Designing a Layout - Determining the Footprint

With a subject identified, my next step is to consider the footprint available for the layout. That footprint is related to the desired architecture of the layout. In this case, I have selected that as well. In my earlier plan, I began to play with a concept I call a double helix layout, which is two nested helices, one climbing up and the other climbing back down. One or both could in theory be sceniced, but it allows the layout to climb to a summit and then descend the other side. For this layout, I plan on scenicing one helix, while hiding the other one, in large part due to the available space for the layout.

The room available for the layout is 11' wide, and 27'-9.5" long. The lower side of the room (in the drawing above is a built in bookcase that separates the room from an adjacent family room/crew lounge, which is not available for the layout. The upper wall is currently occupied by a set of 3 double door closets, and a built in sink. Within the upper right hand closet is an entrance to a crawl space which has to be maintained. The right wall also has the intake for the basement AC/heating system. The left wall has a door which exits into the garage.  Located on the opposite side of the wall in the garage is a utility sink and a washing machine, so that left wall is full of plumbing.

The closets and sink can be relocated - however, the closets represent a significant amount of storage for the house. However, at 32" deep, they could be reduced in depth if not eliminated. Similarly, the sink could be relocated, and consolidated with the sink in the garage - which I would also like to get out of the garage, along with the washing machine (if it cannot be eliminated altogether as it is not the primary washer in the house). Since the left wall contains plumbing, it makes sense to locate these elements on the left wall. Doing this, and replacing the closets with 16" depth shelves, results in the room plan below.

I've also extended the shelves over the new sink and washing machine. Allocating between 30" and 36" for aisles around the layout produces a possible footprint that is about 20'-9" long and between 44" and 56" wide.

My next concern is to determine how the hidden helix will affect the necessary width of the layout. Because the down helix will be a substantial run and will need to handle bidirectional traffic, it will need to be double tracked. In addition, there will need to be a third track in the helix, at least from the lowest staging level to the bottom of the visible portion of the layout. So, the helix is at least three tracks wide. Since two tracks have to run from the top of the helix to the bottom, those two will need to be the innermost tracks. Based on the Model Railroader Canadian Canyons layout, which used an N-scale Helix with a radii of 16.5" and 18.5", I plan to use an inner radius of 17" and to make this track the default downhill track. Based on my experience in Ntrak, I plan to use 1.5" track spacing in the helix, thus, this makes the radii of the helix 17" - 18.5" - 20". I'll add 2" to the inside and outside openings in the helix, so the inner access space of the helix will be a 15" radius curve, and the outer edge of the helix at a 22" radius. This makes the helix 44" wide, which with the scene wrapping around the outside of the helix, would imply a scene of no more than 6" deep and probably less. Perhaps the aisles should be narrowed a little more around this blob so as to allow for a slightly deeper scene?

Of course, another alternative is to locate the top-to-bottom helix at one end of the layout, and the shorter helix between staging and the lower end at the other end. This would mean that the visible portion of the layout would have an odd number of long runs. In my prior plan, I had six long runs, which equated to the following scenes:

• Minturn;
• Lower Eagle Canyon;
• Belden-Red Cliff;
• Camp Hale;
• Dean-Mitchell-West Tennessee Pass; and
• East Tennessee Pass.

Having an extra side is interesting, as it would allow me to insert a bit more of a run between Red Cliff and Camp Hale, as well as spacing Dean and Mitchell a bit further apart. But would that also work vertically? More specifically, can I make the grades work with that concept? I see a couple of potential advantages. First, the helix component drops from 44" side to 41" wide, and I could have 7.5" deep scenes around the helix without impinging on the aisles at all. Second, the length of the layout could be used to help make the entrance to the staging yard from the lower end of the layout be more visible by reducing the run in the helix. Third, as mentioned before - spacing is key. The railroad disappears between Red Cliff and Camp Hale and there was time between a train passing at Dean to get to Mitchell, so it would be nice to space those scenes more appropriately. So, at this point, I want to look at this at least as a promising concept.with this concept, these would be the following scenes for the long sides of the layout:

• Minturn;
• Lower Eagle Canyon;
• Belden-Red Cliff;
• Silver Gulch;
• Camp Hale-Dean;
• Mitchell-West Tennessee Pass; and
• East Tennessee Pass.

Just as a thought, if I could get an eighth side added, I would actually add a bit of the Avon/Eagle area in to the west of Minturn to add some additional switching, but because I would like to have Minturn be fairly level, I do not see exactly how I would do that, but I can keep that in mind. The tough part of this plan is that it places East Tennessee Pass above Pando siding (at Camp Hale), Camp Hale above Belden and Red Cliff and those above Minturn, all on one side of the layout. On the other side of the layout is the Mitchell-West Tennessee Pass above Silver Gulch which is above the Lower Eagle Canyon. All of which are scenery elements. Thus all the operational elements are on one side of the layout and all of the spacing elements are on the other side. This is not going to be good for the traffic of the operators, who are all going to be busy adding helpers in Minturn, switching at Belden and Red Cliff, passing in Pando, and taking helpers off at East Tennessee Pass. No, that idea is not a good one.

So, Silver Gulch needs to be dropped off as a full side. Instead, what if I placed Avon/Eagle in to the West of Minturn. This would lead to the following scenes:

• Avon/Eagle;
• Minturn;
• Lower Eagle Canyon;
• Belden-Red Cliff;
• Camp Hale-Dean;
• Mitchell-West Tennessee Pass; and
• East Tennessee Pass.

This works much better I think. While Belden-Red Cliff are still located over Minturn, they are at least located more towards the ends of the layout, and should be able to work with the Minturn yard master to share space. Avon/Eagle, Pando Siding (again at Camp Hale) and East Tennessee Pass are all above each other, but it is hard to imaging that all three will be busy at once, so this is at least a better balance of aisle traffic. So, that is the plan at the moment - at least assuming that I do want to separate the helices. And with that in mind, that led me to develop an initial proposed layout boundary.

In defining the shape, I did narrow the aisle on the left to 27", but it immediately gets wider past the corner of the sink, all other aisles only neck down to 30" right now. This means that I have some narrow scenes, at 7.5" depth, but those can likely be deepened by narrowing the aisles further for short stretches. I currently have the top-to-bottom helix on the right side, and as the layout progresses upward in a clockwise fashion, this places East Tennessee pass on the upper side of the layout. This too works, as any expansion would likely extend over onto the shelves (first) and would create a walk under area in the upper right hand corner. Minturn Yard is thus on the lower side of the layout, and I widened the aisle and benchwork to accommodate yard operations. The lower helix on the left would climb clockwise, while the top-to-bottom helix could climb in either direction. That will need to be determined. But, that is my thought process, at least to this point. 

Designing a Layout

Model Railroader recently started a new video series where they are walking you through the process of designing a layout. I watched it, and enjoyed it. So, I thought since I am starting a new plan, I might blog my own version of the same. Well, it isn't a new plan, but a new version of an older idea, which I previously put in the blog, here.

One of the comments on this plan resonated greatly with me, as I had adopted a particular perspective on the layout based on the view from Highway 24 in Minturn. Unfortunately, this view puts several of the later scenes in perspectives that are not those typically seen by railfans, and that always bothered me. The suggestion was simply to flip the view in Minturn to look at the yard with the town in the background rather than the foreground.

As part of the planning for this effort, I went back to Google Maps and captured a set of images of the relevant areas oriented in the viewing perspective. So, here is the view at Minturn.

So, in this configuration an eastbound train moves from right to left through the scene. There are obviously a few nice advantages already apparent. The wye would extend into the backdrop, the town would be in the background and on the backdrop rather than lost off of the front of the layout.

Further to the east (geographic south) of Minturn, we reach the point where Highway 24 will cross over the railroad and will climb off the front of the layout.

Continuing eastward, we quickly arrive at the area around Belden. Here, we see a very different perspective than what I have normally scene modeled. The mine here is now in the foreground, and the tunnels are in the back of the scene. Most models I have scene have the mine in the background. However, here we get a really nice effect, as the mine is located in a turnback curve that will look right at the end of the layout. Previously, I had to soften this curve, and sacrifice much of the bend to maintain the geometry of the layout. Further, the mine will now be closer to the front, so the NJ Zinc Mine will need to be modeled with increased fidelity compared to a model in the background.

Redcliff now also shows the same sort of geometry, and can be at the other end of the layout where I need to turn the corner again. Much of the town is now in the foreground, but there are a few significant buildings that will be modeled in this arrangement.

The next view is around Silver Gulch, as the line emerges from the Eagle River Canyon. Notable here is that while the Eagle River is in the foreground, behind the track is a small pond whihc now can easily be modeled.

The next two images are of the Camp Hale area. Highway 24 will come from the background, cross over the railroad on an overpass, and the remains of Camp Hale become a foreground scene.

East of Camp Hale, we encounter Yoder Gulch, and the railroad climbs onto the hillside. Again, the perspective works very well.

As the railroad continues east, it clings to the hillside and passes through Dean Tunnel at Pando. Again, Highway 24 is in the foreground, so the perspective of the railfan is preserved. 

Between Pando and the next major element is Bennet Gulch, which is just a more nondescript area where the highway and the railroad close the elevation difference.

We finally arrive now at Mitchell, where the railroad makes a series of sweeping curves across the valley to gain elevation. This element is now true to the classic railfan view.

There is not a lot left to see before we reach the summit. After Mitchell, the line and the highway climb with the valley between them. The key feature here is that the valley rapidly narrows as we approach the tunnel.

The western portal of Tennessee Pass is at the end of the valley. on a sweeping curve that would lead into another end of the layout. 

I plan to model the highway climbing over the pass, while the railroad goes underneath, so this switchback will fit into the end of the layout and with a little more bending, will bring the viewer to the other side.

On the eastern side of the pass, the East Portal of Tennessee Pass exits into the Tennessee Pass siding. This is the farthest east siding I think I will model because I want to model the helpers cutting off. One challenge that now comes to play is that the wye would exit onto the front of the layout, and the wye is on the west end of the siding. I may model it with the wye to the eastern end of the siding so as to help the transition of the mainline into the helix which will need to rotate in the opposite direction as to the direction of the climb. I did just this on the prior layout. I also took the wye, and used the tail track to connect to Leadville. While not quite correct, it allowed me to give the Leadville/Malta turn somewhere to go. It actually is closer to the configuration when there was a connection to Leadville west of Malta, but this is a minor compromise to move this connection from Malta to the siding at Tennessee Pass. The operational benefits are just too good. 

So, all in all, it looks like this configuration will work quite well. I've taken these captures and built a little guide file so that as I work on the different areas I have some references to study for the design of each scene. So, that is Google Maps as a tool to conceptualize the design of a layout area.

One of my other concerns from the previous plan was that the space was a little tight to make everything work. But I have been considering that issue as well. My next post will focus on the plan to resolve those issues, and will define the footprint of the layout plan. 

Update: Just to maintain a little continuity here, I am still working on the T Layout, the HOn2Z cars, and the FMC4200 Hopper prototypes. I haven't forgotten those projects, and will post updates periodically in between layout planning updates.

A Sidebar Project

I have some side interests. And sometimes they take my attention away. I had been planning to finish the bridges for the T Layout, but somewhere I misplaced a set of 3D printed detail parts. So, just order replacements right? Well, I decided that I didn't want to do that until I had a few other files ready for printing, and these files involved cars for an HOn2Z Gilpin Gold Tram layout I have been toying with as a short term project. The key to the layout, and whether or not I pursue it though is the ability to make the cars/engines the way I want to, namely, not in HOn30, but in what I have taken to calling HOn2Z.

Yep, rather than use N-gauge track which is really about 30" in HO, I decided that I really liked using Z-gauge track. A quick calculation showed that the 6.5 mm gauge is a little less than true HOn2 (which would be 7 mm gauge). I was even a little surprised to find out that there is an HOn2 set of NMRA standrads for 7 mm gauge, so this is not going to be HOn2. In fact, technically, it is 22.3" gauge, so really, it might be HOn22, or as I coined it HOn2Z.

By using Z, I can use existing track if I want, particularly turnouts like the new Atlas Z scale turnouts. Plus, I already have done a lot in Nn3 which also uses a 6.5 mm gauge, so it makes sense. Add in wheels and trucks from Microtrains, and a bunch of mechanisms - well, I like HOn2Z. So, now for the cars.

Well, I got a bunch drawn up for printing to pass the Shapeways requirements. The Gilpin has 3 ore car types as shown below.

Gilpin Half Cord Ore Car (Type 1)

Gilpin One Cord Ore Car (Type 2)

Gilpin One Cord Ore Car (Type 3)

I designed them around Andrews Style Z-scale trucks from Microtrains. The wheels are close, but a little small for most of the Gilpin cars, although the truck style is wrong. I decided to live with the wheels so that the cars operate well, and have a consistent coupling height (Gilpin cars actually had different coupler heights over the years). With a little work, MT couplers could also be installed, but I went with a Link-Pin Style coupler that I designed and had previously tested. I figure if you are going this far, might as well do that well too. Seeing these three car types together really shows the differences.

Left to Right: Gilpin Half Cord Ore Car (Type 1), Gilpin One Cord Ore Car (Type 2), Gilpin One Cord Ore Car (Type 3)

I've also drawn up a set of flat cars. I have drawings for the flat car, and the flat car with stake pockets, but the flat car with stakes is presumed. After all, if you had stake pockets, why not stakes too.

Of course, the stake pockets were likely used to make the gondolas, just as I did in the model.

And, they were also used on the excursion cars.

And, while the Gilpin didn't have any, it is fairly easy to imagine a box car as a combination of a flat car with stakes, and a similar (albeit slightly longer) box car found on some of the Maine 2-foot roads.

In addition, another set of modifications led to the longest car on the Gilpin, at 23 feet, the water car. A few elements of this are plausibly imagined, as the data and pictures simply don't provide answers, but I think it will look reasonable.

Test prints of each are on their way, and I have a few more cars to finish, before tackling the engines. But with some cars coming, track was next. I did find some data about the ties used, and from that, designed a track jig. And using a home grade 3D printer, proceeded to make some tie jigs. 

Each section has a pinned connection to align them. And once they are assembled, I am planning to mount them on a painted poplar board from the local lumber yard. 

I am currently painting the tie gigs to help locate the PC clad tie locations, but once that is done, I will have a tool to lay ties for a couple of feet of track to display my cars, and to include a prototype handlaid stub turnout (just like the real ones used on the Gilpin). So, I may not need the Atlas turnouts, or I may simply use stub turnouts in certain places, and model the line as slightly more modern and upgraded than the actual. If I do build a little layout, it will almost certainly be a protofreelance line, rather than a Gilpin model. After all, I am already going there with the box car. But more on that in a later post.

A New Bridge

It has been a while since I have posted, but there has been some slow progress made. Previously, I had shown the pedestrian bridge for the town. I finally had cause to order some parts from Shapeways, and I ordered the bridge too.

On the left is the FUD version, on the right is the professional plastic version. This is a material produced by HP for 3D printing. It is fairly robust, but not quite as fine as the FUD material. The grains you see are not the photograph, but the material. So, for T, not quite good enough for me, but for other scales....

Anyways, the bridge needed a bit of painting to give it a nice brick color, and then some concrete washes to bring out the individual bricks, and the concrete portions of the bridge. I tried several approaches, and actually found that burnishing the paint on with the tip of a toothpick seemed to work best.

One more feature of the bridge is the installation of a mirror under the bridge. In this case, I found a mirrored plastic panel on Amazon.com. This can be easily cut with normal scissors. For a backdrop mirror, I think it works well enough.

You can see the bottom seam of the mirror, but in this case, I have simply set the bridge of a paper printout of the road. I think when I install the real thing, and have the road butting into the mirror, as well as the sidewalks on the sides to add a little 3D definition, and as a result, I expect that the seem will be much more difficult to see.

Against the backdrop, this will be a nice little trick to add depth into the backdrop. Just compare this picture, to the top view of the bridge.

I think it works very well.

Detailed Trees in T

Yes, I may be crazy. I've been thinking a lot about trees for the little layout, As part of that, I have been looking at different ways of making trees and seeing how well they work in T. Most recently, I have been working on some deciduous trees that would stand up to foreground scrutiny. The basic idea is to use stranded wire, in this case 18 gauge speaker wire as the basis for the trunk and branches. In my case, I picked up a roll of 18 gauge stranded wire which has 10 strands of 28 gauge wire. Lengths of this wire, with the insulation removed were cut and used to form tree-like structures.

The process was simple enough. I took a stiff piece of flower wire, about 3/4" long, and inserted into the stranded wire for the trunk. The flower wire extends past the bottom of the trunk to provide an anchor into the scenery. The speaker wire is twisted, and as you go up the tree, you can form smaller limbs and branches from smaller groups of wire twisted together, until yo get to individual strands. There is no particular pattern, just trying to get something that is tree like. A few drops of super glue, strategically located will hold everything together. A few branches may also need to be clipped to length as the tree takes its final form.

Some of the trees use one length of wire strands, others use two or more bunches. The lengths do not need to be identical, but probably need to be close to look right.

After the super glue has set, I use a little bit of caulk adhesive for projects to fill in the trunk and eliminate the stranded look. The key is that whatever material is used, should be paint friendly. At this point, you have a bunch of shiny metal trees. Next up is to paint the trunks.

I first painted the trees with a brownish color, and then followed up with a dark grey. The exact colors don't matter, as long as the look like trees. Most trees have much more of a greyish he to the trunk, but most also have some dark brown or black features. The key with the second grey coat is that this coat focuses more on the trunk, and does not need to cover everything. I follow those paint coats up with a little bit of another grey acrylic paint, brushed onto the trunk and major branches. I used a lighter grey paint, but again, the exact color is up to you, and again, the goal is not full coverage, just adding some color.

This is the result. The trees have some texture due to the colors used int he paint. For a dead tree, or a leafless tree, this is a decent model. Most of the first set of trees I made are in the 25-50 foot range in T. That is about 5/8" tall to 1.25" tall. This approach is much easier in larger scales.

The next step is to add some foliage. I used some polyfil fiber that I had previously painted a dark green color. Darker green is better here than lighter green, as we will add more texture and color. The key is not to make a big piece of polyfil and wrap it all over the tree, but to make a bunch of little balls, that tie into the ends of one small group of branches. Again, a little super glue to help things stay in place is useful. Note also, that sometimes, you can get a spread of roots at the base of the tree from the twisted wire trunk.

Next up, I spray the tree with adhesive spray glue, and sprinkle ground foam on the tree. Try to avoid glue or ground foam on the trunk. Fortunately in this case, the tool used to hold the tree will cover the trunk. Also, make sure to sprinkle ground foam on not only the top, sides, but also the bottom of the tree to get the undersides of the branches. Allow the tree to dry and shake the tree to remove any loose ground foam.

The neat thing about the finished trees is that in the right light and angle, you can actually see through the tree a little bit in places. But at a distance, and not backlit, the trees are solid, just like the real thing.

At this point, a few errant strands of polyfil may need to be trimmed away with scissors, but we are almost done. The last step is to spray just a little yellow spray paint over the top of the tree. The goal is to slightly lighten the upper leaves to achieve a highlighting effect. Generally though, the goal is not to turn the tree yellow. It is easy to overdue the effect. If that happens, a little ground foam on the wet yellow paint solves the problem.

Two nearly finished trees. I need to prune one or two errant fibers still, but these trees are almost ready for planting on the module. Overall, I did a dozen trees in a week. The first night was about two hours of work getting to wire trees, and the remaining nights were 10-30 minutes. The slow part is allowing things to dry between steps. If done in an assembly line fashion, I could probably make 60-70 trees a week this way, with an investment of about 15 hours of work. A little slow, but the results are fantastic.