue since no one called? How long before Fitzgerald would have been able to ask for help themselves? Crystal would have had a lot of explaining to do on arrival in port with damage to the Bow, especially if a US Navy ship turned up missing.

      This could have been much, much worse.

      Delete
  14. Looking at the courses/changes, about 10-15 min before the abrupt starboard course change (collison?) there was a course change to port (86 to 69) possibly turning into Fitzgerald, and crossing her track earlier than she expected. If that course change was programed into the auto pilot, and if Fitzgerald had marked the track with the old course, they may have thought they would pass astern on their current course (Crystal passing across their bow). They may not have realized that Crystal had turned to port, putting them on a collision course until it was too late.

    If nobody was on the bridge of Crystal (or not trained well/paying attention) they wouldn't have heard or responded to bridge to bridge) With 10-15 min, depending on when Fitzgerald recognized they were turning and the original plotted course had changed. They may not have had much time to react, or until it was too late. Will be interesting to see the investigation and time/track laydown.

    ReplyDelete
    Replies
    1. Agreed. I'm very much looking forward to more information from the investigation. Hopefully it will be public.

      Delete
    2. Ken From looking at the Crystals track and the damage to both ships my SWAG is that the Fitzgerald was overtaking the Crystal. The OOD and CIC should have been plotting the Crystal course. They thought they were safely passing Crystal. If they were plotting a course of 86 degrees and as they drew abreast of the Crystal she changed course to 69 degrees. The course change could have them on a collision course. And the ships would have hit at an angle around 20 degrees. The right side of the Fitzgerald would have been torn up much like it was until the ships parted. And the Crystals left side would received the kind it did including pulling on the anchor out of the hauser. And the bulbous bow may have not received any damage. If the ships hit perpendicular the Fitzgerald would have been much worse and she may have been cut in half or at least been cut into for many feet. And the Crystal's bow would have damage on both sides on just one, and the bulbous bow would have sustained damage. And to claim that the crew slept through a major accident would like the passenger of the self driving car sleeping through a major accident. Every crew member knew when the two ships hit each other. They heard it and they felt. I would also assume none of the Navy sailors slept through the accident for the same reason. To even think that in crowed restricted waters no one was on the bridge would be like claiming pilots on a 747 were talking with the stewards in the passenger compartment while landing the plane. It is almost laughable.

      Delete
    3. Really seems like TR's writing style. Laughable isn't the word I've used; criminally negligent and "they're going to go away for a long time" is more likely. I'm sure you won't explain how they didn't call it in and accelerated away from the accident for 15 minutes with computer precision; your version has them standing on the bridge the whole time sipping tea watching it all happen for a good 15 minutes?

      Delete
    4. TR here - not me. I'm never anonymous on here. I'm happy to prove you wrong using my own log in here, or on my own blog.

      Delete
  15. A couple of thoughts. Between 1615 and 1619 the ACX Crystal made a small course change, from ~086 to ~ 070. This was slightly more than 10 minutes before the (assumed) time of collision. If the CPA (closest point of approach) was determined before the course change (most likely), the Fitzgerald would have thought themselves to be well clear of the Crystal's path. If this course change had not occurred, there would have not been a collision. I also noticed mention of a perpendicular impact between the Crystal's bow and Fitzgerald's starboard side. I believe if that had occurred, we'd be talking about a truly catastrophic event in which the Fitzgerald was cut in half, much like what occurred in the Melbourne and Evans disaster. The Melbourne was 20k tons and a few knots faster the Crystal while the Fitzgerald was longer and heavier than the Evans. The force of 30k tons moving at 18 kts striking a DDG broadside would most likely resulted in the sinking of Fitzgerald similar to the Evans. Interesting is the amount of change in course of the Crystal after the (assumed) collision - I wonder if the Fitzgerald was somehow hung on the bow of the Crystal, acting like a large rudder, driving the Crystal's bow in a starboard turn, until it finally broke free, allowing the Crystal's autopilot to then attempt to return to heading. If this were the case, it would suggest that the Fitzgerald was turning away (turning to port) and gaining headway (accelerating) attempting to avoid the Crystal but ran out of time and space.

    ReplyDelete
    Replies
    1. re: Perpendicular; yes, I agree, it's probably not the right term. My point is, while it likely wasn't a 90 degree impact,it was more than a shallow impact or they would have deflected off each other or had long scars along the Fitz. 70 degrees? 60 degrees? I'm not sure, but it seems fairly perpendicular from the scar, I think. I agree completely in your assessment that Fitzgerald was hung on the bow of the Crystal from the "spearing" for a short time after impact.

      Delete
    2. re Fitzgerald "hung on Crystal's bow" acting as a "rudder" or "bow thruster":

      I’ve wondered if the Fitzgerald rode up or was lifted up during the collision. Look how high the damage is on ACX Crystal, the top gunnel is wrapped over. The Fitz damaged area doesn’t sit up that high above waterline, does it?

      Crystal has one of those cattle-prod bow-bulbs just below water maybe that helped lift the Fitz in the apparently oblique collision. Certainly the Fitz would have been heeled over into the Crystal by the physics, but that bow bulb may have helped lift the Fitz and indeed "hang it" as you guys are speculating.


      Helluva big bang in any case.

      http://www.shipspotting.com/photos/middle/9/2/6/2682629.jpg

      Photo by Akira Uekawa at this page

      http://www.shipspotting.com/gallery/photo.php?lid=2682429

      Delete
    3. ... and another good photo, credited to AP

      http://i.dailymail.co.uk/i/pix/2017/06/19/14/418AF63B00000578-4617742-image-a-1_1497878301473.jpg

      Delete
    4. Thoughts about radar, relative position of the ships, Crystal's course change and detection of the course change.
      One thing that hasn't been addressed (at least I haven't seen it) is the physical proximity of the two ships. Times and speeds have been identified but not the physical distances between the two ships. Obviously, no information about the actual lat/long for the Fitzgerald, however, we can make some assumptions about relative positions.
      As reported, the speed of the Crystal was at 18+ kts. For discussion, assume the Fitzgerald was making ~12 kts. Crystal made a small (~15 degrees) course change 10-15 minutes before the collision. The Crystal was near, and closing on, the Fitzgerald.
      To illustrate (using the speeds identified above), if the two ships were on parallel courses before Crystal changed course, the relative speed differential was about 6 kts.
      If the collision occurred 10 minutes after Crystal changed course, distance between the two ships would have been roughly 1 nautical mile; if the Crystal's course change occurred 15 minutes before the collision,
      distance between the two ships would have been about 1.5 nautical miles.
      If the Fitz's radar operator had the range on his PPI set to 25 miles, it is unlikely he would notice the course change on his scope, the change in location/displacement of the 'blip' located toward the center of the scope would be very small.
      It would appear the Crystal was very, very close to the Fitzgerald when the course change was made. I'd suggest this also contributes to the theory that there were no look outs on the Crystal and the bridge basically unmanned.
      At 1 to 1.5 miles distance from another ship, it would not be prudent to change to a collision course.

      Delete
    5. Usual arrangement on the bridge of a merchant vessel is two radars, one on distant range (e.g. 24 miles) and another on shorter range (e.g. 4 nm). I assume a destroyer has same or more. The change should have been pretty obvious on shorter range scale, assuming anyone was watching. Third mate on Crystal may have been doing chart corrections, cargo discharge & loading plan for next port or other clerical work and only checking his radar occasionally, apparently not often enough. Destroyers can go a lot faster than container ships, and frequently do. Anyone's guess at this point which ship was going faster, and what the bridge watchstanders were doing.

      Delete
    6. This was a busy time of day, in a busy channel. If auto-pilot corrections were unnoticeable than I wonder why we don't hear more often about collisions. Given the ACX had a crew of 20 I think I heard, and the USS Fitz close to 300, plus a ship of war, and fast engines, it seems to me it should not be caught by a Container ship. Even if early hours, I have read that there is a lot of traffic to make port by morning, and I would not think you would put your greenest sailors on duty? Easy for me to say I know.

      Delete
    7. An Arleigh Burke DDG has an array of radars and a number of watch standers. Surface search radar, air search radar and beyond - multiple radar repeaters both in CIC and on the bridge, each usually can select the radar of choice (not dedicated).
      I'm certain the Fitzgerald's CIC watch crew had tracked the Crystal and had, at least at one point in time, determined the Crystal was going to safely pass beyond the allowable closest point of approach.
      As far as speed of the Fitzgerald, considering that the morning of the 17th was a Saturday,
      I suspect the Fitzgerald was loitering outside of port, waiting for daybreak before approaching and entering the harbor, getting the lines over, ship squared away, and then the off duty personnel hitting the town on liberty.
      So, the question is, was it making 12 kts, 15 kts 18 kts or was it barely making headway? It certainly wasn't going 30 knots! And, unless it were disabled, I'm certain it was not DIW: being stationary in the ocean is not a good ride for a sleeping crew. Being stationary is also dangerous since you cannot maneuver without headway.
      One contributor mentioned the optional speed would have been 15 kts. If that were the case, using the previous scenario changing the closure speed to 3 kts (18 vice 15), Crystal changing course 10 minutes prior to impact, Crystal was a scant 1/2 mile (1000 yards) from Fitzgerald when it turned to the collision course. If the course change were 15 minutes prior, the Crystal would have been 3/4 of a mile (1500 yards) from the Fitzgerald.
      The Arleigh Burke DDGs have 4 gas turbines, 100,000 hp total. A pair drive the port side shaft, the other pair drive the starboard side shaft. A variable pitch screw is used on each shaft. (All this information is available on line.)
      What I don't know is if the Burke class can run just a single gas turbine/single shaft/screw when loitering (to save fuel). A DE I served aboard had 4 Fairbanks Morse Diesels which we could run in any combination of 1-4 engines, to save fuel. Of course, running on less than all 4 limited our speed and ability to accelerate. If the Fitzgerald were running on less than all GTs, it would have impeded her ability to accelerate/maneuver away from the collision.
      Several mentions of the Crystal running without lights. The (Tokyo) moon rise was at 2330 (16th) and was slightly more than a half moon (56% illumination) and at 0130 was approximately 22 degrees above the horizon (using Tokyo as a location) Weather (again in Tokyo)) was for the most part partly cloudy, high relative humidity of 73%, visibility of just over 6 miles, temp in upper 60s, wind of around 13 mph.
      Considering both vessels were headed generally east (toward the rising moon), with the Crystal closing on the Fitzgerald, the topside look outs on the Fitzgerald would have been very lucky to even see the Crystal (again without lights) until it was very close in. And, even if Crystal was running with navigation lights, it would have been very difficult to pick up a subtle course change toward the Fitzgerald.

      Delete
    8. Thanks for the detailed, informative reply. Will re-read it. The only issue I have with that is what the Navyman in a post near the top added (name: "Unknown") -- they kept all ships 2K feet away -- and this is not even on a ship of this class. In a very busy channel, I am not sure why the destroyer would loiter. At this point I am beginning to believe a catastrophic failure of system or engine occurred on the destroyer.

      Delete
    9. There would be standard Captain's orders on when to notify. In addition, they post night orders that will cover specific circumstances or locations. 2000 yards CPA is common but it's up to the CO. It would also depend on location. Open ocean where there is little traffic may be further, shipping lanes may or may not be closer depending on type/amount of traffic, experience of the bridge crew, and confidence of the CO. That will be key in the investigation.

      Most OOD's will follow those night orders, but it is not unheard of for a ship to change course to avoid falling within that CPA (although often course changes off base are reportable in night orders as well) Also if the course/speed/CPA was calculated by the watch team and reported to the CO, he could acknowledge and/or check. Since another Crystal course change occurred 10-15 min before the possible collision, did they have time to recognize it and re-calculate a CPA in that time to notify the CO of the change? Again, they would need a couple of fixes, and there was probably a lot going on. Crystal wouldn't have been the only ship out there they were watching.

      Navy ships very infrequently sit stationary. Normally we would run racetrack or other courses to maintain station (how fast, how big, and whether we cut holes in the ocean or just went any direction within a box depended on what we were there for) I doubt they would be doing that in a busy shipping lane area. Most likely they were moving area A to B in an slow easy sail, planning to arrive at a station for exercises or return to homeport the next morning.

      Delete
    10. Not sure I understand Stan Rooker's math: The closure speed depends on the angle between the two ships. If the ships were traveling "head on" and one is traveling at 15kn, and the other at 18kn, the closure speed will be 33kn; only if they were traveling in the same heading (one behind and overtaking the other) would their closure speed be 3kn. At 33kn over 15,000 yards would be closed in 10 minutes. Based on the pictures a head-on collision seems unlikely, but since we don't have the destroyer's track, we can only speculate as to what the closing speed actually was.

      Delete
    11. All my assumptions are based on the two ships being on generally easterly headings. Crystal's headings and track are fairly well defined. As you say, we don't know what the Fitzgerald's actual track was.
      So, a few more thoughts and observations. A head on collision at 33 kts would have resulted in a sinking. Assuming any amount of headway by Fitzgerald, just about any head on collison with a 30k ton ship running 18 kts is going to be catastrophic. Additionally a head-on or near head-on collision would result in port side-port side or starboard-starboard damage, not the port side bow and starboard side damage present on the two vessels involved.
      I believe it fairly safe to say the Crystal was closing on the Fitzgerald from somewhere aft of the Fitz's starboard beam. Had it been a perpendicular hit with the Crystal making 18 kts, we'd be talking about the sinking of the Fitzgerald. The exact angle of closing, based on ships courses, will determine whether this was a crossing or closing incident. At the point of the collision, the Fitzgerald was most likely trapped in the "V" formed by Crystal's bow and submerged "bulb protrusion", Fitz's headway in conjunction with the collision angle caused the Crystal to turn to starboard after the collision. More than likely, the Fitz was attempting to get out of the way but ran out of time and space. I'd hazard a guess that the reason there is not more scaring on the Crystal's bow is that when the Fitz finally broke away, it did so very violently to port.

      Delete
    12. Another point to consider is that, based on an interview with a mother of a sailor on the Fitz, sailors were thrown off their bunks by the impact, which would mean a collision alarm was not sounded in anticipation of a collision. Does this mean the Fitz was not aware of the impending collision until it happened? I think safe to say the Crystal did not know about the Fitz, though I don't know turning off the AIS system by the Fitz puts them at the peril of auto-pilot systems.

      Delete
    13. I didn't hear the interview, but it is very interesting that there was no collision alarm. Or, did it sound simultaneously with the collision? Or did it sound at all? Needless to say, it would have been prudent to sound the alarm if there was any possibility of collision. However, the OD may have thought the danger of collision had been averted by maneuvering or altering the Fitz's course. And, of course, there is the possibility they didn't see the Crystal. Any number of shipboard systems could have been down.
      Again, until we actually know the Fitzgerald's track and system's status, we won't know.

      It is difficult to believe that the Crystal could actually collide with the Fitzgerald with absolutely no warning completely undetected until impact. This gives some credence to the possibility of Crystal's last course change going undetected by Fitzgerald. What it also does is raise the question of what the Fitz's topside lookouts did or didn't see, and whether or not the Crystal was running with or with out navigation lights.
      As far as an AIS on a Navy ship, it would not be used to disclose the position of a warship. It would be a monitoring device only.
      If the Crystal had radar and had it on, and if someone looked at the scope, they saw the Fitzgerald. A 505' ship is a heckuva radar target. Whether or not they paid any attention to their scope, is another question.
      More than likely, they didn't determine course or speed of the Fitzgerald, and simply turned to their new heading, which resulted in the collision.
      That is of course, assuming the Fitzgerald was maintaining course and speed. IMO the two vessels were relatively close to each other when Crystal changed course.
      As far as the sailors being thrown out of their racks at impact, I'd imagine the near instantaneous acceleration imposed by a collision from a 30k ton ship with a relative speed differential of just a few knots could very easily cause that.
      As stated by another, it was a helluva bang.
      To illustrate the actual magnitude of that small 16 degree course change. Without having the Fitzgerald's time stamped course, speed and lat/long data , we can't plot it's track relative to the Crystal's track.
      However, we can obviously assume Fitz's position at the point of impact, and the relative position of the Crystal 10 or 15 minutes prior to impact, based on it's course and speed.
      If the Crystal changed course 16 degrees (from the AIS information), making 18 kts, at 10 minutes prior to the collision, the difference in Crystal's position at that point in time (impact) would be about .82 miles from the position it would have been in had the course change not occurred. If the Crystal's course change had occurred 15 minutes prior to the collision, the position difference is even greater, being about 1.25 miles.
      In other words, had the course change not occurred (both ship's speed/course had remained constant), there wouldn't have been a collision.
      When all is said and done, IMO, there will be a number of events that contributed to the collision, anyone of which had not occurred, there wouldn't have been a collision.
      However, the last small course change made by the Crystal will be one, if not the, key factor contributing to the collision.

      Delete
    14. Thank you. This blog and comments have been an education.

      Delete
  16. could somebody discount the fact that the crystal had a heavy ass (due to cargo), thus, was unable to turn away in a timely manner?

    ReplyDelete
    Replies
    1. She did alright making the "U Turn" a half hour after the hit, in 5 min, and she has bow thrusters too, but I'm not sure if they're used while at speed or only when they're in port. There is no indication the Crystal "saw it coming" at all.

      Delete
    2. Bow thrusters generally are ineffective above 5kn though some cruise ships claim they are effective up to 9kn.


      Do we know what the visibility was? If poor consider that the impact could have been mistaken for something else, no radar contact, no AIS target, what else could a ship hit.... an uncharted mount / rock? If they thought they had hit a submerged object then the contingency plan for that would have been followed. Part would be for anchors to be cleared, crew go fwd to clear anchors and discover damage, report back, ship turns to head back to where impact was felt. As the warship is damaged it may show on radar thus crew can find it and steer towards it.
      BTW some AIS units drop transmit power to 1W once speed drops below 3kn, this could explain the loss of signal from the ship being picked up by shore side.
      The ~30min until turning back could be explained by the thought they had hit a submerged object. Master called 5min gone, general alarm activated, crew mustered (no internal lift many crew have to climb up or down to station) 5min more, plan dug out, crew ordered fwd, 5min, crew on focsle 5~10min, Mate ordered fwd to confirm discovery of damage another 5min and so on...

      Delete
    3. Above post sounds plausible in the absence of real info. Modern warships are generally designed to reflect as little radar signal as possible back towards ship that is scanning, and more of it in other directions, than ships less concerned about hostile actions, countermeasures for missile attacks, etc. Radar image may have looked like a small vessel, and in earlier observations it probably appeared to be staying out of the way. Shipping containers are dandy radar reflectors, so Crystal probably had anti-clutter turned up to suppress return from own containers, and once Fitzgerald was close, she probably did not show up at all on the radar screen. View from the bridge of anything close is mostly blocked by stacks of containers, and once Fitzgerald was within a half mile, hull, sidelights, sternlight and superstructure, except for the mast-tops, probably could not be seen from the bridge. At time of collision (and during the long minutes of closing if Fitzgerald was being overtaken), Fitzgerald's white lights, facing forward, would not have been seen from the bridge of the Crystal. Bridge not necessarily unmanned, but probably just checking the radar ever several minutes, while doing paperwork that robs night vision. Crystal's bridge watch likely had no idea what they hit, and looking out the window, might not have seen the Fitzgerald even while pinned to the bow. Collisions at sea are infrequent occurrences, not always recognized as such in the dark, and submerged object strike not an unreasonable thought, especially with hard and prolonged turn to starboard suggesting a submerged object jamming the rudder, and a period of uncertainty thus likely. After calling master, next call is to wake up an engineer, have him/her go below, check for damage in the engine room, then start an auxiliary generator, switch main engine from heavy oil to light fuel oil for maneuvering, and stand by for orders such as slow or stop. Fifteen minutes from collision to turn-around readily explained.

      Delete
    4. It wan't 15 minutes to turn around. Have a look at the timeline again; it was 15 minutes of acceleration before cutting autopilot. Another 10 before they turned around. There would be no need to wake the master or engineer; the ship was flung 65 degrees with the collision. Can't imagine anyone would sleep through that.

      Delete
    5. Auto-pilot no doubt had rudder at limit trying to return to set course, and ship was still turning away from set course. Audible alarms and error message displaying on auto-pilot for "rudder limit" and "off heading". Full rudder on a ship at sea speed creates a lot of cavitation, vibration and noise. Here's the psychology: The auto-pilot alarm messages and the noise and vibration from the stern were a form of misdirection. Denial is a big part of thinking, and no one wants to entertain the possibility that they steamed into a collision cluelessly. Attention was probably focused on the wrong end of the ship. If I stepped onto the bridge (or started paying attention) a few seconds after the collision, my first thought probably would have been that something was in the rudder and jamming it. I would have sent someone to check the after steering compartment and the engine room for holes, leaks or other visible damage before thinking to send someone to the bow to check for damage or another vessel pinned against it. Meanwhile, Fitzgerald's radios were knocked out, so no calls or answers from them. 15 minutes or 25, it still is not surprising it took a while to recognize and accept possibility that they had been in a collision, and to follow duty to return to scene and offer assistance.

      Delete
    6. Anonymous I appreciate your insight into the Crystal side of the equation. Can you expand on a couple of points?

      "Shipping containers are dandy radar reflectors, so Crystal probably had anti-clutter turned up to suppress return from own containers, and once Fitzgerald was close, she probably did not show up at all on the radar screen." Does the anti-clutter effectively result in a close-in range limitation so they can only detect targets beyond a certain distance.

      Is it safe to assume the 65 degree turn caused the Fitzgerald to rapidly disappear into a large blindspot caused by the containers behind the bridge? Given a crew of maybe 20 I have heard about what other watch positions were likely manned? should have been manned?

      "while doing paperwork that robs night vision." Would they not have red lighting on the bridge to preserve night vision?

      Delete
    7. The three minutes between data points leaves a lot to the imagination. 65deg course alteration, in the course of piloting I can get a rate of turn of over 40deg a minute with that size box boat. A rapid turn will scrub speed off, it then takes a while to build.

      Manning at night. Bridge one to three on watch, engine control room most likely ums (unmanned machinery space) with duty engineer resting and system set to alert / wake him. Might have an oiler doing rounds...

      Delete
    8. Some radar signal is reflected by nearby waves and wavelets and own ship's structure and cargo, etc. on deck. Anti-clutter or sea clutter adjustment tunes radar so that reflected signal typical of waves is suppressed on screen, and hopefully somewhat stronger signals from targets of interest, e.g. from sportfishing boat or sailboat, still come through and aren't lost in the clutter. The tuning is usually automatic, and something of a compromise. Sometimes the signals of interest get suppressed, too.

      My guess, and it's only a guess without the Fitzgerald's data, is that Fitgerald got off the Crystal's bow by bringing engines up to full power and turning away to port. The stern of the Fitzgerald probably didn't smack the aft port side of the Crystal because the bow of the Crystal started swinging to port, and her stern to starboard, with the auto-pilot trying to get back to the set course. If someone happened to be out on the port bridge wing, they probably would have seen the Fitzgerald peeling off.

      The ship going nearly sideways while the Fitsgerald pushed the bow around certainly would scrub off a lot of speed in a hurry. Red light is commonly used while making rough log entries, plotting on paper chart (increasingly rare), and tasks usually done as part of watchstanding duties. I suspect that some watch officers try to catch up on other paperwork perhaps not readily done under a red light. Also, a lot of paperwork these days is done on compute.

      Delete
  17. I wonder with the CO asleep, if those on duty did not want to enact evasive enough maneuvers to waken, or summon the Commanding Officer. I have a hard time believing that the container ship, with AIS on, would not assume smaller ships would stay out of the way. With additional radar on the destroyer hard to imagine how it got hit. I wonder if other traffic played a role.

    ReplyDelete
  18. Regarding the "sea jihad" scenario: I'm captain of an articuated tug-barge unit operating on the US East and Gulf Coasts. I deal with a lot of foreign watchstanders and USN in crossing, meeting and overtaking situations. If the Fitzgerald's AIS was transmitting, this blog would have shown us both vessels' tracks, and a lot of questions would be answered. Warships frequently run without AIS transmitting, and watchkeeping with visual observation and radar plotting have always sufficed to keep me from colliding with them. So I don't think the stereotyped third mate was aware that he was near a US warship. Secondly, the Fitzgerald could probably go twice as fast or faster than the Crystal, like the 20 knot or faster cargo and naval vessels that I encounter on my 10-knot tug. There is no way I could turn around, stern-chase and catch one of them, and neither could the Crystal, unless the Fitzgerald was running at slow speed and the bridge watch oblivious. It sounds like the undermanned, overworked, sleep-deprived, underpaid and poorly motivated bridge watch of the Crystal had stepped out for a sandwich or other non-navigational activity and left the bridge unattended, perhaps assuming that the vessel showing them a green light would stay out of their way. A small change in course and speed during that time could be due to currents, not deliberate calculation. Damage to the starboard (green running light) side of the Fitzgerald suggests they attempted to cross the bow of the Crystal from left to right, but does not rule out other possibilities. The Fitzgerald probably acted more like a bow thruster than a rudder while pinned on the bow of the Crystal. Regarding lack of comms post collision: surely someone on a naval vessel has a satellite phone, which has its own batteries and antenna, and would not be disabled by the collision. Maybe someday we'll get the whole story.

    ReplyDelete
    Replies
    1. Thank you for your input; the JP CG indicates they were contacted at 2:25 by the Crystal, which was when they first became aware of the collision. I would hope that the Fitz had someone with a sat phone, but it doesn't seem like it was on hand, or it wouldn't have been news to the JP CG... I'm looking forward to the investigation and what it draws out...

      Delete
    2. you assume that someone on the Fitz would "immediately" radio and report the collision. They would not, the ships first priority would have been sounding general quarters to set condition Zebra(all watertight hatches secured) and begin damage control operations to keep the ship afloat. Once the ship was stabilized, flooding contained, etc only then would the Fitz have attempted to radio in to report the collision. Whether the collision took place at 1:30am or 2:30am local really does not make much difference on the Fitz, because the majority of the crew(including the majority of the senior officers and enlisted) would be in their racks asleep, with only watch standers up.

      Considering that it appears that the Fitz was holed below the waterline, the crew did a rather amazing job to keep the ship afloat, with the loss of only 7 sailors. I spent 10 years in the USN,and was a qualified helmsman and Lee Helmsman.

      Delete
    3. Well, "immediately" might be a little too immediate, but in less than 55 minutes from the collision, yes. The Crystal called in the collision 55 minutes after the impact, after returning to the scene. So yes, I do believe if the Fitzgerald had been able to make a distress call they would have done so in 55 minutes from the initial catastrophic event, that could easily have sunk her.
      FYI,
      https://www.voanews.com/a/navy-identifies-sailors-killed-in-collision/3906043.html

      Delete
  19. Regarding the Crystal continuing at previous speed after collision for 15 minutes, it should be kept in mind that merchant vessels of this type do not have direct engine controls on the bridge and frequently do not have anyone on watch in the engine room at night. The engineering department generally needs or wants 15-30 minutes notice if they will have to stand by for orders from the bridge for speed changes or maneuvering. Just wait until the fully autonomous ships are here.

    ReplyDelete
    Replies
    1. I'm not buying what you're selling. The autopilot was steering the ship and there was no off button on the bridge, just a bunch of people running in circles who couldn't find he radio either? Nope. There was nobody on the bridge of the Crystal, or if there was, they were not qualified to be.

      Delete
  20. Probably had women on the deck looking in their hand mirror primping their eyelashes and putting on lipstick just like they do behind the wheel of a car! Next thing you know... BAM!!!!!!

    ReplyDelete
    Replies
    1. This comment would not be funny even if no one had died.

      Delete
    2. Oh please ��

      Delete
  21. There is a line between being the overtaking vessel, thus the burden vessel which must give right of way and the vessel having the burden vessel on her port quarter. It is not an easy line to determine and in this case it appears the Navy vessel believed themselves the privileged vessel from the over-taking container ship while the container ship considered themselves the privileged vessel due to the Navy vessel on their port bow. Any change in course or speed by either vessel should be sufficiently large to be apparent/evident to the other vessel. Bottom line a tragic accident that personal on neither vessel wished for. Bridge to bridge VHF radio communication should have been established and intentions by both vessels established and agreed to. In busy shipping lanes there is all too much chatter un-necessary chatter and for this reason sometimes VHF is turned down.

    ReplyDelete
    Replies
    1. Another problem with VHF comms in high traffic areas, especially at night or in limited visibility, is confusion as to who is calling or answering on the VHF. For example, "Calling the vessel four miles on my port bow, this is the ship on your starboard bow" might be applicable to more than one combination of vessels in radio range. Having the AIS transponder turned on helps avoid confusion.

      Delete
  22. If the AIS track is correct and usually so close to the coast, in a well trafficked area, it's usually pretty accurate.
    therefore, for a container ship moving at 17+ knots, it's hard to understand how it's track can change at such a right angle, without a collision at that time, 16:30Z. The momentum of the Fitz would have spun the bow of the Crystal as depicted in the track.
    Sorry to say, because the USN will not look good in any of this, but probably when the dust settles, the comments by Anonymous ( 18 June 2017 at 12:52:00) above will be pretty close to what happened.

    ReplyDelete
  23. "An hour passed before Japan authorities were notified of Fitzgerald collision"

    "The collision happened at around 1:30 a.m. but it was not until 2:25 a.m. that the container ship informed the Japanese coastguard of the accident, said coastguard spokesman Takeshi Aikawa told Reuters."

    http://www.reuters.com/article/us-usa-navy-asia-idUSKBN19913U

    ReplyDelete
  24. This all has been wonderful to read. I'm not military nor a sailor (been on a few cruise ships), but I'm fascinated by this incident. Thank you.

    ReplyDelete
    Replies
    1. Thank you Patti! Glad my blog post found its way to you :)

      Delete
  25. Great analysis. My guess is that you could start it about 10 minutes earlier when the Crystal was sailing at 18 knots on a course of 90 degrees (approx.). At this time it would have appeared to the Fitz that she would clear Crystal by a significant margin. Crystal then turns about 20 degrees north to clear Tishima Island, probably an automatic turn based on a proximity to a waypoint if the rest of your analysis is correct. This brought her onto a collision course with the Fitzgerald. Shortly afterwards Fitz turned to port to avoid the collision, delaying the inevitable for a few minutes. Earlier recognition of the turn by the Fitz might have avoided the accident. However no automatic system would have predicted the accident until after the turn by Crystal was completed.

    Normally the vessel on the right has the right of way. (Colregs rule 15). That does not give them the right to turn into the path of the Fitz at the last minute unless there is an obstruction or other special situation. To avoid collisions, both vessels may maneuver.

    ReplyDelete
    Replies
    1. "Shortly afterwards Fitz turned to port to avoid the collision, delaying the inevitable for a few minutes."

      We can only speculate on the maneuvering of the Fitzgerald, but if the Fitz. turning to port led to the collision, then likely any other maneuver would have avoided it. 18 knots over just a few minutes should create enough distance for near miss at the least. If they had turned to starboard, they could have passed behind the Crystal.

      They may have simply accelerated, hoping to pass just in front (maybe that's what they did, pushing the Crystal even more starboard after the impact). But if the reports are accurate that no warning sounded, the captain was asleep, and no alerts were sent to the Navy or Coast Guard, then the simplest explanation is that they weren't aware of the Crystal and so didn't react at all until the collision happened.

      Delete
  26. From the NYT: "Japanese CG said on Monday that a deadly collision ... had occurred nearly an hour earlier than previously believed ... investigating whether there had been a delay in reporting". Seems to nicely match Steffan's hypothetical timeline.

    ReplyDelete
    Replies
    1. Sorry, hit publish too soon. It seems like the JCG originally reported the collision happened around 2:20, which was what the USN press release stated. After further investigation the shipping company said the collision happened around 1:30, and JCG amended its information. From the looks of it, the USN public statements were simply relying on JCG statements, and not necessarily on reports from the Fitz. This is not completely out of the question. The PR chain of command in the USN may not have been given all the relevant information from the operations side of the house.

      To be fair, it is also possible that the shipping company "updated" its remarks after reading a post such as this, and *their* PR people don't have all the right information.

      You'd think communications would be better, but in large organizations that is often not the case. I could see any of the organizations involved making half-baked statements without access to all the available data.

      Delete
  27. No naval knowledge or "when I was" stories from me but I started following VesselFinder and MarineTraffic right after the incident. Your narratives dumb it down for me. I've shared your blog on news sites, even sent it to my Naval Academy friend. First and foremost are 7 sailors lost. My heart breaks for their families.

    ReplyDelete
  28. The US is fighting ISIS rebels in the South Philippines for the moment. Is it totaly insane to think that "some" Philipino personnal on the cargo ship might want to revenge and deliberately hit the Fitzgerald. And ,afterwards, make it appear like an accident...

    ReplyDelete
    Replies
    1. Regardless of what anyone on the Crystal might have wanted to do, Fitzgerald had AIS off, so could not have been identified as a US warship on dark night either visually or electronically. Not insane, just ruled out in this case by the available facts.

      Delete
    2. Yup. What they said. Not having AIS on didn't help the matter, since AIS is (or can be) overlaid on the radar view and might have triggered different actions by the autopilot / alarms. As it was, Fitz was just a blob on the radar with no correlating AIS signature.

      Delete
  29. What is the name of the captain of the ACX Crystal? Can't find it anywhere.

    ReplyDelete
  30. First, thoughts and prayers go out to the crew and their families. As a former surface warfare officer and OOD with over 1500 hours of bridge time, this incident raises a lot of questions. There have been many well thought out theories posted but it is impossible to reconstruct the collision without the track information for the Fitzgerald. Let's leave the forensics to the pros.

    ReplyDelete
    Replies
    1. I understand the concern about speculative theories. However, if the forensics are left solely to the pros, the public will likely be left in the dark. And the public does have an interest here, beyond even the loss of life and beyond questions about the competence of the way the US Navy implements safe navigation with its vastly expensive ships. For example, it appears that the oceans are at every moment populated by large zombie-like container ships, thinly staffed, and patently unable to follow the mandated rules of navigation (e.g. the overtaking vessel must steer clear of the vessel being overtaken). There is a big possibility here that Crystal, overtaking the Fitzgerald, on autopilot, made a port turn to close the Fitzgerald only ten minutes before the collision, in direct violation of maritime rules of the road. The dangers posed by these ships extend far beyond the particulars of this very tragic accident (an accident I believe might well have been avoided if the container ship had not adjusted course toward the Fitzgerald, regardless of the state of awareness aboard the US Navy combatant).

      Delete
  31. Former Navy LCDR here, Contact Coordinator on Leahy class cruiser and SSBN. Clear to me that the ACX was at least 50% at fault (on autopilot, with "last clear chance" duty to avoid the collision), but the merchant does not appear to know it was steaming right into the Fitz (that particular course change was not prudent in the first place and ACX made no last minute course corrections nor reduce speed). Fitz is 50% to blame as well. Focus will be on OOD, Starboard lookout (and rest of watch for that matter) and whatever radar, sonar, was being used. We cannot talk about the navigational stuff, except to say that some equipment may be to blame (again, cannot discuss). Starting point in the collision analysis is the starboard lookout. I estimate that Fitz and ACX were on a parallel 090 course, and Fitz was not too concerned because she was being overtaken by ACX who was abaft her right beam with two or three miles of separation (assuming Fitz was steaming at 10 knots). The OOD was probably feeling good. It appears he had recently navigated around two large merchants, and the horizon was relatively free of contacts. The lack of forward contacts could have lulled the watch team into a false sense of security. Plus, they were returning home, they were tired from the previous day's exercises, and some were going on liberty soon.

    I have little doubt that the Starboard lookout was tracking the ACX (first picked up on radar), but depending on whether ACX had her running lights on, the lookout may not have had a good view of ACX (plus, lack of moonlight hampers visibility). The key is that the CO was not called (presumably), so that tells us the OOD did not think ACX was a threat. At this point, the OOD is thinking the Fitz has the right of way. He was being overtaken. There was another similar merchant on the left beam too that was fairly close and was also overtaking the Fitz (the merchants were steaming faster because they had to travel further than Fitz). All of a sudden, the ACX turns 15 degrees to port. Did the lookout notice that course change? Apparently not. It was very small. In this scenario, the OOD does not realize that the Fitz is no longer the stand on vessel. The other watch-standers probably did not notice the course change either because it takes time to develop a new contact solution (course and speed), and here, only 12 minutes elapsed between the time of course change and impact. Thus, there was little time to develop a new solution. So the OOD is not aware of the constant bearing decreasing range scenario with the ACX. Plus, the watch is focused on contacts in front of them, not behind them. Meanwhile, the ACX is bearing down on Fitz. The OOD must have assumed that no merchant in their right mind would turn towards him and essentially ram him. But the ACX was on autopilot and may not have had anyone on the bridge. If there had been a lookout on ACX, we are probably not talking about it now.

    ReplyDelete
    Replies
    1. What about the non-use of AIS by the Fitz? Why would they assume the Crystal has seen them unless they are in radio contact?

      Delete
    2. Having several years worth of experience looking at contacts at night, I am sure, absent some very unusual circumstances, that a vessel in ACX's position would have seen the Fitz's starboard running light as well as masthead light (assuming they were on). If anyone was watching on ACX, they would have easily seen Fitz off their port bow, well before her course change to port. Essentially, Fitz was rammed.

      Delete
    3. Plus, ACX likely had Fitz on radar.

      Delete
    4. Why wouldn't the auto-pilot navigation system detect it? Thanks in advance.

      Delete
    5. Thank you very much Greg, that's extremely insightful.

      Delete
    6. Auto-pilot keeps the ship on a constant heading, or on a trackline between two waypoints. Detection of vessels and other targets and calculating closest point of approach and danger of collision is done by other systems (radar and AIS, or the sailor's eye) and usually involves some human input. Automatic alarms can be set, but in heavily trafficked areas are usually not turned on because they would constantly be set off by the close vessels that we already know about. AIS-based anti-collision systems would not have helped Crystal, since Fitzgerald was not transmitting AIS data. There seems to have been a failure on the part of both vessels to make full use of radar anti-collision features.

      Delete
    7. Greg, A couple of questions -

      "It appears he had recently navigated around two large merchants, and the horizon was relatively free of contacts. The lack of forward contacts could have lulled the watch team into a false sense of security." and "There was another similar merchant on the left beam too that was fairly close and was also overtaking the Fitz (the merchants were steaming faster because they had to travel further than Fitz)."

      I haven't seen any info on other vessels in the area. Are those sentences hypotheticals or have I just missed that?

      "All of a sudden, the ACX turns 15 degrees to port. Did the lookout notice that course change? Apparently not. It was very small. In this scenario, the OOD does not realize that the Fitz is no longer the stand on vessel."

      It's been 30 years since I left the navy but doesn't the Fitzgerald remain the stand on vessel since she was originally being overtaken?

      Delete
    8. "I haven't seen any info on other vessels in the area. Are those sentences hypotheticals or have I just missed that?"

      My statements are based on my "solution" for the Fitz's course and speed in the 30 minutes prior to the collision. It is based on AIS data, the location of the TSS, rules of the road, and my belief that the OOD and the watch were taken by surprise. I have Fitz on a course of approx. 090 and 10 to 12 knots.

      "It's been 30 years since I left the navy but doesn't the Fitzgerald remain the stand on vessel since she was originally being overtaken?"

      They were in the classic overtaking situation (by two vessels, one to port and one to starboard). However, at some point, the overtaking is deemed complete (according to rule 13d, it's when the overtaking vessel is "past and clear."). Here, the ACX Crystal will argue that it was "past and clear" when it made its course change to port. The ACX Crystal will use the old, "green means go" rule and argue that since the overtaking was complete (it may have been slightly past the Fitz) and it saw the Fitz's green running light, it was within its right to turn to port so that it could stay in the sea lane (and avoid running into Toshima island). In court, this case could turn on this issue.

      Delete
  32. This collision is very similar to the USS Radford (DD 968) and Saudi Riyadh collision in 1999. In that incident, the Radford was doing doughnut holes around a buoy and the merchant bore down on them. The watch standers were very inexperienced and did not recognize the danger until it was too late.

    ReplyDelete
    Replies
    1. This is a great point that I didn't know; while I believed this wasn't unprecedented, I didn't have a concrete incident to refer to. Thank you!

      Delete
    2. Or the USS Porter in 2012 - the damage even looks similar -
      http://gcaptain.com/intense-bridge-conversation-porter/

      Delete
  33. Soon after this occurred I have seen some discussion about whether the USS Fitzgerald was at fault. Most early comments noted that the Fitzgerald was struck on the starboard side so she must have made a mistake. The discussion on here has been more accurate in my view. Using the tracks Steffan posted here I tried to figure out where the Fitzgerald might have been headed and whether it was a crossing situation (Fitzgerald's fault) or an overtaking situation (Crystal's fault).

    I created a picture to show a couple of options for the Fitzgerald's course. https://ibb.co/esgUdQ

    Because of the islands in the area it seems that the Fitzgerald was going to pass north of Toshima or near Kozushima. The first results in an overtaking situation. It would be consistent with the Fitzgerald returning to her home port in Yokosuka. The second results in a crossing situation. It would be more consistent with the Fitzgerald leaving Yokosuka and heading south from the heavy traffic lanes out of the area. Any critique is welcome.

    ReplyDelete
  34. I was in these exact waters on Reeves many moons ago. Like you said, there is a Traffic Separation Scheme (TSS) where the collision occurred. By looking at AIS data of the ACX Crystal's track, I was able to determine that she was headed eastbound but was in the westbound lane. She should have been approximately five miles south (i.e., further from land) than she was, so she was in the wrong lane. I am sure this course was chosen for expediency. With this context, if Fitz was heading south, she would have been cutting directly across the sea lanes. While the rule for course changes while in a sea lane is to make large course changes so that other ships can see your broad side, I don't think your typical U.S. warship is a fan of that rule, especially when it means you are the burdened vessel and are exposing your beam to oncoming eastbound traffic. It also creates close CPAs. Plus, the Wan Hai 266 was on a course parallel to ACX but two miles to the north. If Fitz was on a course perpendicular to these two merchants, she is cutting some close CPAs. My opinion is that she was on a parallel course for at least the 30 minutes prior to collision.

    ReplyDelete
  35. I know that the reports say that it was a clear night, but could fog have been an issue? From a Japanese safety report on fog. "It is remarkable that 71% (108 cases) of casualties in fog occurred in the three months between May and July...the occurrence of fog-related accidents is concentrated in the hours between 01:00 and 10:00. This is because the fog came on in the chilled air around midnight and dispersed in the morning." There were 111 collisions due to fog between 2001 and 2005. Most involved cargo vessels. June is the worst month. The area where the collision occurred is a known problem area for fog.

    ReplyDelete
  36. I have a few questions of my own. I recently retired from the Canadian Navy. I have been on the bridge for transits of crowded waterways (suez, entry to Singapore, entry to Istanbul, entry to Toyko, etc) at night. We have watchmen on both bridge wings, radar, cameras which have a bird’s eye view of the scene, communicated directly to another bridge watchman and to the Control Room. There is an experienced OOD plus a junior, helmsman, communicator (and someone near on the brow ready to drop anchor quickly). It would be about the same in the USN. Lots of people, lots of tech, lots of eyes of all kinds. They might be tired, but at 0130 or 0230, not that tired, as the duty watch would have switched not long before. This is what these guys (and girls do). I cannot buy that they were distracted and tired. Not that I have any solutions, but rather that this element does not fit. Secondly, since this is a busy channel, why did not another nearby ship radio in the collision? Third, in the redundancy of the comms, there was a sat radio. Every warship has this. Stuff was flung about, for sure. The OOD was 100% engaged in damage control. But someone on the bridge had the delegated duty to radio in such incidences. That they never did is rather hard to believe. Even if they lost electricity, the bridge and the control would have emergency power.
    I was on a navy ship that collided with another ship, and on another navy ship that had a massive engine room fire. I understand the confusion of emergency at sea. But these sailors would have trained extensively for these scenarios. Everyone had a role to play, and knew their role. This is not to say that in the real world, that all goes according to plan, but NO ONE would have forgotten to radio this to the Naval authorities or the JCG, or even to call out to nearby ships, esp. if they though men were in the water. The lack of comms from the Fitz is a big mystery….

    ReplyDelete
    Replies
    1. I wonder if there was a failure of systems on Fitz, and further, if it will be exposed. But I would think it would get out one day.

      Delete
  37. Any idea of what was going on with the Crystal hours later, around 21:30 UTC, when just northeast of Oshima Island, they spent half an hour decelerating from 13 knots to almost standstill, then turned southwest and inched along for several kilometers before again correcting course on toward Tokyo Bay? Was this just an awkward way to resume their original course (I assume they had planned to pass west of Oshima Island)?

    Were they avoiding another collision, or possible being asked to change their destination? Did they think they hit something again?

    ReplyDelete
  38. All analysis are one sided so far. Why ? Crystal's route has been highlighted. What about for uss Fitzgerald?

    ReplyDelete
    Replies
    1. The only publicly available data for the course/speed/etc is from the AIS transponder aboard the ACX Crystal, so that's been my focus. The direction/speed/etc of the USS Fitzgerald are being extrapolated from additional crumbs of info, but most of the irrefutable data is about the ACX Crystal.

      Delete
  39. Sailorman, so you know who is talking. My background retired, living 1/2 year in a sailboat, cruising with Wife, we did 5,000 miles last year. Double handed sailing is solo sailing with sleep.

    Our boat has AIS, VHF, RADAR, HF, IRIDIUM, plotters.

    Last year. Through inattention, I wacked a big steel buoy and did some damage. I did my own after incident report and came up with a list of things I should have done but didn't. So I get what it feels like to be on the hot seat.

    I accept the original proposed chain of events on the Crystal, not sure but good enough for now, makes sense. Nothing about the Fitz makes sense.

    AIS: Forget RADAR, watch standers, etc for the moment. That is largely irrelevant when in a commercial traffic way. RECEIVE ONLY AIS will tell you COA, TCPA, speed, heading, RATE OF TURN, and tons of other data about the vessel. It also has a proximity alarm. Presumably the Fitz had AIS, therefore they had the means to avoid this incident.

    Traffic Control: Is there a shoreside traffic control zone that extends this far out? I'm thinking of something like Halifax or S Johns, NL. They should have civilian track information.

    VHF: Wonderful stuff. My VHF is tied to the AIS and the GPS. So I can make a direct call bridge to bridge knowing the vessels name. It has a distress feature that when activated sends a distress call in luring my lat/long. My BATTERY OPERATED handheld VHFs do the same thing. So, were I in trouble, and lost all power, these units would still work. In fact we take them, one for each, with us in the dingy, just in case we swamp.

    I see no reason why the $1.2 billion destroyer should not have even better, more resilient, redundant capabilities. And with a 300 man crew surely someone would have thought to call for help.

    While I don't want to speculate or lay blame there has been much negative consideration of the Crystal and her 20 man crew in the hour following the impact. Were similar critical analysis used on the Fitz one could equally wonder what they were up to and why they remained silent.

    ReplyDelete
  40. The Captain of the Crystal speaks: http://news (dot) trust (dot) org/item/20170626101937-6xsul

    Greatly appreciate your thorough analysis and resources, and all the comments by experienced naval navigators.

    ReplyDelete
    Replies
    1. I'm thrilled with the high quality comments too.
      The captain's statement isn't worth anything; it's easily dismissed by the AIS data that beacons every ~3min. No 10min gap to fit his story in. He's going away for a long long time.

      Delete
    2. I thought the mandated transmit interval would be much less than 3 minutes, in traffic and traveling at Crystal's speed. Three minutes' delay could make a big difference. Do you think Crystal was actually reporting on a shorter interval?

      Delete
    3. When you're right, you're right; the data I used is from MarineTrffic.com, and is not actually "raw"; I mean, that's my source data and "raw" in that sense, but technically there is raw AIS data collected by nearby AIS-T receivers that could be more frequent than every 3 minutes. In the US those receivers are run by the USCG, so in Japan I'd *guess* the receivers are run by the JP CG. I've requested that raw data from the USCG before under FOIA before, but it took months to get. Maybe the JP CG would release the raw data too, I'm not sure.
      Good catch!

      Delete
    4. https://help.marinetraffic.com/hc/en-us/articles/205426887-What-kind-of-information-is-AIS-transmitted-

      Dynamic Information (such information isautomatically transmitted every 2 to 10 seconds depending on the vessel's speed and course while underway and every 6 minutes while anchored from vessels equipped with Class A transponders)

      I work for a company that sends it's AIS data stream to marine traffic … they get updates more often than at 3min intervals per vessel

      Delete
    5. Short of doing a FOIA (Do they have something FOIA-like?) to get the data from the Japanese Coast Guard's receivers, I don't know how to get any higher resolution; you're completely correct tho, it isn't just every ~3 min, but that's all I see from this end of the stick.

      Delete
  41. It's not clear the USN has fully embraced AIS as a safety of navigation system. Naval combatants are disinclined to rely on information provided by potentially unfriendly parties, information that might easily be spoofed. The AIS on an older ship like Fitzgerald could be a stand alone laptop installed on the bridge, available to the OOD but not integrated into the ship's navigation system. By policy it would operate in receive-only mode. Radar and visual lookouts are more likely to be the primary safety of navigation sensors. But even an excellent radar (SPS-67/73 ?) would take some time (not sure how long) to assimilate a moderate course change by Crystal if it were on a nearly parallel course to that of Fitzgerald. In this scenario a recognition delay of even a few minutes might put the Fitzgerald OOD far enough behind the problem that a misjudgment in recognizing the threat, or an incorrect course change in response, would be non-recoverable. I'm sure that this event will provoke an intense internal discussion by the USN about the utility of AIS. However, much of this discussion might not be visible to the public even though stand-alone AIS is not itself a classified system.

    ReplyDelete
  42. Please explain: "Shipping data in Thomson Reuters Eikon shows that the ACX Crystal, chartered by Japan's Nippon Yusen KK, made a complete U-turn between 12:58 a.m. and 2:46 a.m." Why such a large time spread?

    ReplyDelete
    Replies
    1. You'll need to ask them. With great respect to Tim's work, I think it's really weird; I provide you with the 3-minute resolution data above that clearly disproves the statement.

      Oh hey, here's another blog post now ;)
      http://www.vesselofinterest.com/2017/06/the-leaked-statement-from-acx-crystals.html

      Delete
    2. Yes, thank you. Glad I found your site.

      Delete
  43. Article from US Naval Institute (USNI) quoting sailor onboard that Fitzgerald fought damage to the ship for almost an hour before the Japanese Coast Guard was notified. The Crystal notified when they returned to the scene, at about the same time the ship was able to get their Iridium Sat Phone up. Also discuss damage to the CO's cabin and how he was rescued as well as some pictures of the cabin.

    Details and pictures lend more credance to the feeling this was a "glancing blow" (maybe 45 degrees or so?) and that the Crystal was not paying attention at the time. Also more likely that Fitzgerald had initially identified the Crystals course as "safe" and the 10-15 min from the Crystal's turn to port, they didn't recognize the course change by lookout or radar until it was too late. The question is how much advance notice did they have before the collision.

    I find it hard to believe they had no notice, but based on speed, night time and time from turn to collision, maybe 5 min? Will be interesting to read when the report comes out. It will come out as the Navy is good about that and posts all such investigations for the public to view.

    https://news.usni.org/2017/06/21/investigators-believe-uss-fitzgerald-crew-fought-flooding-for-an-hour-before-distress-call-reached-help

    ReplyDelete
    Replies
    1. I think this is about right. The port turn, which was not a big course change, would take time to recognize if Fitzgerald were relying solely on radar or lookouts and might take longer to recognize than the time it took the ships to close on each other. One question I have posed above is how much the USN is willing to rely on AIS as a safety of navigation "sensor" as well as an intelligence gathering system. The course change would be immediately transmitted if Crystal were transmitting AIS at the required rate commensurate with the local traffic density (and this should have been much more rapidly than every three minutes). But the USN has good reasons not to rely unconditionally on AIS (creates a dependency on unreliable sources of information since AIS can be spoofed), so we'll have to see if a discussion about this emerges from the USN deliberations about the crash. It's quite possible the USN may regard its AIS policies as being sensitive, so the public may never know.

      Delete
  44. Sailorman speaking.
    Do you have a copy of the Porter investigation you could share?
    To add to your above comment "hard to believe" is an understatement. With AIS it's impossible to believe. With AIS all necessary data is there on the screen, the COA alarm is sounding. You have to actively ignore it, as they apparently did in the Porter incident (from the leaked voice recording.)

    ReplyDelete
  45. COA (Closest Point of Approach) NOT COA.


    Duhhhh!

    ReplyDelete
  46. Fault? Fault is not the issue from the standpoint of FITZGERALD's CO, OOD, JOOD, and CICWO. "Accountability" and "Responsibility" inevitably fall upon their shoulders, jointly and severally, no matter the actions of the container ship. The old maxim, "It happened on your watch!" inevitably applies.

    ReplyDelete
  47. A report on the damage control efforts was released today.

    https://www.nytimes.com/interactive/2017/08/17/world/asia/document-navy-report-on-uss-fitzgerald.html

    It doesn't address actions before the collision but a timeline on page 22 of the PDF does report the Fitzgerald was on course 230 true at midnight.

    ReplyDelete
  48. With the course of 230, and speed of 20 knots, assuming they were constant (or close) for 15-30 min prior to the collision, and using assumed point of collision as a marker, it should be relatively easy to get a better picture of the relationship of the two shops in the time leading up to the collision including the earlier course change by Crystal.

    ReplyDelete
  49. A "preliminary inquiry" was published yesterday by the Navy.

    http://www.secnav.navy.mil/foia/readingroom/HotTopics/USS%20Fitzgerald/Supplemental%20Inquiry%20USS%20Fitzgerald.pdf

    ReplyDelete
  50. The U.S. Navy has issued a combined report on the Fitzgerald and John. S McCain collisions.

    http://www.navy.mil/submit/display.asp?story_id=103130

    Something I find puzzling about the report is the geometry of the Fitzgerald collision. The Fitzgerald and Crystal were on roughly complementary courses, set to pass port to port (or perhaps collide nearly head-on), but just before the collision the Fitzgerald turned sharply to port, right into the path of the Crystal. The diagram showing the relative positions of the ships at the time of collision suggests that the Fitzgerald had turned almost 180 degrees. This seems unlikely.

    Also puzzling is the turn to port; the OOD first ordered a turn to starboard, but rescinded that and instead ordered the turn to port, seemingly right into the path of the Crystal. He presumably hoped to get across in front of the Crystal.

    Was there another ship to starboard that would have interfered with a course change in that direction? The report leaves a lot of questions.

    The report on the John S. McCain, OTOH, seems very clear (though certainly damning...).

    ReplyDelete