The Transcontinental PRR

[eric220]

While I ponder the bridges, and slowly lose the internal debate that will eventually lead me to admit that Vollmer is right ( ), an interesting question came up over on NScale.net.  I tried to place the signals here as I've seen them in PRR interlocking diagrams.  Specifically, on each side of the crossovers, facing outward only, thus:



I've assumed that for such interlockings, the section between the signal bridges that includes all the turnouts isn't really a separate signaled block, because it wouldn't be very long.  I assume that the separated signals around the interlocking above would function the same as these, they're just in different physical locations:



To put it another way, as a train moved through them, the signals would react something like this:








Does anyone know if that would be correct?

[Bob Bufkin]

Something on the fourth one down doesn't quite look correct.  Looks too restrictive to me.  I think the second signal from the right would be a slow not stop aspect.

[eric220]

Bob - It would be if these were all independently signaled blocks.  That interlocking is only 32 inches long.  It seems awfully short to make it an independently signaled block.  Also, as I said before, on all of the diagrams that I've seen, there are signal bridges on each end of the interlocking with signals only facing away from the turnouts.  This arrangement cannot be treated as a full block because if they were, there's no way for a train coming out of the interlocking to see the occupancy of the next block.

The way that I've got it set up here is that occupancy in the interlocking causes both adjacent blocks to show occupied.  The effect is the same as a train straddling two blocks.

[Dave V]

Eric,

Colorado Midland as you know used steel trestles in Eleven Mile Canyon and at Aspen.  If you're looking at "rapid construction" circa 1915, steel deck girder and truss on concrete piers would probably be correct.  This is still Pennsy, not the Rio Grande Southern.

[eric220]

Dave - As I mentioned earlier, the more I think about it, the less I'm able to justify not going with stone bridges.  It's going to look very european, but like you said, this is the Pennsy, not the RGS.  My PRR was building rapidly, yes, but they were still building for the long term, hence the double track mainline and tunnel to accommodate.  I just picked up an Atlas viaduct to play around with.  I should be able to make it work.

[Dave V]

 

U mad, bro?

[eric220]

U mad, bro?

My delicate emotions have been dashed.  Damn you for keeping me honest to the prototype!  (Or at least trying.)   

[Dave V]

"In the name of the Curve, the K4, and the GG1, Amen" <makes the sign of the Keystone> 

[eric220]

The Gandy Dancers came over tonight, and we got a lot done.  I've now got more track feeders than I know what to do with, and some other mind-numbing tasks were completed.  Also, the upper level benchwork has been expanded around to Loveland Pass.



As you can see, I busted out the Atlas viaduct to take a look at it in context.



The final version will have longer piers on at least two of the legs.  That is all for now.

<makes the sign of the Keystone> 

[eric220]

I've been out of town recently, hence the slowing of the updates.  I'm back now and getting a little work in.  I've been musing over the signaling diagram that I presented earlier.  I've got several sections of track similar to the interlocking that I showed before.  Some are crossovers, others are tunnels.  In each case, I've got a block that has logical boundaries (tunnel portals, limits of interlocking) but the distance between those boundaries is quite short.  In some cases, the distance between these features is also very short (such as a crossover near a tunnel portal).  I don't want to have short blocks, as they would hamper signaling.  I've come up with an idea to use overlapping blocks, as I show above.

To take another example, let's consider a short tunnel.  I want signals on each side of the tunnel that show "stop" when the track inside the tunnel is occupied.  This means that the tunnel needs to be an electrically isolated block to detect a train inside.  Consider the following diagram, with the tunnel in block "B".



Conventional wisdom would say that each electrically isolated section should be signaled as an independent block (shown in red), thus:



There are a couple of problems with this arrangement.  If the tunnel in block "B" is short, and the limits of block "B" are defined by the location of the tunnel portals, block "B" will also be very short.  Short blocks tend to reduce the effectiveness of multi-block signaling.  Also, in this case the signals facing block "B" would be very close to the tunnel portals, making them very hard for operators to see.

My thought is to use block "B" as a detector section that is part of both block "A" and "C".  Effectively, the depicted section of track would be two overlapping blocks instead of three.  Something like this:



There are several advantages to this arrangement.  First, signals on either side of the tunnel show "stop" when a train is in the tunnel, regardless of which direction the train is going.  Second, both block "A" and block "C" are longer, because both include the length of block "B".  This added length becomes very valuable in a couple of situations where the tunnel in "B" is a bit longer, but block "C" is fairly short because of a crossover near the tunnel (in the next block after "C").

I'm pretty sure that this arrangement is not at all prototypical, but I think that it will be useful.  I'm interested in reactions. 

[cv_acr]

Strangely enough, Canadian Pacific actually has some places a bit like your tunnel example where block signals are staggered, forming a slight overlapped block like above. (There's an example on the line my club is modelling where the signals are set apart with a slight overlap and in between there is a rail overpass, overhead pipeline and highway overpass.) The overlap should be pretty short. However, you wouldn't just have signals at each end of the tunnel simply because there's a tunel; but if a block boundary naturally should be located at or near the tunnel, and the tunnel is short, the RR might move one or both direction's signals slightly for visibility. Staggered signals could also be a trick to lengthen approaches between interlockings that might be a little closer together than 2 standard length blocks, but too far for just one.

As far as interlockings go, the section(s) of track inside the interlocking is definately a separately detected section. Your fourth image in the progression isn't quite right, as the last signal before the interlocking should show approach, not stop as another train can definately safely roll up to the interlocking. (The interlocking signals could be "Stop" with no train within 100 miles of it. In fact, unless something is specifically lined through the interlocking, all the interlocking signals MUST be held at their most restrictive indication ("Stop"). Technically, signalling is about the status of the current block and the next SIGNAL, not the occupancy of the next block.)

Approaching the interlocking, you treat that approach signal and block the same as any other block; check the block occupancy and the signal indication at the interlocking to determine the indication for the approach signal.

At the interlocking, IF the dispatcher or signalman has tried to clear a signal, you evaluate the route through the interlocking, the occupancy of the block(s) within the interlocking, the next track block (if there are no exit signals) and then the indication on the next signal (interlocking or intermediate) to determine the indication for the interlocking signal. (Let's say for example you have a 4-track main, and you cross all the way over from one outside track to the other, you'd actually have to look at 5 separate blocks (1 for each track in the interlocking and the next open track block based on the route) to determine occupancy between this signal and the next.)
Each [straight] track through the interlock needs to be it's own block; as long as routes don't conflict, it should be possible for trains on different tracks to be passing through the interlocking at the same time. The point of the interlocking setup is to physically not allow any conflicting movements to be cleared.

[eric220]

Instead of working on the physical layout like I should be, I've been playing around in FileMaker.  I've already got my inventory in a FileMaker database, and I'm planning to build a series of databases that will support operations.  My first crack has been at the Waybills application.



This is a screenshot from my iPhone.  The database is being shared from my desktop and dynamically displayed on the iPhone.  Most of the fields are self-explanitory.  I've coded six "regions" on the layout that appear next to the origin and destination.  There are two on the lower level, two on the upper level, and one for each staging yard.  Each has a two-letter abbreviation that corresponds to a color-coded background that appears behind the letters on the waybill for easy routing.

My eventual plan is to acquire a few iPod Touches and distribute them to crews for ops.  Additionally, FileMaker can publish database layouts as web pages, so anyone with a wi-fi enabled smart phone will be able to pull up the databases in a web browser.  Eventually, crews will be able to view waybills, train consists, and switch lists.  I'm also planning an iPad formatted version for yard crews.

Yes, I am a nerd.

[seusscaboose]

Holy crap... That is sweet... Not sure how it'll all work out yet... But you have peaked my interest... Ed... Are you watching this?

[eric220]

Sorry for the slow updates, but I just started a new job, and work on the layout has slowed to a crawl.  Prior to starting the new job, I got the Gandy Dancers over for a couple of work sessions.  We extended the benchwork on the upper level all the way around to the helix, and the cookie-cutter followed, bringing the ROW to the outskirts of Idaho Springs.

[Philip H]

new job, huh?  Chasing baddies wasn't enough excitement for ya?

Seriously, good luck.  And keep us posted on that Filemaker thing - I'm a slow but coming along nicely Apple convert.