If you’re interested in having a train in your own backyard, it’s more affordable than you think.
The costs of the raw materials are comparable to that of some of the swing set / fort / slide combination play structures seen in so many backyards. If you’re handy with tools and willing to invest your time, you can build one of these trains and run it in your own backyard.
You’ll need a fairly level patch of ground for the track layout. The track curved sections should have a radius of no less than thirty feet. Any curves that are so tight as to have less than a thirty foot radius will probably contribute to the engine climbing the rails and derailing.
You’ll need to store the engine to protect it from the weather and keep it secure. The easiest approach to secure storage is to let the engine roll into a weather-proof house with a door that can be locked. An alternative might be to somehow transport the engine and move it into a garage. But the engine weighs about three hundred pounds, so this might be difficult to do.
Much of the construction of the engine is straight woodworking. If you have the skills to work with wood, then you probably have the skills to build the engine and add the trim pieces.
You’ll also need to build the wheels and drive train on the chassis, plus the electronic circuitry to control the motors. If you have the skills to build the circuits and program the computer, then you could do this part as well.
I Can Help You
If you want to build a train but don’t feel that you have all of the necessary skills (or don’t have the time), then I can help by augmenting the build as much or as little as required at standard shop rates.
I also can provide any of the component assemblies, including:
Computer and Motor Control Unit
For more information, you can contact me via email at firstname.lastname@example.org.
What a surprise to see a near-identical train showing up in made-for-tv movies! I’ve seen this train in at least two Hallmark Channel Christmas movies. These movies were released in the years 2004 and 2005 and are still being shown on the Hallmark Channel during the Christmas season.
I’m guessing that, when filming these movies, the prop master needed a train for the set and performed the same research that I did when trying to locate such a train. I’m assuming that he came to the same conclusion that I did that there’s nothing readily available on the market, and that he would need to build one. He probably found the same plans from the 1965 issue of Popular Mechanics and built one from that.
I would love to get some detailed specs and photos of this train for comparison. But I have been unable to find who to contact at the Hallmark Channel. I’ll bet their version is stored away somewhere in a warehouse and waiting for its next movie appearance!
The first movie is “A Boyfriend for Christmas” released by the Hallmark Channel in 2004. The train appears in the opening credits during the first two minutes of the movie. It can be seen in the following clip from the movie:
The second movie is “Single Santa Meets Mrs. Clause” released in 2004. The train first appears at the 1:41 mark in the following clip:
The next movie is “Meet the Santas” released by the Hallmark Channel in 2005. The train is part of the backdrop of Santa’s North Pole Workshop. It can be seen at the 20:29 mark in the following movie video:
If you look closely, you’ll notice that the train retains its same color scheme and markings in both movies.
I’m close to having the first phase of this train-building exercise completed. Here is a video of the (mostly) completed engine running on the track:
I’m in the debugging phase now. I have a mostly complete engine and set of tracks. I expect to find a number of problems that will need addressing and tweaking, but I’ll work through those as they become apparent.
Staying on Track
The first set of problems that became immediately apparent have to do with keeping the train on track – literally. There are some locations on the track where the engine jumps the track on a consistent basis.
I’m tuning the track so the engine can run in a clockwise direction, mainly to avoid having to climb a rather steep section of the track if it were going in the other direction. What I’ve noticed is that the tendency of the engine to jump the track is dependent on that front left wheel.
As the engine approaches the curved section of the track, the front left wheel is forced to turn when the wheel flange contacts the inside of the rail. Since my track is not completely level, there are some sections where there is more weight on the front wheels than there is in other sections. If there’s not enough weight on the wheel, it can easily climb the rail and go off track.
I suppose one solution would be to replace the flange on that wheel with a flange that’s larger in diameter. But I’d have to completely replace the wheel and find a v-pulley with a larger flange. Another solution might be to add weight to the engine on the front to help keep the wheel from climbing.
With a little experimentation, I found that I could adjust the track by shimming sections to help keep a consistent weight at all sections of the track. That seems to work, although I’ll learn more as time goes on.
Another problem is that the drive chain has a tendency to jump off the sprocket. This could be caused by incorrect chain tension or by the drive and slave sprockets not being properly aligned. It also can be caused if the drive sprocket is too small.
The 350 Watt drive motors I used have built-in gear reductions. I picked these motors in part because I needed the motor shaft to turn at a certain speed to drive the engine. The motors came with a chain sprocket already attached. Unfortunately, the attached sprocket has only nine teeth. This doesn’t provide enough contact with the chain to keep the chain from jumping the sprocket.
I found some replacement sprockets that match the chain size that contain thirteen teeth instead of nine. I’ve ordered these sprockets. When they arrive, I’ll swap out the sprockets (maybe having to add links to the chains to make them longer). I fully expect that this will solve the jumping problem. More on this when I’ve made the changes.
The thirteen-tooth sprockets finally arrived. I removed the nine-tooth sprockets from the motors and replaced them with the thirteen-tooth sprockets. The sprockets fit just fine, and the change in the chain path did not cause any interference with the chassis. That was one of my concerns – that by changing the chain path it would interfere with the chassis. If that had happened, then I’d have to reposition the motors somehow to restore the required clearances.
As expected, however, the existing chains now were too short to accommodate the larger sprockets. I had to add a few links to the chains to make them long enough. Fortunately, I had anticipated this and had ordered a few extra links when I ordered the sprockets, so that wasn’t a problem.
On the first test of the upgraded drive train one of the chains immediately came off the drive sprocket while the neighbors were watching. But all it required to fix was an adjustment of the motor to get the alignment and chain tension correct. After that, it seemed to work fine.
In the two weeks that followed, there have been four different children who have logged maybe fifty laps around the track with almost no problems. The only two problems were when a small limb fell from a tree onto the track near a track joint. The pilot wheel caught the limb and climbed over the rail, causing it to come off track. Another lesson learned: always inspect the track before every run.
The other problem was when I opened the train house doors, but didn’t latch them in the open position. When the train came through the back door (under remote control and out of line of sight), it caught the edge of the partially opened door with a resounding thunk. ‘Nother lesson learned.
Next Steps: Add amplified speakers and electronics to the on-board computer to play sound files to create the chug-chug sound and maybe a train steam engine whistle.
Ok. So it’s really more like “Chugs and Whistles”, but this title sounds better.
I’d like to add sound to play through some on-board speakers to give the engine the “chug-chug” sound. And maybe add a sound file for a steam whistle.
I found some sound files on the internet that I think would make a nice addition. Here’s the sound file for the “chug-chug” sound:
And here’s a couple of the steam-engine whistle sounds I’d like to use:
I’ll need to figure out how to integrate these into the on-board computer to play these sounds through some speakers.
The first step is to make the sound files available to the computer. Since I already have an Arduino computer installed to drive the motors and handle the remote control, I should be able to have it handle the sound as well. But it’s not as easy as you might think.
To store the sound file I’ll need to install an SD card and add the circuitry and programming to read a computer file. Then, I’ll need some sort of DAC (Digital to Analog Converter) to turn the digital sound file into an analog signal that can drive a speaker. And I’ll need to amplify the speaker signal, since the Arduino emits only about fifty milliamps of power – not nearly enough to power a speaker.
Since the circuitry is so complex, I looked for an alternative where I could buy some sort of component that already has this built in. After some searching, I settled on the Adafruit “Music Maker” MP3 Shield for Arduino w/3W Stereo Amp. This component includes an SD card slot, the DAC, and has a built-in amplifier.
The card requires some assembly and soldering, but is made to fit neatly with the Arduino computer. Here’s the Arduino computer and the components of the music card:
After assembling the Adafruit card, I attached it to the Arduino computer to see how it performs. And immediately realized I have a problem.
The Arduino computer card has a number of ports for attaching wires to control motors or to read sensors. But, while the Adafruit Music Maker card fits the Arduino computer like a hand in a glove, it uses all of the Arduino ports for its interface. The reason this is a problem is that I’m already using several of the ports to read sensors for the throttle and remote control unit, as well as for sending signals to drive the motor.
So to solve this problem I just bought a second Arduino computer to dedicate to the sound module. After getting it installed, I was able to read the sound files for the choo-choo and whistle sounds and play it through the speakers.
Adding a separate Arduino computer to dedicate to the sound function raised another problem. I want the choo-choo sound to begin when the train starts moving, and to stop when the train stops. So I somehow need to have the main computer controlling the throttle to signal the second computer controlling the playing of the sound. So I tied the two computers together in a Master-Slave configuration. Now, when the throttle advances beyond zero, it signals the sound computer to wake up and play its sound.
The speakers came from some old computer speakers I had laying around. I took the speakers out of the case and wired them into my circuits, using the Adafruit amplifier to get the maximum volume of sound.
Unfortunately, the sound still wasn’t very loud. It gets drowned out by the noise of the whine of the electric motors and drive train.
So then I purchased a set of car speakers and an amplifier used in car sound systems to further boost the volume. My computer speakers are rated at five watts, while the car speakers are rated 35 watts RMS with a 140 watt peak. And the amplifier includes a volume control knob that should provide some adjustment as required. Here’s what the amplifier and speakers look like:
So that’s where the sound enhancement stands at the moment. Next steps are to install the new speakers and amplifier and see (hear) what happens!
I built a crossing ramp to let me get my lawn tractor across the tracks to the infield. As long as I have a crossing, I decided I need a crossing signal. What I’d really like is a set of the twin red lights that blink in an alternating pattern when the train comes, just like a real railroad. Since I can’t find anything suitable anywhere (not even eBay or Amazon), I’ll have to build my own.
The first step was figuring out how to make this happen, and determine what parts I needed. The red lights were pretty easy. Auto parts stores have generic red lights that run on twelve volts DC that can be used for my lights. I found quite a selection on Amazon. But about nine out of ten of the lights sold today are the LED lights. I wanted the older style incandescent lights just to make it look more authentic. The LED lights are instant on and instant off, while the incandescent lights glow in and glow out. These are the lights I ordered.
The next step was to make the black round background with the black visor. I had some aluminum flashing left over from the train house. So I cut the round background with a center hole sized to the housing for the tail lights. Then I cut a piece for the visor. Everything was attached to some scrap exterior plywood left over from the train house floor. A coat of flat black paint (I used barbecue grill paint that I had left over), and the lighting housing was ready to go.
To power the lights I purchased (again on Amazon) an exterior weather box to house the components. In order to have the lights come on when the train approaches and to blink in the proper pattern, I needed to build another computer circuit. I purchased yet another (my third) Arduino computer to handle the lighting controls.
I needed to somehow have the computer sense when the engine was in close proximity so it would know to start the lighting sequence. There are InfraRed (IR) sensors available for use with Arduino computers. But there were two problems: first, IR sensors are for measuring in inches, while I needed to get a reading when the engine was several feet away. Also, in an outdoor application, I was afraid that the ambient sunlight would include the same spectrum of IR light that the sensor was trying to read, so it would be washed out.
The alternative was to use an ultra-sound sensor. I found this HC-SR04 Ultrasonic Sensor on Amazon. It emits ultra-sonic sound waves from one of the ports, and reads the reflection from the other port. Based on the time interval from send to receive, it’s easy to calculate the distance. So I wired this sensor into my Arduino computer.
Next step was to make the lights blink in an alternating pattern. The Arduino doesn’t emit enough power to run the lights. so I needed to add a couple of relays. The computer can trigger the relay, which acts as a switch to turn a separate and more powerful circuit on and off to work the lights. I was going to use an automobile relay similar to the one I used on the engine, but then found that there are relays made for Arduino projects that are easily incorporated into Arduino circuts. This is the component I used. Note that it has four relays, while I need only two (one per light). I thought it might be prudent to have some spares, as I’ve fried a few of these components while trying to get them to work.
I programmed the Arduino computer so that it’s constantly watching for something on the track within a few feet. Once it senses that it’s there, it wakes up the relays and alternately flashes the lights for a few cycles. Then it goes back to sleep.
I put all of this, including a small twelve volt battery, into the weather housing and mounted it on the crossing sign pole. I cut two small holes in the side of the box for the sensor to look through. This makes the sensor exposed to the weather, so I don’t know how long it will last before I have to replace it. We’ll see over time. Then I plugged the Arduino into my laptop computer to calibrate the settings and hoped that it would work.
Once I had everything working, I closed up the box and ran the live test. Here is a video of everything in service:
If I can find the right type of bell, I’d like to add it to the pole and get the ding ding sound while the lights flash. But most of the bells I’ve found so far are like the school bells that ring continuously. I need one that runs on twelve volts DC that can ring with individual dings. And it needs to be weather resistant. I’ll keep looking – maybe I can get that added one day.
I’ve researched bells for my application. It looks like the closest thing readily available would be either a school bell or a fire bell. But these bells ring continuously. I need a bell that I can have ding just once. Then I can control the dinging with my Arduino computer.
I posted an inquiry on a hobby forum and received a response from across the country in which someone offered a bell salvaged from an old Kodak x-ray film developing machine. The bell was powered by 120 VAC, but should be able to be modified to run on a 12 VDC by removing some of the wire windings in the armature. I’m waiting on the arrival of the bell. Then we’ll see if it can be modified to run in my crossing signal. More when it develops…
I need to build an engine house to have a place to store the train engine and keep it out of the weather. Most trains that I’ve seen have a separate switch spur that allows the engine to roll into the house. For now, I’m going to take a simpler approach and just build the engine house on top of the main loop of the track. This way, I don’t have to add more track and build a switch. Of course, this means that the engine house will need doors at both the track entrance and exit to allow the train to roll through.
I looked on Google at a number of pictures of engine houses, trying to get a flavor of what the architecture might look like. Unfortunately, most of the pictures of engine houses are of structures that are rather plain and utilitarian. So there doesn’t seem to be a lot of latitude in how to design the structure to look “railroadish”.
I did get a few ideas, though. By incorporating these into a design, this is sort of what I have in mind for my engine house:
I’m going to make it bigger than necessary for housing the engine. That way I’ll have some storage capacity for parts and spares related to maintaining the railroad.
I initially was planning on a shed that was ten by twelve feet. But the proximity of the track to some trees that I’m unwilling to cut means that the largest I can make it will be ten by ten.
The shed floor can’t span the entire width of the house, as the track needs to be exposed inside the house. So to maximize storage capacity, I’ll have the track running along one side of the house to leave as much area as possible for the floor in the remainder of the shed. That’s why you see the track going into the side of the house instead of into the middle.
To preserve a look of symmetry, I’ll have a real door over the track entrance and exit, and a faux door on the other side of the structure that won’t be functional. And to keep it from looking too plain, I plan to have a “chicken coop” along the roof ridge. I’m sure there’s an architectural term for this, but I don’t know what it is. But I did see this feature in a number of the pictures of engine houses, so “I’ll have one of what she’s having!”.
Now that I have an idea of what the engine house will look like, it’s time to start. First order of business is to make a list of the building materials and get that on order. I took delivery of the material and set it in the garage for now. I’ll move the pieces to the building site as needed so that they don’t sit in the weather and warp. Here’s what the engine house looks like at the moment. Note that it comes “Some Assembly Required”:
Selecting and Clearing the Site
The track oval sits on a slight grade. I’m going to position the engine house at the high point of the track so that the track is at the same relative height (to the doors) on both the entrance and exit. There are some small trees and brush that’ll need to be cleared away. This is what the site looks like before and after clearing:
Preparing the Foundation
I could prepare a foundation involving grading and pouring a concrete foundation. But I’m going to keep things simple (and cheap) by sitting the house on concrete cap blocks. It won’t need to support much weight, and the house won’t be getting much traffic. So the cap blocks should work just fine for my purposes.
I’ll position nine blocks, spaced at five foot intervals, to support the 10′ x 10′ engine house. It’s critical that the top load-bearing surfaces of each of the blocks is exactly level with the other eight blocks. If even one of the blocks is out of level, it will affect the levelness of the floor, which in turn will cause the framing to be out of square. So I’m taking extra care to ensure that the foundation blocks are set properly.
Building the Floor
Now that the foundation blocks are set, I can build the floor. The shed is ten by ten feet, but I won’t be flooring the entire area. The track will go through the shed, so I’ll be flooring only that portion of the interior where there’s no track.
Here is what the flooring looks like as it progresses:
Now that the foundation and floor is set, the framing can begin. The walls comprise eight-foot studs to match the eight-foot size of the siding panels. You can see in the gallery of photos the progression of the building, from the framing of the walls and roof to the installation of the siding and doors. The roofing with asphalt architectural shingles completed the structure.
Click through the pictures in the gallery to see the progression of the completion of the engine house:
Since I’ve never built a train track, I’m sure that I’ll learn more than I’ll ever wanted to know about train tracks by the time I have it completed. I’m also sure that I’ll eventually look back and realize that I would have done some things differently once I’ve gone through the process.
That said, I plan to research as much as I can to reduce the risk that I’ll have to scrap major pieces of work after the fact because I didn’t have the knowledge to do it right the first time.
So, from my reading, here’s some things to be considered when I design and build the track:
The width between the rails is critical. If the distance between the wheel flanges is 14.5″, then there needs to be another quarter inch added for the distance between the rails so the wheels don’t bind.
The width between the rails in the curved sections needs to be about one eighth of an inch larger to keep the wheels from climbing the rails and derailing.
The curved sections can’t be too tight. The degree of curve is expressed as the distance of the radius that would scribe the curve. From reading, it seems that the tightest curve that the train will be able to handle is about a thirty foot radius. If you think about the geometry, the minimum radius of the curve would have to be increased if the wheelbase of the engine increases or as the track width increases. I’m going with the thirty-foot radius and hoping it works out.
The grade of the track can’t be too steep. It seems from reading that a three or four percent grade is pretty much the maximum that a train should encounter. Because of the slope on the property where the track will go, I may have as much as a twelve or fourteen percent grade. Based on the hill-climbing tests I’ve already conducted, I know the motors have enough power to handle a fifteen percent grade, assuming I can get enough traction to keep the wheels from slipping.
So my general approach to building the track will be to create track sections that can be attached to each other on the track bed. The curve sections will need to reflect a thirty foot radius, so I’ll build some sort of template for those. In addition, I need to ensure that the distance between the two rails is consistent. So I’ll need some sort of template to ensure that the distance between the two rails doesn’t vary as I assemble the track sections.
Once I get some curve sections of track completed, I’ll want to run a test to make sure the engine can handle the curve without derailing. Assuming that test works well, then I can use this approach for the remainder of the curve sections.
I’ll also need to figure out how I want to set the track on the ground. The simplest way would be to just lay the sections on the ground. There may be several problems with this approach, however. The ground is uneven, so the track would be uneven as well. If the track tilts to the side, then the engine will also tilt on that section of the track. Too much tilt and it may fall over. That would be a train wreck. Not good. Also, the grass and weeds will grow between the cross-ties and will be a nuisance to control.
The best approach would be to set the track on a bed of ballast (gravel) to give it the best support and enable me to make leveling adjustments. I don’t want to have to deal with removing tons of gravel at some point in the future, so I’ll start with no gravel and see how it goes. Maybe later, if necessary…