Reached another milestone yesterday–the main helix between the main and upper levels is now complete! I have track reaching to 58″ off the floor. Like the helix between staging and the main level, I used what I call the “double pinwheel” method of building a helix (click here for a full write up on the double pinwheel helix). It requires only straight-line cuts, it’s very forgiving of non-precise cuts, and it’s rock solid.
The helix joins the town of St Charles to the towns of Turners Siding, Kemmergem, Monarch and Benedict on the upper level. The grade (3%) starts at the RR-west end of St Charles yard and continues up three turns of the helix. There will be another 15 feet of hidden track before the line emerges, so that’s effectively a fourth turn’s worth of elevation gain after the helix. Of course, when I built the base supports for the helix, I miscalculated something that had the initial grade into the helix closer to 4%. At first I was like, “well, maybe it will work.” I test ran a few locomotives up the first loop (all I had constructed upon discovery), and while my lightest locomotive could still pull 11-12 cars up the grade, I finally wised up and decided to fix it before it became unfixable. After dropping to 3% (lowering all the supports by 3/4″), the lightest locomotive could again haul 14 cars and a cab from a dead stop up the grade, the same limit as the helix from staging. I’m sure I will be glad I made this adjustment in the future! Measure twice, cut once… thankfully it just required loosening and reattaching some screws and only a few new cuts of supports to fix it.
It will be a little while now before I start the upper-level benchwork, but finishing the helix will allow me to install the lower-level backdrop, so the beautiful insulation pads will no longer be visible on the main level. I know that will be a big disappointment to many of you who feel the insulation just adds an extra layer of realism… ok, enough sarcasm. Building helixes isn’t fun for many modelers (including me), so I’m very glad this portion of the layout is now complete!
This week’s project was completing the fascia for the lower level. I love the look of the curved black fascia and track diagrams. I’ve detailed fascia elsewhere, so I’ll stick to what’s unique here. While the switch mechanisms can be partially installed prior to fascia, it takes the facia being in-place to install the manual switch control knobs and push rods. While most of the mechanisms were pretty basic, there are three switches more than 30″ from the fascia on the “RR east” end of St Charles wye where the tracks emerge from the helix and staging. I wasn’t sure if I’d be able to use the push rods for long distances, especially since two of the switches are beyond 36″, the length of the .062″ steel rods I use. The trick with the push rods is the longer they are, the more they tend to flex and bend (and, in turn, not throw your switch mechanisms). This can be partially rectified by using additional brass tube guides in wooden blocks along the rod’s path, about every 12-15″ or so. That was good enough for the first mechanism that was <36″ from the fascia.
For the longer rods, I decided to try connecting 2 steel rods using a 6″ piece of 3/32″ brass tubing and Gorilla Glue. I use the Gorilla Glue to attach the wooden knobs to the steel rods, so I know it’s got at least SOME game with metal. Since these rods will be hidden by scenery, I decided not to trust glue alone, so I lightly bent both the tube and steel wire about 1″ from the end of the tube on both sides–if there’s one thing I’ve learned, even a slightly bent .062″ wire does NOT want to pull through a 3/32″ piece of brass tube! Once I added the bends, the mechanism is solid as a rock! I’ve now verified that the manual push-rod controls are viable to at least 48″ from the fascia–not bad at all, and all remaining switch controls should be well under this length.
Another unique feature of the St Charles fascia is the addition of two SPST toggle switches that isolate two of the tracks from the wiring bus. A while ago, I detailed how I did something similar for my staging tracks so I could easily silence sound locomotives when they’re not actively involved in the operations. St Charles was often home to a mine run, so the pair of mine-run engines hung out on either the “house track” or aptly named “engine track” adjacent to the depot. Since these are the only tracks on the levels with scenery where I anticipate parking locomotives, I decided to give them the same insulation and toggle setup as the staging tracks. While I will likely rarely use these, I figured it’s SO much simpler to add them now than decide I need them after-the-fact.
Hit a major milestone yesterday: the lower level tracks are complete! I completed the first scene, the end of the line at the Mayflower tipple, several months ago, and the last few months I’ve been working on the long scene at St Charles. St Charles – the branch’s namesake – was home to a wye, a depot, a couple small tipples, and a three-track “yard.” St Charles was also home to a mine run when the tipples were busy, so this is the central scene on the layout. All told, the St Charles scene is about 26 linear feet long and required 14 hand-laid switches (about 1/2 of them curved) and a couple of bridges.
Electrically, I made the entire Mayflower branch an auto-reversing zone for the wye. Can I just say the On-Guard AR solid-state DCC auto-reversers are awesome? The switching of polarity on the wye is absolutely seamless and unnoticeable–I highly recommend them! I also isolated a single rail of both the “house track” and “engine track” in the middle of the wye and will place a switch on the fascia. This will allow me to turn the power off to the two tracks that would hold idling power in case any sound locomotives get annoying. Next step is fascia and switch controls.
Celebrated something of a mini golden spike this week. I completed the mainline track for the Mayflower leg of the wye in St Charles. With this bit of track installed, trains can finally run up from the staging level and to a tipple (Mayflower). It’s not much in the grand scheme of things, but its a milestone nonetheless. Next will come the Monarch leg of the wye with its house track and engine track followed by the two yard tracks above the wye. Getting pretty close to having all the track on the main level installed.
I learned a good lesson on hand-laying track. The frog needs to be electrically isolated by cutting the rails on either side of the frog. To keep the frog and rails aligned, I spike both sides of the cut before cutting. When I wired everything up and applied track power the first time in this new section, it immediately shorted out. I could hear a soft click coming from one of the switches. Usually this means one to three things 1) I wired a feeder to the wrong bus, 2) my cut didn’t make it all the way through or 3) the point rails somehow slid back together with the frog. A check of all three of these things was good, but when I dropped the feeders from the point rails, the short cleared–it was still somewhere on this switch. After a few minutes of jiggling random things and staring at the problem, I finally figured out that two of the spikes that were holding separate point rails down near the frog cut were touching even though the rails weren’t. Go figure, metal spikes attached to metal rails conduct electricity… a quick repositioning of one of the offending spikes did the trick.
After a nice and productive summer focusing on building models (thank you St. Louis RPM meet for the motivation), I’ve returned to working on the layout. The next step is a big one, the scene of St. Charles, VA. This is the central scene of the entire layout. This is where trains emerge from staging, the site of the local station and engine tracks, the small yard for the branch, and the wye taking trains either up to Mayflower or Kemmergem, Monarch and Benedict. While I could have modeled these tracks in pieces, I decided to do the entire scene at once to avoid further stalling and to limit the need for multiple coats of stain on the ties (takes forever to clear out the smell!).
I’m using basic cookie cutter construction. I lay down 1/4″ door skin ply and draw out the tracks using templates and a piece of flextrack with thumbtacks to hold it in place. After cutting the upper subroadbed, I then trace the pattern only 1/2″ plywood staggering the seams. For this scene, I cut out all the 1/4″ ply to make sure everything fit, then cut the thicker ply. Construction so far is proceeding according to the plan, though I did make a modification to the track plan–instead of having the two small coal-loader tracks butting into one another, I reversed the siding on the wye to place the switch along the benchwork instead of in the back corner where it would be tough to reach. The prototype siding in this area appears to have been double-ended, so it’s not a big deal either way. As a bonus, the new arrangement allowed me to get a few more cars’ capacity on these sidings.
The toughest part of this scene is the pair of bridges across Bailey’s Trace at one end of the wye. On the prototype, there was a plate girder bridge just over 50′ in length on the main, and a shorter plate girder bridge (about 25′) which adjoins a curved bridge portion which appears to have been joined with a short wooden trestle on the curved portion (it’s all been redone with ballasted decks now). I used a 50′ Micro Engineering bridge for one section and cut down a 30′ ME bridge by one section on the other. I hand-laid the bridge ties using a template I drew up on a piece of paper. Next will come the concrete supports so I can attach the bridges and lay the rails across.
It’s encouraging to see progress in the layout room again, and I will definitely have my work cut out for me between now and Christmas hand-laying 14 switches!
Working on the fascia is very mundane and unexciting, but as a human factors engineer by education, I take my fascia very seriously! On most layouts, the fascia is the primary interface between operators and the layout. It’s often where we place controls and place names, and it’s also a blank canvas we can use to help our operators better understand the scene they’re interacting with. Like many who perform switching operations on their layouts, I like to use the fascia to help operators understand as much as possible about towns, tracks and industries to aid in making their switching moves.
As far as construction goes, my fascia is just sheet Masonite fastened to the benchwork with drywall screws. I cut it in thick strips that will account for all the vertical scenery contours along the front edge. Where there are noticeable gaps or indentations from the screws, I touch things up with lightweight spackling compound and wipe it smooth with a damp cloth after letting it harden for about an hour–this saves a lot of time sanding later. Then I draw in the ground contour with a pencil and cut it with a jigsaw. A little black paint, and things are ready for the operator features.
On my last layout, I drew the track diagram onto the front of the fascia, and it worked so well I decided to continue it on my new layout. It’s essentially an elongated track diagram like you’d find in a railroad track chart that lines up with the adjacent track and switches on the layout. I draw the lines parallel to the bottom of the fascia 3/8″ apart. To draw them, I’ll set a combination square so the end of the ruler is at the exact height I want when the square is pressed against the bottom of the fascia, then I’ll run the combination square around the bottom of the fascia while dragging a pencil along the top to draw it onto the fascia. I draw the switches onto the fascia at 45 degree angles with the convergence drawn where I want the switch control rod to be placed (directly perpendicular to the switch mechanism usually). One lesson learned from my last layout I was able to incorporate into this one is that the “straight” line through a switch control is the “normal” position of the switch while the “divergent” line represents the “thrown” position. It required a little “wiggling” of the track lines in the yard ladder to make this work, but it’s intended to help operators understand how to set the switches before departing a town.
I installed the switch control rods (see full article on switch control mechanisms here) after the lines were drawn but before placing the graphic tape on the fascia to represent the tracks. I use two sizes of white graphic tape for the tracks, 3/32″ to represent the main, and 1/32″ to represent sidings. This offers another visual clue for operators so they know which tracks not to block and what to leave clear after leaving town. Next come the labels. Every town on the layout has both a name and a unique 2-letter destination code for routing cars. In this case, Mayflower is represented by “MY.” Within each town, each track has a unique number, for example, MY 1 is the tipple track for the Mayflower Tipple closest to the main. For tipples, I also designate where the empty (MTY) and loaded (LD) cars go, so an empty hopper destined to be loaded on Mayflower’s outermost track would need to be placed at MY 4 MTY. Towns are labeled with large dry transfer lettering with the town name above a line and its destination code in smaller letters below. I also cut a small gap in each track tape on the fascia near the switch and use dry transfer numbers to label each track by number. Industries are given white labels with black lettering that gives the industry name, town initials, track number(s) and MTY | LD track dividing line if needed (see picture of the Mayflower Tipple and its labels). If two or more industries are on the same track, I use a “B,” “C” or so forth to denote each spotting location. In the case of Mayflower, I’ve identified a spot on load track 4 where I might have operators occasionally spot cars of supplies and a spot on the tail track of the main where I might have them spot covered hoppers of AN–each of these spots has its own industry label adjacent to the correct track on the fascia track diagram to make it clear where to spot the cars.
There’s nothing really novel here, but I do think this method of using track diagrams on the fascia and integrating switch controls, town names and industry track designators with unique operational codes goes a long way toward making the layout more intuitive and operator-friendly.
I’ve completed laying rails on the Mayflower section of the railroad, and I’m just about ready to run the first mine run–exciting times! Everything is hand-laid, and I’ll share a little about my technique. I lay everything in place instead of using a jig because I find the track flows better. I use code 83 for the main and code 70 for sidings, and I spike everything with small spikes, about every 5th tie except for switches. I like to work outside-in as you’ll see in the pictures of the load yard ladder. This means the first few switches take a while because you’re notching for the points of ALL the switches on that piece of rail, but it speeds up as you go because you’ve already filed pieces up-front. I don’t pay much attention to frog numbers or curves–I just lay them as the ties dictate (and remember, the ties were laid down based on outlines traced around flex track, so that’s where I check radii and things).
You also might notice that I’m not putting in guard rails alongside the frogs for some switches which is prototypical for sidings in some areas of the country. I use them on both sides of every mainline switch, and I use them on the tightest curve side of every curved switch. For the rest, I test cars and put them in as needed. If your frog is straight, you probably wont need them. If your frog is curved at all, you probably will. I like the look of the switches with no guard rails because it emphasizes “siding.” I also use the track to tell part of the story, so you’ll notice that I laid an extra switch’s worth of ties for a 5th track under the tipple that has been removed (try doing THAT with commercial track)–Mayflower had a spot to load on a 5th track, but track diagrams in my era only show 4 tracks. I’ve modeled it as if they just pulled up the rails and laid through the switch instead of pulling up all the ties.
All the feeders (a gazillion) got dropped and attached yesterday, so I just need to set out some cars and recruit my engineer for the first-ever mine run on the layout. Woohoo!
Made some progress this week on the Mayflower Branch section of the layout. All the ties have been laid! This is a tedious but simple process made a lot easier by the outline of track left over from tracing flextrack onto my subroadbed for cutting. I just place the material (1/4″ door skin for me) on top of the layout, place thumbtacks in the holes of a piece of limber Atlas flex track, fasten the track down with the pins using turn radius templates and “eyeballing” the rest, then use a pencil to trace down both sides of the track. For marking switches, I leave a portion of the track fastened and move the loose section to trace the divergent track. After tracing, where the pencil marks diverge is the location for the turnout points and longer block ties. After the subroadbed is secured in place with risers, screws and glue, I’ve got a perfect template for the tracks on the layout for laying ties (more about making and laying ties below).
Now that all the ties are in place and I have the mock-up of the Mayflower Tipple, I can really start to visualize the entire scene. I’m really liking how the track snakes into the scene, and I think the gentle curve into the tipple will really look cool with strings of hoppers hanging out. You’ll also notice the tracks run into the wall–I didn’t have enough room to model the empty yard, just some space for empties above the tipple. This was an easy compromise to make for space because the empty yard at Mayflower was a stub-ended affair, so crews still had to run around and shove cuts of empty hoppers, just as they’ll need to do here. I can’t wait to get the rails down and operate that first mine run! There’s a lot of rail-laying between now and then, but it’s good to see it coming together.
Making Ties. Rather than buy bags of ties, I cut my own from sheets of basswood, 1/16″ for siding ties and 3/32″ for mainline ties. It’s not that ties on the prototype were different heights, but using different height ties on the layout keeps siding tracks a little lower than the main (very prototypical). I use a “spacer tie” to line up a metal straightedge the proper distance from the edge of the basswood board, then cut it with a couple strokes of a sharp utility blade. Most sheets come in 2′ length, and I’ve found it easier to cut it to 1′ length first so I don’t have to move the straightedge in the middle of a cut. With a bunch of sticks in hand, I then use a Northwest Shore Line “Chopper II” (amazing tool) and a scale rule to cut the ties in .5′ increments from a scale 8.5′ to 16.5′ with a separate, well-marked ziploc baggie for each size and length. Standard ties are 8.5′, and switches require a few of each longer size as you progress up the switch with 16′ ties for block ties at the points.
Laying Ties. Laying ties is a simple matter of putting down some wood glue on the subroadbed and placing them. I work in sections of about 8-12″ at a time to make sure the glue doesn’t dry before the tie gets there. Ties tell a story about the kind of track you’re modeling, and it’s one of the reasons I love hand-laying track. Mainline track should be in good working order with closely spaced ties perpendicular to the rails and just a little side-to-side variation. Well kept sidings are similar but with perhaps a bit wider spacing between ties. For old, well-used sidings like you’d see at coal tipples, I’m pretty haphazard with my ties, allowing some of them to kink off perpendicular and lots of variation in spacing and alignment from side-to-side. It looks absolutely disgusting before the rails go on, but the effect is more subtle once the ties are stained and the rails are in place. I love disgusting looking track that still runs well, so I can be pretty aggressively messy when laying siding ties!
I’ve finally started laying subroadbed onto the main level of the layout, and I’ve chosen the section between Baker and Mayflower as my first scene (see track plan). Mayflower is the first of four large tipples on the layout, so to make sure I’ve got proportions and track spacing right in the plan, I decided to build an HO scale mock-up from foam core and cardboard first. I’ll just warn you, this is not a time-saving process, but it certainly helps you visualize a scene and make adjustments before building more permanent structures and track. Because this was the first and the only large tipple on the lower deck where vertical spacing would be important, I decided it was worth the effort.
Step 1 – Research. I’ve made a commitment to modeling the tipples as closely as possible (rule: no Walthers New River Mine allowed–it’s not a bad kit; it’s just over-used and too recognizable). The trouble is this area is not well documented by photos, so there’s a lot of guesswork and extrapolation involved. The first resource I had was a track chart that showed four tracks under the tipple, a medium-sized operation. Next, I found an aerial photo from the 1960s that showed the basic footprint of the tipple and the distinctive portion of the tipple connecting the tipple to the mine that sits 45 degrees to the tracks. Finally, I found a single grainy photo of the front of the tipple and a painting of it from a post card that someone had posted to Pinterest. With these resources, I had enough to rough-out a drawing.
Step 2 – “Scale” Drawing. Regardless of whether or not I’m building a mock-up or the final model, I need some sort of scale drawing to build from, and for tipples that were one-of-a-kind, actual drawings are very rare. I use the word “scale” here loosely because none of the research offered any clear dimensions. Still, distances between tracks and height above hoppers can be estimated, so I did my best. To make the drawing, I used my favorite drawing program, MS PowerPoint. No, PowerPoint isn’t designed for this, but it’s easy to use, and you can draw lines to specific dimensions and angles. I would work on one side, matching it as best I could to the photos. Then I’d copy relevant bits from the one side to use as size references for the next side until all the parts were drawn. Some sides were longer than a piece of paper, so I drew these as two separate drawings I could glue together later. The tricky part was the photo and postcard showed two different time periods for the tipple. The photo showed a section added to the front over two tracks, so I had to figure out how this might have worked in concert with the tipple represented in the postcard.
I decided to draw windows and corrugated metal siding on it as well. This was a simple matter of making a generic window and copy/paste it in place. The siding was just a texture I found online and copied into scale sheets to place behind the drawing (use the “send to back” function on right click). When everything looked decent, I printed it out.
Step 3 – Building the Mock-Up. For material, I used black-on-black 1/4″ foam core picked up at the local big box hobby store. It’s easy to cut and work with and has enough rigidity to make a durable structure. I cut out pieces of the drawing and then glued them to the foam core using normal white glue. Where I have a corner, I needed to pick one wall to recess the width of the foam core so they won’t overlap (i.e., when gluing the drawing would overhang the foam core by one foam core width). As you can see from pictures, the glue caused a little discoloration and warping of the paper, but hey, it’s a mock-up. Next I cut along the edges using a metal ruler and X-Acto blade. The trickest part by far was cutting the intricate frames of the leg pieces (5 total). A mock-up doesn’t really require this level of detail, but I decided it was important enough to judging the look of the tipple that I spent an extra 3 hours or so cutting these out. The glue and paper caused some of the sides to bow a little, so for these I cut out 1/2″ wide strips of foam core to attach perpendicular along the back and bring them back into shape and provide lateral strength. This required laying a heavy book on top while it dried, but it worked well.
With the pieces cut out and strengthened, I started gluing them together. I worked on one corner at a time and used little pieces of square foam core to keep the corners square. I added structural pieces of foam core wherever needed to keep things sturdy. Again, the trick was to work one joint at a time (no more than two) instead of trying to build it all at once. The glue dried sufficiently in about 30 min, so I’d just come by every 30-60 min and glue another side on. Finally, I made the roof out of corrugated cardboard salvaged from a box–it’s a little thinner than the foam core but still has sufficient rigidity. After test fitting the roof pieces and making adjustments with the blade, I glued on the roof pieces–this is tricky because you have to worry about a lot of edges at once.
Step 4 – Evaluating the Scene. With the Mayflower Tipple mock-up complete, I could now mock-up the scene on the layout and see where the tracks would go, how tight the spacing would be for hoppers (adequate), how far I could bring it out or push it back in the scene, how much space I would have with the upper deck, etc. As it turns out, the tipple fits perfectly and doesn’t need any adjustment, but it would be easy to make cuts and repairs to the mock-up to try different things that would be a lot tougher to do on a final model. Any changes would simply be added to the drawing to use for the final model.
Conclusion. While I could have been halfway done with a permanent model in the time I built this mock-up, I now have the confidence in my drawing and in the scene to build the final model. Besides, it will be a while before construction on the upper deck in this area will be complete enough to install a permanent model, so in the meantime, the mock-up will give me and other crew members a good stand-in to enhance operations that I don’t mind getting a little roughed up. Beats having to use your imagination when switching out empty and loaded hoppers!
This post is less of an update and more of an effort to chronicle a minor but important part of layout construction: DCC track bus planning. Installing DCC has two major components, the first is wiring up the DCC system with its command station, booster(s), throttle panels, computer connections, wireless throttle receiver, and network cables–I covered the installation of the DCC and logical connections of the DCC system here. The second component of DCC is installing the wiring for the track which includes electrical buses, circuit breakers, and dividing the layout into zones which I’ll cover here. Even if you don’t plan to use circuit breakers, its still a great idea to “future proof” your layout by wiring for zones–you can always tie the zones together at the booster, but you don’t want to be installing new buses and feeders under finished scenery. These zones will also help with electrical troubleshooting if you can’t find a short–you’ll at least be able to isolate it to a smaller section of track.
My layout is small enough to easily be powered by a single 5A DCC booster, in my case, an old Digitrax DCS100 Chief acting as a booster (a DCS51 Zephyr is the command station). At most, I’ll have 3 trains and perhaps 6 locomotives running at any given time. The DSC100 can handle this current easily, but with circuit protection provided by only the booster, if any one locomotive runs against a switch and shorts on the frog, power for the entire layout and all trains shuts off. The solution? Dividing the layout into zones and using solid-state circuit breakers!
I already needed at least three zones on the layout because I have two reversing loops which need to be isolated from the main bus and powered through an auto-reversing circuit breaker. In my case, I use the On-Guard AR reversing breakers from DCC Specialties. Although a tad on the expensive side, the solid-state AR circuits are rock solid and work more reliably than their analog counterparts, especially with the current draw of sound locomotives. Aside from the reversing zones, I decided to divide the remainder of the layout into 3 additional zones, one per level. How many zones you use is a balance of cost and functionality–too many, and the cost quickly gets out of hand along with the wiring (each zone needs its own wiring bus), too few, and operators causing short circuits will quickly ruin the fun of other operators in the same zone.
With my max of three trains in mind, I drew zone lines primarily to ensure each train would be in its own zone most of the time. Consequently, the zones are different sizes based not only the amount of track it powers but the operating locations for the trains. For example, zone 3 is the largest on the layout and powers not only the entire upper deck but the helix between the lower and upper decks as well. That’s a lot of track, but only one train at a time will venture to the upper deck, so it can be large. The main level is two zones with one covering the wye and yard at St. Charles, and the second (a reversing zone) covering the branch to Mayflower–there will often be a train at St Charles while another works the Mayflower Branch. Finally, I made the staging level its own zone (along with a second reversing zone) so that operators moving things in and out of staging won’t impact operators on the visible portion of the layout.
Once I knew where the zones would be, I ran an electrical bus from the vicinity of the booster all along the benchwork where the tracks of that zone would go. I’m all about overkill here, so I use household Romex 14 gauge copper wiring that I pull out of the sheath. I drill holes and run the black and white wires through the benchwork about 4″ apart. One important step is to LABEL THE BENCHWORK with a Sharpie every few pieces of wood, especially if you have multiple buses running side-by-side, so you don’t become confused as to which track you’re hooking up to what zone–if you mess this up, your track could be wired to multiple zones simultaneously and eliminate any benefit of electrically isolating the zones. At the ends of the run, I just wrap the lines around a drywall screw. If I need to run a bus in two directions, I just make a junction and join the three wires together with a wire nut, just like I would do with household wiring.
I strip about 1″ of insulation off each wire about every 2-3′ where it will connect to track, and I wrap a 4″ section of the uninsulated ground wire from the Romex around the bare spot about 3 times leaving 1″ or so hanging off both ends. A little solder keeps this pigtail in place. I’ll offset the white- and black-line pigtails by about 4-6″ horizontally in addition to the distance between the wires to minimize the risk of shorts. These pigtails become the connection points for track feeders using wire nuts. Of course, you’ll need to physically separate the tracks between zones by either cutting gaps in the rails or using a plastic insulating joiner. Be sure to overlap your gaps by 3/4″ – 1″ for zones connected to an auto-reverser!
Between the bus for each zone and the booster is a circuit breaker, in my case a PSX-3 (essentially 3 PSX-1s) from DCC Specialties (I covered how to program these with a DCS51 here). Like the auto-reversing breakers, the PSX breakers are solid-state and are well worth the money over analog breakers, especially if you’re running sound locomotives. I have been extremely impressed with these units so far! The PSX design makes it easy to daisy-chain multiple breakers with just one set of wires to the booster. The only split I had to make was after the PSXs when I had to run the last booster connection off the PSX to the two AR circuits. To make the connection to the buses easier, I made a simple “panel” on part of the benchwork where the 14 gauge wires for each bus come through, get wrapped around a drywall screw, and have the ends exposed for connection to the PSX via smaller wires connected with wire nuts.
A final step for me was figuring out a way to monitor each zone to know what’s active and what’s shorting out. It’s more difficult to detect a short with the PSX than the DCS100 because the booster makes a distinct noise when it’s reacting to a short, but the PSX is silent. For monitoring, I turned to the Digitrax UP5 universal interconnector panel. I have five of these panels at various spots along the fascia of the staging level for connecting throttles, and each has a “track status” light that can be wired to the track bus. I chose one in proximity to each zone to be the “zone indicator”, labeled with a sticker for the zone number to help me remember, and wired it to pigtails of the corresponding track zone. This way, if a short occurs, only one light corresponding to the affected zone will go out to aid troubleshooting. I wired the UP5 adjacent to the command station and booster directly to the booster wires (no circuit breaker in between) and labeled it to be the master monitor–if it’s on and everything else is out, it tells me the problem is somewhere in the circuit breaker wiring.
Finally, I drew pictures (the ones seen here) of the zones on top of a layout diagram to help me remember exactly where the zones go, and I placed this in my layout binder with all the other helpful information on the layout. This wiring took a good bit of time to plan out and install, but now the layout has the robust electrical backbone to make for smooth connections, easier troubleshooting, and ultimately more fun for operators.
I haven’t added an update for a while because the layout doesn’t really look any different. That’s not to say I haven’t been working, it’s just been all the boring stuff – wiring. It may be boring, but if you want your layout to run well, it’s doggone important, so it shouldn’t be left as an afterthought.
First, I corrected an issue I first noticed when running trains around the staging level – sound! I never ran sound locomotives on my previous layout, and now that I’ve got a few, I noticed how annoying it is to have diesels sitting in staging making idling noises that are easily heard throughout the room. These trains are supposed to be dozens of miles away, and I don’t want to hear them while they’re in staging. Of course, there’s the option to “mute” a sound locomotive temporarily, but I don’t want to force operators to go through an entire consist muting and unmuting every time they pick up or drop off a train in staging (and every time a short occurred, all the sound would come right back on anyway). For a solution, I went “back to the future.”
I installed a small SPST toggle switch (picked up 20 for cheap on eBay) for each staging or locomotive track (9 total) on the fascia where the track diagram will be. Unfortunately, this meant pulling dozens of feeders from the staging tracks and running a secondary bus for one rail under each track that’s connected to the switch. Thankfully this didn’t involve any de-soldering because all my feeders are connected via wire nuts. Now I can turn off the power to any staging track with sound locomotives until they’re needed, and all the operator has to do is flip a single switch. Old school “electrical block” solution to a DCC problem.
While I was at it, I broke down and ordered a DCC Specialties PSX-4 solid-state circuit breaker to go with my On-Guard-AR auto-reversing circuit breakers. I’ve been planning to do this all along and wiring for multiple zones, but after reading through the PSX documentation, I discovered my understanding of the PSX-4 and how it interacts with AR breakers was a little lacking. I didn’t know that the AR breakers should not be wired off a PSX zone but directly wired to the booster themselves. Turns out with the two reversing zones already protected by an AR circuit breaker, I really only needed a PSX-3 because I only have one non-reversing zone per level (1=staging, 2=lower level, 3=upper level, 1R=staging reverse tracks, 2R=St Charles wye to Mayflower). “Snap” – “hey look, now I have a PSX-3 and a spare PSX-1”. . . smart DCC Specialties! Not gonna lie, it was a little confusing to program the PSX to work well with the AR breakers using the Digitrax Zephyr, but I finally figured it out:
Set the Digitrax booster to .5 sec short circuit trip using booster instructions (CV 18 to “c” on my DCS100)
Solder the Digitrax jumper on each PSX zone per PSX instructions
Set the PSX to programming mode via the jumper per the PSX instructions
Ignore the part about setting the PSX address unless you need the PSX to respond to “on/off” or other special commands from the DCC (If you just need it to be a circuit breaker, you don’t)
Connect a single PSX zone directly to the Track A / Track B from the Zephyr
Turn the track power on on the Zephyr
Put the Zephyr into “OPS Programming” mode
Select CV55 (“CV”, “55”, “CV”)
Press “1” and “CV-W” (add delay to the PSX so the AR zones will trip/reverse first)
Select CV65 (“CV”, “65”, “CV”)
Press “80” and “CV-W” (set delay to 10ms which works for my setup)
If Zephyr shows “Busy,” exit programming mode and try CV65 steps again
Set PSX to ops mode via the jumper per the PSX instructions
Repeat steps for additional PSX zones
I encourage you to read all the instructions first and choose your own adventure–just sharing what worked for me.
Finally, before I lay subroadbed and track for the lower level, I had to add the wiring bus, or in this case, two wiring buses. Two are needed because the St. Charles wye needs a reversing circuit, and the reversing district carries over all the way to Mayflower. Lots of drilling holes, cutting open Romex, and pulling heavy gauge wire through. I still have to make the little pigtails for feeders, but it’s mostly done. Shouldn’t be long before I’m laying track on the main level!
Well, the train garage now has a roof. Or in other words, I reached a major milestone today by completing the benchwork for the main level. What made this particular project tricky was I’m trying to match the fascia contours of the staging level exactly so they look like they’re one thick deck with staging under and trains over. The upper level won’t need to match, so parts of that project will be slightly easier.
All the benchwork is open box, so most pieces were fairly easy to build. However, building the peninsula benchwork was tedious because 1) I needed a pretty big, single piece of benchwork to give it strength to reach from the wall to the edge of the peninsula, and 2) I needed a very specific octagon pattern at a very specific place in the middle of the peninsula to serve as the foundation for the second helix which will take trains from the lower (main) to the upper deck. I’m very glad I have an open basement area just outside the train room where I can build big pieces of benchwork. I’m also glad my wife announced she’d like to replace the carpet in the basement soon, so I’ll take advantage of this window for making messes on the carpet without repercussion!
Now that I can really see it, I really like the space I’ve got between staging and the main deck. I could have brought the staging up a couple inches and still be ok, but it’s nice being able to see trains in staging just by backing up (instead of squatting down), and I like the functionality the fascia cubbies (now painted black) add – something not possible with less vertical space. I also like how the cabinet corner is turning out. I’ll still be able to reach all the drawers and the DCC components next to the cabinet easily, and it will make the corner under the Mayflower tipple look really finished when everything’s done.
Next step is to build the hidden connecting track between St. Charles and Mayflower (already started), and then comes the first subroadbed and track that will be part of the visible layout with scenery–exciting!
I spent a few hours on the layout this past week primarily getting the staging level ready to have benchwork for the main level laid on top of it. There will only be 8″ or so between the decks, so it will be tough to do things like swing a hammer to nail down track once the main level goes in. I pounded the 100s of track nails down well to avoid trouble later, and I finished the final 3 fascia pockets for holding clipboards and other operator paperwork. As you can see, I tried to pick places where nothing is going on behind the pockets so I’m not covering switches or key viewpoints for spotting trains or clearing fouling points and the ends of tracks. Not sure exactly what I’ll keep in the lower pockets, though the one in the back corner is big enough to hold 8.5×11″ papers, so I finished it out with a masonite floor. I will say, I grossly underestimated the amount of sawdust MDF makes!
While I was at it, I decided to fill up some of the extra space on the staging level with a few car storage tracks. Who doesn’t need more places to store cars, right? I’ve got plenty of staging, but being a coal railroad with little non-coal traffic, I wanted a place to store some of the extra non-coal cars that won’t be used every session, and I wanted to store them where I could easily add them to trains without taking up an actual staging track. I just nailed in 3 tracks directly onto the subroadbed in a spot that’s easily accessible and doesn’t block anything critical behind it like switches or fouling points for the active staging tracks. It’ll hold about 15-18 cars which will help. I’ve already got a coupe extra short staging tracks for the locomotives. These are connected to the railroad via a switch because I didn’t want to be picking up locomotives every time I swapped them out–I’m ok with the extra handling of freight cars.
While I’m giving an update, here’s a picture of the toughest benchwork on the layout so far–it’s the corner by the door that goes around the staging helix. What made it so tough is I wanted to match the curvature and location of the staging-level fascia below, and it happens to be a series of complex curves with NONE of the edge on a 90 or 45-degree line. I also wanted it to be secured well enough to walls, cantilevers and the helix that it wouldn’t need legs down to the staging level. I ended up building it in-place around the helix, and I’m satisfied it will do the trick.
Just [almost] finished the first helix. Man, am I glad not to have to work around a floating wall for the second one! As promised, I’ve written an article on how to design and build a helix that’s strong, reliable, easy to build and an efficient use of space and material–it’s a design I call the “Double Pinwheel Helix.” I’ve built two helices this way now, and I haven’t seen another design that comes anywhere close to being this simple to build using nothing but plywood and a circular saw (no jigsawing for hours), and it’s very forgiving if you don’t cut the pieces exact or your space is a little wonky.
Rather than post it here, I put the article on Appalachian Railroad Modeling where more people would be able to find it and hopefully be inspired to overcome their fear of building a helix. You can find the article here. Here are a couple of photos so you can see the progress.
My undergrad is in Human Factors Engineering, the science of designing things so people can use them easily and intuitively. Though I’ve never been a Human Factors Engineer in my work, I enjoy applying the principles I learned to my model railroad. The first step of human factors for a layout is usually the space for operators–designing aisles wide enough to accommodate your people and give them access to the areas of the layout they’ll need. The next step is figuring out what other needs for space they’ll have while operating. One of those space needs is a place to store their “stuff” they use while operating–enter the fascia pocket!
I’m continuing my practices from my former Interstate RR layout. Upon being assigned a train, each operator will be given a clipboard with their train information (location, engine number(s), throttle assignment, etc.), instructions on how to work the train, any applicable train orders, a map/diagram of the layout, a switchlist, a pencil, and a decoupling tool. The operator will need this clipboard throughout the session, and since I DON’T want operators setting their clipboard on top of the layout, it’s important I design in convenient spaces for these clipboards.
Once I put in the first section of main-level benchwork above the staging level, I had a nice ~8.5″ space between levels that might work for this purpose, but I needed some way to keep the staging tracks safe. I decided to build clipboard pockets over the staging tracks since trains only need about 3.5″ of the 8.5″ for clearance. The clipboards are 9″ wide and 13″ long, so I’m making each clipboard pocket 10″ wide and 12.5″ deep–this will ensure the clipboard sticks out about 1/2″, just enough to ensure the clipboard is easy to grab without impeding on the aisle or snagging on operators.
To support each pocket, I need a base about 3.75″ tall between the staging base and the pocket. I’m using these as storage pockets for useful things for operators as well. The pocket depth depends on how far the staging tracks are set back from the fascia. In some areas, I can make pockets 12″ deep for papers. In other areas, I can make pockets 4.5″ deep to hold extra switch lists, timetables or hand-outs. Some areas where the track is close only allow a ~1″ pocket. . . hmm. . . I know, that’s perfect for storing extra 9V batteries for the wireless throttles!
The next step was figuring out where to put them. I wanted them in spots useful for operators, but they also needed to be away from the yard ladders in staging and spread apart enough to allow easy access to all staging areas. I decided on five locations that line up well with the action on both decks. A sixth area is formed by a counter top a few inches under the benchwork for Mayflower. There will be a pocket within 2-3 feet of an operator for about 90% of the layout.
I made my pockets from 1/2″ MDF–strong and smooth. To secure the base to the staging level, I mock fit it in place, drew an outline, and used the outline to drill pilot holes through the subroadbed. After gluing the base pieces in place, I went back and drilled up through these holes into the MDF to lock it in place. Next, I secured the deck for the clipboard pocket to the base using glue and countersunk screws since the back would be suspended over the staging tracks. Finally, I glued the side pieces on top of the clipboard deck to finish the pocket. While I could have made the side pieces solid and mounted the clipboard deck inside them, I figured this method would be more durable as there’s no way for the deck to break away from the sides.
These pockets are certainly nothing fancy or extraordinary, but I find its often the small, ordinary things that make a difference. In this case, I’m hoping the ability to easily find a space to put their “stuff” will make operating on this layout just a little more enjoyable and stress free.