I was watching TV show 9-1-1 when they responded to a nasty train wreck. It showed the tracks at a switch moving back & forth and the engineer put it in emergency with many cars on the ground. Had never seen anything like this before and asked some railfan friends on FB what happened. Here is one reply: "That’s called a floating switch when it does not lock in when thrown automatically. "I have never started a forum before and am attaching pics I took from TV, the inner 2 tracks were the ones moving back & forth. Wondered what some of you will have to say about floating switch, does it happen very often ?
The “inner two tracks” are the points; that’s what would float.
I would have to assume that there was a signal connected to the switch if it were in mainline territory. Said signal should not have cleared in either direction, so any derailments would have been the result of running a signal.
If this switch were an automatically-thrown switch in the yard (I threw these by the hundreds every day I worked), again, the signal displayed should be dark, since you shouldn’t get a signal until after the switch has locked itself in position. Also, in yards, where movements are to be made at “Restricted Speed”, one of the things you must be able to stop short of (within half the range of vision, the rule now says) is a switch not properly lined. So again, it’s on the engine crew if something untoward happens.
That is true. There was one time in my career that the switch did not throw when I wanted it to, and the lights on my board were indicating that it had thrown. No disasters, but a very flummoxed operator and a brand-new throw rod when the problem was found.
My bad. [#oops] Thanks to Comcast’s on demand, I was able to see the episode and it did play on May 11. The switch points are seen by the engineer in his headlights as he is approaching them and they are in a constant back and forth motion which is not to me, anything I think would ever happen. Most switches I have ever worked on had considerable friction even when well graphited and in good condition. No way could they oscillate as shown in the story. But that is the device to show the train derailing.
The screenwriter probably had a Lionel set when growing up. The electric switches had points that sometimes would move back and forth rapidly when there was a wiring problem.
To me, an automatic switch (AKA variable point switch) is a hand throw switch that can be run through. The points stay lined for the last route used by a trailing move. Remote controlled or dual controlled power switch would be a better term.
Once we had a power switch that wouldn’t lock up, the points kept moving. It was said there was a bug in the control system when too many changes had been ‘stacked’ into the system. I don’t know if that was really the problem. I don’t believe the signal cleared up because the switch didn’t lock up.
Tentatively this sounds like a control problem, the kind of thing that can happen if you have too many people tinkering with circuitry or code without systems testing.
If you have a power switch that is intended to ‘default’ to being fully locked in one position, and the limit switch or sensor that indicates full position in the opposite hand is defective in some way (or ‘mandatory’ engagement of locking does not complete) a ‘transaction-based’ programming model might try to cycle the switch back to starting position and then ‘repeat the exercise hoping for different results’. The same action might be true for CTC or PTC switches that must confirm ‘route integrity’ remotely. Now at least some of the time this might actually work, so it makes sense to repeatedly cycle some number of times (well clear of any potential switch occupancy, of course) but some idiot enamored of modular code might write the ‘excursion’ actions into one module that could become divorced from other parts of the signal system, or have higher execution priority or whatever. The described action is just what I’d expect from a switch trying to lock in ‘commanded’ position, failing to confirm, and return-cycling.
Another possibility might be water or mechanical damage to some of the wiring or components, leading to low-level cycling of the hardware.
Why the whole of this action would not always and everywhere be accompanied by the most restrictive aspects in the rule book, I can’t say. But weird things can happen with computers when too many cooks have been tampering with recipes.
Of course the actual actuation time will be bone slow for anything but a pneumatic switch, and the lockout timer on many of these things is specifically hardwired to preclude just the sort of cycling that c
Other than pneumatic switch operators, does anyone know of fast operating switch machines. Most I am aware of take multiple seconds to operate. The video had the points moving cack and forth in a blur.
My first reaction to what was discussed was - RANDOM DRUG TEST immediately.
Does not seem to fit the real world.
Most switch machines that I’ve been into or around (GRS/National/etc) are like giant Swiss mechanical watches. Lots of internal motion to generate forces to move the points at agonizingly slow speed. With most switch points in main tracks in CWR territory being 39’-0" rascals held in place with hook plates instead of massive heel blocks, it takes considerable force to “bend” those rascals with the gearing in the switch machine. (no longer a hinged joint at the heel block)
I recall in the movie “Silver Streak” a couple of switches very quickly moving - but the pacing of the movie at those points wouldn’t really allow a leisurely swing. Other movies have shown the same thing. Minor detail to the average viewer.
Pneumatic switches–ah, yes! As I drank in the sight of the J that would take N&W #42 out of Bristol, I heard the switches being thrown as the headend cars and Pullmans were added to the train.
This story sounds iike another case of a TV show where the real story has been “dramatized” to make a mundane real-life event something that can fill a 30 or 60 minute TV time slot (22 or 45 minutes with advertisements).
variable point switch, automatic switch …is this different than a …spring switch = Spring switches are used by some prototype railroads to save the crew from having to re-align the switch points after passing through the switch. The switch is sprung in one direction (typically the mainline), allowing any train to pass through the switch from the frog end without manually aligning the points.
A variable point switch (aka automatic switch) stays lined for the last route used by a trailing point switch. The trailing point movement can run through a switch lined for the other route and the points stay lined. They don’t spring back.
Many yard track flop over handle type switches were of this design, even if they weren’t “officially” acknowledged as such.
