03-31-2018, 01:17 PM
(03-31-2018, 12:15 PM)Canard Wrote: I can understand why they did it this way here.
It’s right at the end of the Willis Way platform. Other areas do not have the Radar sensor and are presumably triggered instantly by the inductive presence detector loops in the ground. That won’t work, here, or when an LRV is stopped at the station, the sign would be illuminated the entire time, which imposes an unnecessary restriction on traffic.
How could the system you propose know when a train is about to start moving and turn on the signal?
That’s why they did it this way.
At a glance it might seem they could just Boolean “AND” the inductive track loop and the radar sensor - train must be present AND moving to turn on the light.
Bit, you are also assuming everything is all integrated and so on as one massive system. For all we know, there’s 24V going up to that mast and that’s it, it could be a self-contained thing operating in isolation. There are a billion reasons why they did what they did, and I’m sure they’ll come up with a simple fix.
Automation is never as simple as “why didn’t they just...”
When the driver presses the button to start moving, the system knows the LRV is moving. No precognition required.
However, my point is that the system isn’t actually a single integrated system: it’s multiple pieces, some of which work completely independently, which I find somewhat surprising.
There isn’t good monitoring with the sign being effectively a totally separate system from the actual LRV and track systems. Sign on and no LRV is an error condition, and sign off but LRV coming is also an error condition. The second is obviously an immediate hazard condition, while the first is a hazard condition in the long run when one considers the effect on people’s respect for signs and signals. It’s not obvious that either of these situations will even be detected, even if a collision occurs. If the signage is reliable enough, it may not matter, which would indeed explain the design. It just seems surprising to me.
So I’ll be interested to see how this and other similar signs work in practice once the system is operating. So far all we’ve seen is testing operation that doesn’t match regular service patterns.
On a related note, have people noticed that during testing so far the LRV regularly violates its horizontal bar signals?