 |
Re: Tourist Railroad Safety
Author: Shortline Sammie
Date: 06-02-2012 - 20:01
Gary Hunter Wrote:
-------------------------------------------------------
> Since every post on this thread seems to focus on
> legal issues and requirements, let me suggest the
> possibility that the train crew did not fully
> understand the technical operation of "automated"
> crossing protection. I don't know the particular
> railroads or specific crossing, but since lights
> and gates were mentioned, it is fairly certain
> that it was not a private crossing. In most of
> these systems there are two approach circuits and
> a center island circuit that is a little wider
> than the auto roadway. Starting from a completely
> reset system, the approach islands will trigger
> within a few seconds after detection (let's not
> complicate this with predictor electronis
> yet--GCP). A timer also starts that looks for
> island occupation. If the island isn't detected
> within a certain time, the gates go up. If that
> happens, and it does if there is a stop right
> before the crossing, the train must stop short of
> the crossing and tiptoe into the island to get a
> gate reset and drop the arms again. The "island"
> is very simple in electronic logic: If the island
> is occupied, the gates come down. The timer for
> the distant approach is just to make sure a dead
> track battery, stalled train or other malfunction
> won't tie up a crossing indefinitely. Some crew
> who don't get briefed on this detail may go
> through the crossing assuming the "automatic"
> aspects of the protection will function
> immediately. Once a timer has timed out, all
> crossings become "stop and proceed" (slowly). Even
> seasoned hoggers from large RRs working on
> branches and tourist operations can do this from
> habit, due to almost all mainline and state
> highway crossings being GCP these days. Crossings
> with grade crossing predictors change this
> behavior by actually sensing approach speed and
> varying the time before gate drop. Most GCPs can
> allow a stop short of the crossing, and resume
> functionality when train motion resumes. Rules for
> the big RRs state that trains must assure that
> gates are down before entering the crossing. This
> may or may not explain the scenario, but it seems
> that tourist RRs may need to make sure that trains
> do not just bolt forward from a gates-up
> situation, assuming the gates will drop before the
> crossing is fouled. Just a thought.
Good description of crossing activation. I might elaborate a bit more from our perspective.
The FRA requires 20 seconds warning time on signalized crossings from the time the crossing is activated until the locomotive fouls (enters) the crossing.
On crossings equipped with gates, FRA requires the gates be in the horizontal position for a MINIMUM of five seconds before the train enters the crossing.
Some crossings that do not have "predictors" consist solely of an island circuit which requires the train to come to a full stop before proceeding once it crosses the insulated joints awaiting the equipment to sense a train. Rusty rail is a gremlin that can cause a partial activation where the crossing activates then clears and reactivates and modern AC-DC systems have helped overcome the problem. The classic "fix" was to weld strings of stainless weld bead on the rails to improve the shunting ability.
Any type of crossing protection malfunction (gate broken, mech defective, system inoperative) requires the train to stop and manually flag the crossing.
In Oregon, whistles are not required at private crossings and for backup moves across a public crossing the locomotive horn is not an acceptable safety device UNLESS the decibel level at the crossing meets the FRA minimum as if the locomotive was leading.
As has been said, any operation (Class I, short line, museum or industrial) that does not over protect a crossing is asking for trouble.
Dick Samuels
www.oregonpacificrr.com
-------------------------------------------------------
> Since every post on this thread seems to focus on
> legal issues and requirements, let me suggest the
> possibility that the train crew did not fully
> understand the technical operation of "automated"
> crossing protection. I don't know the particular
> railroads or specific crossing, but since lights
> and gates were mentioned, it is fairly certain
> that it was not a private crossing. In most of
> these systems there are two approach circuits and
> a center island circuit that is a little wider
> than the auto roadway. Starting from a completely
> reset system, the approach islands will trigger
> within a few seconds after detection (let's not
> complicate this with predictor electronis
> yet--GCP). A timer also starts that looks for
> island occupation. If the island isn't detected
> within a certain time, the gates go up. If that
> happens, and it does if there is a stop right
> before the crossing, the train must stop short of
> the crossing and tiptoe into the island to get a
> gate reset and drop the arms again. The "island"
> is very simple in electronic logic: If the island
> is occupied, the gates come down. The timer for
> the distant approach is just to make sure a dead
> track battery, stalled train or other malfunction
> won't tie up a crossing indefinitely. Some crew
> who don't get briefed on this detail may go
> through the crossing assuming the "automatic"
> aspects of the protection will function
> immediately. Once a timer has timed out, all
> crossings become "stop and proceed" (slowly). Even
> seasoned hoggers from large RRs working on
> branches and tourist operations can do this from
> habit, due to almost all mainline and state
> highway crossings being GCP these days. Crossings
> with grade crossing predictors change this
> behavior by actually sensing approach speed and
> varying the time before gate drop. Most GCPs can
> allow a stop short of the crossing, and resume
> functionality when train motion resumes. Rules for
> the big RRs state that trains must assure that
> gates are down before entering the crossing. This
> may or may not explain the scenario, but it seems
> that tourist RRs may need to make sure that trains
> do not just bolt forward from a gates-up
> situation, assuming the gates will drop before the
> crossing is fouled. Just a thought.
Good description of crossing activation. I might elaborate a bit more from our perspective.
The FRA requires 20 seconds warning time on signalized crossings from the time the crossing is activated until the locomotive fouls (enters) the crossing.
On crossings equipped with gates, FRA requires the gates be in the horizontal position for a MINIMUM of five seconds before the train enters the crossing.
Some crossings that do not have "predictors" consist solely of an island circuit which requires the train to come to a full stop before proceeding once it crosses the insulated joints awaiting the equipment to sense a train. Rusty rail is a gremlin that can cause a partial activation where the crossing activates then clears and reactivates and modern AC-DC systems have helped overcome the problem. The classic "fix" was to weld strings of stainless weld bead on the rails to improve the shunting ability.
Any type of crossing protection malfunction (gate broken, mech defective, system inoperative) requires the train to stop and manually flag the crossing.
In Oregon, whistles are not required at private crossings and for backup moves across a public crossing the locomotive horn is not an acceptable safety device UNLESS the decibel level at the crossing meets the FRA minimum as if the locomotive was leading.
As has been said, any operation (Class I, short line, museum or industrial) that does not over protect a crossing is asking for trouble.
Dick Samuels
www.oregonpacificrr.com
