New Garden Railway!

A new railroad project has begun at our house! The Loudonville Botanical Railway is official.

I have wanted a garden railway ever since I saw one outside a train store as a child, and after a visit to Rich Chiodo's magnificent Isle of Shoals Tramway in 2019 I was convinced. For years I have cast an eye on our backyard wondering where we could locate one, but the prime spaces were taken up with other things such as my wife's vegetable garden and a Japanese maple tree. A swing set it on the way too for Harrison. Our lawn isn't very large and our driveway cuts through it, making free space a premium. But, two things have recently changed.

First, our neighbor took down a very large tree that cast a lot of shade onto our lawn which was nice in the hot summertime but less appreciated in the spring and fall seasons. With that gone, the area below one of our large trees now seems much brighter and more attractive. I frequently sit in a chair under the tree and read train magazines or play with Harrison, and it has become my little outdoor sanctuary.

Second, as a family we have a lot less free time then we did in the past. When we do get up to our local live steam club, the Adirondack Live Steamers, I don't have the energy or space to load up all of my G scale Thomas the Tank engine trains, my 16mm trains including live steam engines, and Harrison's new G scale trains. Plus, I have a ride-on 7.25" gauge train that I want to run. It truly is an embarrassment of riches when it comes to trains, but in the course of perhaps spending 3-4 hours at the club I need to prioritize what to bring and "play with". Less is more. So my large scale trains are rarely used.

With those two thoughts in my mind, I began thinking about whether I could put a train and garden under my favorite tree. My wife was supportive of the idea (though our plans were a bit different) and Harrison would absolutely adore it. And it might be a good family building and learning project.

So, while it wouldn't be my dream garden railway, it would be something fun for NOW. What were my priorities?

1.) It had to be simple to build and maintain. There were three ideas for construction that my wife and I discussed: concrete retaining wall blocks; landscaping timbers; and cedar planks. The first was ruled out because a block wall was too permanent (though for my "dream" layout at my next house they will be what I use). Landscaping timbers with nails and rebar to hold them together looked too "log cabin". That left 1x6 or 1x8 boards. I have built similar raised gardens for my wife and they have lasted 5-6 years already (including several relocations of them, which no doubt hastened their demise prematurely). Cedar boards require no maintenance, and if they eventually rot out I can remove them and replace them with new ones. 

2.) If we decided we didn't like the layout, it had to be easy to remove. We might not be in this house forever. We might decide the location of the layout doesn't work. For whatever reason, it might have to go. It is easy to pull off out the boards, remove the track, and let the dirt and plants recede into the yard. A block wall... not so much.

3.) It had to be quick to build. I wanted to have the layout installed by the end of the summer, with the tracks running by Memorial Day. This whole project seemed like a nice birthday gift to myself. I could have the wood box built in a couple of hours, and then I had a couple of weeks to fill it up with dirt. Laying track over the holiday weekend would go quick, and it would up and running. Throughout the summer, we could focus on plants. 

4.) It had to be easy to repair. Harrison and Shadow (our puppy) will still be playing in the backyard, and both will take an active interest in the layout. Harrison might chew and break stuff, and Shadow might run around on top of it and mess it up (and chew and break stuff). So even though I have experience handlaying G scale track (a switch, a crossing and other such stuff) it had to be easy to repair if necessary. That meant reliable commercial track. Structures, figures and other details will be minimal.

With all that in mind, I took a tape measure outside and tried to determine the diameter of the tree based on the circumference so that I could figure out how large a circle of track I needed. Thee tree is about 40" diameter. I wondered if an 8' square would be large enough, or would a 10' square layout be better. My wife actually wanted an octagon shaped layout, but I felt that would look too artificial and rigid. We both preferred a circular garden bed but there was no easy way to create one without using stone blocks. 

There are many problems with building a layout under a tree (like possible damage to the roots by covering them with dirt, and all the leaves which will fall and cover your tracks come Autumn) but the worst in my opinion is the fact that the tree creates a hill. The area under the tree is sloped, thus to have any sort of flat area you need to add a lot of dirt AND you need to extend far out from the tree. But I was restricted by how large I could build. Compromises had to be made.

Like indoor layouts, the rule for garden layouts is to go with the largest diameter possible. Since I don't plan to run Big Boys or heavyweight coaches I could use something smaller. LGB basic curves (R1) are 4' diameter, and all their equipment fits on them. I wanted something like 5-6' diameter curves, which left me LGB R2 (nominally 5' diameter) and Piko (also 5' diameter).

LGB track is stronger (they used to advertise an elephant standing on it) and more is available used, so in the end I selected their R2 curves. To help expand it I added some straight sections to the circle. This not only made it bigger, but later I can cut in some switches if I want a siding. I purchased a loop of LGB R2 track on Ebay at a great price and was on my way.

 

Then... for Mother's Day my wife wanted some new raised garden beds built. After briefly discussing my plans, I decided to jump in and pick up the lumber for my layout too. I went with cedar 1"x6" deck boards, which unlike cedar siding boards have rounded edges. I thought they would look nicer, and they are a good 1/2" thicker too. The boards were 10' long, and I assembled four walls consisting of two boards each plus some cut up cedar 2"x4" boards for braces in the corners and in the middle.

They were hauled out to the driveway and assembled into pairs of "L" shapes. Then, I arranged them around the tree and screwed them together. As expected, the corners hung above the ground and the middle portions were on solid ground. That is what happens when you drop a square onto a cone shape. No problem... I planned to dig out the dirt in the middle of the walls and all would be well.

But I couldn't! It was solid roots. Even if I could get through it, it would likely damage the tree. So, plan B involved filling the inside with so much dirt that it squeezed out of the corners and buried them into the ground. 

My wife pointed out that one of the corners was dangerously close (18") to the driveway. She had a point (see what I did there?). So, I took some measurements and essentially cut off the corner. It isn't quite a square (what I wanted), and it isn't quite an octagon (what she wanted), but it is a compromise. I guess it is technically a pentagon.

The existing corner was cut away, using a jigsaw and a clamped board for a nice straight cut. A new patch board with angled ends was built up and screwed in diagonally at the joints. I found some inexpensive metal octagonal brackets on Amazon that were perfect. I am pretty happy with the result.

So was my assistant Harrison.

Thinking ahead, I also built a little platform in the corner by my chair which will safely hold the transformer off of the ground and out of the garden.

The next step was to fill it with dirt. I calculated that it would probably take between 2-3 cubic yards to fill it up, and I didn't want to bring home dozens of bags of dirt from the store. Thankfully, a couple of houses down my street is a construction company that sells direct in bulk. I arranged for 3 cubic yards to be dropped off, and then I went to work. (Some of the dirt is for my wife's gardens).

It had rained a bit and the dirt was extra heavy. We had the truck dump it off right near the layout so I only had to shovel it into a wheelbarrow and move it about 20 feet, but it was still back breaking. I could only do about 15 shovelfuls at a time into our lightweight plastic wheelbarrow, and I had to make about 25 trips. I did most of it during my hour-long lunch break, and when it started to rain a little bit again I pressed on. I really piled it on at the corners and let it settle and flow down the cracks to solidify them.

After I got out of work I tamped it all down with a homemade tool. The soil just collapsed under my feet along the edges, and some squirted out at the corners. I knew it would happen, but it was disheartening to see everything drop at least 3-4 inches. So, another dozen wheelbarrows of dirt went in, and then another round of tamping. That was followed with more dirt applied to just the low spots. Finally, my body was aching and I had to call it quits for the night.

The next morning more dirt was added inside, as well as all along the outside perimeter. The corner areas got a healthy application of dirt to support the bottom of the walls, as well as form a dam to hopefully hold back the dirt inside which wanted to slip out. It looks messy now, but once grass grows it will look better. 

The circle of LGB R2 track I ordered arrived a couple of weeks before I needed it, and I just had to open the box and check to see if it was okay. Naturally, my assistant wanted to help again. He was a bit disappointed we didn't run a train on it that evening.

A week later, a trip to the local hobby shop (well, 90 minutes round-trip) produced some straight track, a few more curved pieces to give me flexibility on the siding, a switch, a bumper, and the rail-clamps which are used to attach the power feeder wires to the rails. I even let Harrison pick out the style of end-bumper for the track, and of course he selected the bright red one. Oh well... 

Harrison and I connected the track into large sections which were held together temporarily with zip ties. My wife helped me maneuver them under the tree and it looked... small. Actually, it was she who pointed it out. She wanted the track closer to the edge, and it was a good idea. So, I ordered some more 1' long straight sections to augment the 2' straight sections on each side. 

During the next several days I watered the garden bed with a sprinkler for an hour. This jump-started the dirt settling process which, in conjunction with my previous tamping, would create a nice, dense base to hold up the track.

When the extra straight track arrived, I added it to the loop and it looked better. My wife suggested we install a passing siding in the front but I don't really want more track than necessary. In the future we can always add it. Note the bend at the end of the siding... my wife put in our first plant making our garden railway official! It is lemon thyme and it looks great, but I had to add a dogleg to the siding to avoid it. Oh well.

To mark the location of the track, I stuck some inexpensive bamboo skewers down between the ties. Then the track was carefully removed in sections. 

Memorial Day weekend was very wet and rainy and I was loathe to do any outside work. But, I wanted some progress to be made so I scribed a rough track centerline with my new, small kid's shovel and then pulled the skewers. A trench was dug and I wasn't too particular how deep it went, as the ballast would fill it up no matter what. I aimed for about 5" deep.

Then, more tamping. Thankfully, Harrison offered to help out.

For ballast, I used "stone dust" which I obtained inexpensively in 5 gallon buckets from the same construction company that sold me the dirt. They usually sell it in cubic yards, which is roughly equal to 40 five-gallon buckets, and that was simply too much for me. A pile like that on my driveway wouldn't do, so they agreed to sell it to me by the bucket. I thought three buckets would be enough (spoiler: it took a little over seven, and I have two full ones in the garage for future maintenance).

The first layer went in real quick. Perhaps I made my trench too deep.

Unfortunately, I ran out of ballast and I couldn't get more over Memorial Day weekend so I had to wait a couple of days to buy it. Three more buckets got me up to level but not enough to finish the track. After I put the second round of ballast in, more tamping ensued. I didn't really level it, but just looked at it by eye. There were no steep grades so I was content.

It was now ready for the track. To secure the sections together permanently I purchased TrainLi brass rail clamps. They aren't cheap but they work really well to secure the track physically (hopefully from errant dogs and kids) and they provide a solid electrical connection, essential when running track power. They screw in the from the top, which I think is a bit uglier than those (such as Split Jaw's) which screw from the sides and look more like real bolted rail joiner plates. But, it is much easier to screw them down from the top than the sides if they ever need adjustment in the future.

I brought the track indoors and started assembling the track into three-piece curve sections or two-piece straight sections. The old LGB rail joiners were removed by twisting them off with pliers, essentially ruining them and then the clamps were screwed on. I applied some Dielectric grease (Permatex brand) in each joint when assembling. The clamps and grease will minimize electrical drop-off around the loop.

Finally, on my birthday the weather was pleasant and I finally got a chance to lay the track down. It went quick. Somehow, my ballast trench didn't perfectly match up with the centerline of the track. That's okay, though, as there is plenty of well packed ballast underneath it.

Harrison helped out with some of the screw clamps, and he got the honor of driving home the last spike, which was the outer screw that connected the siding to the switch.

It only needed its final layer of ballast to "set" the track. This would help fill all the voids beneath the ties and really shore it all up. As time goes by, everything will settle and I will need to re-level the track and add more ballast. I applied it heavily and then spread it out with a cheap chip paintbrush. 

I needed to later tamp between the ties in places with the handle of the brush to really pack it in. 

I then went section by section and gently shook the track to fling off all the ballast on the ties (like hitting HO track with a metal spoon when ballasting), and then tamped it down in place. Another thin layer of ballast over the top finished it off.

At last, a gentle mist of water over everything helped it set up. It it stiff like concrete but easily broken apart when necessary.

That evening, we were pretty excited to try it out. Harrison's LGB train made an appearance, and despite being a toy it fit right in with what we were trying to accomplish. He was thrilled to run it, and chase it around and around tree. 

My wife got into the act and planted six or seven little things in the dirt. Over time, they will spread and survive... or die. It is a learning adventure we will discover together. And I can't wait to run my live steam engines on it. All in all, it has met my expectations and will only continue to do so as it develops as a garden railway. 

Published: Boston & Maine Milk Reefer article in NER Coupler magazine

My scratchbuilt HO scale Boston & Maine express milk reefer the NMRA Northeastern Region's (NER) Spring 2026 The Coupler magazine. The first part starts on page 27. 

For the full construction blog posts see here and here and here and here and here and here. 

T-Trak: Chester Railway Museum module (part 1) - Railroad Equipment

I am a member of the local N scale T-trak club in Albany and operate my Japanese Tokyo module with them. However, sometimes my wife wishes I would run some of "our" (she has an extensive collection of Western Pacific N scale equipment) trains at the shows. Since I am always looking for new projects I thought building a new T-trak module would be fun.

My "triple" Japanese module is nearly 36" long. That is a foot shorter than a regular N-trak module, but it still is hard to store something that big between shows. So I wanted something smaller. I brainstormed a British themed module a while ago and that might eventually be built. But for now, I am sticking with American. So, what should be the theme?

The first idea that jumped out at me was the Chester Railway Museum in Chester, Mass. We have camped there twice in their caboose and loved it. Running right by it is the double-track mainline of the former Boston and Albany (now CSX) railroad. The museum has a station, a caboose, a tank car painted for Baker's Chocolate, a speeder, two old wooden boxcars, and an old steam engine on display. It is compact, has sentimental memories, and wouldn't be too difficult to model. 

One thing I decided was that it wasn't going to be a perfect model of the station or the equipment. I was going for "close enough".

The general arrangement at the museum is interesting. To get there you drive in from the road shown at the bottom of the picture below, swing under the tracks via the bridge on the right, and then pull up to the museum. The ground is slanted, rising from right to left, and it all looks like it was carved out of a hill. Very nice, but not easy to model on a flat top module. So, I decided to have the right side of the module be right before the bridge to avoid that issue. 

In the picture above, northwest (towards Albany NY) is to the left and Southeast (towards Boston) is to the right.

Here is a shot of the dirt access driveway leading directly to the caboose, as well as the grade in general.

The museum grounds feature a disconnected track running along the statino platform and connecting the station area to the caboose. A small speeder is parked near the tank car, and two older wooden boxcars are at by the station platform, nicely painted on the side facing the museum. They use one for storage and the other as a children's play area with toys, games, a model train along the ceiling, etc. The track ends at a display of a wooden passenger car truck, which sadly needs further restoration.

On the other side of the station platform is a bay window caboose painted blue for the Baltimore & Ohio railroad that is used as a kitchen for catered events on site. When we were there last fall they were digging out the area to hold a full size passenger car, and it was finally delivered in late December 2025. I don't know how much of this area I plan to model.

The station property and CSX mainline are separated by a chain link fence. On the other side of the the mainline tracks is a maintenance of way siding that is occasionally used. There is no room on the front of the T-trak module to include it completely, though I might add a track in the weeds with a dummy switch. Or not, I am not sure.

I decided to do a quick search online to see what I could find for a tank car, caboose, and station. Without those key components, the project just wouldn't work. 

Finding a suitable tank car was easy. The tank car at the museum has a dome but if you search online for prototype information on Baker's Chocolate tank cars (such as here and here) many of the pictures of other prototype cars don't have domes. I don't recall the history of the museum's railcar, but it might not have been used in actual chocolate service.

There are several companies that make models of them in N scale with domes (Model Power, Intermountain, Micro Trains, Atlas Industrial Rail) but I picked Intermountain because it had a white platform around the dome. It might not be a perfect match but it was close enough. 

Upon receiving it I trimmed the coupler "air hoses" off, painted the wheels brown, and gave it a light weathering using drybrushed brown acrylic paint. I also rubbed the handrails with the sides of a pencil to darken them a bit. Though it probably could use another layer of overall grime, I didn't want to be heavy handed as they try and keep it in nice shape. Within minutes of working on it I realized that N scale is tiny, getting the wheels into the trucks isn't fun, and it is very easy to break delicate details!

The caboose was a bit tricker. The one at the museum (where we sleep at night) was an old Rutland car built by the New York Central when the Rutland was their subsidiary, currently painted up for the Boston and Albany. 

I couldn't find anything exact in N scale except a laser cut wood kit by American Model Builders. A friend of mine who models the NYC in N scale advised me that this kit was not for the faint of heart, so I passed. Instead, I chose an Athearn model (ATH14447) that featured three important characteristics: it was wood-sided, it had the right number of windows in approximately the right locations, and the cupola was also in the right location.  

I disassembled it by removing some truly miniscule screws, and then removing the cupola. This allowed me access to crack the clear window glazing from the inside. The roofwalk was also scrapped, though I kept the smokejack. Then, I used metal and nail files to smooth the roof and remove any roofwalk support nibs.

I used two small jewelers files to open up the windows to a larger size. There was barely any room to slide the file in and out of the opening before hitting the opposite wall. But, the plastic cut pretty easily. Then, I used a small chisel blade to remove the exterior window frames. Next, the windows were framed with Evergreen #8102 styrene on three sides, and some 0.060" Plastruct L-angle on the top. These windows and gutters were likely later additions added by the museum when they restored the caboose for camping service on the property.

The cupola was butchered by cutting most of each of the walls away, leaving just the corner posts. These were then shortened in height.

Finally, new window "frames" were added by taking some scribed styrene and cutting away portions to leave only one vertical row of siding (to represent the frame) with the rest gone (to represent the window opening). Fiddly, because everything was so small, but hardly noticeable from a foot away.

Lastly, the roof was given a layer of single-ply tissue paper and flooded with thin superglue.

Unfortunately, when I woke up the next morning it was a disaster. Excess tissue soaked with glue bent down and attached to the sides of the caboose. I spent the next hour doing my best to scrape, chisel, and sand away excess tissue paper stuck to to the sides. Two windows had to be rebuilt, and some wood siding was scribed with a knife to try and restore the look of wood boards. I was disheartened but pushed on, figuring I wouldn't really know how bad it looked until it was painted.

I used red primer to paint the caboose, which was a good match for the final color of the car. It was sprayed with clear gloss on the sides and then custom decals by Precision Design Company were applied. They aren't inexpensive but the company is great to work with. Mine went on fine, though I botched them a bit on one side and had to redo them (thank goodness they include spares!). Because of the window spacing I had to relocate the "NYC" oval emblem to the end of the caboose instead of between the windows, but I can live with that.

I didn't reattach the body to the body to the frame as I might want to add lights, and this will be simpler with it disassembled.

Also on site is a small 0-4-0 steam engine #74 named "LUCY". Trying to find an exact model of this in N scale was impossible, but again I only needed something close enough. 

I don't think it has an actual connection with the museum or the granite quarry across the street. It was looking a little rough around the edges, but she is cute and colorful.

In my collection I had a pair of small steamers which might work: a Bachmann "Docksider" 0-4-0 and a LifeLike saddle tank 0-6-0. Neither ran well so I didn't mind sacrificing them for the project. The Bachmann engine's length and wheel arrangement was correct, but the LifeLike's saddle tank assembly was closer to what I needed. So, I started with that.

I disassembled everything, and since it never had to run again I didn't have to be too careful. I removed the valve gear by pulling the crank pins, which allowed me to remove the center axle and the gears on the axles. I glued the chassis back together and had an 0-4-0. The frame was too long, so I took it over to my bandsaw and removed about a quarter-inch from the rear. I used a belt sander to shorten the cab to match.

Then, the boiler shell was modified. I carved away some grab irons molded on the saddle tank, the two extra sand domes, the water filler hatch, and the headlight. I used files to smooth over the areas, and also plugged the two dome holes some styrene rod that I had on hand that fit perfectly. 

After only about an hour's worth of work, this is what I had. Is it perfect? No. But the real thing looks a bit rough around the edges too. If you compare it with the prototype above you will note that I didn't replicate the large headlight on the front. I found it a bit ugly so I omitted it, though I may go back later and add something.

I then painted it up with black, green, silver, and white paint. There was a lot of touch ups using toothpicks to get the black, white, and silver exactly where they needed to go. And I didn't paint the wheel rims white as I knew I couldn't do it cleanly, but I did do the side rods and crossheads. 

Alas, it wasn't meant to be. Though it looked resplendent in its green and black scheme, both my wife and son pointed out that the prototype engine was all black. I had painted it from memory using a black-and-white printout of the engine, and my memory was faulty. 

So, I had to repaint it black. With a red roof (another detail I didn't notice with my black and white picture. Then I looked at more pictures online and saw that the front buffer had wood boards painted green (perhaps that was what I was thinking of?), and a number plate that was red. The window frames also should have been green but I decided to quit while I was ahead.

Finally, I ordered some decals and again I turned to Precision Design Company. Though I was planning on just putting the number 74 on the cab, they drew up the artwork for the name "LUCY" and a white box around the number as an option. And did they ever go on easy. Though the entire model is only about 1.5" long, it came out great.

I also went and repainted the cab roof bright red to match pictures I found online, and I also added a black oversize headlight on the front made from a cut piece of paintbrush handle.

For the speeder, I first bought an old styrene or resin kit from Durango Press. However, after destroying the walls just trying to get them separated from the sprues I punted and went with a 3D resin casting that I purchased from Anthony Kochevar. For those with 3D printers, the file can be found here. Anthony sent me a painted one and his work was fantastic. It is roughly the size of a marble yet has so much detail packed in. 

With the rolling stock done (except for the two wooden boxcars and the B&O caboose), I decided that the project was indeed feasible and moved on to building the station itself.