Commuter & Freight Trains Collide North Of Los Angeles

[Railway Man]

It could still easily be that the UP train passed a clear (green) signal last.   If he has passed a signal before the Metrolink train passed the red it would be too late and too short a time probably to stop.  The line is curved sharply so even if signals are spaced closely they could easily be out of view.  ...

Since the point of impact is about 0.25 miles north of the switch for the siding in question, it would seem that the UP train had already passed the most recent signal before the MetroLink train ran the siding (if indeed that is what heppened). In fact, based on what I know of that area, the southbound signals for that control point (north end of the siding) may have just been coming into view or were already already in view for the UP train. I know the SB signals are to the left of the tracks for southbound trains along that curve. It may even be possible that the UP crew saw that signal as red or green dropping to red and was already taking action, although time will tell. All speculation and "educated guesses" at this time. Jamie

Think about how signal aspects work and signal braking distances work.  Generally braking distances between signals are sufficient for a train with the heaviest possible tons per operative brake to brake to a stop from the most restrictive signal aspect other than red.  (Generally, because there are instances in signal design where signal spacing is less than that, but the explanation is complicated.)

I don't know the specific aspects on this line (and there are many different possibilities) but it is four-aspect signaling from G to R and train operation is similar regardless of whether the second aspect is Y/FG, Y/G, FY/R, or FY.  Let's assume that the aspects from proceed to stop are G-FY-Y-R.  Let's assume two trains are going to meet at a siding in single-track territory.  One train will hold the main and the other take the siding.  Neither has arrived at the siding yet, so one train is lined in and not cleared out the other end, and the other lined down the main but not past the far end of the siding. The aspects the take-siding train will receive are G-FY-Y-RY.  The aspects the hold-main train will receive are G-FY-Y-R. 

Suppose the train holding the main track disregards the speed information of the aspects and proceeds at maximum authorized track speed.  That will not change the aspects delivered to the other train until it passes the red aspect.  If the train entering the siding is complying with its aspects, by the time it reaches the siding it will have slowed to the information required by the aspects, or the maximum authorized turnout speed, whichever is more restrictive.  Only after the train holding the main disregards the red aspect at the control point will the aspects delivered to the train taking the siding drop to red all the way to the next control point.  If the train taking the siding is beyond its Y aspect at the moment when the train holding the main passes its R aspect, the train taking the siding will never receive a signal indication that the track conditions ahead have become unsafe.  Even if the train taking the siding is between its G and FY and sees the FY drop to red in its face, the braking distances are probably too short to make much difference in the closing speed, if the train that's holding the main continues at maximum authorized track speed.  Only in the case where the train taking the siding is two or three blocks away from the unauthorized train will it have sufficient time to call the dispatcher and the dispatcher have sufficient time to call the other train.  Assuming the other train answers the radio then there might be time for both trains to stop short of a collision.  But probably not.

The point is that in the case where one train is obeying its aspects and another is not, the scenario is a meet at a siding in single-track territory, and the meet is as usual as tight as possible to minimize delays to both trains, a wayside signaling system absent a PTC overlay will rarely deliver information to the train that is obeying the aspects to take any sort of meaningful preventative action.

RWM

[Railway Man]

I don't it would be an easy for the UP to have his last passed signal showing clear (green). This exact type of scenario as been well thought out in signaling engineering. I've got to think some more about this, but I am starting to think UP would have seen a downgraded aspect, had begun slowing down, but like you said, not enough time. Gotta think this out some more.

This diagram does not reflect U.S. practice.  It looks like someone's idea of U.S. practice who is thinking from a European or British perspective.  Most of the track cuts are in the wrong places.

RWM

[Southwest Chief]

I agree with Jaime (Pasadena Sub).  The location of the collision tells me that UP had already passed the last block signal.  Remember there are several tunnels that UP had just traveled through.  And as far as I can remember there are no signals in the tunnels.  They are spaced between the open spots. 

According to all reports thus far, UP was lined to proceed.  Metrolink should have held.  Due to the short distance between the control point where Metrolink should have held and the point of collision there was no way UP would have received a red indication after Metrolink overshot the control point.  It was an unfortunate perfect timing accident.

UP would have seen the Metrolink train before they got to the next signal indication. 

Another disturbing report I just heard this morning on a local news radio station (KNX 1070), was a bystander near the Chatsworth station reported the signal indication for Metrolink was green (clear).  However I'm not sure if this person knew exactly what they were looking at.  It's even possible they were looking at a signal in the opposite direction, which likely would have been green for the UP train. 

But this statement about a clear signal along with reports of the Metrolink engineer's experience and professionalism troubles me that perhaps the signals were malfunctioning.  Or even more troubling, the thought of sabotage has crossed my mind.  The tunnels and a regularly scheduled UP freight and Metrolink do make potential targets.  But to shunt the signal wires (much like what was done when the Sunset Limited derailed in Arizona many years ago) to give a green indication would need some time to do and I don't think there was enough time.  Also something like this would have shown up on the dispatcher panel.  In fact a news report mentioned that the dispatcher radioed to Metrolink indicating they had run a red signal.  And allegedly the Metrolink conductor reported back they had just wrecked.

And to further clarify, I'm fairly certain (based on news report photos and my own experience traveling the line) the signal in question that Metrolink went through was an SP style tri-light signal.  This style is very unique, and I believe exclusive to former SP lines here in Southern California.  They look like this:

SP Tri-Light

Oh and some have mentioned automatic systems on this line.  There are no forms of Automatic Train Control on this former SP line.  The only system in place in the local area is Automatic train Stop (ATS) on the former Santa Fe San Diegan line.

[Railway Man]

But this statement about a clear signal along with reports of the Metrolink engineer's experience and professionalism troubles me that perhaps the signals were malfunctioning.  Or even more troubling, the thought of sabotage has crossed my mind.  The tunnels and a regularly scheduled UP freight and Metrolink do make potential targets.  But to shunt the signal wires (much like what was done when the Sunset Limited derailed in Arizona many years ago) to give a green indication would need some time to do and I don't think there was enough time.  Also something like this would have shown up on the dispatcher panel.  In fact a news report mentioned that the dispatcher radioed to Metrolink indicating they had run a red signal.  And allegedly the Metrolink conductor reported back they had just wrecked.

Let me assure you from experience that the amount of knowledge of signal systems that is required to intentionally deliver a false-clear aspect rules out about 99.9999% of the human race.  There are millions of vastly less technically demanding ways to deliver harm to a railway system.  No one has mentioned terrorism as a possible cause, have they?

All that a CTC dispatching display will indicate is that a train has proceeded beyond its authorized limits.  The dispatcher cannot "see" signal aspects -- ANY signal aspect.  The dispatcher's display merely shows track occupancies, train IDs, and "signal requests" initiated by the dispatcher.  The signaling system in the field acts on those requests if it is safe to do so. 

If you were looking at the display in the right place at the right moment that a train went by a red signal at speed, you would see an occupancy appear in the OS (the area between the insulated joints at the control point), and, if the block beyond the OS had no occupancy, an occupancy would appear there.  If the block beyond the OS was already occupied it would not change but the second train ID would probably jump to that block.  The display is not-quite real time because it takes time for the signal system to transmit the information to the dispatching office.  Sometimes it's within a second but in some locations it may be as much as 10 seconds lag time.  In many cases the dispatcher display won't show anything meaningful.  Some signaling systems send an alert to the dispatcher when an unexpected occupancy occurs.

And to further clarify, I'm fairly certain (based on news report photos and my own experience traveling the line) the signal in question that Metrolink went through was an SP style tri-light signal.  This style is very unique, and I believe exclusive to former SP lines here in Southern California. 

This is not unique to SP; it's a US&S style that was standard on Rock Island and New York Central.  There are some other similar signal heads that have been made by other manufacturers and used on other railroads, too.

RWM

[ChuckCobleigh]

my step-dad works for amtrak and is a huge railfan and has alot of friends in high places and he thinks that the engineer of metrolink 111 had either a siezure or heart attack and so does another metrolink conductor.

possibly. but we won't know if that happened unless they do an autopsy.

I'd bet the farm that they do an autopsy given the horrible results of this crash (it may even be the law that an autopsy be done in such cases)...so we will definetly find out if the issue was one related to his health or if it was caused by substance abuse.

In California, unless the decedent is under the immediate care of a doctor who can sign a death certificate, it is a coroner's case, so an autopsy will be performed. 

[SFbrkmn]

This story did not receive national coverage Fri night acct of the hurricane. I listened to live continouis coverage on KNX radio website. They stated that close to 20 dead. I don,t know the LA area that well, where was this in location to LAUPT and how miles away?

[Southwest Chief]

No one has mentioned terrorism as a possible cause, have they?

It's been hinted to on some local media outlets and can't be ruled out at this time.  Partly why the FBI is investigating as well.  And I agree with you that very few individuals would know how to do it, and it would be next to impossible to do in this location.  Also if there was any evidence of signal shunting it would have been reported early on given the fast reporting and leaks of information thus far.  But again it hasn't been ruled out (or at least reported as being ruled out) as I type this.

[Anonymous]

If the engineer missed a red light due to distraction, how long would that distraction have needed to be? 

[eolafan]

If the engineer missed a red light due to distraction, how long would that distraction have needed to be? 

If I were riding down the road at, say about 45 mph and the traffic light were green and I took my eyes off of the signal to send or read a text message (especially sending one) of about two lines in length, the signal could easily turn red in my face and I would not even have the time to react as traffic signals can turn red in a virtual instant...as can a track signal if another train (in this case the U.P. freight) crossed in front of the Metrolink train...in other words, my guess is that it could have very easily have happened in about fifteen seconds or less. I may be wrong though, so others may have opposing ideas to offer.

[MP57313]

This story did not receive national coverage Fri night acct of the hurricane. <snip> I don,t know the LA area that well, where was this in location to LAUPT and how miles away?

According to an Amtrak timetable, Chatsworth is 28 miles from LAUS (as it is called now).  Chatsworth, like Van Nuys and Northridge, is not a separate incorporated city but rather a district within the City of Los Angeles.

[jackrmeyer]

my step-dad works for amtrak and is a huge railfan and has alot of friends in high places and he thinks that the engineer of metrolink 111 had either a siezure or heart attack and so does another metrolink conductor.

possibly. but we won't know if that happened unless they do an autopsy.

I'd bet the farm that they do an autopsy given the horrible results of this crash (it may even be the law that an autopsy be done in such cases)...so we will definetly find out if the issue was one related to his health or if it was caused by substance abuse.

In California, unless the decedent is under the immediate care of a doctor who can sign a death certificate, it is a coroner's case, so an autopsy will be performed. 

Okay, so here's my ignorance showing:  What is the physical mechanism (i.e. accelerator pedal, hand throttle, etc.) of train operation on a MetroLink train?  I thought all such devices are "dead-man" (sorry) operation, that stop operating immediately when positive pressure is removed (like a Bic lighter that goes out when you stop depressing the thumb valve or drop it).  If so, it seems unlikely that an engineer would be physically disabled in such a manner as to continue operating the control(s). 

Secondarily:  Could braking be tied to such a device, so that, when not positively activated, braking would automatically ensue?  I.E., let go and the thing stops (at least as fast as is reasonable with something that massive), or at least slows (bleeds off speed)?

[BaltACD]

If the engineer missed a red light due to distraction, how long would that distraction have needed to be? 

If I were riding down the road at, say about 45 mph and the traffic light were green and I took my eyes off of the signal to send or read a text message (especially sending one) of about two lines in length, the signal could easily turn red in my face and I would not even have the time to react as traffic signals can turn red in a virtual instant...as can a track signal if another train (in this case the U.P. freight) crossed in front of the Metrolink train...in other words, my guess is that it could have very easily have happened in about fifteen seconds or less. I may be wrong though, so others may have opposing ideas to offer.

From Metrolink stating their engineer was at fault it is evident that the signal for the Metrolink train was Red and had never been cleared for the movement of the train.  If a lined signal is to be changed ahead of the train that it has been lined for, the control operator (operator or dispatcher) must contact the train it was lined for and question them to see if they can make a safe stop (not placing train in emergency) to get train stopped prior to the signal in question.  If the train responds that they can make a safe stop prior to the signal, then the signal can be taken down....when a signal is taken down there is a timer that automatically runs to prevent lining another route (depending on the territory the timer could be anywhere from 5 to 15 minutes depending upon the territory involved) while the timer is running the signal in the field is at STOP.  Signals cannot be taken down and immediately lined for a different route

From the currently available information, this is nothing more and nothing less than a Man Failure incident.  Human Beings are fail able, and in this case one failed with catastrophic consequences.  While we all would like to think that this is a very rare occurrence; the human fallibility with Stop Signals is much more common than rail management would like to see.   Efficiency Testing.  Efficiency Testing.  Efficiency Testing!

[Railway Man]

Okay, so here's my ignorance showing:  What is the physical mechanism (i.e. accelerator pedal, hand throttle, etc.) of train operation on a MetroLink train?  I thought all such devices are "dead-man" (sorry) operation, that stop operating immediately when positive pressure is removed (like a Bic lighter that goes out when you stop depressing the thumb valve or drop it).  If so, it seems unlikely that an engineer would be physically disabled in such a manner as to continue operating the control(s). 

Secondarily:  Could braking be tied to such a device, so that, when not positively activated, braking would automatically ensue?  I.E., let go and the thing stops (at least as fast as is reasonable with something that massive), or at least slows (bleeds off speed)?

Metrolink locomotives are controlled like all others, with throttle and brake levers operated by hand.

Dead-man pedals are long gone -- at least on every railroad I've worked for or with.  Instead, an alerter is installed (and integrated with the braking system) on all locomotives used in passenger service (FRA requirement).  There is no FRA requirement for an alerter on freight locomotives but most (all?) locomotives built within the last 10 years are equipped with such.

Here's the relevant section from 49 CFR on how alerters work: 

"Alerter means a device or system installed in the locomotive cab to promote continuous, active locomotive engineer attentiveness by monitoring select locomotive engineer-induced control activities. If fluctuation of a monitored locomotive engineer-induced control activity is not detected within a predetermined time, a sequence of audible and visual alarms is activated so as to progressively prompt a response by the locomotive engineer.  Failure by the locomotive engineer to institute a change of state in a monitored control, or acknowledge the alerter alarm activity through a manual reset provision, results in a penalty brake application that brings the locomotive or train to a stop."

RWM

[TH&B]

If the engineer missed a red light due to distraction, how long would that distraction have needed to be? 

If I were riding down the road at, say about 45 mph and the traffic light were green and I took my eyes off of the signal to send or read a text message (especially sending one) of about two lines in length, the signal could easily turn red in my face and I would not even have the time to react as traffic signals can turn red in a virtual instant...as can a track signal if another train (in this case the U.P. freight) crossed in front of the Metrolink train...in other words, my guess is that it could have very easily have happened in about fifteen seconds or less. I may be wrong though, so others may have opposing ideas to offer.

 

That is not likely the scenario.   The light the Metrolink train supposedly ran would have been RED several minutes while the UP train was in the blocks ahead travalling.   So that type of distarction you are describing shouldn't apply.  But.... for the length of time the train was stopped at the station the light could have changed, and then combining it with text messaging then there would be enough time for the light to change, and now we are talking several minutes.

[jackrmeyer]

Okay, so here's my ignorance showing:  What is the physical mechanism (i.e. accelerator pedal, hand throttle, etc.) of train operation on a MetroLink train?  I thought all such devices are "dead-man" (sorry) operation, that stop operating immediately when positive pressure is removed (like a Bic lighter that goes out when you stop depressing the thumb valve or drop it).  If so, it seems unlikely that an engineer would be physically disabled in such a manner as to continue operating the control(s). 

Secondarily:  Could braking be tied to such a device, so that, when not positively activated, braking would automatically ensue?  I.E., let go and the thing stops (at least as fast as is reasonable with something that massive), or at least slows (bleeds off speed)?

Metrolink locomotives are controlled like all others, with throttle and brake levers operated by hand.

Dead-man pedals are long gone -- at least on every railroad I've worked for or with.  Instead, an alerter is installed (and integrated with the braking system) on all locomotives used in passenger service (FRA requirement).  There is no FRA requirement for an alerter on freight locomotives but most (all?) locomotives built within the last 10 years are equipped with such.

Here's the relevant section from 49 CFR on how alerters work: 

"Alerter means a device or system installed in the locomotive cab to promote continuous, active locomotive engineer attentiveness by monitoring select locomotive engineer-induced control activities. If fluctuation of a monitored locomotive engineer-induced control activity is not detected within a predetermined time, a sequence of audible and visual alarms is activated so as to progressively prompt a response by the locomotive engineer.  Failure by the locomotive engineer to institute a change of state in a monitored control, or acknowledge the alerter alarm activity through a manual reset provision, results in a penalty brake application that brings the locomotive or train to a stop."

RWM

 Thank you, very informative!

[Railway Man]

First, there is no evidence anyone was text-messaging. Only a report of dubious authenticity.

Second, about the only thing in common between wayside signaling systems and traffic light systems is they both use colored lights to display information.  Analogies between them are usually false.

Third, the operating rules, if complied with, provide ample time to safely operate a train.  You cannot approach at a high rate of speed a signal indicating stop on a railway, a signal at which you cannot stop your train short of the signal, without previously having proceeded through an aspect progression which you have ignored. 

RWM

[TH&B]

 

Third, the operating rules, if complied with, provide ample time to safely operate a train.  You cannot approach at a high rate of speed a signal indicating stop on a railway, a signal at which you cannot stop your train short of the signal, without previously having proceeded through an aspect progression which you have ignored. 

RWM

 

The staion stop could be considered delayed in the block.  Then forgotten, passed out, or destarcted could be some possiblities. 

[arkansasrailfan]

The numbers are now at 25 dead, 135 injured. They think that will not find anymore survivors.

[CSXDixieLine]

If the engineer missed a red light due to distraction, how long would that distraction have needed to be? 

If I were riding down the road at, say about 45 mph and the traffic light were green and I took my eyes off of the signal to send or read a text message (especially sending one) of about two lines in length, the signal could easily turn red in my face and I would not even have the time to react as traffic signals can turn red in a virtual instant...as can a track signal if another train (in this case the U.P. freight) crossed in front of the Metrolink train...in other words, my guess is that it could have very easily have happened in about fifteen seconds or less. I may be wrong though, so others may have opposing ideas to offer.

Also realize that in this situation, it is highly unlikely the signal changed from green to red. If the UP train was indeed cleared southbound, this would have probably happened some time before the collision occurred and probably some time before the MetroLink train even arrived at the Chatsworth station. Seems likely that the northbound signal prior to Chatsworth station would have been approach or diverging approach (I don't know the exact track configuration there or which track the Metro train was on at the station) and the signal at the north end of the siding would have been stop for the entire time the train was in the station and while the train was approaching the signals from the south. It does seem plausable that the engineer could have assumed the next signal was green and became distracted somehow when the actual signal should have been observed.

Again, as with all my other posts on this subject, this is all speculation on my behalf based on what I have observed via media reports. The only reason I am even commenting on what could have happened is that we have MetroLink on record stating that their engineer ran a red signal--I can not recall any such statements ever being made so quickly following a major rail, aviation, marine or other transportation accident.

Jamie

[beaulieu]

If you want an overview of the location where the accident occured plug

"Heather Lee Lane, Chatsworth, CA " 

into Google Earth. The curve right above the lane is where the collision occurred. The UP train had just come out of the tunnel and the collision occurred right in the middle of the big curve.

There was only one signal between the Metrolink train and the location of the collision after it made its station stop at Chatsworth Station, which is within the length of the siding. The northbound (RR West) home signal for CP Toponaga (North switch of the siding) is right about where the oval shaped pond is on the right of the tracks.

[Awesome!]

I think based on the information we've heard so far, it appears that Metrolink may have to implement three safety changes:

1) Replace the passenger car with cab controls with a true cab car, perhaps a retired EMD F40PH with its prime mover and tractor motors removed and replaced with a standard wheels, ballast weight on the frame and a reinforced cab for better survivability in an accident. I believe they're starting to do this on Chicago's METRA system.

2) Put in a warning system on the engineer's controls so that if the train approaches or goes into a Red block it gives off a very loud warning signal.

3) Implement some sort of automatic brake application system if a train approaches and goes into a Red block.

NYC Subway has the automatic brake application system and is very effective. But i think is very costly to apply.

[BaltACD]

 

Third, the operating rules, if complied with, provide ample time to safely operate a train.  You cannot approach at a high rate of speed a signal indicating stop on a railway, a signal at which you cannot stop your train short of the signal, without previously having proceeded through an aspect progression which you have ignored. 

RWM

 

The staion stop could be considered delayed in the block.  Then forgotten, passed out, or destarcted could be some possiblities. 

Historically, Passenger trains making scheduled station stops are not considered as being delayed in the block.  CSX after the 1996 Silver Spring incident changed its rules to require Restricted Speed after a station stop until the next signal comes into view, then proceed in accordance with that signal indication.

The 'black box' data in concert with the signal system data showed, that the MARC train had departed it's station stop as if it had not passed the preceeding 'Approach' Signal, accelerating back to track speed immediately from the station stop, rather than approaching the next signal prepared to stop as required by the signal indication

[K. P. Harrier]

A few hours ago I heard a very ominous word of mouth report:  Many of those still hospitalized are not expect to survive!  That's really bad news!

[arkansasrailfan]

yeah, I heard that too. nobody really could survive when the car you are in is compacted partaily by a diesel that can't fit inside.

[TH&B]

This is not acceptable.  s*** is gonna hit the fan on this one.  And a question is how are other comuter trains controled ? How are they safer in other cities ? How is GO Transit safer for example? They were the first to use the this equipement.  How is Chicago safer ?

 

These are fair questions now. 

[arkansasrailfan]

Metra and VRE use differnet cars(at least they used to or still do)

[EJE818]

Metra does not use the same kind of commuter cars as Metrolink. Metra has had trains derail at higher speeds than this Metrolink train was going and the damage to the cars was not near as significant. This accident will not only make headlines due to the possibility of text messaging, but it will also bring up if there are any safety flaws in Metrolink's cars. One thing is for sure, for going only 40MPH, it is amazing how much damage occured. Imagine what could've happened if the train has doing 70-80MPH. I think Metra also has only one person in the cab, so other than the type of commuter cars, Metra is run very similar to Metrolink.

[Flashwave]

This is not acceptable.  s*** is gonna hit the fan on this one.  And a question is how are other comuter trains controled ? How are they safer in other cities ? How is GO Transit safer for example? They were the first to use the this equipement.  How is Chicago safer ?

 

These are fair questions now. 

 Fate hasn't gptten to their names on the list yet. Seriously. Accidents happen. The only major differences are signalling, drivers, and make of cars, depending on the rrs being compared. Signals are all checked by a fedaral inspection. Which pitches those out. car makes that are the same are just that, the differences between a commuter bilevel and a Bombardiar are the same as a CVhevy vs. a Ford.  Guess what's left? And guess what's universal in planes, boats, and taxis

Metra does not use the same kind of commuter cars as Metrolink. Metra has had trains derail at higher speeds than this Metrolink train was going and the damage to the cars was not near as significant. This accident will not only make headlines due to the possibility of text messaging, but it will also bring up if there are any safety flaws in Metrolink's cars. One thing is for sure, for going only 40MPH, it is amazing how much damage occured. Imagine what could've happened if the train has doing 70-80MPH. I think Metra also has only one person in the cab, so other than the type of commuter cars, Metra is run very similar to Metrolink.

They may be different cars, but derailment is different than the head-on collision. I sincerely doubt that flaws are Metrolink specific, we could drag the whole of commuters down.

[greyhounds]

A few hours ago I heard a very ominous word of mouth report:  Many of those still hospitalized are not expect to survive!  That's really bad news!

I really pray and hope that's just another unfounded rummor.

Here's some hope.  If a wounded soldier makes it to a US Army hospital alive, he/she has a better than 98% chance of living.  The LA hospitals will do as well, hopefully.

[MJChittick]

One thing is for sure, for going only 40MPH, it is amazing how much damage occured. Imagine what could've happened if the train has doing 70-80MPH.

It's been reported that BOTH trains were going approximately 40 MPH!!!  That's the equivalent of an 80 MPH single-train derailment, and that's the kind of damage seen, plus the effect of the engine telescoping back into the first coach.