Block Detection

Railroad signaling, regardless if scale or size, works on one principal:  block detection.  Simply put, the signal system must be aware when a certain block is occupied.  Once that is known, the system can be designed to display signal aspects that protect movements between blocks. 

On the prototype, since trains either generate their own power or get it from a catenary wire (or third rail), the detection works by changing the status of power running to a relay, which is held open by the power.  Trains, with solid steel wheels and axles, then complete a circuit between the rails, dropping the relays (or solid-state sensors) that in turn trigger a track occupancy light (TOL) on the dispatcher/operator board (or screen). 

In model railroading, with DC power over two rails, the power in the rails is what drives the locomotives.  Therefore, another method of detection is necessary.  This topic is well-covered on other sites on the internet, so I will only discuss the one I chose to employ:  current-sensing block detection, using the BD-20 detection unit manufactured by NCE.

Essentially, the BD-20 works by detecting current flowing to a track circuit, generated by the presence of a piece of equipment that draws current.  This works immediately for locomotives, but also lit cabooses or passenger cars.   When the BD-20 senses the current drop, it activates, and sends a signal to the computer that the block is occupied.  When powered, the LED on the detector is illuminated, as seen below.  Here, Track 2 is occupied, where Track 1 is not.

The two middle terminals will be wired to the NCE input unit, which has yet to be installed.  One step at a time!

One issue is that this current-sensing detection does not work with standard freight cars, since they do not draw current.  How do we solve this?  The most expedient method is to use special wheelsets, equipped with a small resistor to allow a tiny amount of current to be drawn between rails.  These can be bought - at approximately $2.25 per axle, which adds up quickly since I have close to 400 cars - or made.  I choose the latter. 

Shown is an Atlas HO scale 33" metal 'caboose' wheelset.  Those pugs at Atlas make 'freight car' wheelsets that do not fit anyone's trucks but theirs, and 'caboose' wheelsets that just happen to fit everyone else's truck sideframes.  Yea, I figured that out the hard way.  That said, they are beautiful wheelsets for operations on a big layout - heavy and properly gauged to NMRA standards.  What I did was order 200 4.7K ohm resistors, costing 2.4 cents apiece.  Again, I learned the hard way that the 4.7K resistors were too big for my use - the detection wasn't picking up any current.  So, I changed to 3.6K ohm resistors, and the results have been much more consistent.  I glued them to the axles with CA adhesive, let that cure, and then connected both sides of the resistor to that respective wheel with silver conductive ink.  Once that cured, I tried it out, and wa-la!  Resistive wheelsets and working block detection.

So, now I am plodding along building wheelsets and installing block detection, much of the latter requiring me to add additional power bus wire and feeders for the second main track.  It's a large undertaking, and it's a bit discomforting to cut into a power bus that is working well for basic operation!  But, with eyes on the prize of a working system, it is what needs to happen.  7 blocks down, 23 to go...

Step by step!

~RGDave

The Edge of a Dream

Ever since I was young, I have been fascinated by railroad signal systems. Of course, the trains came first, and those remain the star attraction, but the idea that an operator or dispatcher could give trains permission to move over track by controlling signals and switches from a remote location always captured my attention. So, you can imagine the first time I witnessed a working signal system on a model railroad – WOW!

A sign of things to come...medium clear!

Therefore, one of my life-long modeling goals has been to have a working, dispatcher-controlled signal system on my model railroad. For me, that means scale, functional, NYC-style searchlight signals.  Exactly like the one above, manufactured by Integrated Signal Systems.  But there were always questions: first, how does that work? Second, once that works, it needs a full-time dispatcher. How then can I have the signals work as an ‘automatic’ signal system for the times when I am running the trains myself?

These two short questions have taken more than a year to answer. And we’re still working on it. It has been a real learning process to this point trying to figure out how all these goals can be accomplished; many thanks go to the guys that have spent so much time on this so far - Nick, and Alex. Thanks also go to Jack, Mark, and Scott, all of whom have made valuable suggestions in the design.  Model railroad signaling is complicated and expensive, especially on a big layout like the Onondaga Cutoff. The signal systems themselves, especially if they are to be dispatcher-controlled, boil down to systems that must work hand-in-hand with the control system.

Digitrax users have the benefit of the fact that Digitrax has a fully functional signal system to choose from that simply couples in to the Loconet cab bus. For us NCE users, we pay for the easier interface that NCE provides by having to use other’s signal components, since NCE does not yet have a working signal control board. In fact, while I am depending on NCE for block detection and layout control, I am depending on Digitrax for future turnout control and for signal control.

The link between the two systems is provided by the Java Model Railroad Interface, or JMRI. JMRI is an open software package available online for no charge, supported by a group of modelers that use it daily. This software allows for computer control of the DCC system and model railroad. It allows a computer interface to program decoders, speed tables, consists, effects, and also provides a pathway to link NCE to Digitrax via a USB cable so that my NCE layout can use the full Digitrax signal system. Plenty to come on this amazing project!

The Island

Work on the new wye, the Island Track, is coming along.  All that remains is to wire in a new PSX-AR unit into the power bus, which will allow the tail tracks of the new wye to automatically reverse polarity depending on which leg of the wye is aligned for the mainline.  The PSX-AR is a DCC breaker/auto reverse unit that, according to the documents and reviews, will allow this to appear seamless to the operators.

The new look of the eastern part of CP294, on the bottom level:

The new turnout is in the center of the frame, with the new 294 lead working east, and with the corresponding 274 lead working west behind the risers.  They come together at the Island Switch.

More to come soon, as the operation integrates this new infrastructure!

~RGDave

Competing For Time

Work on the Onondaga Cutoff competes for available time, and that is a precious commodity indeed these days. With several home projects also on the table, the layout is not always where I get to focus my effort. Further, even when I’m able to spend time in the basement, there’s choices to make: backdrop, car construction/upgrade, locomotive construction/upgrade, signal wiring, or something else? Lots of chores and not much time.

That said, a lesson I learned early in my model railroading is that it’s important to try and get your hands on the model railroad each day that you’re home. Even if it’s just a few minutes between errands or chores, you might have time to adjust a car, or sand one spackle patch of backdrop, or weather one side of a locomotive. Each item brings you closer to your goal, and each item done with care is something you can put behind you as you move forward.

An example is a recent addition to the roster - C40-8 6039, a locomotive I bought about 5 years ago, and one that I just detailed, weathered, and equipped with a DCC decoder recently.  I finished this project in 15- to 20-minute blocks of time over several weeks.  Now it's done!  Here's the big GE awaiting service at Onondaga Diesel:

Conrail 6039 at Onondaga Diesel

The hostler was aboard, and moved the unit slowly up to the shop entrance.  Onondaga Diesel was a busy place this day!  SD60M 5544 was in 3 Bay ahead of the 6039, and a variety of other locomotives surrounded them.


I am working with several friends to develop an implementation plan for the new signal system, which will allow the railroad to stay functional during that time. It takes a lot of extra thought now, but this way I hope to make less work later. Coordinating the design of the system and the signals themselves, too, is time-consuming, but it is something that is vastly improved by access to the internet and email. This sort of system would simply not be as feasible without those tools.

Thanks to several recent operating sessions, a new need has become clear - the ability to move locomotive consists from one end of staging to the other, without going through staging.  The solution is an 'Island Track' reversing wye, located between the two entrances to staging.  Here's an image of CP294 with several noteable changes.

First, in the foreground, you can see the new crossover that I installed to allow trains access to Track 1 coming down from 282.  That new crossover allows universal access to staging, which became necessary due to trains fouling 282 during sessions. 

Second, you can see the new island track going in in the distance.  At the far end of the crossovers, you can see new roadbed and benchwork that stretches back into the distance; this will be the west leg of the wye, with the east leg visible behind the risers.  All of this to make operations more efficient - and fun.

Speaking of which, I am working to do an operating session shortly, with more to come before springtime really sets in and brings outdoor work back to the daily schedule. And, well- there’s always plenty to keep me busy on the OC. I don’t remember the last time I was bored!

~RGDave

Scenery for the Short Line

With signal design underway and new wiring to follow on the OC, the only area where I can begin to install fascia board and scenery is on the M&E.  Fascia is being fabricated from 1/8" masonite hardboard, the same material as the backdrop, and will be installed before scenery so as to help define the land contour in front of the tracks.  A combination of foamboard and plaster castings, as well as some traditional paper mache, will be utilized as base scenery.

Here we see the Skaneateles Creek area on the M&E with newly-installed facia:

I love the look of model railroads with scenery and fascia - to me, it is a very 'finished' look and lends a professional feel to the look of the entire layout space.  I am eager to get the rest of the railroad to this state, too, but signal wiring must come first.

Foamboard will be stacked, glued, and shaped to represent cliffs, and then I have several ideas for how to model the distinctive Finger Lakes-area shale that is so prevalent in Central New York.  It exists in several shades of gray and dark purple, so it will be an adventure to replicate that look in HO scale!

~RGDave

Next Steps

Today, I have a a few action shots for you all, and a general discussion of the current progress on the Onondaga Cutoff.

First, though, the images.  Be sure to click on each for a full-size view: 

WAON-14 spots boxcars of scrap paper to be used for paperboard at the big Iroquois Paper facility east of Camillus, NY.

SEBU-7 sweeps around the curve and slams through the crossovers at CP280 with the Onondaga Engine Facility in the background.  Leading three other units is Conrail C39-8 #6000, recently upgraded with an LED headlight.

Now that the trackwork is in place and there are less and less issues with track leveling, alignment, and design, the Onondaga Cutoff is moving towards the next real challenge:  an operating signal system.

Ever since I was very young, I have been interested in signals on railroads, and in how they function.  It follows that to me, the crown jewel of a quality model railroad is a prototypical, working signal system.  From an operation as well as a photographical standpoint, nothing compares to watching a train on the layout 'knock down' the correct signal aspect for the move at hand!

Thanks to the efforts of several operators, namely Nick and Alex, I am formulating a plan and budget for the purchase of components that will be used to replicate block detection of trains which will work with computer-based CTC, including the operation of a working signal system.  While I am fully commited to Northcoast Engineering (NEC) components for the DCC system, I will be using an older Windows-based PC to run free software known as "Java Model Railroad Interface" or JMRI for signal control.  JMRI will work with a Digitrax signal system to drive signal heads and display aspects for operations based on block occupancy.  I will do my best to document the process for you here.

Since block detection and signal wiring will require extensive additional wiring work beneath the roadbed, I have elected to minimize progress on layout fascia and scenery until the signal system is installed.  It will be much easier to run new wire and make new connections before the fascia blocks easy access!  However, this approach means it will likely be more than a year before scenery begins in earnest on the main line.  However, since the M&E will not have signals, work on scenery there can begin sooner.  Plenty more to come as the winter wears on!

Into the New Year

A new year, and slow, steady progress on the Onondaga Cutoff continues! Best wishes for 2012 to all. Thanks to assistance from several regular operators, most of the backdrop is now installed, awaiting tape, spackle, and paint. It’s actually nice not to see the foundation walls, at least, behind the railroad.


One recent development on the OC is that through operations, a limitation has become clear. If you click on this ‘straight-line diagram’ of the railroad, which shows both original and new crossovers now at CP 294, it becomes clearer.

Though there are 5 tracks in staging, there was no way for a train in either direction on Track 1 in staging to come out of staging and cross over to Track 2 until that train reached an interlocking with turnouts that allow that move. For westbound trains, this is resolved at CP 277, where a universal crossover exists just for this purpose. This allows a dispatcher to route trains on either track in either direction, and at a nice distance from Onondaga Yard.

For eastbounds, however, there was no universal interlocking from staging all the way to CP 282, which is also the west entrance to Onondaga Yard. Trains leaving staging on Track 1 working east could not cross over to Track 2 until they were at 282. When a train is working the yard and fouling Track 1 at CP 282, then, Track 1 was essentially unavailable from CP 295 all the way to CP 280, more than half the railroad.

Therefore, it was necessary to build a crossover to enable trains leaving staging eastbound on Track 1 from CP 295 to cross over to Track 2. The existing geometry allowed me to install a new #8 right-hand crossover at CP 294, creating a new universal interlocking there – and eliminating a major bottleneck for the dispatcher.

However, the installation of that crossover meant cutting into existing track and roadbed, drilling new holes for Tortoise machine throw rods, mounting Tortoises – challenging tasks, especially after the fact. It took time to align the new turnouts with existing track, and with each other. This allowed me to locate the throw rod locations and drill the holes, and finally to install each component while ensuring the geometry worked out as intended. Finally, both turnouts had to have feeders installed for track power.


The next major development is a biggie: design and construction of a working, dispatcher-controlled signal system. Much more to come soon as this gets underway!

~RGDave