Metrolink Crash in Chatsworth, CA - 9/12/08

My question was about how the NTSB, specifically, investigates incidents involving trains. I'm sure diff gvt agencies have diff procedures, and I am sure there are issues unique to train accidents that would not figure into, say, a plane crash. I wonder if I can find out more or if they are just too cagey abt exactly what they do. Prolly the latter... oh well

Just go to the BPL and read one of their reports. They're not cagey. They keep their investigation under wraps while it's in progress. That's not being cagey - that's being professional and not ruining or contaminating your investigation. Once they're done, the NTSB is very detailed in their investigations. So is the FAA.

Concordgirl the NTSB is thorough and exhaustive. As I mentioned, you can Google "NTSB" get a link to their home page and then follow the cues to a huge number of NTSB Railroad Accident Reports. By reading a couple of the reports you'll learn their methods and I think you'll agree they're thorough. They're not infallible though. I've even found mistakes, some pretty serrious ones in my opinion, in some of the reports.

I believe they've already re-staged the accident, have they not? Using equipment as close to duplicate as they could get, to establish when the Metrolink engineer would've first seen the approaching UP freight. I think they said it was 4 seconds.

I have to wonder about why the engineer failed to apply brakes. Four seconds goes by pretty fast. Do you think the engineer might've been so stunned, so shocked to look up and find he was staring at an oncoming freight, knowing it was too late to really take any effectve action, that he may've been shocked into inaction? If he'd been distracted by texting he would've known in a heartbeat the enormity of the mistake he'd made and the almost certain fate he was going to suffer as a result. God what a horrible feeling that would be!

Still want to see what the drug screen shows.

Tommy Meehan wrote:Concordgirl the NTSB is thorough and exhaustive. As I mentioned, you can Google "NTSB" get a link to their home page and then follow the cues to a huge number of NTSB Railroad Accident Reports. By reading a couple of the reports you'll learn their methods and I think you'll agree they're thorough. They're not infallible though. I've even found mistakes, some pretty serrious ones in my opinion, in some of the reports.

I believe they've already re-staged the accident, have they not? Using equipment as close to duplicate as they could get, to establish when the Metrolink engineer would've first seen the approaching UP freight. I think they said it was 4 seconds.

I have to wonder about why the engineer failed to apply brakes. Four seconds goes by pretty fast. Do you think the engineer might've been so stunned, so shocked to look up and find he was staring at an oncoming freight, knowing it was too late to really take any effectve action, that he may've been shocked into inaction? If he'd been distracted by texting he would've known in a heartbeat the enormity of the mistake he'd made and the almost certain fate he was going to suffer as a result. God what a horrible feeling that would be!

I was just thinking the same thing a bit earlier and am suprized that no one mentioned the possibily of him looking up to see a freight train coming towards him and just freezing up! I'm also wondering if he even looked up at all before the impact which will never be answered unless they can zoom in on the camera from the freight engines and see if he was looking up or not.

3rdrail wrote:Still want to see what the drug screen shows.

Had the drug screen come up positive, it would have been all over newspapers by now.
A drug screening only takes 96 hours at most, and even NTSB would have released such information.

DutchRailnut wrote:The Metrolink engine did not have a camera, only the UP engine did.

I wonder if we'll ever see video or image of that!

With all the modern technology I was wondering if there is a possibility that in addition to recreating the crash if they could recreate his run coupled with the times he sent/received text messages? It would be interesting to see if by fluke he was sending/receiving texts around the same time he was passing within sight of the signals.

Tommy Meehan wrote:Concordgirl the NTSB is thorough and exhaustive. As I mentioned, you can Google "NTSB" get a link to their home page and then follow the cues to a huge number of NTSB Railroad Accident Reports. By reading a couple of the reports you'll learn their methods and I think you'll agree they're thorough. They're not infallible though. I've even found mistakes, some pretty serrious ones in my opinion, in some of the reports.

Cool, thanks I will check that out. I'd be curious what the first day being called to investigate a wreck is like. I'm sure they have procedures to follow, just like archaeologists on a dig start by mapping out the area in a grid. I'm curious about how they deal with issues for which there will be no definitive evidence one way or another. How thoroughly do they look into less likely scenarios like signal malfunctions? Do they just restage the accident and say Yup, the signals are the correct color when we went through them, and that's it? Or do they redo the accident multiple times over different days in different weather?

I have a question for the engineers on here. How often do you guys encounter signals displaying the wrong color, or otherwise malfunctioning? Is it like once a week, once a month, once every few months? I'm starting to wonder how reliable they actually are, I obviously don't work in that field so I have no idea

I will add one thing about the signals... Signals are very reliable for the most part, just because this accident has brought the fact up doesn't mean that railroad signals should be seriously taken into question on whether they are reliable or not because they are and like all things they do have issues at times and that is why we have rules for malfunctioning or improperly displayed signals. There have been no serious accidents as a result of the signal system and I don't think now is a time to panic. The NTSB does a very thorough investigaion which is why it takes a long time for them to release a full report on the accident, they hit every corner possible and for the most part come up with very intersting findings. regarding the signals they do downloads from the black boxes and runs multiple tests along with checking them inside and out so if all signs point towards them working and there final word is they work then that is what we have to go on unless we hear otherwise. In this case if the signal somehow showed clear then yes there would be no rule for that regarding a signal malfunction because there would be no way to know that it was wrong unless told or being able to see why it's wrong but otherwise in the case of a signal malfunction example it being dark or all lit up you wouldn't just proceed normal as if it's nothing you are governed by the rules for a Improperly displayed signal. Mostly all R.R accidents are caused by HUMAN ERROR and I am going to stick by that for this one as well!

I just want to throw this theory that I thought about out here. I know that a signal test was done by the NTSB and the correct aspects that should have been displayed were, but as we have seen and heard in these last few pages, signal malfunctions, although very, very rare, do occur. Someone said many pages back the engineers on Metrolink have gotten into the habit of not calling clear signals on their runs (If I remember correctly unless I read the post wrong)......well during class, I was thinking, if the report of the 3 people on the platform is true and the signal was indeed green, (which we all know should not have been) then the engineer (May he rest in peace) would have not bothered to call the clear signal at the CP and therefore continued on running down the line acting on a clear aspect until it was too late

DutchRailnut wrote:

3rdrail wrote:Still want to see what the drug screen shows.

Had the drug screen come up positive, it would have been all over newspapers by now.
A drug screening only takes 96 hours at most, and even NTSB would have released such information.

Not so sure about it's release, Dutch. Considering the apparent lack of concentration, I would think that a clear drug screen result would be just as newsworthy. I haven't seen mention of either.

As of September 30, they had not released any of their findings from the tox screen done during the autopsy. Idk if it just takes longer than people think, or if they are holding the results back.

This explains in a little more detail to what extent the NTSB addresses the issue of false signals:

"In the days after the crash, investigators conducted three tests to check whether the signals were functioning properly.

The first was an electronic test of several signals before and after the Chatsworth station. The second was a "sight distance" survey that, among other things, checked whether Sanchez would have had a clear view of the signals. Investigators also examined computer data to see if the signals were working correctly on the day of the crash.

NTSB board member Kitty Higgins has said all the signals -- including two south of the Chatsworth station and one north of the station near the track switch mechanism -- were visible and working. Not only was the final signal red, Higgins said, but the light just before the station was solid yellow. "That indicates that the third signal is red, that you stop," she said.

"We can say with confidence that the signal system was working," Higgins said at a news conference after the tests.

NTSB officials have said a final test was planned to confirm that the two yellow signals and the red stop signal were all functioning properly. Results have not been announced."

concordgirl wrote: I'm curious about how [NTSB] deal with issues for which there will be no definitive evidence one way or another. How thoroughly do they look into less likely scenarios like signal malfunctions?

The NTSB are nothing if not thorough, they're experts when it come to looking at an accident. They're technicians and they don't hesitate -- based on the reports I've read -- to admit when they could not make a conclusive finding.

I believe the NTSB is largely responsible for the locomotive and signal box data recorders we have today. In earlier times they lobbied hard for them. (Something the railroads resisted at first on cost grounds.)

Btw, I have read NTSB reports where malfunctioning signals were involved. One in Texas was due to a maintainer's error. Another involved a freight train being unable to stop in time to avoid hitting another train at a junction. The student engineer claimed he never saw the distant signal -- that it must've been dark -- and by the time he realized he'd passed it (when he saw the home signal displaying 'Stop') there was not enough distance to stop the train short of the collision. There were other factors, but NTSB witnessed the signal go 'dark' and discovered there was a defect, not just with that signal, but that entire model of signals. There was a case in Pennsy involving a Penn Central freight passing a stop signal that, when the sun hit it a certain way, did appear to be yellow when it was actually red. I don't want to say more -- I don't want to be reprimanded by the mod for going off-topic -- but all of these were documented by NTSB. I have the reports.

But when you think of the thousands of trains that operate everyday, and the hundreds of thousands if not millions of correct signal indications they receive, the systems are very reliable. I too seriously doubt malfunctioning signals played a role at Chatworth. But you hope the investigators keep a open mind so on those rare rare occasions when it does happen they will uncover it.

The NTSB is still a very political organization, subject to political and practical issues. They will identify a myriad of factors leading to the crash without pointing at any one or two individually. Even with the NTSB's prior determination that Sanchez was texting just prior to the Metrolink train crash, they will look at lots of other causal issues for the train crash. While they are certainly thorough (the investigation file will be massive), they will often take a look at the issue of what incremental safety advances can be achieved if recommended. The NTSB has been pushing for automatic stop technology for 10 years, and even more so these last 5 years, but have come up against train industry executives and their lobbyists. The big question is whether they will use this event as a platform to make a big push for such technology or whether they will push is some other 'lower cost' safety direction.

RR Safety wrote:The NTSB has been pushing for automatic stop technology for 10 years, and even more so these last 5 years, but have come up against train industry executives and their lobbyists. The big question is whether they will use this event as a platform to make a big push for such technology or whether they will push is some other 'lower cost' safety direction.

Congress already made their push - the new legislation that passed, and I believe was even signed, requires PTC on lines meeting certain requirements (passenger rail, being one of them), by 2015 or so.

Now we get to see if a standard technology gets adopted, or if every RR does their own, thus duplicating the mess that lead Europe to ETMS...

Bold prediction: After a few years of kicking and screaming, the industry will move forward, then suddenly realize the safety advantages far outweigh the initial costs, and adopt whatever standard everywhere. Especially given how the cost will certainly drop fast - electronics just don't stay expensive very long these days.