Trains.com

Rear end collision on the UP west of Cheyenne WY. 10/04/18

10954 views
136 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Saturday, November 10, 2018 7:36 PM

Initially it was reported that the crew had gotten off the engine of the struck train.  This raised the question in this thread as to whether a crew ever gets off of the engine to avoid injury from being struck from behind by another train. 

However, it was not immediately announced whether this was the motive for the crew to get off of the train struck in this U.P. collision.  They might have just gotten off for some other reason, so it was just reported merely to indicate they were not involved in the collision. 

But then, we learned that the dispatcher had ordered the crew to get off their engine because they were about to be struck from behind by the runaway.  So this would seem to indicate that the crew was told to get off for their own safety. 

But there is one other point that needs clarification.  After learning of the runaway, was the struck train already stopped for some other reason, and then the crew told to get off for their safety?  Or after learning of the runaway, was the struck train moving, perhaps at track speed, ahead of the runaway; and then ordered to stop and for the crew to get off?

If the struck train was ordered to stop because of the runaway, what would be the reason for that order? 

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Wednesday, November 7, 2018 12:40 PM

Ten cars deep fits the scenario I see as related to blockage of the brake pipe.  I speculated above that it would be no deeper than 25 cars.  If it were say 75 cars deep, the engineer’s emergency application would have likely piled up the train as the un-braked hind end of maybe 75 cars pushed against 75 cars in emergency.

  • Member since
    March 2003
  • From: Central Iowa
  • 6,901 posts
Posted by jeffhergert on Wednesday, November 7, 2018 11:44 AM

I heard this morning that they had 10 cars on the head end that were responding to the brake application.  Those cars' brakes showed excessive heat build up.  Allegedly, the 10th or llth car, part of a pick up in Laramie, had been recently worked on by a car man for a train line problem.

Jeff 

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Wednesday, November 7, 2018 7:28 AM

zugmann
 
Euclid
point (with its large total amount of natural leakage and reduced pressure) was suddenly added back and sensed by the air compressor system of the locomotive, so that system began once again to deliver many cfm of compressed air to the brake pipe.

 

If the pressure on the trailing portion was reduced by more than a few PSI, and the blockage is suddenly removed, your train is most likely going to go into emergency. 

 

I was wondering about that, but I have no way to draw a conclusion.  If the blockage suddenly opened up, the head end would experience a sudden loss of pressure.  I guess whether it dumped the air would depend on the rate of reduction on the first car ahead of the blockage.  For at least the first instant, that rate of reduction would probably be quite high.

However, maybe the blockage did not instantly clear, so maybe there was enough lag to slow down the rate of reduction to be less than what it would take to cause an emergency application. 

Under the circumstances of this runaway, I understand that the engineer made an emergency application while the airflow into the brake pipe was zero cfm.  If the zero airflow condition was a symptom of the brake pipe being blocked, I speculate that the emergency application would have propagated to the entire train had the emergency application been made when the slack was stretched. 

It seems that the failure of the EOT to make an emergency application was a problem unrelated to the cause of the head end emergency application being ineffective. 

  • Member since
    March 2003
  • From: Central Iowa
  • 6,901 posts
Posted by jeffhergert on Tuesday, November 6, 2018 12:55 PM

zugmann

 

 
petitnj
And once again, the signal/PTC/Cab/etc. system was not working. Since the crew was always awoken by these things before, now when they fall asleep, the outcome is disaster. More reason to remove the crew from driving the train and only run when PTC like systems operate.

 

Going by the initial report, I doubt the crew was sleeping.

 

I originally posted that I was told (by a manager) that PTC and/or cab signals were not working.  I later said that I thought that might not be true.  That I felt the railroad may have been throwing out things to divert as much blame as possible on the crew.  The report says PTC was operative, but with the other problems it really didn't matter.

The report doesn't say whether the head end dump switch was activated.  I'm inclined to think it was.  If they can't prove it was operated, look for the crew to get most of the blame.  Just my opinion.

Jeff 

  • Member since
    March 2003
  • From: Central Iowa
  • 6,901 posts
Posted by jeffhergert on Tuesday, November 6, 2018 12:46 PM

zugmann

 

 
Euclid
I have no idea how the slack run-in may have blocked the brake pipe and they unblocked it as it ran out.

 

Could be a trolley line on a cushion drawbar getting hung up. Sometimes made worse when people put the wrong length of hoses on them.

 

That's exactly what happened to me a year or two back.  Someone put on the wrong (too long) hose and when the slack came in, the hose got pinched between the draft gear and the car body.

Once we were stopped, a car man came out and fixed it.  He wired it in place and said we were good to go.  

We were on a short grade, I had clear signals and what brakes (air and dynamics) I had were slowing us down and bringing us to a stop.  Because of that I didn't pop the rear end.  I held that in reserve if conditions changed.  Once stopped and before the car man arrived, the conductor started walking back.  He called on the radio and asked if the brake pistons should be in or out.  I said out.  He found the trouble spot in that way.  Out of 102 cars, the head 18 were responding to the brake applications.

Jeff

  

  • Member since
    January 2002
  • From: Canterlot
  • 9,575 posts
Posted by zugmann on Tuesday, November 6, 2018 12:16 PM

Euclid
I have no idea how the slack run-in may have blocked the brake pipe and they unblocked it as it ran out.

Could be a trolley line on a cushion drawbar getting hung up. Sometimes made worse when people put the wrong length of hoses on them.

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
    January 2002
  • From: Canterlot
  • 9,575 posts
Posted by zugmann on Tuesday, November 6, 2018 12:11 PM

petitnj
And once again, the signal/PTC/Cab/etc. system was not working. Since the crew was always awoken by these things before, now when they fall asleep, the outcome is disaster. More reason to remove the crew from driving the train and only run when PTC like systems operate.

Going by the initial report, I doubt the crew was sleeping.

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
    March 2003
  • From: Central Iowa
  • 6,901 posts
Posted by jeffhergert on Tuesday, November 6, 2018 12:10 PM

Regarding comm loss.  The head end device (stand alone or integrated into computer screens) wait 16 mins and 30 seconds without comm before they register the comm loss.  You could be without comm for 15 minutes, have comm reestablished for a minute and then again lose comm.  The count down would restart.

For DP operation, comm loss with the DP(s) is immediately registered.  There is no built in wait time.  For both systems, you can operate for 16 mins and 30 seconds without the capability to intiate an emergency application from the rear of train (comm loss) before you need to take action.  Slowing to 30, etc.  So when the EOT head end shows Front-Rear Comm Loss, you have to take action.  Your grace time is up.  With the DP, you have to actually watch the clock.

My earlier post about the change in air flow had some cut and paste changes, hence the different fonts.  One part that I had added, but didn't make the change was that to me, usually air flow goes up when slack is bunched and goes down when it is stretched.  Just the opposite of what the report said.  It does make me think that when bunched up it may have pinched a hose.

A slip joint is where the change in the position of the glad hands, possibly made worse by worn/damaged gaskets or not properly seated gaskets.  It allows air to escape at the joint.  

The other night I had a 14851 foot manifest with a single DP at 8500 feet back from the lead engine consist.  Air flow off the headend was 0, the DP fluctuated with changes in slack between 14 and 18 cfm.  Not bad for something like that.

Jeff

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Tuesday, November 6, 2018 12:04 PM

BigJim
Did I not say that there was no apparent blockage at the time the train was going through the dip?

 

In the dip, while the slack was bunched with dynamic braking, the flow rate was zero.  Isn’t the zero flow rate an anomaly?  I have no idea how the slack run-in may have blocked the brake pipe and they unblocked it as it ran out.  I cannot see a way that slack could have caused that.  So I see no direct evidence that the brake pipe was ever blocked.  It is only the air flow into the brake pipe going from something like normal to zero and then back again that fits the cause of the train line being blocked and then unblocked. 

 

  • Member since
    April 2001
  • From: Roanoke, VA
  • 2,019 posts
Posted by BigJim on Tuesday, November 6, 2018 11:40 AM

A "slip joint" is where two sections of trainline telescope, one into the other. These are used on some types of cushioned cars and cars with long drawheads and are positioned at the end portions of the car. There is a lot of slack on these types of cars.

Did I not say that there was no apparent blockage at the time the train was going through the dip? The only time that I have ever seen a trainline blocked internally is from ice build up. And, the only time that I have ever seen an air hose pinched shut is when the train was cut and the air hose flew back and got caught by something essentially folding it two.

AGAIN...This up & down pressure reading tells me the trainline was continuous. And, with the CFM numbers in the twenties, as was reported, this is nothing I would have any concern over.

My concern is that the pressure reading on the rear didn't drop when the automatic brake was applied. Somehow something went wrong and it may be that an anglecock somehow closed. I have had that happen!

.

  • Member since
    January 2002
  • From: Canterlot
  • 9,575 posts
Posted by zugmann on Tuesday, November 6, 2018 11:10 AM

Euclid
point (with its large total amount of natural leakage and reduced pressure) was suddenly added back and sensed by the air compressor system of the locomotive, so that system began once again to deliver many cfm of compressed air to the brake pipe.

If the pressure on the trailing portion was reduced by more than a few PSI, and the blockage is suddenly removed, your train is most likely going to go into emergency. 

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
    January 2014
  • 8,221 posts
Posted by Euclid on Tuesday, November 6, 2018 10:51 AM

BigJim
The way the air pressure was flucuating, it tells me that it could have very easily been the very common afformentioned "Slip Joint" causing this.

Can you or someone else please explain what a slip joint is and the problem it creates?

My belief is that the air flow into the brake pipe dropped to zero cfm during the slack run-in because the run-in somehow pinched off an air hose and thus isolated a large portion of the brake pipe from the locomotive. 

The small portion of brake pipe that was left connected to the locomotive did not experience much total pressure drop because there were relatively few hose joints in it to leak. 

So with the compressor controls sensing no pressure drop due to leakage in that small head end section of cars, the system found no need to provide replacement air to compensate for the leakage. 

So that is why the rate of air flow into the brake pipe dropped to zero cfm when the slack ran in and essentially created two brake pipes; one connected to the locomotive and one isolated from the locomotive.

Then when the slack ran back out, the blockage in the brake pipe was removed.  So then the very large portion of brake pipe trailing the blockage point (with its large total amount of natural leakage and reduced pressure) was suddenly added back and sensed by the air compressor system of the locomotive, so that system began once again to deliver many cfm of compressed air to the brake pipe. 

  • Member since
    April 2001
  • From: Roanoke, VA
  • 2,019 posts
Posted by BigJim on Tuesday, November 6, 2018 7:47 AM

Euclid
When the slack ran in, the rate of air being pumped into the brake pipe to compensate for leakage dropped to zero.  For the leakage to drop to zero during the slack run-in suggests to me that the run-in was causing a blockage very near the head end.


Nope!
The way the air pressure was flucuating, it tells me that it could have very easily been the very common afformentioned "Slip Joint" causing this. It also could have been a leaking air hose gasket. Both a result of the bunched slack.

That small of CFM flow was not uncommon and certainly does not indicate any kind of blockage. On the contrary, the fluctuation of pressure as the train proceeded through the dip indicates to me a continuous trainline.

.

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Tuesday, November 6, 2018 7:31 AM
When the slack ran in, the rate of air being pumped into the brake pipe to compensate for leakage dropped to zero.  For the leakage to drop to zero during the slack run-in suggests to me that the run-in was causing a blockage very near the head end.   
 
To block the brake pipe by pinching an air hose, and have the air replenishment drop to zero, the leakage rate must have dropped to zero as well.  That suggests that very few cars were involved with the unblocked portion of the brake pipe.  I would speculate that no more than 25 cars were ahead of the blockage.  If that is true, the car ahead of the blockage and the one behind it would both be in the wreck of the derailment.  Both cars would have a high probability of damage to the components responsible for a pinching of an air hose. 

How common is the problem of an air hose being blocked, but not broken or punctured when the slack runs in, and then returning to normal function when the slack runs back out?

  • Member since
    December 2001
  • From: Northern New York
  • 25,021 posts
Posted by tree68 on Tuesday, November 6, 2018 5:50 AM

Electroliner 1935
Larry, you are a ham radio op, any thoughts on this. 

So many variables.  I would suppose that the EOT would be more likely to physically fail than the locomotive end.  

I wondered about, and it's been answered, how long it would take before a failure of the EOTD would be noticed by the equipment.   We see variable conditions on the UHF bands, but that's usually over many miles, not the mile or two involved with an EOTD.

Variable conditions (signal blockages and other interference) would be one reason to allow a certain amount of time before sounding a loss-of-signal alarm. 

The fact that the engineer saw a gradually decreasing pressure from the EOTD would suggest that at least one-way comms were in place.  The question then becomes whether a command to dump the brakes was, indeed, sent to the EOTD, as has been discussed.  

We'll learn more when they update the report.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    March 2003
  • From: Central Iowa
  • 6,901 posts
Posted by jeffhergert on Monday, November 5, 2018 8:45 PM

Electroliner 1935

 

 
blue streak 1
More important is why the loss of communication to the EOT occurred ?  According to the preliminary report communication from the EOT was maintained ? 

 

Communications is normally bidirectional. Can one direction fail? Loco hearing EOT but EOT not hearing Loco? Larry, you are a ham radio op, any thoughts on this. 

 

There can be Front-Rear No Comm and Rear-Front No Comm.  With Rear-Front you may lose a reading off the rear and if it lasts, usually in my experience changes to a Front-Rear No Comm status or Radio Break status. 

The Front-Rear and Radio Break status require reducing speed to 30 mph on most territories until comm is restored.  On mountain grade territories the train must not be operated until comm/ability to activate a rear end emergency dump is restored.  Trains may move enough to clear obstructions that could be causing comm loss. 

With Rear-Front No Comm, it is thought that an EOT will respond to an emergency dump command.  No action is required unless the status degrades.  

Jeff

  • Member since
    March 2003
  • From: Central Iowa
  • 6,901 posts
Posted by jeffhergert on Monday, November 5, 2018 8:30 PM

Air Turbine EOTs don't normally take enough air to cause air flow to register.  That air flow was changing with the change in slack condition is not unusual.  We call that a slip joint.

On another site, someone with more experience with that area said that loss of comm with the EOTs is common.  You can lose comm and not know it for 16 mins 30 seconds before the head end device registers comm loss.  (If you think you are out of comm, you can press the comm test key.  It will come back with currents status.)  An EOT not registering a pressure change after a brake application/release is usually a good indicator that you don't have comm.

Our rules call for operation of the eot emergency toggle switch, even if showing a no comm status.  Just in case comm momentarily is established.

Jeff

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Monday, November 5, 2018 8:17 PM

Wherever the air flow got obstructed - the actions of the train that were described - would make it appear that the obstruction was very near the head end of the train and that little if any train braking was being had from the portion of the train that was ahead of the obstruction and would have had brakes being applied.  

It is not mentioned, what if any observations that the crew may have communicated about seeing brakes applied on any portion of their train - activated brakes should have developed some level of smoke (that would have been hard to see in the dark) or what was observed of the applied brakes creating 'rings of fire' around the wheels that were being actively braked.  Investigation of the scene by NTSB should be able to identify among the derailed cars, those that had wheel tread bluing from the excess heat generated from the applied brakes.

Never too old to have a happy childhood!

              

  • Member since
    September 2010
  • 2,515 posts
Posted by Electroliner 1935 on Monday, November 5, 2018 8:14 PM

blue streak 1
More important is why the loss of communication to the EOT occurred ?  According to the preliminary report communication from the EOT was maintained ? 

Communications is normally bidirectional. Can one direction fail? Loco hearing EOT but EOT not hearing Loco? Larry, you are a ham radio op, any thoughts on this. 

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,919 posts
Posted by blue streak 1 on Monday, November 5, 2018 8:02 PM

Will the investigation be able to determine the number of cars that had the emergency application ?  That location should be able to determine where  the  apparent brake pipe pinch occurred ?  Maybe it even happened on a loco ?

More important is why the loss of communication to the EOT occurred ?  According to the preliminary report communication from the EOT was maintained ?  That has many possible worries for operation of these extra long trains ? 

 

I'm presuming the EOT had an turbine in it - how many CFM would that normally consume? The number will need to be compared to the volume of the trains brake pipe especially if any pinch occurred.  

  • Member since
    December 2001
  • From: Northern New York
  • 25,021 posts
Posted by tree68 on Monday, November 5, 2018 7:12 PM

I'm presuming the EOT had an turbine in it - how many CFM would that normally consume?  

If a hose was, indeed, pinched, that would tend to explain why the pressure displayed by the EOT dropped as it did, why there was an air flow of zero, and why the reduction, as well as the emergency application, did not propogate through the entire train.

 

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Monday, November 5, 2018 7:08 PM

Euclid

If everything else was working properly, and if the engineer made an emergency application, but the end of train device failed to duplicate that emergency application at the same time the engineer made his, would not the whole train still go into emergency braking? 

It should have.

Euclid

It sounds like they are saying that the engineer’s emergency application did not propagate to the entire train, or it may not have applied on any of the train. 

Yes, exactly.

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Monday, November 5, 2018 6:56 PM

If everything else was working properly, and if the engineer made an emergency application, but the end of train device failed to duplicate that emergency application at the same time the engineer made his, would not the whole train still go into emergency braking? 

It sounds like they are saying that the engineer’s emergency application did not propagate to the entire train, or it may not have applied on any of the train. 

 

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Monday, November 5, 2018 6:38 PM

BaltACD
SD70Dude

Very perplexing - the head end was reading the air pressure being sent by the EOT but the EOT didn't respond to the Emergency activation from the head end.  Evidence that there was at least one way communication between the EOT and the head end.

Upon reading the report again, it doesn't say if the Engineer tried to use the emergency toggle switch on the HOT/IDU console.

Even though placing the automatic brake valve in emergency should have also triggered an emergency application from the EOT, in reality those systems don't always work together perfectly, especially on older locomotives. 

The apparent loss of brake pipe continuity while the train was moving - quite possibly every single time the slack was bunched up - is far more disturbing.  It reminds me of Jeff's story from not so long ago about an air hose being pinched by a drawbar. 

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Monday, November 5, 2018 6:21 PM

SD70Dude

Very perplexing - the head end was reading the air pressure being sent by the EOT but the EOT didn't respond to the Emergency activation from the head end.  Evidence that there was at least one way communication between the EOT and the head end.

Never too old to have a happy childhood!

              

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Monday, November 5, 2018 4:12 PM

jeffhergert

Other than it being some kind of braking issue, not much has leaked out.  FWIW, someone on another site said the rumor going around his area was that the crew did nothing wrong.  This person also pointed out it was close to a spot where a closed anglecock led to a similar collision in 1979.  In that one, the crew on the caboose didn't open the emergency valve.

IMO, they have a good idea of what went wrong.  They aren't going to say much until they can find a way to place as much blame as possible on the crew.  It's been done before where the crew lived.  Should be that much easier when the crew didn't.  I won't be surprised if the images from the inward facing camera aren't retrievable.  

Jeff 

NTSB preliminary report:

https://www.ntsb.gov/investigations/AccidentReports/Pages/RRD19FR001-preliminary-report.aspx?fbclid=IwAR1Pk0ox2CdOIhCudL1SJMCw9Eq56quQ2mkC9LgeGmTjleXi6DbcCTl_NyY

I would like to know why the EOT didn't dump the tail end. 

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Saturday, October 20, 2018 11:46 AM

Shadow the Cats owner
As brakes where out on trains I am sure they adjust the brakes we do the same.  

Part of the Terminal Air Test performed by Car Inspectors is to measure Brake Piston Travel with the brakes applied (I believe in excess of 9 inches is a failure) and to replace worn out brake shoes that they observe.  When a car has all new brake shoes (nominally 8 per car - 4 shoes per truck, ie. one brake shoe per wheel or two brake shoes per axle) the brake piston travel is well under 9 inches.  This is how brakes get adjusted when a train is built - this will also happen when a train undergoes the required 1000 mile inspection.

Never too old to have a happy childhood!

              

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Saturday, October 20, 2018 10:15 AM

Shadow the Cats owner
As brakes wear out on trains I am sure they adjust the brakes we do the same.

Of course they do; the key is to examine the way that is done on railroad cars and translate that into what would be required of a spring-brake mechanism.  A good starting point is slack adjusters.

One consequence in the 'old days' was that the actual spring have very long travel (e.g. many turns for a coil or helical spring).  If I recall correctly some designs had a spring running a significant part of the length of the car.  When I was a kid and first found out about cold-gas generators in rocketry, I thought some kind of pressurized accumulator with gas generation was an answer to providing reservoir pressure or effective application quickly in emergencies; I still think that represents a "better" safety approach than mechanical springs pulled off by active service pressure.  But it retains many of the objections associated with those approaches.

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