UP Trains Collide in Goodwell, OK

Since I'm not entirely sure that the horse we're beating is dead yet, and various questions of signal failure have been raised:

What do Union Pacific rules at this location say about a conflict between signal indication and switch position? (In 2008, I wondered the same about Chatsworth, Calif.) Is a train allowed to run through a trailing points switch set against it if a signal indication permits the move? I know that a switch target might also be a signal, but I'm not sure if this powered switch had one.

Articles in the trade press had railroad spokespersons bragging that for reasons of fuel economy, their Dispatchers do keep crews informed of the situation ahead, to avoid running trains on the "hurry up and wait" system. For the benefit of stock market analysts, they predict how much money the Arabs will lose due to the practice. Other posts from engineers on this Group claim that their Dispatchers don't go with that program.

JayBee wrote:The Alerter works off a variable timer, at very low speeds the time between Alerter activations is fairly long, at speeds in the 60 mph range it is about every 15 seconds.

So if you are cruising at 60 mph, you are having to respond to alerter 240 times per hour? No wonder it becomes something you can do half asleep. JS

On the original Metroliner Multiple Unit cars from 1967, touching grounded metal in the cab would also satisfy the alerter.

JimBoylan wrote:What do Union Pacific rules at this location say about a conflict between signal indication and switch position?

GCOR (which UP is part of) says:
"Do not run through switches, other than spring switches or variable switches."

Of course IF the signal was seen as clear (or anything more permissive than stop or restricting) and the train was running at speed not running through the switch is more difficult. But the signal should not have been permissive if the switch was not aligned correctly. The signals and track were reported to be in working condition after the accident ... which precludes a run through (which would have damaged the track switch). The bad signal could have been an intermittent problem, but running through a switch aligned for the other track should have caused damage.

Jim,

UPRR rules don't allow running through any switches, except spring switches, at least not intentionally. And main track power switches (sometimes called dual-control switches) do not normally have targets. Even if they did, a train usually can't see the target until it is very close.

Sometimes the TD will tell crews of opposing meets to save fuel, but they are usually so busy handling the rest of their territory, answering the radios, phones, giving out Track and Time or Track Warrants that that they don't have time to do this. UPRR has automated so much of it's Dispatcher processes, that it has an "automatic" mode for it's CAD (computer assisted dispatching system). CAD-zilla as it is not-so-affectionately called, can plan moves, request switch and signals, and head lower-priority trains into a siding to allow higher priority trains to meet or pass. But remember, CAD is just a computer program overlaid into a CTC system, which should not be able to override the ABS signal system and allow conflicting movements.

justalurker66 wrote:

JimBoylan wrote:What do Union Pacific rules at this location say about a conflict between signal indication and switch position?

GCOR (which UP is part of) says:
"Do not run through switches, other than spring switches or variable switches."

Of course IF the signal was seen as clear (or anything more permissive than stop or restricting) and the train was running at speed not running through the switch is more difficult. But the signal should not have been permissive if the switch was not aligned correctly. The signals and track were reported to be in working condition after the accident ... which precludes a run through (which would have damaged the track switch). The bad signal could have been an intermittent problem, but running through a switch aligned for the other track should have caused damage.

However, if you look at the picture NTSB posted of the ESS at Goodwell, clearly something is not right with it (linked above). One point is aligned for movement through the diverging route, while the other (which should be firmly against the outside rail I think?) is pushed away a couple of inches. The "throwbar" (or whatever it's proper name is) appears to have some bent parts in it. Whether this means anything or not, I'm not prepared to say. But it certainly doesn't look like a turnout in good working order to me.

up8677 wrote:However, if you look at the picture NTSB posted of the ESS at Goodwell, clearly something is not right with it (linked above). One point is aligned for movement through the diverging route, while the other (which should be firmly against the outside rail I think?) is pushed away a couple of inches. The "throwbar" (or whatever it's proper name is) appears to have some bent parts in it. Whether this means anything or not, I'm not prepared to say. But it certainly doesn't look like a turnout in good working order to me.

That damage is the result of the East Bound Stack train running through the ESS switch at Goodwell, a switch that was obviously lined against them. At 68 mph I'm surprised the damage isn't worse that it is, but I suppose that's a testament to the engineering that goes into our infrastructure.

What it means is that there's even more credibility to the fact that the crew of that East Bound train, made some very serious errors. I'm sure the final NTSB report will be quite telling as well.

I asked about the rules for running through a stiff points switch set against you because of other head on collisions (like Abbington Twp., Pa.) where an engineer claimed signal permission to proceed onto single track, but did not discuss the damaged switch.

I know the classic definition of restricted speed included being prepared to stop short of switch not properly lined, so if that's still in there even a restricting indication would not permit a move through a switch (other than a spring switch) that's lined the wrong way for your move.

ExCon90 wrote:I know the classic definition of restricted speed included being prepared to stop short of switch not properly lined, so if that's still in there even a restricting indication would not permit a move through a switch (other than a spring switch) that's lined the wrong way for your move.

Correct, by rule you are supposed to be prepared to stop before a misaligned switch. However, if the move doesn't stop before the misaligned switch that's rule viol;ation #1, which would eventually snowaball into more rules violations, such as occupying main track without authority, and of course all too often, and collision and or derailment.

JimBoylan wrote:I asked about the rules for running through a stiff points switch set against you because of other head on collisions (like Abbington Twp., Pa.) where an engineer claimed signal permission to proceed onto single track, but did not discuss the damaged switch.

Here's the NTSB report for that one. http://www.ntsb.gov/doclib/reports/2008/RAB0803.pdf Very similar to what MIGHT have happened in this UP case. Good thing SEPTA wasn't going 70 mph on that stretch of track.

Not to start a fight, but if management ran the risk of death every time someone had a brain fart, or some electronics screwed up, you can sure bet they would have triply redundant, effective, back-up safety systems without griping and delay because of their cost.

JS

To the various comments about what if you're given a proceed indication into a switch lined against you, no properly functioning interlocking (and each control point of a CTC system is an interlocking) is going to give you any sort of proceed indication (including Restricting) into a trailing point switch that is not lined for you. You could get Restricting into a block and find a switch lined against you outside interlocking limits but that's not the case here.

I'm not aware of the rules I worked under even addressing the issue of a proceed indication better than Restricting (Restricting, unlike every other proceed indication, does not guarantee you a clear block) and encountering a switch set against you. I goes what would cover it is what for us (and I think still us under NORAC rules) was Rule 108: "In case of doubt or uncertainty, the safe course must be taken". Encountering something the signal system should not permit certainly meets my definition of "doubt or uncertainty".

lstone19 wrote:I'm not aware of the rules I worked under even addressing the issue of a proceed indication better than Restricting (Restricting, unlike every other proceed indication, does not guarantee you a clear block) and encountering a switch set against you.

The rule I noted above would apply: "Do not run through switches, other than spring switches or variable switches." But adding some rational thinking, how would a train know that a switch was set against them unless the signalling system told them (via a restricting or stop signal) or they had another form of warning? Assuming a train was operating under control of a better than Restricting signal there would be little chance of avoiding running through the switch.

lstone19 wrote:Encountering something the signal system should not permit certainly meets my definition of "doubt or uncertainty".

And *IF* an engineer had time to note the conflict stopping would be appropriate and following the rule of not running through switches. If a signal was more permissive than Restricting the switch problem would not be expected and would be harder to avoid.

(Not forgetting that this particular switch is part of an interlocking where there should not have been a permissive signal unless the switch was lined properly and the track was clear to proceed, both physically and with the dispatcher's permission.)

justalurker66 wrote:

lstone19 wrote:I'm not aware of the rules I worked under even addressing the issue of a proceed indication better than Restricting (Restricting, unlike every other proceed indication, does not guarantee you a clear block) and encountering a switch set against you.

The rule I noted above would apply: "Do not run through switches, other than spring switches or variable switches." But adding some rational thinking, how would a train know that a switch was set against them unless the signalling system told them (via a restricting or stop signal) or they had another form of warning? Assuming a train was operating under control of a better than Restricting signal there would be little chance of avoiding running through the switch.

Well, of course. That should never happen. At an interlocking, a rule about not trailing through switches should never even be in play. To get to that point, one of two things had to have already happened: either the train violated a signal or the signal system failed. If the latter, you don't go hanging employees for rule "violations" that occur due to some other failure.

I'm sure we had a rule about not running through switches but it's not something I ever even thought about applying to a situation where by other rule (signal rules), you were "guaranteed" a properly lined up route. Rather, it was something to think about (well, actually to me it was always just plain common sense) in yard or other unsignalled operations.

Somehow I doubt, if they did get by a red signal and then run-through the switch back into single-track, that the NTSB will write-up the reason as running through a switch. That would be secondary to the real failure.

There are 2 "talking " detectors near Guymon,I wonder if they have data loggers as well?
From what I have seen in the past along UP rails, UP uses 2 "approach" type signals to warn of a switch ahead, then
the "Home" signal at the switch.Most siding switches are good for 30-40 mph "diverting" moves at best.

Would the 2 "talking" detectors be considered "distractions" by being only approx 4 miles apart?
Did the crew acknowage the detector's "No Defects" message on the radio??