Cessna crashes onto tracks; struck by Metrolink train

Help Support Amtrak Unlimited Discussion Forum:

caravanman

Engineer
Joined
Mar 22, 2004
Messages
4,617
Location
Nottingham, England.
Here in the UK, railway employees know how to turn green electric train signals to danger by use of track circuit clips. If these work in the US too, then issuing these low tech items to emergency service vehicles might be one idea?
 

jis

Chief Dispatcher
Staff member
Administator
Moderator
Gathering Team Member
Joined
Aug 24, 2003
Messages
32,722
Location
Space Coast, Florida, Area code 3-2-1
Here in the UK, railway employees know how to turn green electric train signals to danger by use of track circuit clips. If these work in the US too, then issuing these low tech items to emergency service vehicles might be one idea?
That would work if there is track circuit in the area, which is not guaranteed in the brave new world of GPS driven PTC, and axle counters to determine section occupancy.
 

Devil's Advocate

🚂〰️〰️〰️〰️
Joined
May 24, 2010
Messages
13,301
Location
🇺🇸
Here in the UK, railway employees know how to turn green electric train signals to danger by use of track circuit clips. If these work in the US too, then issuing these low tech items to emergency service vehicles might be one idea?
Even leaving a lit flare on the tracks a mile down the line in both directions could do wonders for reducing severity.
 

west point

Engineer
Joined
Jun 9, 2015
Messages
3,452
Location
SW ATL airport
If arriving or at scene of incident-----------
1. If crossing call number on blue sign if possible then 911 if possible.
2. If not at crossing call the RR if you know which one or make an educated guess calling probable RR.
3.. Render first aid if needed.
4. Remove anyone if possible from tracks perpendicular away from tracks.
5. Place flares if available either side of incident.
If you hear train walk toward sound on with flare and / or shirt.
6. If not needed at scene walk toward any mile marker (MM) and call RR with the mile number and if RR ID is on marker. Remember any block signal box will have identification of location. Other structures may have ID as well.
7. Otherwise walk, if possible, toward closest curve with operating flare or white shirt to flag down a train. Go around curve for some distance noting any MM. Place flare or red or white shirt on stick in middle of track.
8. Return to scene and walk in other direction of track repeating above.
 

Cal

Engineer
Joined
Jan 23, 2021
Messages
3,847
Location
Socal
I'm curious how long it actually takes 6-car length commuter and intercity trains to stop. I believe they enter stations at around 25-30, and they stop with room to spare (or slow down and crawl for a little before coming to stop to ensure they're at the right place). A platform is around 600-700 feet? So if they take around 700 feet to slow down 30mph, it'd take about 2100 feet to slow from 90? And brakes need to kick in so.... 2500-2700?

Is my math right or am I totally off?
 
Joined
Jan 15, 2012
Messages
3,651
Location
LAX
I'm curious how long it actually takes 6-car length commuter and intercity trains to stop. I believe they enter stations at around 25-30, and they stop with room to spare (or slow down and crawl for a little before coming to stop to ensure they're at the right place). A platform is around 600-700 feet? So if they take around 700 feet to slow down 30mph, it'd take about 2100 feet to slow from 90? And brakes need to kick in so.... 2500-2700?

Is my math right or am I totally off?
Don't know about your math. From the Metrolink Timetable, the "Maximum Authorized Speed" (passenger) is 60 in the vicinity of the Sylmar/San Fernando Station (north of the crash site). Beginning just south of the station the speed is 79 which includes the vicinity of the crash site.
 

WWW

No real RR Job
Joined
Nov 18, 2018
Messages
472
Location
MSP
MATH - 60 mph is a mile a minute - 5 miles covered in 5 minutes with no speed adjustment

With more than a mile long coal and grain trains and the modular and mix bag freights I have often it heard -
A train of that length weight and kinetic movement takes a mile to stop without derailing (jack-knifing) cars -
So after an accident maybe 40-50 - more - cars will run over the debris scene of the accident -
something to consider if in the vicinity of that scene - stay away and stay safe especially with hazardous cargo

With this passenger train - that airplane was demolished to light weight airborne pieces -
There was no loading platform evident in the vicinity - so the train was at speed no slowing to stop and
the train set did not stop until w-a-y pass the accident scene (no reports of how far - no reports of passenger
injuries on the train making an emergency stop)

The accident reports and findings will be months away -
Will there likely be a repeat of this accident - NO -
But will the authorities be keen on how to timely stop a train from compounding an accident scene ? ? ?
There has to be a solution beyond the basic PTC elements - - -
 

Rambling Robert

Service Attendant
Joined
May 30, 2021
Messages
169
Location
Boston
The Cessna pilot declared a Mayday which is very serious and air traffic control obviously knew it. Apparently by the time the Metrolink dispatcher and train engineer heard of the emergency .... something got lost in the translation.

The protocol that would work is the Metrolink dispatcher confirms a Mayday distress and that makes future communications and protocols very important.
 

Trogdor

BURNiNATOR
Joined
Aug 3, 2004
Messages
5,927
Location
Here
I'm curious how long it actually takes 6-car length commuter and intercity trains to stop. I believe they enter stations at around 25-30, and they stop with room to spare (or slow down and crawl for a little before coming to stop to ensure they're at the right place). A platform is around 600-700 feet? So if they take around 700 feet to slow down 30mph, it'd take about 2100 feet to slow from 90? And brakes need to kick in so.... 2500-2700?

Is my math right or am I totally off?

Stopping distance is not linear with speed. The rate of deceleration isn't necessarily linear, but depending on the situation, it could be close enough.

If we assume the time it takes to go from 30 to 0 is the same as the time it takes to go from 90 to 60 or 60 to 30, the distance covered will certainly not be the same.

I don't have specific data available, but one quick Google search has suggested some trains have an emergency braking deceleration of 1.4 m/s2, which translates, very roughly, to 3 mph/second. In other words, to go from 90 mph to 60 mph would take 10 seconds, 60 to 30 mph would take 10 seconds, and 30 to 0 would take 10 seconds.

The total braking time would be 30 seconds (plus a bit of a factor as the brakes build up to maximum effectiveness). However, the distance traveled when you are going from 90 mph to 60 mph in 10 seconds is considerably farther than the distance traveled when going from 30 mph to 0 in 10 seconds. Remember, after the first 10 seconds of braking, you're still traveling twice as fast as you are on a train going 30 mph, meaning that you've probably covered roughly 2.5-ish (rough approximation) times the distance of a train that was going a constant 30 mph (let alone one that was slowing down from 30 to 0).

In the next 10 seconds, you've gone from 60 to 30, you've covered yet again another 1.5 (ish) times the distance of a train traveling at a constant 30 mph. Going from 30 to 0 covers 0.5 (ish) the distance of a constant 30 mph. I say "ish" because the actual distance covered is more of a curve than linear. In reality, you're covering a higher distance because of the time you spend at a higher speed while you slow down. Even using these rough numbers, a train slowing down from 90 mph would require 9 times the distance of a train slowing down from 30 mph.

Apply that to a train making a controlled stop on a 600-foot platform, and the engineer needs to start braking at least 5400 feet (i.e. over a mile) in advance of the stop. Certainly not 2100-2700 feet.
 

basketmaker

Lead Service Attendant
Joined
Apr 19, 2018
Messages
336
Location
Brighton, CO (DEN but FMG-Preferred)
Okay a Cessna 172 uses Avgas 100 or 110 which is highly flammable! The fuel tanks are in the wings of the aircraft. There is minimal if any fuel leakage shown in the video after the plane's wings clipped the various poles (crossing guards, lights, etc). And the fact that even after the train hit it and it pretty much disintegrated there was no fire? My guess is that the original airplane emergency was lack of fuel?

As for the delay in stopping the train I would venture that the officers on scene radioed their dispatch and then it took a minute or so to locate the phone number for the railroad and then a minute or so for the railroad to verify exact location before contacting any trains in the vicinity. As for the blue emergency signs at crossing I would say the officers were probably more focused on getting the pilot out and medical attention for him than looking around for a phone number. Not to mention was the blue sign even posted at the crossing?

The train speed also seems off as the track in dead straight for 3.24 miles prior to the crash scene. Was the train operator not paying attention (not the first time MetroLink drivers weren't)? He/she had a perfect line of sight for a long distance to throw the train into emergency. Or at least slow it down.

As someone stated we hopefully will get all the details in the NTSB report.
 

basketmaker

Lead Service Attendant
Joined
Apr 19, 2018
Messages
336
Location
Brighton, CO (DEN but FMG-Preferred)
Indeed. So maybe hold off on speculating that the pilot or engineer were negligent.

A simple look at Google Maps shows that the sign you question is present:
View attachment 26725
Call it speculating if you desire just as others have been doing in the thread. As for the blue sign yes in that snapshot (April 2021) it is posted as well from the other direction. Just if it was there at the time of the crash?
 

willem

Conductor
Joined
Aug 17, 2014
Messages
1,188
Location
?
Oh, thank you. I needed an exercise in applied math.
[...] So if they take around 700 feet to slow down 30mph, it'd take about 2100 feet to slow from 90? And brakes need to kick in so.... 2500-2700?

Is my math right or am I totally off?
As Trogdor noted, "Stopping distance is not linear with speed." I wonder if deceleration is constant, because braking is changing kinetic energy to some other form of energy (often heat energy), and kinetic energy varies with the square of speed. Something going 90 mph has nine times the kinetic energy of something going 30 mph. But a varying deceleration makes computation more difficult, so let's assume a constant deceleration.
[...] some trains have an emergency braking deceleration of 1.4 m/s2, which translates, very roughly, to 3 mph/second. [...]
I have no better number, so let's use 3 mph/second as deceleration.

The time dependent location of an object (train) moving on a straight line (track, with no curve) in the direction of travel at constant acceleration is the starting point (milepost, assumed to be zero) plus the change due to the initial speed plus the change due to the acceleration. In other words,
s(t) = s0 + v0*t + 0.5*a0*t^2 where s0 is the initial location, v0 is the initial speed, and a0 is the initial acceleration (assumed to be constant).

Consistency in units is important. Let's use miles and hours. As I said, I assume the initial location is at zero. I'll assume the initial speed is 60 mph, although it was probably higher, based on TinCan782's comment. (Plug in a different number if you want to assume a different initial speed.) I'll assume constant acceleration of -3 mph/second, or -(3*3600) mph/hour, based on Trogdor's comment.

The location of interest is where the speed is zero. If the initial speed is 60 mph and it drops by 10,800 mph every hour, then it will be zero after 60/10800 hours, or 0.00556 hours. (Three digits should provide sufficient accuracy, but feel free to use more digits if you like.) Plugging that value into the formula,
s(0.00556) = 0 + 60*0.00556 - 0.5*10800*0.00556*0.00556

Evaluating the expression, we find the train comes to a stop after 0.1667 miles, one sixth of a mile.

Whenever conducting a thought experiment like this, it is useful to ask, "Is that answer reasonable?" With an initial speed of 60 mph and a deceleration of 3 mph/second, it would take 20 seconds to slow to zero. Happily, 0.00556 hours is 20.016 seconds; the difference is a rounding error because I didn't carry more digits. A train going 60 mph will cover 1/3 of a mile in 20 seconds, so certainly the answer should be less than 1/3 of a mile. Since the assumed deceleration is constant, it seem reasonable that the distance covered would be half the distance covered with no deceleration. Happily, 0.1667 miles is half of 1/3 of a mile. Whatever mistakes I made, I seem to have made them consistently. Corrections and clarifications are welcome.
[...] I say "ish" because the actual distance covered is more of a curve than linear. In reality, you're covering a higher distance because of the time you spend at a higher speed while you slow down. Even using these rough numbers, a train slowing down from 90 mph would require 9 times the distance of a train slowing down from 30 mph. [...]
I'm gratified to see that a 90 mph train takes 9 times longer to stop than a 30 mph train, exactly in line with the change in kinetic energy.

Because the deceleration is constant, you needn't say "ish". If you plot the speed versus time, you get a triangle. The distance covered is the area under the triangle. You are correct that "the distance covered is more of a curve" but the change in the rate of covering the distance is constant, as indicated by assuming a deceleration of 3 mph/second.

However, I think you slipped up somewhere in your calculations. If the train is going 90 mph, then it travels 1.5 miles every minute or 0.75 miles every 30 seconds. If it takes the train 30 seconds to slow to a stop, it cannot go more than 0.75 miles as comes to that stop.
 

west point

Engineer
Joined
Jun 9, 2015
Messages
3,452
Location
SW ATL airport
Just a thought. Did the airport control tower have the RR emergency number? With the 2 or more persons in the control tower, one could have called 911 and maybe they did and one call the RR. Since the control tower has view of the RR you would think they would have the number in case of a derailment?
 

Dutchrailnut

Conductor
Joined
Jul 4, 2006
Messages
1,195
Location
Brookfield, Connecticut, USA
the tower may not have had a clue as to were plane crashed , if it was on radar they would only know were it disappeared from screen .
for crossings there is no general emergency number only the number on ENS sign and ID posted is relevant. even a road name is not relevant as road name a railroad uses is not always same as official name and sometimes a road crosses tracks multiple times so which crossing is it .
 
Last edited:

Cal

Engineer
Joined
Jan 23, 2021
Messages
3,847
Location
Socal
Stopping distance is not linear with speed. The rate of deceleration isn't necessarily linear, but depending on the situation, it could be close enough.

If we assume the time it takes to go from 30 to 0 is the same as the time it takes to go from 90 to 60 or 60 to 30, the distance covered will certainly not be the same.

I don't have specific data available, but one quick Google search has suggested some trains have an emergency braking deceleration of 1.4 m/s2, which translates, very roughly, to 3 mph/second. In other words, to go from 90 mph to 60 mph would take 10 seconds, 60 to 30 mph would take 10 seconds, and 30 to 0 would take 10 seconds.

The total braking time would be 30 seconds (plus a bit of a factor as the brakes build up to maximum effectiveness). However, the distance traveled when you are going from 90 mph to 60 mph in 10 seconds is considerably farther than the distance traveled when going from 30 mph to 0 in 10 seconds. Remember, after the first 10 seconds of braking, you're still traveling twice as fast as you are on a train going 30 mph, meaning that you've probably covered roughly 2.5-ish (rough approximation) times the distance of a train that was going a constant 30 mph (let alone one that was slowing down from 30 to 0).

In the next 10 seconds, you've gone from 60 to 30, you've covered yet again another 1.5 (ish) times the distance of a train traveling at a constant 30 mph. Going from 30 to 0 covers 0.5 (ish) the distance of a constant 30 mph. I say "ish" because the actual distance covered is more of a curve than linear. In reality, you're covering a higher distance because of the time you spend at a higher speed while you slow down. Even using these rough numbers, a train slowing down from 90 mph would require 9 times the distance of a train slowing down from 30 mph.

Apply that to a train making a controlled stop on a 600-foot platform, and the engineer needs to start braking at least 5400 feet (i.e. over a mile) in advance of the stop. Certainly not 2100-2700 feet.
Ah I knew I had to be off. Thanks for the quick lesson.
 

John Santos

Lead Service Attendant
Joined
Jun 24, 2018
Messages
287
the tower may not have had a clue as to were plain crashed , if it was on radar they would only know were it disappeared from screen .
for crossings there is no general emergency number only the number on ENS sign and ID posted is relevant. even a road name is not relevant as road name a railroad uses is not always same as official name and sometimes a road crosses tracks multiple times so which crossing is it .
The pilot declared an emergency as he was taking off, too late to stop before running out of runway (after V1, if my memory of all my "Air Disaster" pandemic binge watching is correct.) He was talking to the tower.

The grade crossing where he crashed was at the 1st perpendicular street after the end of the airport's only runway. There are no buildings or trees or other obstructions except the airport perimeter fence between the runway and the street.

The railroad tracks are parallel and about 200 feet away from the runway. It was (or should have been) in easy sight of the tower, which is located between the tracks and the runway, about 1500' to 2000' from the intersection. There are no trees, buildings or hills between the airport and the tracks, street or intersection. Maybe just a low fence and a lot of flat dirt.

No radar involved, it was all VFR.
 

Trogdor

BURNiNATOR
Joined
Aug 3, 2004
Messages
5,927
Location
Here
Oh, thank you. I needed an exercise in applied math.

As Trogdor noted, "Stopping distance is not linear with speed." I wonder if deceleration is constant, because braking is changing kinetic energy to some other form of energy (often heat energy), and kinetic energy varies with the square of speed. Something going 90 mph has nine times the kinetic energy of something going 30 mph. But a varying deceleration makes computation more difficult, so let's assume a constant deceleration.

I have no better number, so let's use 3 mph/second as deceleration.

The time dependent location of an object (train) moving on a straight line (track, with no curve) in the direction of travel at constant acceleration is the starting point (milepost, assumed to be zero) plus the change due to the initial speed plus the change due to the acceleration. In other words,
s(t) = s0 + v0*t + 0.5*a0*t^2 where s0 is the initial location, v0 is the initial speed, and a0 is the initial acceleration (assumed to be constant).

Consistency in units is important. Let's use miles and hours. As I said, I assume the initial location is at zero. I'll assume the initial speed is 60 mph, although it was probably higher, based on TinCan782's comment. (Plug in a different number if you want to assume a different initial speed.) I'll assume constant acceleration of -3 mph/second, or -(3*3600) mph/hour, based on Trogdor's comment.

The location of interest is where the speed is zero. If the initial speed is 60 mph and it drops by 10,800 mph every hour, then it will be zero after 60/10800 hours, or 0.00556 hours. (Three digits should provide sufficient accuracy, but feel free to use more digits if you like.) Plugging that value into the formula,
s(0.00556) = 0 + 60*0.00556 - 0.5*10800*0.00556*0.00556

Evaluating the expression, we find the train comes to a stop after 0.1667 miles, one sixth of a mile.

Whenever conducting a thought experiment like this, it is useful to ask, "Is that answer reasonable?" With an initial speed of 60 mph and a deceleration of 3 mph/second, it would take 20 seconds to slow to zero. Happily, 0.00556 hours is 20.016 seconds; the difference is a rounding error because I didn't carry more digits. A train going 60 mph will cover 1/3 of a mile in 20 seconds, so certainly the answer should be less than 1/3 of a mile. Since the assumed deceleration is constant, it seem reasonable that the distance covered would be half the distance covered with no deceleration. Happily, 0.1667 miles is half of 1/3 of a mile. Whatever mistakes I made, I seem to have made them consistently. Corrections and clarifications are welcome.

I'm gratified to see that a 90 mph train takes 9 times longer to stop than a 30 mph train, exactly in line with the change in kinetic energy.

Because the deceleration is constant, you needn't say "ish". If you plot the speed versus time, you get a triangle. The distance covered is the area under the triangle. You are correct that "the distance covered is more of a curve" but the change in the rate of covering the distance is constant, as indicated by assuming a deceleration of 3 mph/second.

However, I think you slipped up somewhere in your calculations. If the train is going 90 mph, then it travels 1.5 miles every minute or 0.75 miles every 30 seconds. If it takes the train 30 seconds to slow to a stop, it cannot go more than 0.75 miles as comes to that stop.

I should clarify a couple of things. First, the deceleration rate was obtained by a quick google search, and not really validated (shame on me). Digging further, it seems that rate was for light rail trains, and a train such as Metrolink or Amtrak could have a deceleration rate 1/2 to 1/3 of that value.

Second, that was just a general example to show the math. The 5400-foot distance was taking Cal’s note of trains going from 30 to 0 in a distance of 600 feet (which I have also not specifically validated) and extrapolating that. But in that case, that refers to a train not in emergency but rather a train doing a normal service stop, and that deceleration rate will be much lower. The basic point was to show how a train going 3x as fast doesn’t stop in 3x the distance.
 

tgstubbs1

OBS Chief
Joined
Mar 3, 2020
Messages
645
The main question for me is, in this day of almost instantaneous communications, what are the weak links? Why the failure?
 

Ryan

Court Jester
Joined
Apr 14, 2008
Messages
17,829
Location
Off looking for his sense of humor
You're trying to pass information across three separate domains - you've got the ATC folks, the police folks, and the RR folks. They're not designed to inter operate with one another, so crossing the streams is difficult (and generally inadvisable). Perhaps you can solve the problem by putting a RR radio in the tower, with instructions to "in case of emergency, use this radio to call this dispatcher and tell him the tracks are fouled at this location", but that's really solving an edge case of and edge case.
 

WWW

No real RR Job
Joined
Nov 18, 2018
Messages
472
Location
MSP
Deleted - - -
You're trying to pass information across three separate domains - you've got the ATC folks, the police folks, and the RR folks. They're not designed to inter operate with one another, so crossing the streams is difficult (and generally inadvisable). Perhaps you can solve the problem by putting a RR radio in the tower, with instructions to "in case of emergency, use this radio to call this dispatcher and tell him the tracks are fouled at this location", but that's really solving an edge case of and edge case.
Maybe sort of like having a "RED PHONE" to call to avoid a nuclear global warming melt-down.
 

Devil's Advocate

🚂〰️〰️〰️〰️
Joined
May 24, 2010
Messages
13,301
Location
🇺🇸
Out of the dozens or perhaps hundreds of train impacts we have discussed on the forum this honestly does not seem all that bad to me. Nor does it seem likely to be repeated without special corrective action. Pilot survived, police got a win, bystanders got a story, someone got video of his own shrapnel dodging, and even the media was happy. Out of all the things to worry about this is some of the highest hanging fruit that can be seen clearly without a telescope.
 

Rambling Robert

Service Attendant
Joined
May 30, 2021
Messages
169
Location
Boston
Out of the dozens or perhaps hundreds of train impacts we have discussed on the forum this honestly does not seem all that bad to me. Nor does it seem likely to be repeated without special corrective action. Pilot survived, police got a win, bystanders got a story, someone got video of his own shrapnel dodging, and even the media was happy. Out of all the things to worry about this is some of the highest hanging fruit that can be seen clearly without a telescope.
You’re joking right?
Five police officers came very close to losing their lives AS WELL AS a retired USAF pilot. It’s been 95 years since MayDay MayDay MayDay was established as a universal EMERGENCY call across the entire aviation world... where English is the common language.

As a result all RAPID ACTION efforts are made to respond to the aircraft emergency including shutting the airport down if needed.

What universal distress call is associated with the railroads?
 
Last edited:
Top