Towers and Passenger Trains

[NW Mailing List]

In those ICC reports, I always look at the "passing times" with a jaundiced eye, especially when they are used for calculating speed between two open stations. 

First off, there was the matter of variation in clocks. Remember this was the day before clocks with synchronous motors... the railroad used wind up, seven day (or eight day) clocks. Some ran fast, some ran slow. And although the operator in every Train order Office was supposed to listen to the Noon Time Signal from the Naval Observatory as transmitted on the Train Dispatcher's Telegraph wire, and correct the clock accordingly, being a practical railroader and knowing how rules and procedures get shuffled for the sake of worker convenience, I know that it did not always happen. "Okay, my clock is 25 seconds fast... I can live with that." The guy at the adjacent tower says, "Okay, my clock is 25 seconds slow... I can live with that." And suddenly you have an error of almost a minute for the transit times between stations just a few miles apart. That minute materially effects the speed calculation. 

Over the years I was responsible for 16 towers and I was in them all the time (and even worked a couple when we were out of men,) and I know the work-flow of these guys. Oftentimes they had a small nail or hook over their desk, on which they would hang their railroad watch. Much easier to glance up at the watch than turn the neck and look at the wall clock. (Especially if you wear bifocals...) This introduces another factor which may effect the giving of an accurate time by the station: By how many seconds from "true" time did the man's personal watch vary ? 

Another factor to consider: There are 60 seconds in a minute. A train passing at 11:52:01 is OS'ed as having passed at 11:52. A train passing at 11:52:50 is OS'ed as having passed at 11:52. How about a train passing at 11:52:58? Is it also OS'ed as having passed at 11:52, or as having passed at 11:53? Who knows what practice an individual operator would "find convenient." The rules did not address it. 

Here's yet another factor: Where must the train be before it is reported "by"? The old rule was "when the markers are 300 feet past the block station." (That rule was instituted in the pre-automatic block days, and the 300 feet gave a little safety margin in case the following train came charging up to the station and encountered a rear end sitting just a few feet beyond the Red Board. By the 1950s, this rule no longer existed, but what was the operator's perception of how far beyond the station the markers must be before he reported the train "by"? Was the train to be considered "by" when the markers went past the window, or when the train "went off the circuit" and the light went out on the model board? If the interlocking was a long one, "time by the window" and "time off the light" would be some seconds apart, and perhaps a minute or more apart for a train running through a long interlocking at less than full speed. 

Another factor: Operators were supposed to be "on the ground" to observe passing trains. How many seconds elapsed between the markers passing the operator, his returning to his desk (which might involve a flight of stairs,) and his glancing at the clock (or his watch hanging on that nail at his desk) and writing a time on the block sheet? 

Another factor: Suppose the operator is busy copying a Train Order, or getting the Train Dispatcher's blessing for a Clearance Card, or "blocking" with the open offices on either side of him, giving times and receiving times. Or maybe he's manipulating the levers on his interlocking machine. The train goes by. He finishes the task in which he is engaged and soon thereafter makes a (hopefully accurate) guess on what time the train "was by." 

One final observation. I have been in towers when thousands upon thousands of trains were reported by. And many, many times I have heard, "Oh, make him by at... er... oh, make him at 10:17." The point is that the passing of trains, and the reporting of same, was such a routine, work-a-day event that it was not treated with scientific precision. The train is by and any reasonably accurate time will suffice. A minute or so either way doesn't matter... until the ICC grabs the block sheets and does a mathematical calculation on how many second is took a train to travel X number of miles, and uses numbers from both ends which may be off by 30 seconds, a minute, or some other indeterminable error. The errors in time at one end are compounded by the errors in time at the other end. The point is this: Nobody knows exactly, precisely, how many minutes and seconds it took the train to pass through the block. The times are approximations. 

One final tidbit, which really doesn't bear upon the present N&W situation but may be of interest nonetheless. In the extremely busy, heavily trafficked, 4-track NYC territory out of New York City, the operators reported trains to the quarter-minute. For recording the information on the block sheet, they used a superscript of "2," "3" or "4." A train passing between the even minute and +15 seconds was recorded as "11:45." A train passing between the 16th second and the 30th second was recorded as 11:45 + superscript 2 (i.e., the second quarter of the minute.) A train passing between the 31st second and the 45th second was recorded as 11:45 + superscript 3 (i.e. the third quarter of the minute.) And a train passing between the 46th second and the next even minute was recorded as 11:52 + superscript 4 (i.e. the fourth quarter of the minute.) Unfortunately, I have forgotten how they indicated the superscript notations when reporting the train to the Train Disaptcher. 

An old timer gave this young fellow some sage advice many years ago: "Don't believe your own hustle." Things ain't written in stone... especially when it comes to the recorded passing times of trains. 

-- abram burnett 


//////////////////////////////////// 
Sent to you from my Telegraph Key... 
//////////////////////////////////// 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://six.pairlist.net/pipermail/nw-mailing-list/attachments/20130921/60e4dea8/attachment.html>