Talk:Castlecary rail accidents

1937 Accident Causes
I rewrote this section a couple of weeks ago to correct several errors, but I am not happy with it. What I will do in this talk section is try to explain the significance of the distant signal, the possible wrong side failure, how a wrong side failure might have occurred, and I'll try to convey my impression of what the railway inspector investigating the accident actually thought. Until I can find some references, some of this relies on general background knowledge and to some extent a reading between the lines of the official accident report, so I'll need to tidy it up before putting it into the main page. Anyone else wishing to edit the main page is welcome to do so.

In the accident report the signalman was blamed because he committed a clear breach of Signalling Regulation 4, allowing a train into the section without making sure that the line was clear to the Clearing Point, a point 440 yards beyond the home signal. Although this is a major breach of one of the most important regulations, in itself it ought not to have resulted in an accident (the Edinburgh train should have stopped at the home signal), and if the Dundee train had stopped another 115 yards further on, the line would have been clear to the Clearing Point and the signalman would have been justified (so far as Regulation 4 was concerned) in accepting the Edinburgh train, but it is unlikely that the accident would have been very much lessened with only another 115 yards of braking distance.

Railway trains take a long time to stop, and the need for advance warning of signals at danger was recognised in the mid-nineteenth century, as was the need to safeguard against a train not stopping at a signal at danger. Before the end of the nineteenth century, two principles had been adopted across all of Britain's main line railways: 440 yards isn't usually enough distance for a train to stop from line speed, but it was generally enough of a safety margin for things like the driver misjudging their brake application, or for reduced adhesion between wheel and rail.
 * 1) At each signalbox, there was a distant signal that would display "caution" if any of the stop signals were at danger. The distant signal was sited far enough away so that a train approaching at line speed and seeing it at caution could stop before reaching the first stop signal (the home signal).
 * 2) Before allowing a train to approach a station, the line needed to be clear at least 440 yards beyond the home signal, in case, for any reason, the train did not stop.

A corollary of this is that a distant signal showing "clear" meant that all the stop signals at that station were also "clear", so drivers could continue driving at line speed without needing to worry whether the following signals were at danger.

Both drivers claimed to see the distant signal showing "clear". The Dundee driver said he could see the signal from 150 yards away, but in his first statements he just said that it was "drooping". The Edinburgh driver didn't see the signal till it was between 100 yards and 50 yards away, but he said that he had a good look at it and was certain it was showing "clear". Both drivers drove their trains in a manner consistent with seeing the distant signal at "clear". Both drivers said they only saw the position of the arm, pointing downwards from horizontal, and not the colour of the light, but this is of no significance.

What could cause the distant signal to show "clear"? The lever in the signalbox was in the "normal" position, which ought to have put the signal to "caution". The signal moves to the "clear" position by using the lever to pull on a chain attached to a wire, and it returns to "caution" by a weight when the tension on the wire is released.

The accident inspector considered whether the wire could have frozen to something or caught on something, or whether a pulley the chain went round could have frozen, keeping tension in the wire when the lever was returned to "normal". He found nothing for the wire to catch on, and thought freezing was only a remote possibility, particularly as the signal only stood for a few minutes at a time at "clear" whereas it stood for long periods at "caution" (this was another important feature of mechanical railway signalling in the twentieth century - see Abbots Ripton rail accident for the consequences of signals usually showing "clear"). In fact, one of the pulleys for this signal had frozen in the "caution" position at a few nights earlier, but the temperature then was several degrees colder than at the time of the accident. The inspector also considered the fact that the signal was "worked" three times in fairly quick succession not long before the accident (at 3:40 pm, 3:59 pm and 4:09 pm; the Dundee train arrived at about 4:30 pm), although he seems not to have thought about the signal not returning to "caution" between these trains. The last time anyone other than the signalman claimed to see the signal at "caution" was at 3:35 pm.

Apart from the wire freezing, there is the possibility of snow landing on the signal arm and weighting it down, but the inspector calculated that this would require considerably more snow than had actually fallen. There is perhaps a theoretical possibility of snow on the wire pulling it down between its supports, but this was not considered, probably because such a thing was clearly impossible. He also considered whether the wire had contracted in the cold or was set too tight, but quickly dismissed both of these.

The inspector mentioned in passing that the signalman could have left his signals showing clear after the previous train at 4:09 pm, and that he only realised his error when he heard the Dundee train approach, when he quickly put his signals to danger and showed the driver a red lamp. This is an odd thing to mention in the report, and it seems highly unlikely that the inspector gave it any serious consideration. In any case, whatever the position of the signal levers when the Dundee train approached, they were definitely all "normal" when the Edinburgh train apprached.

The signalman claimed to have seen the distant signal arm move to "caution" when he returned the lever to "normal" after the previous train, at 4:09 pm, but the inspector threw doubt on all his statements about looking at the signal, and believed that falling snow would have made his seeing the signal impossible. The signalman also claimed to have seen the backlight at 4:22 pm when he accepted the Dundee train and at 4:32 pm when accepting the Edinburgh train, but even if he did (which the inspector doubted), the inspector proved that the backlight would show even when the signal arm was drooping to a significant degree (the backlight was meant to indicate that the signal was showing "caution").

After the accident, at about 5:40 pm, a rescue engine arrived, with several people on the footplate, all of whom said that the distant signal was "definitely in the horizontal position". At 6:20 pm, a railway official walked up to the signal to check its aspect. It was "fully in the 'on' position". Both of these descriptions mean the signal was showing "caution".

The inspector appears to have been of the opinion that neither driver, and the Edinburgh driver in particular, got as good a look at the signal as they claimed, and that the signal was in fact showing "caution". He criticised the Edinburgh driver's speed, given the weather conditions. The driver was charged with culpable homicide, but charges were later withdrawn. Zipperdeedoodah (talk) 22:28, 26 July 2023 (UTC)

Information on the 1937 Accident
I was recently able to gain a copy of the Report on the 1937 Castlecary accident, and although I have no other sources, I'd like to know if it would be okay to perform a mass Edit to include a lot of information that this Article lacks on said Accident. --5.66.241.71 (talk) 22:14, 20 August 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Castlecary rail accidents. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20110717051449/http://www.cts.cv.ic.ac.uk/documents/publications/iccts00410.pdf to http://www.cts.cv.ic.ac.uk/documents/publications/iccts00410.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 00:03, 1 August 2017 (UTC)

Train crew deaths?
I don't understand the reference to "7 train crew" among the dead. The official report states that all 35 dead were passengers (8 in first class and 27 in third class). Two of these were railway employees but were presumably off-duty at the time.

The report details injuries or shock to both train drivers and the fireman and guard of the Edinburgh train, but mentions no deaths among either crew. Paulwilliam2 (talk) 09:36, 10 December 2017 (UTC) ==