It all went down west of Opelousas in the hamlet of Lawtell.
Map Lawtell: Click to expand.
Mark wrote early Tuesday morning. I could tell he was frantic.
2/14/2012 o5:46
The dispatcher messed up, the train is 10 feet longer than the siding and still has cars on the main and the train on the main can't pass! LOL.
Waiting to hear what they do.....
Me: From the start I had read "10 cars". Possibly he wasn't frantic, but enjoying a bit of dark humor?
Mark added:
2/14/2012 05:57
Dispatcher thought he had 140 feet to spare on the train in the siding....ch 27 if you want.
Me: I was too late, Channel 27 was quiet as the company was now communicating "off air".
Mark:
2/14/2012 06:35: By manually switching the switches, they got by.
I am guessing there are some safety lockouts to prevent close calls, but you can gain some feet by over riding them.
I excitedly responded: (I generally don't use "oh goodness". It is used here in an effort to promote civility)
Oh goodness, you mean that it came down to a visual situation? How'd he stop? I guess he's slowed down for a pass but still. Bet that doesn't make Union Pacific's Great Moments of the Month newsletter.
I added: Dang that was exciting. Thanks for the report. Imagining the tail of a protruding train being wiped out by an oncoming one is stuff Hollywood lives for. ("Dang" is only used in public writing in an attempt to promote civility.)
2/14/2012 09:13 Mark responded:
It was still dark and I was not awake enough for a drive.
Good thing the east bound saw his {tail} end hanging out on the main!!
"tail" is not the usual word used in such circumstances. And, he hadn't. I changed it to promote civility.
I added:
I bet ya someone is in hot soup this morning.
"Hot soup", again, a civil gesture.
Mark wrote 2/14/2012 09:42 :
The dispatcher in Spring said his computer had the siding length wrong.
I wrote back, always the skeptic:
Blame it on the computer, that works.
In the following I refer to comments and answers given by an "Agent 00-L", He wishes to be identified as an informed laymen, not a professional railroader. I think his story is deeper. The CIA has come to mind. If Jewish, "Mossad". Got the picture?
I had related my version of what I believed happened at Lawtell to 00-L. His reply:
No, I haven't heard of whatever happened at Lawtell. It sounds like they probably performed the rare and ancient ritual called "sawing by," which allows two trains to pass each other at a siding that is too short to hold either one of them. It sounds impossible, but here's how it works:
Consider an east-west line with a passing siding, like the situation at Lawtell. Train E is easbound, and Train W is westbound. Neither will fit into the siding for the other to pass.
One of the trains (Train E in this example) cuts enough cars off the rear of its consist so that the rest will fit into the siding, and it abandons the excess cars on the main to the west of the west switch, while the remainder pulls fully into the siding and stops.
Train W comes along the main and passes the front part of Train E that's stopped on the siding. When the nose of Train W reaches Train E's abandoned excess cars, it couples onto them and keeps pushing westward until its own tail clears the west switch. Then, it stops.
The main part of Train E now heads eastward out of the siding onto the main track and continues until its tail end is more than a train length east of the east switch, whereupon it stops.
Next, Train W (with Train E's excess cars still coupled to its nose) backs itself all the way through the siding eastward, stopping briefly to cut off the excess cars from its nose in the siding. After doing so, it continues to back eastward until it is completely through the east switch and back out onto the main. It then high-balls westward toward its destination.
Finally, the main part of Train E backs up into the siding, where it reconnects to its orphaned cars, then high-balls eastward.
Probably only a few in every hundred modern railroaders have ever sawed-by, and most don't even know how to do it. Some would swear that it can't be done.
I wrote Agent 00-L back:
I must be a reincarnated old railroader, I had planned that in my head (a lie) but was not sure about a very very long train backing up. Maybe it wasn't very very long. Now, what do you think about a visual being the only warning to stop?I don't buy that. The rear camera would have told the sided engineer that he had not cleared the switch. There would have been screaming on the radio.
I will send this to Mark to let him know that he missed history reenacted and a very rare show. Thanks for the excellent explanation. It was visual.
OO-L wrote back:
Steve,
The box on the rear coupler of a train's last car contains no camera. That box is called a FRED (flashing rear-end device). Each one has a reporting mark and car number so that it can interchange among railways just like a railcar. That usually gets done on run-through trains.
Besides flashing a red light whenever it doesn't sense daylight, a FRED communicates by radio with a corresponding electronic unit called a "Wilma" (think Flintstone) in each locomotive. The FRED is attached to the air brake line on the last car, and the Wilma has a digital readout of the air pressure as sensed by the FRED. This tells the engineer when he's pumped up enough air to release the brakes on the far end of the train. (Pressure does not immediately equalize through a half-inch pipe that can be well over a mile long.) During an emergency brake application (when the brake-line air is dumped from the locomotive end), the Wilma tells the FRED to dump the air from the rear end as well, thus greatly increasing the speed of brake application in the rear half of the train. If a train breaks in two, the FRED tells the Wilma that air pressure has been inexplicably lost. The FRED has an inertial motion sensor to tell the Wilma and the engineer when the tail end starts and stops (which differs from the head end due to the cumulative effects of coupler slack and cushioned draft gear). A FRED has a battery to power all of these functions, but it also has a small turbine generator driven by air pressure to recharge the battery.
Every locomotive and railcar in North America is listed in the UMLER (Universal Machine-Language Equipment Register), and an integral part of that registration is the maximum length over its couplers. When a conductor receives the printout of his train's manifest, the train's calculated length is there in black and white. He also has a computer list of all pick-ups and set-outs that he is supposed to make along the way--and what the new train length will be after each of these. The crew is responsible for keying this length into the Wilma and keeping it updated each time it changes. When the head end enters a siding (or passes any other point that it will be useful to know when the tail end has cleared), the engineer pushes a button on the Wilma. Using data from the locomotive's own odometer and the known maximum length of the train, the Wilma gives the engineer a digital readout in feet until his tail end has reached the same point. This is how a crew is supposed to know if its tail end is into a siding or not. Of course, if the track is straight and level, and it's daylight, they can often just look.
A dispatcher's computer also knows the calculated length (and even weight) of every train on the line and the length of every siding. That computer will prevent the dispatcher from ordering a too-long train into a too-short siding. Problems occur when the real-word train is longer (for whatever reason) than its length data in the railway's computer system or the Wilma. This happens if the yard crew that assembled the train mistakenly put cars into it that weren't supposed to be there, if the conductor picked up more cars than he was supposed to at some point, or if the crew erroneously entered the calculated train length into the Wilma.
Any of these three situations can result in the tail end of a train hanging out on the main when everybody thinks it's in a siding. In CTC (centralized traffic control) territory, there is a sensor circuit in the main line near a siding switch that will not clear the signals for the train doing the passing unless the way is clear. I would think that there is CTC at Lawtell, but I don't know for sure. If so, the approaching train would have gotten a surprise red signal when it was expecting to have a green one at the siding and would have applied brake in time. Even then, I can imagine that some Cajun French got spoken!
As for back-up moves by a long train, there would have to be a man on the rear of each train to flag any unprotected grade crossings and to communicate with the engineer by radio. In the case of a saw-by, the dispatcher would shut down the line to any other trains and wait for the two crews to get everything sorted out. That's not the sort of thing that could be managed from a computer screen in Spring, Texas.
I wrote that I still liked the idea of a camera and.. I was probably pushing him, but nevertheless, he remained civil.
He added:
Steve,
If there were cameras in FREDs, they'd be useless at night. Besides, the lenses would get dirty, fog up, get covered with ice, etc. The transmission of a camera image would require more bandwidth, battery power, and antenna size than is available to a FRED. As it is, the data transmitted is very compactly encoded into a periodic digital pulse that's encoded so as not to interfere with other nearby FREDs and Wilmas. All FREDs and Wilmas in North America use the same single pair of frequencies--for two-way communication between the two units (except on Norfolk Southern--because NS does everything differently--where they use a different single frequency for both ways.) Since FREDs and locomotives with Wilmas are interchangeable among railways, many of them can be switched between the NS and the everybody-else frequencies as needed.
Mark wrote, having kept up with OO-L's offerings:
Ask Lowell about Fred and EOT,one of them tells the crossing gates to lift after it passes.
Also ask about the detectors that report via radio axle temps,etc.
I am seeing a lot of trains with radio problems,as in no range.
Lawtell is about 4 miles as the crow flies and they were weak!.
Me: Again I was acting as a conduit. Why can't we all just be friends? Civil friends.
The following may be a bit of ruff talk. If not a ruff talk person, please detour. I say this in an attempt to promote civility.
OO-L proceeds.
Mark wrote, having kept up with OO-L's offerings:
Ask Lowell about Fred and EOT,one of them tells the crossing gates to lift after it passes.
Also ask about the detectors that report via radio axle temps,etc.
I am seeing a lot of trains with radio problems,as in no range.
Lawtell is about 4 miles as the crow flies and they were weak!.
Me: Again I was acting as a conduit. Why can't we all just be friends? Civil friends.
The following may be a bit of ruff talk. If not a ruff talk person, please detour. I say this in an attempt to promote civility.
OO-L proceeds.
Steve (and Mark),
A FRED and an EOT are exactly the same thing. EOT/EoTD/ETD stands for "end-of-train device." At first, when conductors lost their cabooses and had to start riding in locomotives with the engineers, they called the FREDs "f---ing rear-end devices" (and it didn't mean "flashing"), so the alternate name was pushed by some in the industry who wanted to avoid the crude appearance. However, I think that "FRED" has won out as the dominant term.
Crossing lights and gates existed for ages before the FRED was invented, and they don't interact with FREDs at all. They sense the presence of a train (or its absence, once passed) by sending a weak electrical signal down one rail and listening for it in the other. The steel wheels and axles of a train shunt the signal over to the other rail when a train is present. The signal is of a type that will travel only a limited distance through steel before dissipating.
Detectors are placed at intervals up and down rail lines. There are three kinds, often combined in any combination at a given location. One is the "hot-box" detector. These consist of a infrared sensor placed on each end of a crosstie to look up at the roller bearings where the truck frames set on the ends of the axles as they roll by. Any failed bearing will heat up from friction and eventually melt the end of the axle off and cause a train wreck; but enough detectors will find them first. The second kind of detector is a "dragging-equipment" detector. This consists of a hinged mechanical flap sticking up a few inches between the rails to detect anything that hits it while a train is passing over. The third kind of detector is the "high/wide" or "load-shift" detector, typically found in advance of bridges with superstructures or in areas where pulpwood or logs are commonly hauled. These have a large frame running upward on each side of the track and across the top. On the frame are mounted instruments that produce and detect vertical and horizontal light beams at the dimensional limits. When something breaks a beam, there is a problem. Some older installations might use tightly stretched wires instead of light beams, but those require repair after each defect detected.
At each detector is a track-side radio link that talks to the passing crews with a computerized voice. It identifies itself and tells them the total number of axles (or the axle number at which any problem might have been detected). It tells them the ambient temperature because the rules require reduced train speeds above certain high temps because of the danger of "sun kinks" and below certain low temps because of the danger of "pull-aparts"--both caused by excessive expansion or contraction of the welded rails. The axle count is supposed to act as a double-check of the train length entered into the Wilma--if everybody is paying enough attention.
I can't really comment much on weak radio range. There are towers at certain intervals up and down the line that pick up the train radio signals and relay them (typically via fiber-optic landlines) to the dispatchers and vice versa. The train radios only need to reach the nearest such tower. It might be that UP has increased the number of towers and lowered train radio outputs to cut down on co-channel interference. This might be an effort to run a huge railway on a limited number of frequencies, but that's just speculation on my part.
Me: I had sent some of 00-L's explanation to Jim in CO. He threw this back onto the table
Your explanation of "sawing by" brings back another railroad term "doubling the hill". I don't know if you have heard of it or not but my brother who worked as a brakeman on the Lufkin division of the SP said it happened a lot. He said on a run from Houston to Lufkin there were several steep grades and sometimes the train would stall on a grade if they didn't have enough engines for the tonnage of the train. When that happened they would cut the train in half and bring the first section the the nearest siding on a level grade. They would then go back and get the second section they had left on the main line, pull it up to the siding, get the first section, couple it to the second section and continue on. He said it was very time consuming and caused the crews to run out of hours sometimes. In those days crews could work sixteen hours but now it is twelve.
OO-L wrote, which I thought was extremely civil:
Your explanation of "sawing by" brings back another railroad term "doubling the hill". I don't know if you have heard of it or not but my brother who worked as a brakeman on the Lufkin division of the SP said it happened a lot. He said on a run from Houston to Lufkin there were several steep grades and sometimes the train would stall on a grade if they didn't have enough engines for the tonnage of the train. When that happened they would cut the train in half and bring the first section the the nearest siding on a level grade. They would then go back and get the second section they had left on the main line, pull it up to the siding, get the first section, couple it to the second section and continue on. He said it was very time consuming and caused the crews to run out of hours sometimes. In those days crews could work sixteen hours but now it is twelve.
OO-L wrote, which I thought was extremely civil:
Steve,
Thanks to you and James. I think that doubling the hill is still done much more often than is sawing-by. It usually happens when one of a train's locomotives has died somewhere out on the road.
Me: Pretty good stuff, I'd say. Thanks Mark, OO-L and Jim, aka, James in some circles.
Me: Pretty good stuff, I'd say. Thanks Mark, OO-L and Jim, aka, James in some circles.