 |
Re: I think that Clem is talking about traffic signal timing integration, not the rail signals...
Author: Dr Zarkoff
Date: 10-23-2022 - 12:57
> > > are the crossings interlocked with the signal
> > > system
> > No, crossing protection and a signal system are
> > separate.
> I think that Clem is referring to the integration of traffic signal timing on adjacent or approaching streets with the gate cycles,
I wanted to make it clear to our readers that x-ing protection and signal systems are two independent systems which, although they might use the same track circuits, really have nothing to do with and no effects on each other. Xing protection can be interfaced with roadway traffic signals when a street or highway crosses the tracks -- this is done on both transit and RRs and the effect is one way: RR crossing devices control the traffic signals, not the other way around. Timing is regulated by the FRA/CFR: the flashers and bells must start 20-30 seconds minimum (my recollection on this timing might be a little off) before the train reaches the crossing. The gates, if equipped, come down shortly after the flashers start (5-10 seconds as I recall). It is to be pointed out that the "starting of the crossing protection devices" are the bells flashers, not the gates.
> Many years ago a now long gone good friend and colleague went to Sweden to see a "doppler" based "train activated" gate timing activation system.
> This used the change in wave lengths of a radio signal emitted by the approaching train (doppler effect), recieved at the crossing gates, to calculate and determine if it was a Swedish iron ore train crawling toward the grade crossing at 40 kph., or a Swedish 200 kph. high speed train barreling toward the grade crossing, and adjust the lowering of the crossing gates, based on the approach speed of the train.
This is no different than the way predictors work: an unoccupied track circuit is a loop with a specific, unchanging inductance. The circuit's "signal" (DC or AC) is transmitted from the crossing end of the circuit. As a train approaches (from the far end of the circuit), its leading axle(s) shorten the length of the loop, which changes its inductance. Since the rate of change corresponds to train speed, logic circuits in the predictor are programmed to activate the crossing devices accordingly. Modern versions of this scheme use AFO-like frequencies in crossing track circuits, which does a way with the need for insulated joints.
> Many years ago a now long gone good friend and colleague went to Sweden to see a "doppler" based "train activated" gate timing activation system.
> This used the change in wave lengths of a radio signal emitted by the approaching train (doppler effect),
This is precisely the same as the way radar works, and in the 1950s, radar's inventor, Robert Watson-Watt, once got pulled over in a speed by a Canadian Mountie who had a radar gun.
>> Since it is based on the train telling the crossing gates how fast it is approaching, it is not based on a fixed (everybody has to wait an extra 10 minutes for the ore train to get there) attivation, because the fixed track based circuit must be set for the fastest approaching train.
Correct, except that the train doesn't tell the crossing protection apparatus how fast it's going, the apparatus analyzes the return signal to figure this out. The length of a "crossing track circuit", no matter what type, is determined by the highest train speed in the territory where the crossing is located.
> He might also be thinking about how Amtrak's ITCS was supposed to include advance grade crossing activation for the higher speed trains, not using track circuits,
No, I wasn't. The FRA will not allow GPS based train location for several reasons: 1) if there is more than one track, GPS isn't accurate enough to determine which track the train is on. The Military's probably is, but the DOD isn't about to allow its use by civilians, and 2) GPS train locating can't detect broken rails; only track circuits can do this. It was designed to be an overlay on an existing CTC system, see: [railroads.dot.gov] .
ERTMS is the European equivalent to PTC in the US. Since European Rys aren't subject to FRA/CFR oversight, it doesn't use track circuits for track occupancy detection. Amtrak uses an adaptation of this, ACSES II, on the NEC. It's an overlay to its 1920s/1930s era cab signal system, which uses track circuits to determine track occupancy -- rails are detected as a track occupancies.
> > > system
> > No, crossing protection and a signal system are
> > separate.
> I think that Clem is referring to the integration of traffic signal timing on adjacent or approaching streets with the gate cycles,
I wanted to make it clear to our readers that x-ing protection and signal systems are two independent systems which, although they might use the same track circuits, really have nothing to do with and no effects on each other. Xing protection can be interfaced with roadway traffic signals when a street or highway crosses the tracks -- this is done on both transit and RRs and the effect is one way: RR crossing devices control the traffic signals, not the other way around. Timing is regulated by the FRA/CFR: the flashers and bells must start 20-30 seconds minimum (my recollection on this timing might be a little off) before the train reaches the crossing. The gates, if equipped, come down shortly after the flashers start (5-10 seconds as I recall). It is to be pointed out that the "starting of the crossing protection devices" are the bells flashers, not the gates.
> Many years ago a now long gone good friend and colleague went to Sweden to see a "doppler" based "train activated" gate timing activation system.
> This used the change in wave lengths of a radio signal emitted by the approaching train (doppler effect), recieved at the crossing gates, to calculate and determine if it was a Swedish iron ore train crawling toward the grade crossing at 40 kph., or a Swedish 200 kph. high speed train barreling toward the grade crossing, and adjust the lowering of the crossing gates, based on the approach speed of the train.
This is no different than the way predictors work: an unoccupied track circuit is a loop with a specific, unchanging inductance. The circuit's "signal" (DC or AC) is transmitted from the crossing end of the circuit. As a train approaches (from the far end of the circuit), its leading axle(s) shorten the length of the loop, which changes its inductance. Since the rate of change corresponds to train speed, logic circuits in the predictor are programmed to activate the crossing devices accordingly. Modern versions of this scheme use AFO-like frequencies in crossing track circuits, which does a way with the need for insulated joints.
> Many years ago a now long gone good friend and colleague went to Sweden to see a "doppler" based "train activated" gate timing activation system.
> This used the change in wave lengths of a radio signal emitted by the approaching train (doppler effect),
This is precisely the same as the way radar works, and in the 1950s, radar's inventor, Robert Watson-Watt, once got pulled over in a speed by a Canadian Mountie who had a radar gun.
>> Since it is based on the train telling the crossing gates how fast it is approaching, it is not based on a fixed (everybody has to wait an extra 10 minutes for the ore train to get there) attivation, because the fixed track based circuit must be set for the fastest approaching train.
Correct, except that the train doesn't tell the crossing protection apparatus how fast it's going, the apparatus analyzes the return signal to figure this out. The length of a "crossing track circuit", no matter what type, is determined by the highest train speed in the territory where the crossing is located.
> He might also be thinking about how Amtrak's ITCS was supposed to include advance grade crossing activation for the higher speed trains, not using track circuits,
No, I wasn't. The FRA will not allow GPS based train location for several reasons: 1) if there is more than one track, GPS isn't accurate enough to determine which track the train is on. The Military's probably is, but the DOD isn't about to allow its use by civilians, and 2) GPS train locating can't detect broken rails; only track circuits can do this. It was designed to be an overlay on an existing CTC system, see: [railroads.dot.gov] .
ERTMS is the European equivalent to PTC in the US. Since European Rys aren't subject to FRA/CFR oversight, it doesn't use track circuits for track occupancy detection. Amtrak uses an adaptation of this, ACSES II, on the NEC. It's an overlay to its 1920s/1930s era cab signal system, which uses track circuits to determine track occupancy -- rails are detected as a track occupancies.
