Trains.com

New story on the Northridge Metrolink crash.

8404 views
86 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Monday, December 22, 2008 5:04 PM

Well I got the answer to my question about how a signal that is supposed to show green could show red.  I conclude that in this case, with the other factors related to the final signal, there would be close to zero probability for that signal to display green.  I would conclude that a false green under the same conditions would likely be unprecedented in the U.S., depending on how it is qualified.  It is probably fair to guess that anything that could happen did happen if you include the entire history of signals, especially considering the fact that earlier signals included more mechanical parts.

 

In more recent times, incidents of false greens have occurred due to a failure to shunt the rails due to sand on the rails or disk brakes on single RDC units.  In this Chatsworth wreck, however, it seems improbable that a failure to shunt occurred with a freight train.  Moreover, there are several other occurrences that do not correlate with a failure to shunt.  So I conclude that a false green would have been almost impossible, however if anyone has an explanation of how it would be possible, I would like to hear it. 

 

There is nothing improbable about text messaging causing distraction, and Sanchez was doing a lot of texting right up to the final seconds before impact.  What is somewhat surprising though, is how long the distraction would have had to be in order to miss the red signal.  If the red signal was visible from the station, I speculate that it was also visible for some distance prior to reaching the station.  So the engineer would have been slowing for the station stop, then standing at the station for a short time, and then leaving the station and accelerating to full speed as he approached and passed the final signal. 

 

This process must have consumed at least a few minutes, all the while with the red signal within sight of the engineer.  And during this time, he had to stop and start the train, plus he also (I assume) had some kind of communication with the conductor in order to leave the station.  So these activities would have required him to devote at least some of his attention to the work at hand.  Furthermore, he had two prior signals warning him that the final light was red.

 

So it seems like he experienced an awfully long span of distraction.  People text all the time while driving cars, and it does create enough distraction to cause crashes.  But I picture the distraction in those cases as lasting only a few seconds because that is all it takes to lose control of a car.  But it takes longer to lose control of a locomotive, so those short distractions that cause car crashes would not seem to hardly interfere with operating a locomotive unless it is during switching.  Therefore, apparently, text messaging and similar distractions lengthen according to the opportunity for them to do so, and a locomotive, not needing to be steered, offers a lot more opportunity than a car does.      

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Monday, December 22, 2008 10:46 AM

edblysard

 

I agree that people sometimes see what they want to see.

During several investigations, I have had eyewitnesses swear events occurred in a pattern or a manner that, for lack of a better description, would be improbable or unbelievable.

Yet these people can, and have passed polygraph test, proving that they believe, completely, what they are sure they saw or "know", when every bit of physical evidence says other wise.

I was not assuming you were defending the engineer...it would be a moot point anyway, considering the outcome of his actions.

I was simply pointing out he had already violated one rule by ignoring the first warning signal aspect...assuming he even saw that signal...for all we know, he was running on his own, internal and personal autopilot....every one of us has, at some point, driven to work, parked, got out of our car, and realized we don't remember a thing about the drive in...simply because we have driven the same route so many times.

As for the competing theories...

I agree, each makes the other weaker...but I have a sneaky suspicion the media simply wants to downplay the cell phone involvement, and find any other acceptable reason for this wreck, because cell phone companies are big advertisers, both in print, radio and TV...I wonder who owns the LA times, and if they have any connections to a cell phone company or service provider.

Think about how much of a status symbol a cell phone is in our culture, and how much money it generates for all parties involved.

Personally, I think the guy simply got used to things being the same day after day, expected no changes in that routine, allowed himself to be distracted, then caught up in a text/conversation, and simply forgot or never saw the signal(s)...he was "driving to work" so to speak.

  This is eerily similar to a number of train accidents which, at least on their faces, seem to have been caused by an engineer seeing what he expected to see because what he expected was what normally happened.  In the case of Chatsworth, my understanding is that the passenger train normally would received a green signal and the freight train (if it were in the area) would be held at the prior siding.  That, of course, didn't happen on the day of the accident.  Other accidents like this include the MARC accident mentioned by one of the other writers, and a head-on collision on the South Shore Line in Gary, Indiana about 10 years ago (the train that normally received a clear signal at a gauntlet bridge received a stop signal the day of the accident).

Still, there are enough people saying that the signal for the passenger train was "green" at Chatsworth that the possibility can't just be disregarded.  My own hunch is that the people reporting this did not closely observe the signal at the time, and that their recollections have been influenced by what others are saying.  This isn't dishonesty - they undoubtedly believe that they saw what they are saying they saw.  You often see this kind of thing with eyewitnesses to accidents.  For example, in a grade crossing collision, you can have mulitple witnesses saying that the crossing signals weren't working even though the camera on the locomotive shows they were operating perfectly. 

I'm sure the NTSB report on the accident is going to address this point in some detail.  NTSB and FRA personnel very carefully inspected the signal system after the accident so, had the signal been displaying a "false proceed", they would have presumably have determined this condition and the cause.    

 

 

  • Member since
    March 2005
  • From: Brewster, NY
  • 648 posts
Posted by Dutchrailnut on Wednesday, December 17, 2008 7:18 AM

Suicide by train does not add up.

 Why???  for several reasons.

 Mr Sanchez messages to his railfan buddies were upbeat and  with plans to meet them.

 Mr Sanchez ordered a luch from  a Deli at his turn around point.

 Mr Sanchez had no guarantee there would be another train after that red signal.

 signals can be held red for other reasons, like track work, signal work or simply time.

 

  • Member since
    August 2006
  • From: South Dakota
  • 1,592 posts
Posted by Dakguy201 on Wednesday, December 17, 2008 5:04 AM

It has been reported that the UP went into emergency 4.6 seconds before the crash, but the Metrolink never braked.   Presumably then, that is roughly the amount of time the trains were visible to one another on the curve.  Even if distracted by something else, I find it hard to believe you would not notice an object the size of a locomotive and painted Armour Yellow coming at you out of the corner of your eye.

I'll wait for the official report, but in my own mind I haven't ruled out sucicide by train.

 

  • Member since
    January 2003
  • From: Kenosha, WI
  • 6,567 posts
Posted by zardoz on Tuesday, December 16, 2008 7:56 PM

dehusman

KCSfan
I have a second question regarding delayed-in-block. Am I right or wrong in thinking that a scheduled station stop is not considered a delay-in-block? If not then that rule is not applicable and there's no need in discussing it.

The delayed in the block rule is silent on station stops (its quoted a couple pages ago).

The requirements under the GCOR rule 9.9 in effect are that in CTC if the train has entered the block on a proceed indication that does not require restricted speed and the train stops or its speed drops below 10 mph, the train must proceed prepared to stop at the next signal.  The train must maintain that speed until can see the next signal, the track is clear to the signal and the next signal displays a proceed indication.

So unless Metrolink has a General Order or special instruction modifying the rules there is absolutely nothing in the rule about a station stop and yes it plays no part in the discussion so there is no point debating it.

Metra addressed the issue back in the early 1990's. There was an actual General Order / Special Order (I do not remember which) that specifically exempted station stops from the "delayed in block" rule.

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Tuesday, December 16, 2008 4:01 PM

Regarding the Metrolink train, what was the means of communication for the conductor to tell the engineer to leave the station?

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Monday, December 15, 2008 6:33 PM

As I understand it, the preceding solid yellow signal, the position of the siding switch, and the location of the U.P. train all would require the final signal to be red.  However, some witnesses say it was green.  From what I understand, the probability of that signal being green would be very small, but still possible.  I understand that there is a fail-safe feature in the control of signals, but it can’t be 100% infallible if a false green is possible. 

 

So here is my question: 

 

What has to happen in a signal mechanism and/or circuitry for the signal to display green when all related factors say it should display red?     

  • Member since
    June 2008
  • From: Salem, Oregon
  • 189 posts
Posted by NP Red on Monday, December 15, 2008 6:16 PM

dehusman
Yes the dispatcher may have been able to tell that the switch had been run through.  Switches have indicators to tell if the switch is lined and locked for a particular route.  When the switch is not lined and locked the OS will flash and the signals will not clear for the route through the switch

 

Thanks for clearing that up for me. I was wondering if there was something I wasn't thinking about.

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Monday, December 15, 2008 2:25 PM

KCSfan
I take it that the purpose of the rule requiring the engineer to call out signals is to provide an audio record that he has observed and acknowledged each signal.

The purpose of the annoucing the signals is to broadcast the position of the train to any other trains or employees in the area.  There is no "audio record" of local radio communications.  The only audio record is of communications between the dispatcher and whoever he talks to or whatever radio/phone he has on at the moment.  If he wasn't listening to the Chatsworth radio, then there is NO record of the radio conversations at Chatsworth.

KCSfan
I have a second question regarding delayed-in-block. Am I right or wrong in thinking that a scheduled station stop is not considered a delay-in-block? If not then that rule is not applicable and there's no need in discussing it.

The delayed in the block rule is silent on station stops (its quoted a couple pages ago).

The requirements under the GCOR rule 9.9 in effect are that in CTC if the train has entered the block on a proceed indication that does not require restricted speed and the train stops or its speed drops below 10 mph, the train must proceed prepared to stop at the next signal.  The train must maintain that speed until can see the next signal, the track is clear to the signal and the next signal displays a proceed indication.

So unless Metrolink has a General Order or special instruction modifying the rules there is absolutely nothing in the rule about a station stop and yes it plays no part in the discussion so there is no point debating it.

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Monday, December 15, 2008 2:12 PM

henry6
NP Red

RWM

So back to my original question. Then there would have been no indication that the switch was run because the block was already "shunted" by the oncoming train, right?

Right.  The dispatcher's board would show the block occupied by the freight and could not descern a second train enterning from the opposite end. 

Yes the dispatcher may have been able to tell that the switch had been run through.  Switches have indicators to tell if the switch is lined and locked for a particular route.  When the switch is not lined and locked the OS will flash and the signals will not clear for the route through the switch.  After a dual control switch at a CP is run through, normall the mechanism is damaged and so it is not "lined and locked" causing the display on the dispatcher's panel to flash.  The switch is considered "out of correspondence".  A switch going out of correspondence under a train is usualy always a very, very bad sign.  In this particular case, by the time the dispatcher could identify there was a problem the collision had probably occurred.

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    November 2007
  • 2,989 posts
Posted by Railway Man on Monday, December 15, 2008 1:57 PM

henry6

NP Red

RWM

So back to my original question. Then there would have been no indication that the switch was run because the block was already "shunted" by the oncoming train, right?

Right.  The dispatcher's board would show the block occupied by the freight and could not descern a second train enterning from the opposite end.  However, when the first block (occuped by the passenger train in this instance) was vacated the dispatchers' board should have shown that.  But remember in this instance we are talking about events that happened in about 20 seconds so it is probably not pertinent as to what the dispatcher could determine or react to; it was too late.

No sir.  Look at my previous post this morning.  A train passing end of siding, whether on main track or siding, first occupies the block that consists of the control point itself, aka the OS -- the distance of track between the facing signals, before it occupies the block beyond the control point.  Each OS block has its own track light (the track schematic graphic turns from white to red) and it turns color on the dispatcher's screen.  It is as obvious.to the dispatcher looking at the screen that the OS is indicating an occupancy.

An occupancy is associated with a train ID.  The ID "follows" the occupancy on the dispatcher's screen as an assumption and convenience.  The signal system of course does not know which train is where, nor does it need to know that.  It only looks at occupancies and uses its logic to display signal indications based on what's occupied or what's not.  When a train passes a control point on the dispatcher's console and travels beyond the OS to enter the next block beyond the CP, the ID jumps to follow it, usually.  But sometimes not.  The ID is just a convenience for the dispatcher and conveys no authority; if the dispatcher really want to know where the train is, the dispatcher calls it on the radio and asks it.   

Thus a train passing an end of siding into a block occupied by another train, has this sequence of events on the dispatcher's screen (on a Digicon console, which is what Metrolink uses; others are similar):

  1. Before the event, the track lights on either side of the OS are red indicating occupancy.  The OS is white.  Both occupancies have a train ID associated with them -- a text graphic immediately above or below the track line.
  2. The train passing the end of siding turns the OS block red.  Now one of the blocks on the siding side of the OS is red (either the main or siding, depending which one the train is leaving from), and the block on the single-track side of the OS is red.
  3. After the head-end of the train travels through the OS and enters the single-track block beyond, its ID may jump over to that block and stack with the ID that was already there for the train already there.

A dispatcher may or may not see this immediately as it occurs; depends a lot on latency of signal system communications and where his or her eyes are at that instant.  And in some instances it may provide time for the dispatcher to radio the train that has exceeded its authority and ask where it is before that train gets itself into trouble, if it really has exceeded its authority.  It could also just be a transient causing the shunt, too.  Signal systems are fail-safe.

RWM

  • Member since
    July 2006
  • 2,535 posts
Posted by KCSfan on Monday, December 15, 2008 1:06 PM

K. P. Harrier
At variance with this thread so far, the conductor, as the sole surviving crew member of the September Metrolink crash, claims the last two signals were green.  And that is why the last two signals were not called out over the radio.
 
As stated by officials of the NTSB, it is a violation of the Metrolink rules to not call out green signals.  But, it appears to be a cultural thing; with most railroaders not calling out clear signals, possibly inspired by previous railroad employment practices.
 
It would seem to me that the tell-tale indicator would be the dispatcher’s and locomotives’ recordings.  If in fact green signals are customarily not called out, there should be few signal call-outs on all the tapes related to the disastrous run of Train 111.
 

Perhaps someone can explain how the conductor knew "the last two signals were green" if they were not called out by the engineer. (I presume the reference is to the two signals preceding the one displaying red at the crash site.) Maybe one, but certainly not both, signals could be seen by the conductor as he stood on the station platform at Chatsworth. I take it that the purpose of the rule requiring the engineer to call out signals is to provide an audio record that he has observed and acknowledged each signal. The conductor would not normally be in a position to see the signals to confirm that the engineer has correctly called out the aspects they display.

I have a second question regarding delayed-in-block. Am I right or wrong in thinking that a scheduled station stop is not considered a delay-in-block? If not then that rule is not applicable and there's no need in discussing it.

Mark

  • Member since
    December 2001
  • 8,156 posts
Posted by henry6 on Monday, December 15, 2008 12:16 PM

NP Red

RWM

So back to my original question. Then there would have been no indication that the switch was run because the block was already "shunted" by the oncoming train, right?

Right.  The dispatcher's board would show the block occupied by the freight and could not descern a second train enterning from the opposite end.  However, when the first block (occuped by the passenger train in this instance) was vacated the dispatchers' board should have shown that.  But remember in this instance we are talking about events that happened in about 20 seconds so it is probably not pertinent as to what the dispatcher could determine or react to; it was too late.

RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.

  • Member since
    June 2008
  • From: Salem, Oregon
  • 189 posts
Posted by NP Red on Monday, December 15, 2008 11:23 AM

Railway Man
Signal systems detect a "shunt" in a track circuit.  A track circuit consists of two rails between two sets of insulated joints.

So back to my original question. Then there would have been no indication that the switch was run because the block was already "shunted" by the oncoming train, right?

  • Member since
    November 2007
  • 2,989 posts
Posted by Railway Man on Monday, December 15, 2008 11:09 AM

NP Red

I'm only quessing but I'm not sure that the dispatcher would see anything different if the red was ran because another train was already in the block just around the corner. How would the system "know" that the switch was run. At this point in time, can the "system" tell if a block is occupied at one end or the other? I know that train speed and direction are calculated in the timing of the gates at grade crossings but does anybody know if that applies to block occupation?

Signal systems detect a "shunt" in a track circuit.  A track circuit consists of two rails between two sets of insulated joints.  Generally a track circuit is the same distance as a block.  A block generally extends between two signals.  A shunt can be a train entering the block, or a broken rail, or a jumper cable between the two rails.  Signal systems are not designed to detect where in the block the shunt occurs.

A grade-crossing signal system using a Grade Crossing Predictor (not all grade crossing signal systems are so equipped) measures change in frequency of a signal compared to time, and from that calculates speed of the train.  If the speed is negative, the train is moving away from the crossing.  Your question implies if that could be done for a block signal system.  The answer is yes, but it would be of no value.

RWM

  • Member since
    November 2007
  • 2,989 posts
Posted by Railway Man on Monday, December 15, 2008 11:02 AM

kolechovski

The signal at the switch the Metro train ran through should have been an absolute signal, right? 

It is.

The dispatcher should have seen that. 

Dispatchers cannot see signal indications.

What would he see if the train runs through that? 

Between 1 and 15 seconds after a train enters a control point (the amount of time depending upon the latency of the signal and communications system), the dispatcher's screen will show an occupancy in the control point.  Occupancies, or "track lights," are shown as a change in color in the track line graphic, usually from white (controlled track, no occupancy) to red (controlled track, occupancy).  After a train enters the next block beyond a control point, that block's track light will also turn from white to red, if it is unoccupied.  But in this case it was occupied so it would already be red.  If the dispatcher happens to be looking at that control point on his screen at that moment, he would see it turn color and know that something was abnormal.  Otherwise during his next normal scan of the console screens he would likely see the color change as it would be an unexpected condition. 

Does anything come up on his screen in that case?  I know many different areas have many different systems, and even what happens after a train runs a red depends on where it's at and teh system in place.

Nothing other than the track light.  The train ID may or may not move with the train in a case like this.  Some computer-aided dispatching systems now have as an option an error message appearing on the screen to alert the dispatcher that an unexpected condition has occurred.  The dispatcher has to click to acknowledge the message before he can do anything else with the console.  The error messages are more trouble than they are worth, in my opinion, as there are hundreds of events in a day that can trigger them, and virtually all of them are normal conditions.

And I still wonder if anyone knows exactly what signals were what color, and whether teh engineer was mandated to be prepared to stop on HIS line, even if he did see a clear signal. 

Information released states the signal system was tested and found to be functioning properly after the accident.

 Even before teh rules debate begins, I still can't understand what the conditions were before the collision, and there seem to be different "reports" on how things were. 

There's no rules debate I am aware of among anyone qualified on railroad operating rules.  The conditions before the collision were normal, according to all information released to date, except one train apparently was not acknowledging signal indications.

Is there anything solid to go on yet?  When's the NTSB supposed to come out, or haven't they given a date yet?  No, I don't consider "media" reports solid, for obvious reasons.

In previous accidents of this magnitude the NTSB completes and releases its report from 12 to 24 months after the accident.  I have heard no information that this accident will have a different analysis timeline.

RWM

  • Member since
    June 2008
  • From: Salem, Oregon
  • 189 posts
Posted by NP Red on Monday, December 15, 2008 10:56 AM

kolechovski
The signal at the switch the Metro train ran through should have been an absolute signal, right?  The dispatcher should have seen that.  What would he see if the train runs through that?  Does anything come up on his screen in that case? 

 

I'm only quessing but I'm not sure that the dispatcher would see anything different if the red was ran because another train was already in the block just around the corner. How would the system "know" that the switch was run. At this point in time, can the "system" tell if a block is occupied at one end or the other? I know that train speed and direction are calculated in the timing of the gates at grade crossings but does anybody know if that applies to block occupation?

  • Member since
    September 2002
  • 762 posts
Posted by kolechovski on Monday, December 15, 2008 9:40 AM

The signal at the switch the Metro train ran through should have been an absolute signal, right?  The dispatcher should have seen that.  What would he see if the train runs through that?  Does anything come up on his screen in that case?  I know many different areas have many different systems, and even what happens after a train runs a red depends on where it's at and teh system in place.

And I still wonder if anyone knows exactly what signals were what color, and whether teh engineer was mandated to be prepared to stop on HIS line, even if he did see a clear signal.  Even before teh rules debate begins, I still can't understand what the conditions were before the collision, and there seem to be different "reports" on how things were.  Is there anything solid to go on yet?  When's the NTSB supposed to come out, or haven't they given a date yet?  No, I don't consider "media" reports solid, for obvious reasons.

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Sunday, December 14, 2008 9:14 PM

gardendance
My eyes also glaze over when I read posts about the various remote control methods and their abreviations. Maybe I need to read the ABC's of railroading to figure CTC, ATO, ATC, PTC and OMGOSH. Anyway, is there, or has anybody thought of a system like this: the engineer punches in a code that acknowledges the signal aspect. If it doesn't agree with what the computer thinks the aspect should be it gets recorded for review. If it isn't the same as, or more restrictive than what the computer thinks then the train stops.

What's confusing is that many people are quoting the wrong rules, rules from different rule books, rules that are associated with different track authorities, obsolete rules, rules for the wrong type of train and then there are all the people who have never been qualified on the rules trying to hold a rules class and getting things anywhere from 5% to 50% right.  By the time its all said and done its pretty confusing missmash of a thread.  Its a lot simpler than it seems, but rumors don't fair well in clarity, so in order for all these other alternative scenarios to flourish people have to get the water as muddy as possible.

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Sunday, December 14, 2008 6:14 PM

Regarding the use of radio as an overlay of operating information:

 

On one hand, the idea of broadcasting operating information seems like it would help communication even though it may duplicate signal indications.  If the communication is as important as this is, why not reinforce it with every means available?

 

On the other hand, the more avenues of information transmission, the greater the chance of them not agreeing, and the emergence of the question as to which avenue has the ultimate authority.

 

I recall a short article in Trains back in the 1970s about a wreck on the PC where a freight train ran a red signal at a Bascule bridge, and struck the counterweight of the bridge.  The article mentioned that the accident would shine light on the new, emerging issue of defining the authority of radio transmissions in controlling trains.  I don’t have the article in front of me, so I am going by memory. 

 

When the approaching train was some distance (perhaps a few miles) away, the bridge operator lined up for the train, and told the engineer by radio that he (the engineer) could highball right along because the bridge was lined up for him.  I am not sure what type of approach authority was in place, but I assume it was a standing rule to approach the bridge and its protecting signal prepared to stop, and that there was no approach signal prior to the bridge.  So the bridge operator was doing the engineer a favor by saving him the need to slow down and approach the bridge being prepared to stop.

 

Then the bridge operator remembered that he had an Army Corps of Engineers boat waiting for the bridge, so he took the route away from the train and raised the bridge for the boat.  Apparently, he was able to take the route away from the train because the train was not yet in the plant for the bridge signals.  The bridge operator did not radio the engineer of this change of mind to give the route to the boat.  He probably figured he did not need to because, under the system of signals and rules, he could open the bridge any time he felt like it with no risk to approaching trains.  Perhaps he thought he could let the boat through and get the bridge back down for the train in time, and therefore, simply did not want to aggravate the situation by telling the engineer to stop right after offering him the convenience of not even needing to slow down.

 

In any case, the train arrived at the bridge with its engineer expecting it to be lined up for him, and it was not.  The engineer could not stop, so he hit the massive counterweight of the bridge.  The momentum of the train, being transmitted through the drawbars, shoved the two locomotives under the gap below the counterweight.  The action sheared off everything above the locomotive frames, and wadded up the wreckage several stories high behind the counterweight. 

 

  • Member since
    August 2005
  • 964 posts
Posted by gardendance on Sunday, December 14, 2008 10:33 AM

SEPTA operates under it's own tracks, AMTRAK's and I believe CSX's and/or Norfolk Southern's tracks and/or dispatching. I heard one SEPTA engineer say that he needed to know everybody's rules, and there were some requirements on one which would be violations on another.

Patrick Boylan

Free yacht rides, 27' sailboat, zip code 19114 Delaware River, get great Delair bridge photos from the river. Send me a private message

  • Member since
    December 2001
  • 8,156 posts
Posted by henry6 on Sunday, December 14, 2008 10:09 AM

gardendance

My eyes tend to glaze over reading these descriptions of rules such as

you may proceed after a stop and proceed signal if the previous signal had bird droppings on it and the 3rd signal is a flashing yellow on alternate Tuesdays

I really can't figure how anybody can keep them straight, yet at least one poster here said the rules are straightforward.

My eyes also glaze over when I read posts about the various remote control methods and their abreviations. Maybe I need to read the ABC's of railroading to figure CTC, ATO, ATC, PTC and OMGOSH. Anyway, is there, or has anybody thought of a system like this: the engineer punches in a code that acknowledges the signal aspect. If it doesn't agree with what the computer thinks the aspect should be it gets recorded for review. If it isn't the same as, or more restrictive than what the computer thinks then the train stops.

I remember a John Kneiling Professional Iconoclast article that had a similar theme for remote control. 2 crew members had the remote control box, it had 4 buttons, slow forward, slower forward, slow reverse and stop. Unless each crew member pressed the same button, or if either control box lost went out of radio range, the locomotive stopped.

Now for the idea that it's better for dispatchers to broadcast all meets, and it's better for every crew to hear these announcements. That's kind of like the idea that you send an email to everybody on the list, even if they're only peripherally involved in the subject. You wind up distracting some folks from more germaine duties, you run the risk of a "cry wolf" situation, "I just heard 20 announcements about a meet on a parallel line" so the engineer doesn't pay attention to the 21st announcement which concerns a move on his line, or at the junction he's coming to. Certainly if the announcement is the dispatcher screaming "Holy crap, trains xxx and yyy are gonna have a cornfield meet, everybody stop immediately" I can see that it's probably a good idea.

 

Yes, it does get confusing.  Several factors make things easier to the employee (whatever craft).  One is that only parts of NORAC Rules are used by any given railroad or operation and thus are the only ones needed to be known if employed in that juristiction.  Two, if your job and the lives of you, your cooworkers, and passengers along with the property of your employer and his customers, are at stake, then you learn and understand the rules governing your job or else  And remember that the NORAC Rules are not the AAR Standard Code which many of us were brought up on, and which, except for a few scattered instances, are not used for operations.  In fact the only Standard Code operations I know of are scattered across the LIRR in the far eastern branches of the system.  Add to this maze the fact that termenology beyond NORAC Rules (and Standard Code) differ from railroad to railroad and region to region.  You have to be a true and total believer to know each and every rule, law, term, and whatever of every inch of railroad in the country or the world...and who is? (show of hands only.)

RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.

  • Member since
    August 2005
  • 964 posts
Posted by gardendance on Sunday, December 14, 2008 8:06 AM

My eyes tend to glaze over reading these descriptions of rules such as

you may proceed after a stop and proceed signal if the previous signal had bird droppings on it and the 3rd signal is a flashing yellow on alternate Tuesdays

I really can't figure how anybody can keep them straight, yet at least one poster here said the rules are straightforward.

My eyes also glaze over when I read posts about the various remote control methods and their abreviations. Maybe I need to read the ABC's of railroading to figure CTC, ATO, ATC, PTC and OMGOSH. Anyway, is there, or has anybody thought of a system like this: the engineer punches in a code that acknowledges the signal aspect. If it doesn't agree with what the computer thinks the aspect should be it gets recorded for review. If it isn't the same as, or more restrictive than what the computer thinks then the train stops.

I remember a John Kneiling Professional Iconoclast article that had a similar theme for remote control. 2 crew members had the remote control box, it had 4 buttons, slow forward, slower forward, slow reverse and stop. Unless each crew member pressed the same button, or if either control box lost went out of radio range, the locomotive stopped.

Now for the idea that it's better for dispatchers to broadcast all meets, and it's better for every crew to hear these announcements. That's kind of like the idea that you send an email to everybody on the list, even if they're only peripherally involved in the subject. You wind up distracting some folks from more germaine duties, you run the risk of a "cry wolf" situation, "I just heard 20 announcements about a meet on a parallel line" so the engineer doesn't pay attention to the 21st announcement which concerns a move on his line, or at the junction he's coming to. Certainly if the announcement is the dispatcher screaming "Holy crap, trains xxx and yyy are gonna have a cornfield meet, everybody stop immediately" I can see that it's probably a good idea.

Patrick Boylan

Free yacht rides, 27' sailboat, zip code 19114 Delaware River, get great Delair bridge photos from the river. Send me a private message

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Sunday, December 14, 2008 7:58 AM

K. P. Harrier
It would seem to me that the tell-tale indicator would be the dispatcher’s and locomotives’ recordings.  If in fact green signals are customarily not called out, there should be few signal call-outs on all the tapes related to the disastrous run of Train 111.

Dispatcher's "recordings" only record the radio towers that the dispatcher is using or listening to because they are recordings of the dispatcher, not general radio traffic.  A dispatcher may have a 100 phone lines, intercoms, radio towers, etc. available to him, but will only have one or two on at any given time.  So unless the dispatcher just happened to be using a radio tower that was within range of the train, there would be no recording of the signals being called out.  The engines do not record radio conversations. 

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    October 2003
  • 7,968 posts
Posted by K. P. Harrier on Saturday, December 13, 2008 10:10 PM

At variance with this thread so far, the conductor, as the sole surviving crew member of the September Metrolink crash, claims the last two signals were green.  And that is why the last two signals were not called out over the radio.

 

As stated by officials of the NTSB, it is a violation of the Metrolink rules to not call out green signals.  But, it appears to be a cultural thing; with most railroaders not calling out clear signals, possibly inspired by previous railroad employment practices.

 

It would seem to me that the tell-tale indicator would be the dispatcher’s and locomotives’ recordings.  If in fact green signals are customarily not called out, there should be few signal call-outs on all the tapes related to the disastrous run of Train 111.

 

----------------------------------------------------------------------------------------------------------------------------------- K.P.’s absolute “theorem” from early, early childhood that he has seen over and over and over again: Those that CAUSE a problem in the first place will act the most violently if questioned or exposed.

  • Member since
    November 2007
  • 2,989 posts
Posted by Railway Man on Saturday, December 13, 2008 3:02 PM

wabash1

Railway man you seem to be back tracking here a bit. Now let me get this right, 1 min after the crash they was pulling the tapes. so the railroad set this crash up? heck the last signal was 1 mile from the station and the crash site was 1/2 mile from the signal anyways You cant even get a trainmaster out from behind his desk in a minute. now if your talking about the radio tapes in the dispatchers office was the nstb standing there? they ussually pull their own tapes so nobody can claim things was tampered with. then you changed your mind about them being able to tell you not to take a signal in that they can make you stop, after you said that the signal superseed anything the dispatcher says. I do agree that the dispatcher can not tell you the signal indication, the only thing they can say is you should have a signal or come looking for a signal or even you can proceed on signal indication. all only tell you that they have requested the signal for your movement. 

I am in agreement with some on here that people are grasping at things thinking that this man could not have been texting and missed a signal that killed so many. in that he cant make that big of a mistake. and looking for loop holes in the rules that might save his reputation, trust me and other railroaders on here in saying that the rules are straight forward air tight very few cracks to slip thru. most of us have been fired over some of these rules in our time on the railroad,  and most times we cant win. just do your time and come back to work.



Wabash, I was referring to the tapes in the dispatching office.  When we had a wreck or injury, we pulled the tapes within a matter of minutes.  The NTSB doesn't pull the tapes, the railroad does.

I think GCOR is clear.  Do you?

RWM

 

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Saturday, December 13, 2008 2:35 PM

wabash1
I am in agreement with some on here that people are grasping at things thinking that this man could not have been texting and missed a signal that killed so many. in that he cant make that big of a mistake. and looking for loop holes in the rules that might save his reputation,

Where do you see that?  I don't see it in this thread except for maybe in the article linked to the first post.

  • Member since
    April 2001
  • From: US
  • 2,849 posts
Posted by wabash1 on Saturday, December 13, 2008 2:26 PM

Railway man you seem to be back tracking here a bit. Now let me get this right, 1 min after the crash they was pulling the tapes. so the railroad set this crash up? heck the last signal was 1 mile from the station and the crash site was 1/2 mile from the signal anyways You cant even get a trainmaster out from behind his desk in a minute. now if your talking about the radio tapes in the dispatchers office was the nstb standing there? they ussually pull their own tapes so nobody can claim things was tampered with. then you changed your mind about them being able to tell you not to take a signal in that they can make you stop, after you said that the signal superseed anything the dispatcher says. I do agree that the dispatcher can not tell you the signal indication, the only thing they can say is you should have a signal or come looking for a signal or even you can proceed on signal indication. all only tell you that they have requested the signal for your movement. 

I am in agreement with some on here that people are grasping at things thinking that this man could not have been texting and missed a signal that killed so many. in that he cant make that big of a mistake. and looking for loop holes in the rules that might save his reputation, trust me and other railroaders on here in saying that the rules are straight forward air tight very few cracks to slip thru. most of us have been fired over some of these rules in our time on the railroad,  and most times we cant win. just do your time and come back to work.

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Saturday, December 13, 2008 8:58 AM

zardoz

Railway Man

(P.S. to Zardoz -- that's IF the dispatcher has time, of course.  These days with huge territories, not always, especially on daylights with all the MOW work.  And some railways discourage this practice.)

RWM

That is very interesting.  I certainly do not doubt you, but I cannot imagine why a railroad would not want to keep everyone equipped with as much information as possible regarding all activities on a segment of railroad. By knowing in advance of a meet, an engineer might want to operate differently.

Its a balancing act.  Engineers may act on information as if it is authority.  The dispatcher tealls the train they will be there for 2 trains.  Two trains go by, but the third train caught up to the first two and the dispatcher decides to bring it on also.  the crew in the siding sees two trains go by and without waiting to get authority depart the station and run into the the oncoming 3rd train.

Providing information can reduce calls to the dispatcher, since the trains know what is going on, they don't ask.  The downside is if the situation is fluid, you could tell the trains one thing and 20 minutes later do something completely different.

Dave H. Painted side goes up. My website : wnbranch.com

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