Amtrak Unlimited Discussion Forum: Replying To WASHINGTON: NTSB hears from Metro on fatal train crash - Amtrak Unlimited Discussion Forum

Green Maned Lion, on Thu, Feb 25, 2010, 04:07 PM, said:

Manual operation would not matter in this case. Manual operation refers to operating with ATO cut out. Train speed, stopping, and starting are handled manually. But ATC, the system that sets speed restrictions and ensures separation, is still active with ATO cut out. So, regardless of ATO operation or not, the speed allowed per ATC would bring that train around the curve and into the stopped train. The operator hit the emergency stop button as soon as she saw the stopped train, but it was too late.

What you are describing and arguing about sounds to me like a system that has had way too many flaws. If I was an operator on that train, and I did not trust the system to safely operate automatically, I might consider it a breaking of regulations to operate it manually- but I would consider it a breach of my moral responsibility for the safety of my passengers to allow the trains control to be placed in the hands of a system I didn't trust to provide for safe operation. I'd either keep operating them manually, or quit.

I didn't say that it wasn't designed to account for it. What I said is absolutely correct, you simply added the "why" the system wasn't adjusted for the slippery track.

Ryan, on Tue, Feb 23, 2010, 10:15 PM, said:

That statement about the 1996 accident is not true. The system was designed to account for slippery rail conditions. As a matter of fact, the controller had that train set for the most restrictive of the 8 levels. However, because the train had run past the end of the platform and the circuit breakers in the front car were opened to enable doors in the other cars to open, the train lost its performance level setting and defaulted to the least restrictive one. Definitely a flaw, but definitely not because the system was not designed to account for slippery rails and increased stopping distances.

Incidents like this one and the one from a few weeks ago with the derail really make me wonder why they are not running trains in automatic. Auto/manual had absolutely nothing to do with the incident in June.

It's not a direct application of FRA rules, but there are a handful of moves in that direction on the table.

http://www.washingto...0022204794.html

Seams it's time the FRA got involved and forced the WMATA to run under FRA rules. For the 1996 crash the family of the killed operator could have sued WMATA for work place negligence. They refused to allow the operator to run it manual and look what happened. This is why we have workers comp.

1. WMATA doesn't have a healthy culture of safety, so operators can get away with this sort of stuff.

2. On the other hand, the 1996 accident at Shady Grove was caused by the automatic system not adjusting for slippery track. The operator asked for permission to run in manual so that he would stop sliding through stations, was denied (more than once, iirc) and was killed when he slid through Shady Grove (a terminal station) and into a train parked on the tracks just past the station. http://www.ntsb.gov/...996/rar9604.pdf

Apparently the struck train in the June 2009 incident "disappeared" from the train control system, so in essence the automated system drove the striking train into the collision.

AlanB, on Tue, Feb 23, 2010, 09:55 PM, said:

I think there are two things I am curious about.

1. How do train operators have a history of 'reprimands'. (I get the feeling to big carriers don't do reprimands, they terminate).


2. What is the quality of the equipment... why are operators choosing to do something different (run in manual). Is the automatic equipment faulty? Do they not trust it?

You are correct that in most cases if someone runs a signal, they'll find themselves on the street.

However, the Metro doesn't operate under FRA rules, so they don't apply here. Not to mention that I didn't see any mention of the operator running a signal. The article talks about one operator running their train in manual instead of automated mode, but never suggests that they ran a signal.

No signal ever got ran, so I don't know what your talking about.

Also, the struck train being in auto or manual should have no bearing on the incident.

However, I came across this document on the NTSB website - it seems to suggest that the striking train should/could have been able to stop.
http://www.ntsb.gov/...R007/434700.pdf