Tuesday, November 30, 2010

Layout Progress as of 11/30/2010

Over the long holiday weekend, I added the remaining grass, bushes and trees to the area around the South Emerson turnout on the layout. Although there are still some details to be added, this scene is pretty much complete.

» Photos

Looking north at the South Emerson turnout:

The next photo is the same area as above but looking in the opposite direction:


Thursday, November 25, 2010

Layout Progress as of 11/25/2010

Happy Thanksgiving to everyone! Aside from consuming mass quantities of turkey and all the trimmings, I did some additional scenery work around South Emerson on the layout. I am mostly adding vegetation behind the tracks to the north of the rock cut and in front of the tracks between the Allatoona Road grade crossing and the I-75 overpass.

» Photos

Looking north at the South Emerson turnout, I have added gravel in front of the tracks where the signal relay shed will be located. The line poles can still be found along most of the right-of-way despite the fact that they were taken out of service sometime in the early 1990s. They are a prominent feature above the rock cut on the prototype as they are on the layout:

The next photo is the same area as above but looking in the opposite direction (south). The newly added scenery includes the cluster of trees and grassy hillside to the left, bushes and vines, and the ballast around the signals:


Sunday, November 14, 2010

Layout Progress as of 11/14/2010

Recent progress on the layout includes adding additional layers of vegetation to existing scenery, adding new vegetation to bare scenes, and adding a few details here and there. The photo above shows a good example of what I have been working on recently. Some of the new "little touches" that really impact the realism of this scene include:

  • The electrical box that houses the circuitry for the grade crossing signals

  • Additional bushes and shrubs around the base of the hill

  • Small tufts of dry grass in a few places

I still have to decide what to do with that bare patch of grass in the foreground. On the prototype, there is dense brush with a side street angling away. While the dense brush would obscure most of the scene, I am still contemplating adding a driveway and a few mailboxes to represent the street that leads to a few homes.

The next photo shows the same scene looking north towards the recently added rock cut that I described in another post. I have included a prototype photo to show how the modeled scene compares to the real thing:

Some of the work I have done in this area includes:

  • Adding static grass to the hill above the rock cut

  • Adding bushes, weeds and vines between the tracks and the trees

  • Adding static grass down the center of the access road

  • Adding the CSX "No Trespassing" sign from a digital photo

  • Adding tress and bushes in front of the tracks

While this scene looks complete, if you look closely you can see bare ground extends past the grass covered areas. I still have lots of vegetation to add to this scene, plus I am working on a concept for adding high tension power lines that cross the tracks at the rock cut.


Monday, November 8, 2010

DIY: Hand Carved Rocks

At the south switch of the Emerson passing siding, the CSX mainline follows the side of a hill that includes a small rock cut that was carved many years ago in order to make room for the roadbed. Staying true to the prototype, I decided to model this rock cut. When I was shaping the basic terrain in this area, I allowed a bit of extra clearance to allow room for the rock cut. I initially planned on using plaster rocks cast in rubber molds, but could not find anything that was an appropriate fit. I then turned to one of Joe Fugate's scenery DVDs which includes a chapter on hand carving rocks. This video series has proved invaluable to me as I have created the layout, and the chapter on hand carved rocks was no different. I have never before done any kind of hand carved plaster work, but armed with the knowledge gleamed from the DVD, I decided to give it a try. The steps below show how I created my very first hand carved rock.

» Steps

Step 1 To get started, I scooped three globs of DAP pre-mixed patching plaster into a plastic cup and added a few shakes (about half a teaspoon) of black tempera powder paint to kill the bright white color. I then used a putty knife to smear a layer of the plaster mix onto the bare rock face. I first sprayed the surface with water so the old plaster would not rob the new plaster of its moisture. I also used a plastic painter's knife (shown) to carve some grooves into the wet plaster that simulate the layers of sedimentary rock:

Step 2 I used a metal painters tool (shown) to pick at the plaster as it setup, creating nicks, grooves and other random gouges in the rock face. If you begin carving before the plaster has setup enough, it will crumble rather than flake. If you get crumbling, stop and come back later to resume your carving. IMPORTANT: Patching plaster contains retarder to extend its working time. For my batch of plaster, it took 8-10 hours before the plaster setup to the point where it could be carved! It was then about another four hours before the center section had setup to the point where I could finish carving. In this photo, taken 12 hours after the plaster was applied to the layout, you can see an area that is still too soft to be carved (the smooth area at the lower left-center side of the rock):

Step 3 Finally, after about 16 hours, the carving is complete. Before the plaster is completely dry, lightly scrub it with a stiff plastic-bristle brush (shown in the previous photo) to give the rock a realistic weathered look. As I scrub, I use a soft paintbrush to clean any small particles off of the rock face so I can see how things are progressing:

Step 4 I begin staining the rock face by using a medium gray acrylic paint thinned to a wash. You will find that the plaster really soaks up the paint, so be sure to give it a few sprays of water every now and then to keep the paint flowing:

Step 5 After the initial gray wash has dried, apply a second dark black-brown wash. In the second photo, notice how this second wash settles into the grooves and cracks--for the first time, this rock is starting to look like the real thing:

Step 6 At this point, I began to blend the rock in with the surrounding scenery by applying my homemade dirt to the painted, textured scenery base. I also dry brushed on some varying shades of grays and browns to introduce some variations into the rock coloring:

Step 7 All done! Well, almost done, since there will be some additional weathering done with pastels as the surrounding scenery is added. Here is a close-up view of the (almost) finished rock and a few wider shots showing how the rock cut fits into the overall scene:


Sunday, October 31, 2010

Happy Halloween!


Friday, October 29, 2010

Layout Progress as of 10/29/2010

Trees, trees and more trees. This week I found out that you can crank out a lot of SuperTrees while watching Monday Night Football and the 2010 World Series. As a result, I added a significant section of forest (about 18" long) behind the tracks at South End Emerson. I also installed working grade crossing signals (above), which will be covered in detail in an upcoming HOWTO post. The photos below show an overview of the progress made this week.

» Photos




Tuesday, October 19, 2010

Layout Progress as of 10/19/2010

I recently finished some scenery work that I started way back in...July? Can that be right? Three months away from the layout? This was a busy summer. Anyway, while looking at some prototype photos of the Old Allatoona Road grade crossing, I noticed the road actually travels through a slight cut after crossing the tracks. Although the cut is low, it is quite noticeable (as you can see in the Google Maps imagery below) and modeling it would be a nice way to tie the layout to the real thing. The photos below show how I modeled this feature.

» Photos

I began building the hill behind the tracks by shaping some plain old green floral foam. I carved the hill to shape before affixing it to the layout using Loctite construction adhesive. I applied the hill right over the top of the existing scenery base, which had already been covered with dirt and some static grass:

I then covered the hill with Sculptamold to blend it in to the surrounding terrain. Since all of the hill except for the cut face along the road will be covered by foliage, I didn't spend too much time on this step:

Next, the hill was painted using my standard textured ground color:

Finally, the usual treatment of ground cover, bushes and trees were added. I rubbed most of the ground cover off of the cut face of the hill to reveal the red Georgia clay just as it appears on the prototype:

This small change resulted in a big improvement to the look of this scene.


Saturday, October 16, 2010

DCC: Block Detection

One of the primary design goals for the layout is to have a fully functional Computerized Traffic Control (CTC) system. Looking even farther into the future, some degree of layout automation is desired. To meet these objectives, a DCC system with fully functional block detection is a must. Since my DCC system is a Digitrax Super Chief, the Digitrax BDL168 was a natural choice for block detection. This board provides for detection of 16 individual blocks and reports block occupancy via messages over Loconet. A personal computer can then be connected to Loconet and receive these messages. With appropriate software, a complete operational CTC system is possible. My software of choice is JMRI, which I will describe below.

» How it works

The following diagram shows how the Digitrax BDL168 is connected to the layout:

The above diagram shows the connections for a typical passing siding track arrangement on the layout. Each turnout is a separate detection section, as is the mainline and the passing siding. Therefore, a passing siding on the layout will consume four of the 16 total detection sections on a BDL168. I anticipate the need for three BDL168 boards on the layout (one for each of the three levels). Only signaled track will be detected; non-signaled tracks such as spurs and industrial sidings will bypass the BDL168.

The following photo shows the first BDL168 installed on the layout:

Astute viewers may note that the BDL168 is connected directly to the DCS100 booster below, which differs from the diagram above. The diagram shows a Digitrax PM42 between the two, but since I only have about 30 feet of mainline in place, I have yet to justify the expense of a power management board.

Digitrax supplies a handy worksheet for you to record information on how your BDL168 boards have been configured. Here is one that I have filled out for the first block detection board installed on the Dixie Line:

In the above worksheet, the detection sections highlighted in yellow represent track that has already been installed on the layout. Other detection sections shown in the worksheet are planned for future use. Basically, the Emerson section of the layout is the proving grounds where I can work on ideas that will eventually be expanded to the entire layout.

I have set the address of this BDL168 to 101. By default, the board is shipped from the factory with an address of 1. I changed the address to a higher number so that there will be no conflicts with other devices such as turnouts that will be in the lower address range.


I have begun to use the Java Model Railroad Interface (JMRI) software to control the layout. The computer is connected to Digitrax Loconet via a LocoBuffer USB, which in turn lets JMRI talk to Loconet. JMRI receives messages from Loconet that tell the software what is happening on the layout, and JMRI sends messages to Loconet to make things happen on the layout such as changing a turnout, setting a signal, or controlling a train.

In JMRI, I use the PanelPro application to setup communication with LocoNet. In PanelPro, block detection is facilitated by using objects known as sensors. A sensor is active when a block is occupied and inactive when a block is clear. In JMRI, you can define a sensor for every detection section on all of your BDL168 boards. The following screen shot shows how I have setup the sensors for my first BDL168 in JMRI:

Ignore the first line in the table; that is some fast clock sensor that JMRI automatically created for some reason that I don't yet know. However, the remaining entries show the 16 sensors that correspond to the 16 detection sections on the BDL168 board that I described above.

Earlier I mentioned that I set the address of the BDL168 board to 101. The board address determines the address for the corresponding JMRI sensors. The formula used to calculate the address of a BDL168 sensor in JMRI is as follows:

(board address - 1) * 16 + detection section

JMRI also requires the sensor address to be prefixed with LS when using Digitrax Loconet, so the 16 detection sections on the BDL168 will be represented in JMRI by sensors LS1601 to LS1616. A second BDL168 with a board address of 102 would be represented in JMRI by sensors LS1617 to LS1632 and so on.

Notice that two of the sensors, LS1606 and LS1607, are shown as active in the above table because there are actually trains sitting on those tracks. Moving one of those trains to another track will cause the appropriate sensors to activate and deactivate accordingly. This is super cool stuff, albeit in a super nerdy way!

OK, so we have Digitrax DCC block detection installed on the layout, we have JMRI talking to Digitrax Loconet, and we have sensors setup in JMRI that are obviously working. This is all fine and dandy, but how do a bunch of invisible computer bits turning on an off as trains move around the tracks actually contribute to the operation of the layout? This is where JMRI Panel Pro really pays off: the ability to build fully functional panels that allow you to operate the layout. Here is the CSX Dixie Line panel so far:

The sensors in the above panel are represented by the red lights: illuminated lights show occupied blocks while dark lights show clear blocks. I added the sensors to the panel by selecting the appropriate entries from the sensor table in the panel editor window. The sensors on the mainline above from left to right are LS1609, LS1608, LS1607 (the siding/lower track), LS1606 (the mainline/upper track), LS1605, and LS1604. Notice that the sensors LS1606 and LS1607 are illuminated since there are trains sitting on both of those tracks.

The graphics are based on the old USS CTC panel designs; although they are quite retro looking and I model the modern era, I actually like they way they look and am sticking with them for now. JMRI provides a graphics library to use when creating panels, so it should be fairly easy to switch to different graphics if I ever desire to do so. This particular panel so far looks very simple since I only have about 30 feet of mainline and one passing siding installed. However, the small size of the panel makes it easy to use and is an ideal learning tool. Plus, there is actually a lot of stuff on this panel, including two turnouts (that I want to control remotely via the panel) and six blocks. A lot of stuff to play around with but still small enough not to overwhelm a novice JMRI user such as myself.

Eventually, this panel will evolve into a fully functional CTC panel. For now, I can run trains and watch the sensor lights turn on and off. Although it seems quite simple on its own, this block detection implementation is actually the foundation that will be used for all future CTC and layout automation.


Thursday, July 8, 2010

Layout Progress as of 7/8/2010

Scenery work continues at the Old Allatoona Road grade crossing at the southern end of the layout. With the wooded area behind the tracks mostly completed, I turned my attention to the hill between the tracks and the fascia. I added static grass, bushes and weeds, then trees. The photos below show the scenery progress in this area.

» Photos

Static grass has been applied along the road and the tracks (short 2mm fibers along the road and medium 4mm fibers along the tracks):

Next, bushes, weeds, vines and other undergrowth was added to between the grassy area and the wooded area:

These prototype photos show the actual scene being modeled. Note the crawling vines that have crept down the hillside and up to the ballast:

Finally, a handful of Scenic Express SuperTrees finishes off the scene. From normal viewing angles, these trees mostly obscure the opening in the backdrop:

One day that family in the red Explorer may actually make it across the tracks!


Saturday, July 3, 2010

Layout Progress as of 7/3/2010

This week I began adding vegetation to the Old Allatoona Road grade crossing scene at the southern end of the layout. I applied static grass fibers in varying lengths using a homemade static grass applicator. I then added trees to build up the woods behind the tracks. Finally, bushes and saplings were added to smooth the transition from grass to trees. There is still more vegetation to be added to this area (a few smaller foreground trees, in particular), but the overall scene is now starting to nicely take shape.

» Photos

In this photo, you can see where static grass has been added to the scene along the road and the railroad ROW. I used short (2mm) fibers along the road and medium (4mm) fibers along the tracks. I chose the autumn grass color because it is almost a perfect match to the prototype photos I took of this area in June 2007:

The static grass fibers look absolutely terrific when applied to the layout. With the texture and depth they provide, the fibers really do look just like tall grass growing on the layout. However, my point-and-shoot digital camera does not do justice when photographing this effect. The following photo is about the best I can do to convey the height and depth of the static grass--look at the far end of the scene where the track passes through the opening and you can clearly see how the fibers stand up in contrast to the dark green background trees painted on the backdrop:

Finally, a quick look at the trees that the family created prior to planting on the layout. These trees are just SuperTrees plant material painted gray, dipped in diluted matte medium and covered with Woodland Scenics coarse foam: