Eastern Railroad Discussion > Crossing Signals - could this be possible

OK we now know there are two reports with different findings as to what happened to the City of New Orleans in March, 1999.

The NTSB report says the trucker whent around the crossing gates and caused the wreck.

The Illinois State Police conclued that the gate came down late and that the trucker did not go around it. According to the State Cops, the crossing lights activated properly, but the gate fell late. They say the trucker disregarded the flashing red lights, but the gate wasn't there as it should have been.

Is it possible for the system to fail as the police say it did?

> Is it possible for the system to fail as the police say it did?


Possible, kind of sorta unlikely. This is not un common for the gate to have a long delay vs. the activation of light pretection. It is unlikely because the circumstances are not understood. It couldn't have been long enough that the trucker could have been a quarter mile or even 2/10 of a mile down the road before the gates activated. And did he see the lights activate or where they on when he was distant from the crossing? Questions that must be awnsered in the investigation.

It is also possible that the gate was on time and that the lights malfunctioned by activating early on a shorted crossing circut. This is even more unlikely though.

Matt
The Roselle Park Railfan
mjryan13@aol.com

Let me tell you something. I maintained signals for 16 yrs and have seen quite a few things that can go wrong with a crossing signal. 99.9% of the time crossing signals work as they were designed. But theres always that .1% that they will do somehing different, from not activating at all, to the lights flashing and the gates not coming down, to one gate coming down and the other not, to the gates picking up right in front of a train and it goes on and on.There are hundreds of variables that can play a part in a signal malfunction.
Yes the lights may have been flashing and the gates had not started down yet, normally gates are designed to be down in no less than 12 seconds after the crossing activates or more than 15 seconds after activation. Most crossing are designed to detect the arrival of a train at a crossing within 20 sec that the lights start to flash in the case of just flashing lights, or 30 sec in the case of most gated crossings. It may vary from RR to RR, but I think the FRA minimum warning time is 20 seconds.
Its a shame what happened. But I'd bet doughnuts to dollars everything was working as intended and like I saw just about everyday of my career as a maintainer and now as a conductor, IMPATIENCE played the part in this accident. Who knows?
I do tell people though to never trust a crossing signal. That they can and do fail. They are not fail-safe! That the crossbuck is a YIELD sign and the flashing red light means STOP!
Simple.

iamtrakn2 wrote:
>
> > Yes the lights may have been flashing and the gates had not
> started down yet, normally gates are designed to be down in no
> less than 12 seconds after the crossing activates or more than
> 15 seconds after activation. Most crossing are designed to
> detect the arrival of a train at a crossing within 20 sec that
> the lights start to flash in the case of just flashing lights,
> or 30 sec in the case of most gated crossings




Any chance that the built-in delay between the lights flashing and the gates dropping is being mis-understood by the State Police ?? That what you (we) know to be normal is perceived by them as a problem...because the gates didn't drop right away ????



Nickatnight

Quite Possible, unless the troopers were trained!
We have had three people end up with a gate on their car hood/windshield on our tourist railway ... and those gates and flashers were working as they should. Once again they are set up that way to allow a vehicle that is past the halfway point to clear the crossing. I was surprised when PUCO inspected our flashers and gates and had us reset the timing and check the operation.

Yes it's possible, but extremely unlikely. The crossing had an event recorder, which indicated that the monitored portion of the system worked as intended: The lights started at the appropriate time, the relay powering the gates dropped out 4 seconds later, and the crossing was occupied by the train after the light started. All of the equipment was tested later, and found to be operating correctly.

The problem is that one witness said that the crossing circuit didn't operate at all, and another stated that the gates definitely didn't drop until later, and then hit the top of the truck's load. Yet another witness said that the truck drove around the descended gates. Thus there are three different accounts. The event recorder on the signal only senses whether the relay that activates the gates worked properly, but not that the gates physically descended, thus there is room for speculation that the gates remained up.

Other evidence was that the tip of the crossing arm on the far side of the crossing was broken toward the east, indicating that something hit it, and the bottom of the crossing arm on the west side had no marks on it that would indicate that it had been dragging along the trailer's load (rebar) before it broke off, as some had suggested.

The bottom line is that the protection system likely worked as designed. The NTSB seems to feel that the gates came down as intended, but the State Police don't. In reading the testimony at the NTSB hearing, I side with the NTSB on the issue.

Even if the gates didn't come down immediately, why did the driver keep on moving over the crossing with the lights flashing? Take a look at a clock and see how long 26 seconds actually is and imagine yourself approaching a crossing with that amount of warning. Maybe the lawyers can plant some doubt in the minds of jurors about the gates working, but only an idiot would drive over the crossing with the lights operating and a train approaching.

Yes to the delay and all valid points gentlemen!

The fact is even if the gates did malfunction the lights were on. As is in most states and I know in Ohio if those lights are flashing you stop! Period! What needs to be answered is why the guy was driving a truck on a suspended licensed.

I, too, am a signal maintainer and I agree with both iamtrakn2 and run8. It is possible for a gate not to descend when suppose to, but it doesn't sound like that is what happened in this case. Run8 obviously knows what the event recorder revealed and facts of the investigation revealed. Unfortunately, it doesn't sound like a modern event recorder. Modern day event recorders use a contact in the gate mech to indicate when the gate is completely horizontal along with what run8 explains about the de-engergizing relay that drops the gate. With the marks, or lack of marks, on the gates as explained, I don't see how the Staters came up with their conclusions. From run8's statement, the relay that begins the gate's decent was de-energized 4 seconds after the activation of the crossing, so we can assume that the gates began their decent four seconds after activation. With no marks on that gate, the only way for that truck to get hit was to go around the gate. If there was something interfering with the hold clear or mechanism of the gate that wouldn't allow it to fall after the relay was de-energized, believe me, FRA or somebody would have found it. FRA crawls all over those things in a situation like that, especially if a witness claims the gate didn't come down when it was suppose to. They would have definitely found a problem if there was one. I side on NTSB's decision.

Xingman
http://home.internetcds.com/~xingman/
Maintaining the signals of the Siskiyou Line

Xingman wrote:
>
> Run8 obviously knows what the event recorder
> revealed and facts of the investigation revealed.

A transcript of the NTSB public hearings is available on their web site. Most of the signaling testimony was on the second day:

http://www.ntsb.gov/events/1999/bourbonnais/details.htm

> Unfortunately, it doesn't sound like a modern event recorder.

It was a Safetran GCP-3000 installation, which is their latest top-of-the-line design. The General Signal Supervisor testified that the gate mechanisms didn't have the necessary contact to monitor the gate position. Perhaps an old mechanism.

The control system tested 100% on site, and back at the manufacturer's plant. All of the relays and mechanical parts tested 100% as well.

From what I read at the NTSB site, the trucks trailer wheels were in the opposite side of the crossing when struck by the train. Mention was made of marks on the wood crossing where the wheels dragged after impact.

When you look at the NTSB simulations, and read the testimony, it seems from the position of the trucks trailer tires that the driver knew what he was doing (i.e. going around the gates), and that the crossing protection was working.

Ken
Mason City, IA

run8 wrote:
>
> Xingman wrote:
> >
> > Run8 obviously knows what the event recorder
> > revealed and facts of the investigation revealed.
>
> A transcript of the NTSB public hearings is available on their
> web site. Most of the signaling testimony was on the second day:
>
> http://www.ntsb.gov/events/1999/bourbonnais/details.htm
>
> > Unfortunately, it doesn't sound like a modern event recorder.
>
> It was a Safetran GCP-3000 installation, which is their latest
> top-of-the-line design. The General Signal Supervisor
> testified that the gate mechanisms didn't have the necessary
> contact to monitor the gate position. Perhaps an old mechanism.
>
> The control system tested 100% on site, and back at the
> manufacturer's plant. All of the relays and mechanical parts
> tested 100% as well.

run8,

The GCP3000 is the controller for the crossing, not the event recorder. Although the 3000 is the latest and greatest for Safetran controllers, it isn't the event recorder. The latest event recorder by Safetran is the SEAR unit (Safetran Event Analyzer Recorder). Before the SEAR's they had the 80025 external interfaces. And before that, just the 80015 record board that mounted inside the 3000. Although none of this is really important if the mech didn't have enough contacts to monitor the gates. I'll read into the signal testimony tonight, thanks for the link.

Xingman
http://home.internetcds.com/~xingman/
Maintaining the signals of the Siskiyou Line

From what I read in media accounts(I have not yet read the NTSB analysis) the reports of the NTSB and the Illinois State Police differ in their findings. A statement summarizing the ISP version was that "The accident did not occur in the manner that the truck driver stated". Another topic of interest is that I read a statement that fatigue, on the part of the truck driver, was believed to play a role in the incident, and that the driver was convicted of Falsification of Federal Hours of Service records and Fasification of Logbook used to record rest and miles driven over a previous 37 hour period. Each count carried a 3 year prison term and probably a hefty fine, although fines were not disclosed in the account I read. What a shame 11 people had to die so that this individual could remain competitive with other modes of transportation. A bigger shame is that this is only a small drop in a very large bucket...

Xingman wrote:
>
> The latest event recorder by Safetran is the SEAR unit (Safetran
> Event Analyzer Recorder). Before the SEAR's they had the 80025
> external interfaces. And before that, just the 80015 record
> board that mounted inside the 3000.

Thanks for the explanation. When you suggested that it might not have been a modern event recorder, I wasn't sure if you knew the control system was a GCP, which would most likely have had an integral solid state recorder. I overlooked the problem of the external interface.

The CN/IC crossing was equipped with the 80015 internal board and an 80025 external interface, according to the testimony.

I think it is a shame. And then to read the testimony from the "Gentleman" from the ITA (ATA) almost made me ill. For years, various arms of the American Trucking Association (ATA) has fought the installation of event recorders in trucks. Yet in the condesending mannor of his testimony, he supported such devices.

Chairman Black in the Hearing, made one pointed comment I wish to present here. Pardon it's lenght, however I believe it is worth the reading.

"CHAIRMAN BLACK: I think I would encourage
you to look at some private companies who run their own
fleets who have very high success rates with safety and
with equipment utilization and equipment maintenance,
costs, fuel costs, and they attribute all of those to
event recorders.
I think it is significant, like I said to the
media several times, in this accident, we have two time
lines very well established in this accident. One of
them is from the railroad signal equipment because of
the event recorder the railroad had on their equipment.
By the way, the reason they had that piece of
equipment was for maintenance issues so it would allow
them to troubleshoot the equipment, but it also turns
out to be a very effective tool in accident
investigation and probably in defense when it comes to
it ultimately to get into that issue. Of course, the
locomotive had an event recorder on it.
Here, this truck obviously involved in
interstate commerce, we don't know anything about it
other than eyewitness statements, and we would not be
having much of a discussion at this meeting had the
trucks had event recorders on them. Some of them
already do."

Mr. Black is right on the money in my opinion.

Ken
Mason City, IA