If you’ve been following the Matson’s Landing in L-Gauge series here on BMR, you’ll recall that I’ve settled on both a prototype, a Series B Climax locomotive, and a scale of 1:33, which works out to roughly 8 studs wide. With the initial high-level requirements defined, it’s time to start working on the actual brick design of the motive power.

When I studied architectural design back in college, one of my favorite professors had a saying: “Form follows function.” What he meant by this was that a pretty design isn’t useful if it doesn’t work. This is especially true when it comes to designing things that move, such as locomotives. If I built a gorgeously detailed locomotive that can’t run on a track, it’s not very effective for a working layout. With this in mind, my first task is to build a functional drive system. Once I know that I have something that performs reliably, I can then work on making it look nice.

The drive system of the real-life Climax locomotive actually lends itself very well to being replicated in LEGO form. A main axle below the locomotive turns gears that drive gears connected to each axle of the locomotive’s trucks, or bogies. Power is therefore transferred from the engine to each of the four wheel sets. For my first attempt at a bogie design, I set out to replicate this setup.

My first step was to take measurements and notes of details of the Climax trucks using the plans that I had found in Model Railroad Craftsman. The side frames of the trucks measured about 7 studs along the top edge, and 5 along the bottom edge. The trucks are assembled from iron bars, angled from bottom to top, with springs on the bolster and both journal boxes. Looking back at the Climax Locomotive Catalog, I found an image of the interior of the truck. It shows bevel gears on each axle, rotated opposite each other, driven by smaller bevel gears along a center axle. With this information, I sat down and started building. Some people work better building virtually at first, then translating to brick. I tend to work in the opposite direction, especially for pieces that have to move. I build first, then document what I’ve built using MLCad.

When I build, I use a process that the website development industry calls “iterative design”. Basically, you create a design, test it, refine it, test it again, and so on, until you come up with a finished product. For this project, I tried to document each iteration for you. This process took a few days, with each new design being slightly better than the last.

For the first iteration, I focused on replicating the prototype truck as closely as possible. I thought the overall design came out well. It was a bit over sized, but it had the basic look of the iron bar trucks with springs, and the gearing also matched the prototype. Testing, however, showed a huge issue very quickly. At 1:33 scale, the locomotive’s base would be about 28 studs long. With a truck on each end, there would not be enough room between the two to fit the axles and universal joints needed to drive the axles, and still allow the trucks to pivot.

For the second iteration, I kept the look of the outer frame, but redesigned the interior of the truck to remove one set of gears. This means that the locomotive would be driven more like a Heisler locomotive, with power to only one axle per truck, but allowing for much more room for the universal joints. During testing, these trucks worked well on straight trucks, but caught on switch points or uneven track. The bottom of the side frame needed to be raised by one plate to allow for more clearance.

Version three of the Climax trucks turned into an almost complete redesign. This version uses a Studs-Not-On-Top (SNOT) approach, which allowed me more clearance at the track level. The change of design also allowed me to shorten the side frames to be closer to the prototype measurement, but still keep the spring detail. This version was also more solid, with no parts falling off while running. It does lose some of the iron bar look, but the overall angled shape remains. I found it to be a good compromise between function and form (remember: Form follows function). Track testing found this design to run well on straights, curves, s-curves and through switches.

Climax truck Version four was a slight redesign of the bolster section, purely for cosmetic reasons. Version three left just a bit too much space between the bottom of the locomotive base and the top of the truck frame. While functionally it worked, I wanted to lessen the space to make it look better. I was able to remove a single plate of height, which brought the measurement between the base and trucks closer to the scaled prototype.

Finally, we have the last iteration, Version five. While testing Version four, I found that the inverted plates on the trucks, when running through curves, were catching on the edges of the locomotive base that I’ve been using. I tried using inverted tiles on the ends of the bolsters, but found that these caught as well. The final solution was to use part 2654, Slide Shoe Round 2×2, to act as slides, keeping the space between the truck and the locomotive base, but allowing the trucks to pivot without catching.

Next up, I’ll start working on the locomotive’s main drive system.