Thursday, December 20, 2007

Train operations: puzzle or spectacle?

I'm currently designing a new layout, and it's got me thinking about how we approach operations. Most of us fall on a continuum somewhere between the "puzzle" operator and the "spectacle" operator.

The "puzzle" operator is interested in realism. He/she likes waybills and timetables, and prefers trains to operate close to prototype, paying rigorous attention to the questions of trains' purposes on the layout. This type of operator will be attracted to big yards and/or lots of industries.

On the other hand, the "spectacle" operator simply likes to watch the trains run. He/she is probably content to spend a lot of time just running their consists around the mainline, with occasional stops to drill an industry. Good scenery and good mainline runs are the order of the day for this type of operator, who can probably get away with smaller yards and fewer industry spurs (and also, possibly, an overall smaller layout size).

If you're in the process of designing a layout, it behooves you to ask yourself what kind of operator you are, and what you enjoy most about running trains.

Friday, December 14, 2007

REVIEW: Trainplayer/Tracklayer 3.0 for Windows and Mac

Trainplayer is a program that's been around for a while, and I'm admittedly late to the party, but the good thing about coming late is that things tend to be in full swing by the time you arrive.

Unlike 3D "virtual model railroading" software such as Microsoft Train Simulator or Auran's Trainz Railroad Simulator, Trainplayer is a decidedly 2D proposition. Specifically, the program starts with a bitmap image, which can be a scan of a drawn or printed trackplan, or output from a CAD or graphics program. Overlayed invisibly on the image is vector graphics information that defines the track plan for the software. This allows you to then drive one or more trains on top of the track plan, throw the switches, couple and uncouple cars, etc. Frankly, the software is an "armchair" model railroader's wet dream.

You might think this approach to computer railroading is less sophisticated than the aforementioned 3D programs. I suppose it is. It's also a hell of a lot of fun, and lately my 3D simulator programs have been gathering dust in favor of this one.

Some neat features have been incorporated into the recent version 3.0, including the ability to run trains automatically, either by recording your moves, or by creating text scripts using a rudimentary programming language. The latter, in my opinion, expands the utility of this software greatly, allowing one to grow from the role of engineer to that of dispatcher.

By itself, Trainplayer allows you to do everything except import your own graphics and create your own track plans. For this, you'll need the more expensive Tracklayer, which incorporates Trainplayer.

The basic package (Trainplayer or Tracklayer) includes ten layouts taken from the classic book "101 Track Plans for Model Railroaders" by Linn Westcott. For more money, you can expand the number of layouts incrementally, going up to all 101 plans. There are also a wide variety of user plans that can be freely downloaded and installed from the Trainplayer Web site.

In sum, if you're particularly interested in operations, or if you're in the planning stage for a real model layout, or if you just like playing with trains, I can't recommend this program highly enough. You'll be hooked.

Tuesday, December 11, 2007

Save space by rethinking operations

When I design and build layouts, I'm disposed toward the small end of the spectrum, and I think a lot about track plans. Operations dictate a big part of what should be included in a plan.

Imagine a simple oval layout; perhaps there are one or two passing sidings, and one or two industry sidings. We also have two sidings representing towns, each with a station/depot and/or a freight house/goods shed. These are the line's two terminii.

A train departs a terminus with the loco at the head end, pointing forward. It travels the loop a few times, then is ready to pull in to the second terminus. In real life, both terminii would include at least a runaround track so the loco could be moved to the other end when going back, and ideally also a turntable or wye so the loco could be facing forward on the return trip.

Runarounds, turntables, and wyes all take up space. They're nice, and they're prototypical, but what if you're building a small layout where space is at a premium, and you don't want to use them? I have three thoughts:

1.) The first is kind of obvious, which is to simply orient the points for both terminii as trailing-point (you can do the same for any industry sidings). This means trains will always travel in the same direction, and will always back into sidings. Simple, but highly unprototypical, and probably a bit boring after a while.

2.) The two terminii points face in different directions, as do the industry sidings. In this case, think about placing your two terminii at or very close to the front of the layout, and think of each as not only a terminus but as visible staging. If you have a kneejerk negative reaction to this, just consider the possiblities. Your train leaves one terminus, switches the trailing-point industries en route, and arrives at a facing-point second terminus. At this point, you can do the runaround and turnaround with your hands, or put a completely new loco on the end, or replace most of the train. It heads back, switching the other industries which are now trailing-point, and arrives home. Repeat ad infinitum.

3.) Include two passing sidings in the oval, and make these your terminii. You can use the passing sidings themselves as runarounds.

Monday, December 10, 2007

A cheap wood frame for us non-carpenter types

I wrote earlier about using pre-stretched art canvas from an art supply store as a baseboard. What if you want to take a frame-and-profile, or "cookie cutter," approach to constructing your base and terrain, and if, like me, you don't like cutting and assembling wood?

How about starting with that same sturdy pre-stretched art canvas, and just unstapling and removing the canvas? Instant wood frame. To this you can hot-glue foam board or styrofoam profiles for the front, back, and sides, as well as for your track bed and any flat areas. Then just add your chicken wire or cardboard webbing, plaster cloth or glueshell, etc.

Save the canvas of course, you never know how you may be able to use it later!

Saturday, December 8, 2007

A different approach to lighting plastic structures internally

Anyone who has attempted to install internal lighting in plastic structures has faced two headaches: the incandescent lamps normally used are bright enough that the structure's walls tend to "glow," and the light also throws into relief the fact that there is nothing inside the building (obviously the latter point is moot for those very few of us who super-detail their structures by placing items inside).

The usual solution for the former is to paint the inside of the structure black. This presents new problems, especially if you're dealing with an already built-up structure, because the edges of windows and doors need to be unpainted so they can be glued to the inside of the walls. However, leaving them unpainted will cause "glow" light around the edges. It's a Catch-22.

My solution is to replace the incandescent light with yellow LEDs right behind each window or door which is to be lit. This looks great in the dark, and the LEDs are dim enough that they won't make the walls glow.

First, determine which windows and doors will always be dark, and attach black construction paper behind the glazing and any other details such as curtains. Easy.

Second, for windows/doors where you want nighttime light to appear, attach regular white paper behind the glazing. Now, using regular 5mm yellow LEDs (not the super-bright ones), place one LED behind the white paper on each window or door, sitting 3mm or so back from the paper. Wire all of them in parallel, making sure the anodes and cathodes are all going the same way (i.e. they should all be wired with the same polarity). To this parallel LED cluster, attach one 1K-ohm resistor in series (it doesn't matter whether this is attached to the anode or cathode side). You can try reducing the resistor's value if the LEDs appear too dim, but don't go below 680 ohms.

The last step is to put a full-wave bridge rectifier between your accessory power (I'm assuming you'll be using 16-20V accessory power to light the structures) and the LEDs with resistor. Make sure you match the + output of the rectifier to the anode side, and the - to the cathode side, with your resistor in series on one side. If you omit this step, the LEDs will only be using half the power cycle since they only pass current in one direction, and they will flicker and appear dim. Since each structure will only be drawing around 15ma or so, one 1-amp rectifier should be good for quite a few structures.

Friday, December 7, 2007

My favorite dirt

When creating ground cover, a lot of people like to use real sand, baseball diamond sand, vermiculite, or other natural sources.

My preference is to use the artificially colored sand that is sold at craft stores, and typically used for sand art, sand in bottles, candles, etc. The bright primary colors won't do you much good, but if you stock up on white, gray, a couple of shades of brown, and maybe a dark red, you can blend 2 or 3 together to get some quite nice shades for ground cover.

It looks even better if you mix in some fine green ground foam, and maybe a little fine ballast or talus to simulate the occasional rock.

Thursday, December 6, 2007

Uses for microcontrollers in model railroad electronics

Microcontrollers are fascinating little devices that have multiple uses for model railroads. I've used them several times -- perhaps in future posts I will drill down to specific how-to's, but for now I'd like to just present a few ideas in hopes of stimulating the creative juices.

First, if you've never used a microcontroller, know that they are not nearly as daunting as you might think. A basic grounding in electronic theory is helpful, but even without this you can buy educational packages that include both the devices and learning tools. In fact, the package I would recommend for most beginners is based on a microcontroller system called the BASIC Stamp, and is available at Radio Shack for $80. This package includes a great how-to book, programming software, and lots of extra electronic parts you can use. (A word about the BASIC Stamp system: it rates low in terms of cost-effectiveness, but high in terms of learning and ease of programming.)

On to the applications:


Custom throttles

Microcontrollers excel at timed, switched, or pulsed electrical signals, so they are a natural bridge between a knob and a sophisticated pulsed-modulated drive signal for locomotives. Or, forget about one knob, and use some combination of switches, sensors, or other electronic components to control the movement of trains.


Animated/timed/sequenced lights or LEDs

A flashing "neon" sign, "crawling" light patterns you see around marquees, arc welders, campfires, train signals, traffic signals, etc.


Block detection

Connect some basic block detection circuitry to a microcontroller, and now the position (and, in DC systems, direction) of your train on the tracks can control signals, crossing gates, or ...


Automatic turnouts

This is not for everyone, but the first layout I built used a microcontroller between block detection circuitry and the magnetic coils on remote turnouts. An example: imagine the mainline approach to a turnout is A, the mainline route after the turnout is B, and the diverging route is C. If a train goes from A into the turnout without stopping, the system makes sure it will continue on to B. However, if the train comes to a stop in A but resumes movement in less than 3 seconds, the turnout is thrown and the train will proceed to C.

In DC (not DCC) systems, you can also differentiate based on the train's direction of travel, or use this in tandem with the timing of starts and stops to intelligently control the turnouts.

You can begin to see the possibilities here; the microcontrolled turnouts can replace a big part of a control panel. Of course, there are limitations to this approach depending on the degree of prototypical operation you wish to retain, but it's intriguing to consider the possibilities.


Automatic uncoupling

Similar to the above, but substitute the coil of an electromagnetic uncoupling magnet for the turnout coils, and now your train motion (again, based on train position and/or direction) can be used to trigger uncoupling events.


Automatic nighttime

Use the microcontroller with a photoresistor (or several of them) to turn on the after-dark lights on your layout, based on the ambient level of light in the room (i.e. whether or not the room lights are turned on), or based on a clock.


Complex animation

Combine your microcontroller with a servo motor, and the animated elements on your layout are no longer limited to simple circular or back-and-forth motion. The BASIC Stamp package described earlier includes one servo motor that can turn 180 degrees end to end; there are also motors available that can turn continuously.

Monday, December 3, 2007

Build or buy a nifty PWM hand held throttle

PWM control is great on DC layouts, particularly for yards and switching districts. It enables truly slow scale speed operation without hiccups or "instant" starts and stops.

This fellow sells a neat little hand held throttle that generates a PWM signal and gets its power for the AC outputs on your transformer. He also shows the complete schematic if you'd like to build it yourself.

Magnetic uncoupling made cheap and invisible

Years ago someone showed me a neat and inexpensive way to plant invisible but highly effective uncoupling magnets under track. I have used this many times in N scale, but HO should work similarly.

The ends of Atlas sectional track have a little recess under each rail; when you connect two sections, these form a little rectangle, which in N scale happens to be the perfect size to accept a 1/8" diameter rare earth magnet on each side. These are sold at Radio Shack in packs of 2; bulk quantities and other sizes can be ordered here. (Again, I haven't tried this in HO or other scales but I'm guessing the 1/4" diameter magnet will work for HO, not sure about the proper thickness though).

Once they're in place, you can move your couplers directly over this track section, and voila, the knuckes are pulled apart. And, no unslightly bar magnet in the middle of your track.

Bear in mind that in order for this to work properly, the couplers must be very close to where the magnets are, and both couplers must be straight. This means that if either or both sides will be a locomotive (or a car with body-mounted couplers), then the section of track on that side of the magnets needs to be a straight section. Cars with truck-mounted couplers are more tolerant if the section is a gradual curve.

Try it, you'll like it!

Sunday, December 2, 2007

Two solutions for lightweight baseboards

I recently constructed two small layouts using two different alternative baseboards, with good results.

The first was a 2' X 3' N scale layout built on a pre-stretched art canvas, obtainable in art supply stores. Overall I was happy with the results; the canvas is lightweight and sturdy, and takes paint and glue really well. Planting trees posed a problem which I overcame by gluing some styrofoam pieces to the reverse side, providing a thicker base for the tree stems. It was also pointed out to me that over time, if the layout is hung on a wall for storage (which seems kind of natural), the canvas may stretch. I populated my layout with very lightweight scenery and structures, but I suppose time will tell.

More recently I committed heresy by building a shelf switcher using the type of styrofoam you're not supposed to use, namely the expanded/beaded white styrofoam (as opposed to the blue or pink extruded kind). The entire 1' X 3' base is two inches thick; I obtained it at a craft supply store. My secret weapon was to coat the top, bottom, and side surfaces with lightweight spackle from the hardware store. After it dried I sanded it flat. This filled in the bumpy surface of the styrofoam and added extra strength. It's also a cinch to cut grooves for running wire, you just lay in the wire then cover with more spackle. This type of baseboard has the advantage of being extremely lightweight; if you're building a small layout that may need to be shipped, I think this is a great solution.

Applying the first layer of ground cover: problems with pock marks

Your terrain is in place, you have your diluted glue, matte medium, or scenic cement (hereafter referred to as “bonding agent”) and you’re ready to apply your first layer of dirt, sand, or ground foam. You sprinkle on some ground cover, then use an eyedropper or spritzer to soak with the bonding agent. You come back a little later, after it’s started to dry, and lo and behold, your soil or grass looks like the surface of the moon.

When the ground cover is dry, it is very sensitive to movement, and the drops from the spritzer or eyedropper create pock marks. Making the bonding agent “wetter” using a drop of dish detergent or a little rubbing alcohol helps to even the flow, but doesn’t do much for the pock marks.

The classic “solution” for this is to forego the bonding agent for the first layer, and instead to paint the terrain first with a mixture of earth-colored latex or acrylic paint and undiluted glue or matte medium, then sprinkle the ground cover over this while it’s still wet. The problem is, even after this dries, not all of the ground cover will have been “gripped” by the paint/glue mixture, leaving you with loose “topsoil” that still needs to be soaked with bonding agent, which can still leave pock marks.

The way I’ve coped with this so far: with or without the underlayer of paint/glue, after the ground cover has been soaked and has been drying for about half an hour (still very wet), I search out every pock mark and carefully “fill in” by sprinkling very small amounts of additional ground cover. This works, but it’s really tedious and I’d like to find a better way. Feel free to comment …