Trains.com

Amtrak 501 Derail in Washington State

74855 views
1887 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    November 2015
  • 1,345 posts
Posted by ATSFGuy on Sunday, December 31, 2017 9:34 PM

Why would an engineer be going 81 mph in a 30 mph zone?

To me, that makes no sense, and it's a recipe for disaster.

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,919 posts
Posted by blue streak 1 on Sunday, December 31, 2017 11:24 PM

If this rail line at one time was straight lets ask a few questions.

1.  What exactly was the agreement between WA DOT and NP to building the first part of curve to span US-99 ?

2.  Did NP require any change in the bridge if the track ever became a high use RR ?

3.  When the road was changed to I-5 and the new northbound lanes built what if any requirements from the RR (NP or BN ?) ?

4.  If the line was straigh would that have indicated a faster turnout at the location of rejoining the BNSF ?

5.  Were these constructions of US-99 and I-5 easements or purchases and what provisions were filed to county for them ?

  • Member since
    May 2013
  • 3,231 posts
Posted by NorthWest on Sunday, December 31, 2017 11:56 PM

I'm not sure the line was ever straight based on old topo maps, but just past the bridge is another 30MPH curve around a hillside where the Lakewood Sub meets the Seattle Sub which would have had the same result. Everyone seems to be making a big deal out of how bad this design was and to an extent I agree, but the presence of the junction two thirds of a mile later meant that speeds needed to be lowered anyway. They could have removed the curve at the junction I suppose, but they would require a massive relocation of a hillside.

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Monday, January 1, 2018 12:27 AM

zugmann

 

 
7j43k
I do wonder if an agency takes an inappropriate length of time to conclude an investigation, whether it would be held accountable for any deaths that arose from the lack of timely information.

 

So, what is considered inappropriate - given that due process is a thing?

 

About the slipperiest slope I've ever read.

 

 

I expect our elected officials will work that out.  When they realize that it might mean saving lives/votes.  Just as they always do.

 

Ed

  • Member since
    November 2015
  • 1,345 posts
Posted by ATSFGuy on Monday, January 1, 2018 1:31 AM

So was rerouting the talgo a bad idea?

  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Monday, January 1, 2018 8:05 AM

blue streak 1

If this rail line at one time was straight lets ask a few questions.

1.  What exactly was the agreement between WA DOT and NP to building the first part of curve to span US-99 ?

2.  Did NP require any change in the bridge if the track ever became a high use RR ?

3.  When the road was changed to I-5 and the new northbound lanes built what if any requirements from the RR (NP or BN ?) ?

4.  If the line was straigh would that have indicated a faster turnout at the location of rejoining the BNSF ?

5.  Were these constructions of US-99 and I-5 easements or purchases and what provisions were filed to county for them

Streak,

The answer to the 'if' predicate of your question fails. There is no evidence that the alignment of the Americn Lake line was adjusted for I-5 or its predecessors.

The circa 1969-70 NP track chart for this line segment shows the line to have been built in 1891. Had there been a relocation for the highways or anything else, the physical limits and date would show on the track chart, as does the aproximately .5 mile long segment from MP 10.8 to 11.3 which was relocated in 1914 to connect with the new double track line from Tenino to Tacoma which was built to move though traffic off the cheaply built 1872/73 Prairie Line.

The American Lake line was built in 1891 to provide a short route from Seattle and Tacoma to Olympia and Grays Harbor. It diverged from the Prarie line at Lakeview. It was built to the branch line standard of closely following the contour of the ground, so on the track chart it is a series of short ups, downs, and occasional near flat spots.

In 1914 the American Lake line was rebuilt from Nisqually to St. Clair and incorporated into the new main line. The new main line was built to the highest standards of the day. The entire 3.5 mile long segment is on .3% grade descending compass northward.

Mac

  • Member since
    December 2017
  • 2,671 posts
Posted by Lithonia Operator on Monday, January 1, 2018 9:12 AM

Perhaps a different track configuration or route profile would be better, but as some have said, this was at a point where trains would shortly need to slow anyway.

But that’s all academic. This was almost certainly a human failure in the engine cab. Why the engineer was lost/impaired/distracted, and what kind of training had been provided, are, to me, the only relevant issues here.

Still in training.


  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Monday, January 1, 2018 9:22 AM

ATSFGuy

So was rerouting the talgo a bad idea?

ATSF,

You decide. Here are some of the facts as I understand them.

Imagine youself as a WSDOT decision maker.

There is an outfit called Sound Transit (ST) that runs commuter trains between Seattle and Tacoma. They are popular and reliable. ST and Pierce County politicians would like to extend service to Lakeview, a suburb about 10 miles south of Tacoma.

The BNSF owns a line from Tacoma, through Lakeview, rejoining its two main track line between Tacoma and Portland at Nisqually. It sees little freight use, but does serve McCord Air Base and Fort Lewis. Both are large military bases that need rail freight access, even if they do not use rail much in the normal course of their operations.

WSDOT spends a bunch of money subsidizing ATK service between Portland OR and Seattle. You have even bought Talgo equipment to cover "your" trains which ATK operates due to its ablity to pay far less than the value of the train slots it uses and the freight train delay it causes. ATK also operates the Coast Starlight, trains 11/14 with conventional high level equipment.

Your trains are slower than driving, and you would like to speed things up. That is why you bought the Talgo trains, they can go faster than conventional trains around some curves. You figure you could save 10-15 minutes running via Lakeview vs. the current route.

When the NP built a new line between Tacoma and Tenino in 1914 it was a double track super railroad, complete with two tunnels at Point Defiance. When rail cars got tall, clearances at the Point Defiance tunnels became inadequate, so the railroad removed one track and shifted the other one to the center of the tunnel. That two miles of single track sometimes causes delays to my passenger trains. I could pay to fix it, but that would be a subsidy to the railroad and I do not want to do that! Horrors!!!

What if I buy the line from Tacoma to Nisqually from BNSF, rebuild the track to current standards for 79 MPH operation, reroute all ATK trains to my new line, which I will call a "high speed route" because it sounds good, AND let ST use it for their commuter trains to Lakeview. It is a tripple play in that the line is a bit shorter, a bit quicker AND I avoid delay at the Point Defiance Tunnels.

There is one itty bitty problem though. The Tacoma end of the line never was, and is not now, designed for that move. To make it work I need to move the ST and ATK depot to the site of the former MILW depot near the Tacoma Dome. That is no big problem, just need a connection between the BNSF main and former MILW line and a new depot.

Beyond the depot things get more expensive. The MILW route goes a couple of blocks west, makes a sharp turn to the south and climbs up a better than 3% grade. I do not want to go south.

I want to continue west and tie into the former NP Prairie line which is descending off the plateau into downtown Tacoma. No problem money wont fix! Fill over the old MILW to raise the grade, then build a long and tall bridge to the hillside where the NP line is, build a bit of new grade and tie into the old NP.

This makes my general profile from former MILW depot site; ascend from about 50 feet to top of plateau at about 300 feet. The NP came off the plateau on a 2.2% grade. I (Mac) suspect that the new line segment is about the same, and that part of the old 2.2% is still in use. Then 15-16 miles of rolling up and down, then about 3 miles of 1.6% descending to the junction at Nisqually. This descent has a 30 MPH 8 degree curve at about the middle of it followed by 30 MPH speed limit to the junction, where my passenger train may have to stop.

The point of this long description is that a trip either way over this new to the locomotive engineers line ends with a steep descent upon which speed must be carefully controlled. This is no great technological feat. The NP, BN and BNSF did it ever since 1873.

So why the wreck of ATK #501 on its maiden trip? Overspeed, 80 MPH, on a 30 MPH curve seems the obvious answer. The question is why.

There are at least two reasonable answers. One, the engineer was lost, did not know where he was in relation to the curve. That is human error. Two, the engineer had not been sufficiently trained on the route, so he did not know where the 30 MPH curve was, perhaps did not know it existed, or thought he could rely on signal indication. If he ws not adequately trained, then that is the fault of WSDOT and/or ATK. That is a management failure.

The fact that the train happened to be Talgo equipment had nothing to do with the wreck, UNLESS, there was a brake failure on the passenger cars, which no one has claimed.

Mac

 

  • Member since
    June 2002
  • 20,096 posts
Posted by daveklepper on Monday, January 1, 2018 10:33 AM

PNW:   I like your analysis.  But is not this also one clear case, like Metro North, and like Frankfort Cuve, a case where PTC would have worked and prevented the tragedy?

  • Member since
    March 2009
  • 2 posts
Posted by cn-22 on Monday, January 1, 2018 10:14 PM

Recording was from the conductor not the engineer.

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Tuesday, January 2, 2018 9:36 AM

The latest I have read is that the second person in the cab was an engineer in training.  Earlier, that person was referred to as a conductor in training to familiarize him or her with the route.  I suspect that with an engineer in training, that person might be running the locomotive under the supervision of the engineer. 

  • Member since
    October 2001
  • From: US
  • 591 posts
Posted by petitnj on Tuesday, January 2, 2018 9:56 AM

One problem with blaming poor training of the engineer is that he didn't know the route. There are clear speed signs along the track and should be called out by the engineer and the other crew in the cab. Never trust your "training" to know the route. That is driving by the seat of your pants and not watching out for every sign and signal. 

  • Member since
    January 2002
  • From: Canterlot
  • 9,575 posts
Posted by zugmann on Tuesday, January 2, 2018 11:37 AM

petitnj
One problem with blaming poor training of the engineer is that he didn't know the route. There are clear speed signs along the track and should be called out by the engineer and the other crew in the cab. Never trust your "training" to know the route. That is driving by the seat of your pants and not watching out for every sign and signal.

Watching out for every sign?  What happens when the sign is knocked down or missing?  When a signal is dark?  You HAVE to know your route.  Being dependent upon signals and signs is downright stupid. 

 

A huge complaint about LEADER/PTC/ACSES/etc. is that guys aren't learning their territory as well as they should.  So when the locomotive or system has an issue and the nanny systems have to be cut out....

It's been fun.  But it isn't much fun anymore.   Signing off for now. 


  

The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.t fun any

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Tuesday, January 2, 2018 3:00 PM

You're an experienced engineer.

You're assigned a route that you know you don't know well enough.

You're told to go 79 MPH with a train full of passengers.

You say: "Yes, sir."  And do it.

 

OR.

 

You're an experienced engineer.

You're assigned a route that you are unaware that you don't know well enough.

Etc.

 

OR.

 

You're an experienced engineer.

You DO know the route well enough.  You just don't bother to do your job.  Well enough.

Etc.

 

Are there other possibilites?

 

Ed

  • Member since
    October 2001
  • From: US
  • 591 posts
Posted by petitnj on Tuesday, January 2, 2018 3:49 PM

This is a matter of how you look at your job. Of course you need to be familiar with the route -- especially with long trains that require complex operations. But you sit in the cab mixing that experience with the attitude that you take in all of the information you have: signs, track profile, signals,... Each one you expect and anticipate. If one is knocked down you have anticipated that it should be there and you check your track profile. And to feed the fire even hotter, the arguements being made are arguments for driverless trains. Ones that take in all of the available information and act accordingly. And familiarity with the job and the route is not an argument here. How you perform your job is the crux of the matter. 

  • Member since
    May 2015
  • 1,836 posts
Posted by 243129 on Tuesday, January 2, 2018 4:31 PM

NorthWest

The engineer is a senior, well respected engineer with plenty of experience. I will not say anything further though I doubt that the crew schedule was altered any for the first run.

 

Senior, experienced engineer and he made 'rookie' errors???

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Wednesday, January 3, 2018 10:43 AM

ATSFGuy

Why would an engineer be going 81 mph in a 30 mph zone?

To me, that makes no sense, and it's a recipe for disaster.

 

One of the things I wonder about in this accident is what was going on in the cab during the 5 minutes prior to the accident.  Presumably, the in-cab recording devices captured this time period, but all that I've seen reported so far is the investigators' determination that the engine crew wasn't using personal electronic devices and the engineer's comment about "overspeed" just before the crash.  Of course, that could simply be because the investigators have already determined there is nothing else on the in-cab recordings that's relevant to the accident. We'll probably find out in due course. 

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Wednesday, January 3, 2018 11:15 AM

We know there were two people in the cab, but we have not been told their identities, nor which one was driving the locomotive.  We have been told that it may be as much as two years before we get the final report.  Back when railroads were in the passenger business, how many years did it take for a class one railroad to investigate a derailment or collision before reaching a conlclusion as to cause? 

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Wednesday, January 3, 2018 11:28 AM

Euclid

We know there were two people in the cab, but we have not been told their identities, nor which one was driving the locomotive.  We have been told that it may be as much as two years before we get the final report.  Back when railroads were in the passenger business, how many years did it take for a class one railroad to investigate a derailment or collision before reaching a conlclusion as to cause? 

I am sure Amtrak has conducted its own internal investigation, but those are not and never have been released to the public in their entirety.  

In my experience internal company-run investigations usually focus on shifting blame to the employee(s) involved, and away from the company itself.  If another problem is discovered steps will likely be taken to quietly correct it, but THAT information will never be distributed even as a memo to employees.  Too embarrassing for those at the top.

There is no doubt that the immediate cause of this disaster was the Engineer's failure to obey speed restrictions, but it looks to me like he was set up to fail due to poor training.  One does not simply run on a mountain-grade railroad without proper familiarization.

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Wednesday, January 3, 2018 11:55 AM

I believe there is similar concern in the NTSB about embarrassment of Amtrak and its public marketing as a competent, safe mode of transportation.  I know the railroads do not release the results of their investigations, but my point is that they are able to figure out what happened in a couple weeks.  They do not need two years.  I wonder how the NTSB is doing with making their arrangements to interview the engineer. 

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Wednesday, January 3, 2018 12:02 PM

SD70Dude
 
Euclid

We know there were two people in the cab, but we have not been told their identities, nor which one was driving the locomotive.  We have been told that it may be as much as two years before we get the final report.  Back when railroads were in the passenger business, how many years did it take for a class one railroad to investigate a derailment or collision before reaching a conlclusion as to cause? 

 

 

I am sure Amtrak has conducted its own internal investigation, but those are not and never have been released to the public in their entirety.  

In my experience internal company-run investigations usually focus on shifting blame to the employee(s) involved, and away from the company itself.  If another problem is discovered steps will likely be taken to quietly correct it, but THAT information will never be distributed even as a memo to employees.  Too embarrassing for those at the top.

There is no doubt that the immediate cause of this disaster was the Engineer's failure to obey speed restrictions, but it looks to me like he was set up to fail due to poor training.  One does not simply run on a mountain-grade railroad without proper familiarization.

 

  The underlying cause could have been lack of "proper familiarization" and/or "poor training" as you say.  But it also could have been distraction.  That's why I'm curious about what the in cab recording devices captured in the 5 minutes prior to the accident.

Note, in this regard, that the NTSB synopsis of the accident says that "a speed board (sign) was located about 2-miles before the curve." That means that both of the engine crew members failed to see the sign.  To me, at least, that suggests that both crew members were distracted by something and failed to see the sign. If so, what?  The in-cab recordings may hold the answer.  We'll see in due course.   

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Wednesday, January 3, 2018 2:28 PM

As I was composing the following, my computer (?) told me this webpage was using significant memory.  And shut me down.  I do note that the letters are lagging as I write.  This happens nowhere else.

 

Now let's try again:

 

It's possible criminal charges are being considered.

NTSB may have been asked by the "potential charger" not to release information.

Let's see what happens if charges are announced or declined.

 

 

 

Ed

  • Member since
    July 2010
  • From: Louisiana
  • 2,310 posts
Posted by Paul of Covington on Wednesday, January 3, 2018 2:33 PM

Falcon48
That means that both of the engine crew members failed to see the sign. To me, at least, that suggests that both crew members were distracted by something and failed to see the sign. If so, what?

   Each other?   In the few seconds that the sign was in view, they may have been discussing instruments, controls, fishing or football.

   This is only vaguely related, but when I used to maintain mainframe computers, I came to the following conclusion:   If a problem takes one person one hour to troubleshoot and repair, it takes two people two hours to fix, and three people will never get it fixed.

_____________ 

  "A stranger's just a friend you ain't met yet." --- Dave Gardner

  • Member since
    July 2010
  • From: Louisiana
  • 2,310 posts
Posted by Paul of Covington on Wednesday, January 3, 2018 2:38 PM

7j43k
I do note that the letters are lagging as I write.

   I just had the same problem with my last post.   I remember this happening ocassionally a few years ago, and that if I "X"ed out and came back, it was OK, so I did and it was. 

_____________ 

  "A stranger's just a friend you ain't met yet." --- Dave Gardner

  • Member since
    October 2012
  • 177 posts
Posted by Jim200 on Wednesday, January 3, 2018 3:49 PM

Prior to the T30/P30 sign, about at the beginning of the multiple tracks, there is a longer sign held by two steel posts which says in capital letters DUPONT. So there were at least 3 signs of possible warning.

One thing not mentioned was that it was rush hour, and there would have been oncoming car lights on Wilmington Dr, sporadically flickering through the trunks of the trees. The NTSB will have to figure out if this was significant or not.

The weather may also have been a factor. From one source, Tacoma was listed as "light rain and fog" with visibility from 5 to 7 miles, from midnight to 2:22 AM, and from 4:00 AM to about 6:00 AM on December 18, 2017. At other times until 8:09 AM, it was listed as cloudy or overcast with 10 mile visibility. Now let's look at a closer report below, from the McChord airbase which is about 5 miles from the 3 possible warning signs. At 7:32 AM it is listed as "light rain" with a visibility of 10 miles and a little more wind speed. Just exactly what the conditions were, needs to be sorted out by the NTSB.

https://www.wunderground.com/history/airport/KTCM/2017/12/18/DailyHistory.html?req_city=&req_state=&req_statename=&reqdb.zip=&reqdb.magic=&reqdb.wmo=

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Wednesday, January 3, 2018 4:42 PM

Paul of Covington

 

 
7j43k
I do note that the letters are lagging as I write.

 

   I just had the same problem with my last post.   I remember this happening ocassionally a few years ago, and that if I "X"ed out and came back, it was OK, so I did and it was. 

 

 

Thanks, Paul

I'll try it if/when it happens again

Not happening now, however.

 

 

Ed

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Wednesday, January 3, 2018 4:46 PM

Jim200

Now let's look at a closer report below, from the McChord airbase which is about 5 miles from the 3 possible warning signs. At 7:32 AM it is listed as "light rain" with a visibility of 10 miles and a little more wind speed. Just exactly what the conditions were, needs to be sorted out by the NTSB.

 

And it was 19 minutes before sunrise.  You can have 10 mile visibility, and not see something 100 feet away.

 

Ed

  • Member since
    May 2015
  • 1,836 posts
Posted by 243129 on Wednesday, January 3, 2018 7:32 PM

When an engineer is considered or pronounced qualified on a piece of territory he has demonstrated that he has indepth and intimate knowledge of that particular terrain. That begs the question as to who qualified this person and what are their qualifications that allow them to examine candidates for qualification?

An experienced qualified engineer when faced with loss of situational awareness due to distraction, fatigue etc. would be inclined to reduce speed until awareness was restored. The 501 engineer did not.

This was by all reports the 'inaugural' run for this service. Where was management? Usually new service is accompanied by the inevitable dog and pony show where management, politicians and sundry 'notables' gather to receive their kudos. At the very least there should have been a road foreman on board when traversing 'new' territory.

As I have stated ad nauseam this and previous disasters point to Amtrak's dangerous hiring and training procedures. They have the unknowing teaching the unknowing. This engineer will be held accountable and fired. Amtrak will emerge   once again unscathed and plod along with their hit and miss trial and error tactics in their attempt to run a railroad. There will be more disasters.

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Wednesday, January 3, 2018 8:08 PM

We once had a thread questioning whether trains needed to slow down in fog.  It proved to be so controversial that it had to be locked.  Generally, it was decided that there was no need to slow down if the engineer knew his territory, and could receive signal indications.  This raised a question of whether an engineer could know his territory well enough to run in totally blinding fog.  Some insisted that this was expected as part of the job of knowing your territory.  It is like a sixth sense. 

So, given that premise, it seems to me that engineers on that Washington line should know their territory well enough to know where that curve is, and they should when to slow down to the speed restriction for that curve; even if the visibility is so bad that he/she cannot read wayside signs.  Unlike signals, the signs are fixed meaning. 

In any case, just a lack of daylight alone certainly should not be an excuse for failing to read a sign.  They run trains in the dark and see things by the headlight.  So I do not see any conceivable excuse for failing slow for the curve because the signs were obscured somehow. 

  • Member since
    September 2017
  • 5,636 posts
Posted by charlie hebdo on Wednesday, January 3, 2018 9:09 PM

19 minutes before sunrise is not very dark.

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy