Figuring out more about figures

I’ve been happy enough with the cannery worker figures that I’ve continued to experiment with making figures over the last few weeks. Here’s a couple interesting observations.

Let the AI propose figures

The current workflow - I think of a description of a figure, let gemini.google.com draw a picture of a person based on that description, and I let tripo3d.ai turn the image into a 3d printer model - is pretty easy. Choosing which figures to make requires a little bit of care. What are suitable figures for my era and place? How do I describe someone in 1930’s clothing? How do I mix up my choices do I don’t have a bunch of similar figures?

As I was doing lots with the AI / large language models to create images, I realized that the generative AI systems are also good at analyzing photos too. So… we could also get the AI chatbot to propose a bunch of possible figures, and I could pick out the ones I want.

The AI chatbots work completely from text, so all I need is the right prompt. I ended up taking one of the photos of the workers at a cannery (such as the panoramic photo from California Canneries in Campbell. I dragged the photo into the chatbot at gemini.google.com, and gave it the following command:

Find the human figures in the attached photo from 1930’s California. For each person, write a four sentence physical description of the person that can serve as a prompt to generate a 3d figure of the person. Each description should start with "Create a front, side, and rear image of a human figure suitable for creating a 3d model of the person where the figure represents a person in 1930’s California. The figure is…" For each figure, the description should indicate details of body and face type: build, height, facial features, etc. inspired by the original photo The four-sentence description should also include a description of movement that would be appropriate for that person: either that the figure is walking (most figures should show some movement), perhaps carrying an item (personal item, fruit lug, tool, bag, appearing in a quarter of the figures), or indicate a reasonable pose for a normal person of the time. The description can also contain facial features or other details that might make the figure unique or animated. The output should be a sequence of description prompts that can be used later to create an image of the figure. Each description should contain a summary or label for the figure (“the lean plant mechanic”), followed by the four-sentence description in plaintext. Do not generate the images at this time.

Note that I started with a smaller prompt, but added detail and restrictions when the AI didn’t quite do what I wanted. For example, the gory detail about “show the output this way” happened because the chatbot would switch between giving textual output and just giving me a picture in a form I didn’t want. The comment about adding movement and animation was needed because otherwise the figures were all just standing with arms at their side.

Running that command on gemini.google.com quickly throws out three or four potential figure descriptions. Pasting those descriptions/prompts into Gemini generates nice images, sometimes with details or roles I hadn’t intended. I was particularly surprised when the chatbot started identifying potential roles.

Here’s a on example Gemini created:

The industrious cannery worker: Create a front, side, and rear image of a human figure suitable for creating a 3d model of the person where the figure represents a person in 1930’s California. The figure is a sturdy, broad-shouldered woman with a kind, experienced face and graying hair pulled back under a simple white head covering. She wears a long-sleeved dress covered by a practical, thick canvas apron, and she is holding a small metal fruit-sorting tool in one hand. The figure is posed standing upright with a slight forward lean, as if preparing to return to an assembly line.

and here's the image that Gemini created for that prompt. Don't ask me what the tool she's holding is.

Now, I don’t want to have figures done automatically: I still want to be choosing what kinds of figures match my era and setting, and what figures help tell the story. But automatically generating these is a neat way to get inspirations about what figures to do next, and give me options I might not have considered.

The most annoying part of AI is how the work doesn’t fit what usually keeps me focused. Each request to generate an image or generate a 3d model can take from 30 seconds to a few minutes. That’s annoying because it involves a bunch of waiting and often bumps me out of a flow state where I’m focused on work. The pauses encourage me to go off and do another task, and then by the time I come back, I’ve forgotten what stage I’m at with each image. Worse, the AI tools often do something screwy - producing images without a background, or disregarding part of the prompt, or generating an incorrect figure. Using the generating AI systems often feels like trying to get a teenager to help - sometimes it does really well, sometimes it misunderstands completely, and some times it does something stupid because it just wants to go play outside. A figure might be shown walking, but in different poses in the front and side images. Another figure may have odd details added, or might appear period-inappropriate.

Dealing with the 3d printing process ends up being in some ways easier. Taking each 3d model, adding support structures, and arranging multiple figures on the build platform is tedious work, but at least it’s constant work that keeps me engaged.

Painting

I’ve usually found painting figures taxing because it takes so long, because I’m not happy with my color choices, or because of delays waiting for paint to dry when painting a few figures.

My first big improvement was painting bigger groups of figures. I’d attach 10-20 figures to a strip of styrene, prime all the figures, then go down the line painting each color - flesh color first, then clothing, then hair (once faces had dried), then shoes, then hat.

For color choices, I started out this time writing down potential color choices for clothing, then made sure I had those colors on my pallette. Having the color ranges figured out meant that I had all my colors ready, and had ideas about variations to try ahead of time. I did one table of colors for the regular worker figures, and another for some of the unusual figures (such as management).

white, yellow

Clothing	Men workers	Women workers
shoes	brown, black	brown, black, tan
pants/dresses	tan, brown, black, blue, olive	yellow, white, light blue, dark blue, tan
shirts/aprons	white, blue, red check, tan
hat	brown, black, tan	white
hair	brown, black, yellow, white	brown, black, yellow, white
shoes	brown, black	brown, black, tan

All this sped up figure painting so I could do a large set in an evening.

Next steps

Once we’re doing figures of random folks, the next question is obviously “can we do figures of people we know?” I’ve also been doing a few figures based on friends. I’ll share some of those results another time.

Structures and Stories: Reproducing Industries of the Santa Clara Valley

If you've been reading this blog for a while, you've already heard how I've crossed over from building models of canneries along the tracks into some full-blown local history research. Last month, I shared that journey at the NMRA Pacific Coast Region's annual convention in Rohnert Park.

My talk - "Structures and Stories: Reproducing Industries of the Santa Clara Valley" - focused on four of the industries along the branch. For each, I gave a history of the building, talked about the business there, and showed the model I built. Then, just as on this blog, I veered off in crazy directions, talking about the deadly elevator at the Guggenhime packing house, or how Higgins-Hyde pissed off the Santa Clara valley, or about the Ainsley cannery's exports to England. The talk seemed to be well-received, though I got more compliments from wives of model railroaders who enjoyed actually hearing about the people, men and women, who helped run the fruit-related businesses in the Santa Clara valley. One listener, after hearing I was using appeals court rulings and accounting records to understand the business, asked if I'd been an investigator in a previous career, because my research seemed very similar to her past work as an IRS investigator!

If you weren't able to attend but you're curious about the presentation, slides are here. I also put together a quick handout on research tips for doing similar historic research.

And fingers crossed I encouraged some of you to similarly do some historic research on the buildings along your favorite stretch of track!

3D Printing a Whole Shift of Cannery Workers

Back when I got my first resin 3d printer, I spent a lot of time trying to figure out what sorts of objects and models were worth printing. Full structures? Took too long, and I only needed one of each. Windows? Hard to print objects with open spaces and unsupported spans. Clutter? Good. Common freight cars? Good. My big lesson was that I wanted to 3d print objects that I needed a lot of, that were mostly a single blobby object, and had enough curvy detail that I couldn’t make them by hand.

One really good find was the idea of human figures. Model railroads do look better with some humans around to help encourage a sense of scale and make our imaginary world seem lived-in and occupied. Although many of the places on a model railroad might be remote and lonely, there’s also a bunch of places where the railroad intersects with lots of people.

Back in 2014, I wrote about some early experiments with figures. I’d just found that another maker, The Great Fredini, had taken a 3d scanning rig to maker faires on the east coast and scanned various faire goers. He’d uploaded the STL files to the Thingiverse, a site for sharing 3d models. These were wonderful models - they were detailed enough for HO figures, I could print lots of them, and the figures looked more American than our favorite fallback of Faller’s 1970s European figures.

The Great Fredini’s models did have one problem: they were all scans of modern, 2010 era people going to an exposition in hot east coast weather. Folks tended to be dressed informally and for hot weather. T-shirts and shorts were common. One of my favorite figures was a big barrel of a guy wearing a football jersey and giving a thumbs up. It’s a great figure for a contemporary layout, but it’s not quite Fred Astaire or Grapes of Wrath.

Some of the faces I wanted on my figures - from California Canneries employee photo, 1920s.

I was also very specific about the kinds of figures I wanted. For the Vasona Branch’s 1930’s setting, most of the people I’d see would be the workers going into and out of the canneries and packing houses. Both kinds of businesses hired hundreds of workers for the summer rush, and the workers would have been dressed in very 1930’s garb. Old photos and histories talk about lines of women walking into the cannery in dresses and white aprons, and coming out with aprons stained from fruit and tomatoes. Men unloading fruit and working in the warehouses would be dressed for hot and dirty weather. I also needed a bunch of variations of the figures - young and old, tall and short, portly and thin. None of the Great Fredini’s figures were appropriate for capturing these crowds.

I tried using a figure modeling program to create my own models, but wasn’t good at setting human poses nor could I figure out how to get 1930’s clothing on the models.

Finally Getting Unstuck

I almost got unstuck last year; at the NMRA Pacific Coast Region’s annual convention in San Luis Obispo, Michael Eldridge gave a talk on using some new 3d modeling software for making human figures. This program, Daz3D, was intended for graphic artists to create realistic human images for advertisements and other purposes. It would also do 3d models, allowed posing, and had a market for buying clothing designed by the company and others. This seemed like a decent solution. In practice, though, the learning curve was steep: I had to learn how to use the software, how to pose a figure, how to get period-appropriate clothing models, and turn it into a solid “watertight” 3d model. This turned out to be beyond my ability for the time I was willing to devote to the project. Posing didn’t come naturally to me, and the 3d model machinations to turn the figures into a printable model required too much knowledge of the cryptic Blender 3d modeling tool. One particular problem was that clothing was treated as an infinitely thin separate object, so combining the figure and clothing into a single solid figure required significant 3d modeling knowledge I didn’t have. I put the project aside again.

Luckily, at this year’s PCR convention, Michael gave a new clinic using new AI tools to make 3d figures. Michael’s new approach got rid of most of the learning curve. Rather than using CAD-like figure software to create a human 3d model, Michael showed a much simpler approach: using some of the generative AI tools (ChatGPT, Gemini, Claude, etc) to generate pictures of a correctly dressed and posed figure, then using a commercial website called tripo3d.ai to create a 3d model from the figure. This approach didn’t require 3d modeling skill at all, but just having one or more photos of the figure to be created; the software inferred anything else it needed (like what a human back looked like), and it would generate a decent model for 3d printing. Yay!

Making a Figure

The first step is to generate a 2d image of the figure; these can be reference sheet images (aka an artist’s drawing of front, side, top, and bottoms of the figure as might be used to get approval from a client for an image or character), or simply a front and side view. These can be generated by the free AI tools such as Google’s Gemini from just an english description of what you want.

For example, you can go to gemini.google.com and type:

Can you generate a reference sheet image for a 3d figure of a middle-aged 1930’s woman in a knee length dress and apron with a nurses cap waling with a small step and without a base under the figure

I got the following, though the form of the image varies wildly on every run.

Generative AI tools are weird - they can generate different results every time, so the figure I generate today from a prompt will certainly not match what I get tomorrow. I added the clause about the “no base under figure” because a previous time I ran it, the figures would often be placed on a circular base. My tests yesterday also made reference images with a slightly different format, and always listed a “drawn by” with a name that I assume was completely made up. Weird.You can put in a different prompt and get another image, demand changes, etc. Note that all the captions and labels won’t matter- the 3d figure generating software seems to ignore it.

Michael had suggested the reference image, but I started asking just for a front and side view - that seemed to be sufficient for the later steps, and was much easier to check over.

For example, type the following into gemini.google.com:

Can you make a front and side image for a 3d model of a 1930’s workman with a thin build and dressed for warmer California weather, and have the figure walking briskly

I got this from that prompt and description.

Male cannery worker figure created from prompt.

Great! So now as long as can type an English description, I can make a figure!

Next step is the 3d model. Michael Eldridge and others suggested tripo3d.ai , a paid service with a $20/month plan that allows you to create models for 60 figures a month or so without paying more. (Signing up for a plan is a challenge - make sure you’re getting the monthly plan. Supposedly, you can use a free plan if you choose to render the 3d figure using their older AI system.) I went to the 3d workspace page ( https://studio.tripo3d.ai/workspace/generate ) , and dragged the front/side view into the “upload” box on the left . I then pressed the “Generate multi views” button just below the image - this automatically splits the front and side view into separate reference images for the model making. Finally, I pressed the “generate model” button at the bottom left to start the process of making the 3d model. It takes about 3 minutes. When the image is generated, I could then go to the "list of assets" to see the model and select it for exporting. (Careful - the models don’t show up in Safari. I had to use the Chrome browser on a Mac to see the model on the tripo3d.ai site.)

Now that I’d downloaded a 3d model as an STL file, I treated it like any other 3d model - I brought it into my 3d program’s slicer program, added a support structure to hold the object when printing and support arms and other overhanging parts of the model, then set up a print job with that figure. The worst part was making a support structure that supported the right parts of the figure, but were easy to cut off. After a quick 3d print, I had my figure printed. The software complained about bugs in the 3d model, but I see those even with the models I create in SketchUp - the slicing software was able to fix the flaws itself just fine.

I ended up doing ten figures from seven models, mirroring the walking figures for variation. I also made multiple prints of each figure, adjusting the scale of each to vary heights. After a bit of printing, I had around 100 figures done, printed in ten minutes on the newer printer.

I’m using this crowd of figures for the workers walking into or out of the canneries and packing houses - I needed a mob of figures, and I needed figures that looked like cannery workers and looked 1930s-like. (See if you can guess the prompt for each.). This workflow gave me the figures I needed, and got it done in essentially one long day.

First set of cannery worker figure designs.

result

I’m very, very happy with this process. With a day or so of work, I had a half dozen figure models, and had 3d-printed the first round of figures. There’s still a fair amount of hand work required to make these figures: setting up the printer, removing and cleaning the printed figures, removing support structure and cleaning up the figure, curing the figure in UV light, then priming the figures. Painting the figures is probably the most time-consuming part of the whole process; I’ve been gluing a dozen or twenty figures to a strip of styrene and going down the line assembly-line fashion. All that’s worth it to me to be able to have some figures that match my era and setting.

Figures arranged for painting

The neat thing about the figures and the 3d software is how easy it was compared to handcrafted models for freight cars. Most human figures that we try doing with the AI tools method take very little thought about how to make something printable; the figures are all sort of blobby, continuous single models that print efficiently. There's not a lot of overhangs or separate pieces to make us have to think about what direction to print the figures, or whether the model can be printed successfully. The software also doesn’t need perfect information; it already knows what a human arm or back looks like, so it can infer something reasonable that isn’t well-specified in the input image. The most challenging part, honestly, was just figuring out support for things like arms and hands so they'd print decently, and then just dealing with printing and cleaning up the figures.

Compared to the commercial figures, the models generated by AI are interesting. They’ve got a surprising amount of detail, such as exaggerated shoes or rolled-up sleeves that both looks good on the model and 3d prints well. The resulting parts remind me more of 1970's O-scale white metal figures cast in rubber molds than the Faller injection-molded figures.

Bags of figures from first day of 3d printing cannery workers. I've got a lot of painting to do.

As I was writing this, I looked back at my past adventures with figures in 2014, and got reminded how much the 3d printing world had changed. Back then, my Form One printer could print nine figures at a time in about 90 minutes - the software had a problem processing more 3d models in the slicing software, and the Form One’s slow movements of the laser beam with galvanometer mirrors meant the 3d printing was slow. With the Anycubic Photon printer and a modern computer, I was able to print 80 figures in 10 minutes because of the faster UV screen that can expose an entire layer of resin at one time. That’s also with a printer that’s one-tenth the cost of the original Form One. It’s really gotten to a point where any modeler can be mass-producing 3d parts.

Stocking Up for the Fruit Season

Part of building a realistic model railroad is building a realistic car fleet that captures the look of the trains and industries that service the places we model.

Back when I was a teenager, my philosophy for selecting model freight cars was easy: if it looked cool, I bought it. I’d pick up a boxcar with a bright paint scheme at the hobby shop after a dentist’s appointment. I picked up a pair of covered hoppers because they looked so different from anything else I had. I built auto rack and modern boxcar kits because I found them at the hobby shop and so I could learn some new techniques or so I could do some modeling some evening.

As I started focusing on modeling prototype locations, I got more picky. I cared more about getting freight cars that would have run in the 1930s, began curating the road names so I’d get a mix of Southern Pacific and eastern railroad cars, and pushed down any interest in inappropriate cars. I sometimes broke the rules - the pink Roma Wine refrigerator car that was eye-catching but the wrong era, the early covered hopper that the Funaro and Camerlengo owner convinced me would be a neat model for my era even if it primarily was used in Florida. I built a pair of Northern Pacific stock cars from the Central Valley kits even though they were likely inappropriate for my layout. As time went on, some of these cars got given away, boxed, or merely kept far away from the layout.

The lack of the stock cars wasn’t a big deal. Although I’d initially assumed that stock movements would be a staple of western railroading, stock movements on the real Vasona Branch were rare. The Northern Pacific cars always seemed like placeholders even for the imaginary cattle movements. I’d bought some kits for SP cars at various times, but they weren’t particularly important priorities when the layout didn’t actually have any stock pens to load them.

But there were always a few rumblings that there might be better uses for the stock cars. I’d seen various photos of stock cars being used to carry fresh (“green”) fruit - photos of workers unloading fruit lugs from a stock car at the Richmond Chase cannery in the 1920’s, Ed Gibson’s find of the crushed apricots in stock cars hit by an errant Daylight passenger train in Bayshore yard in the late 1920s, stories about watermelons being shipped in stock cars well into the 1950’s.

One find, however, convinced me that stock cars might belong on the Vasona Branch. Back during the pandemic, the Campbell Museum shared a bit of Ainsley cannery letterhead summarizing the peaches received by the cannery form the Treat Ranch in the summer of 1914. Looking over the list of incoming cars, I noticed that many of the cars delivering fruit in those days were actually stock cars. That was enough for me: I knew that other canneries were receiving fruit in stock cars into at least the late 1920s, and the accounting of the Ainsley cannery receiving fruit from distant orchards suggested that bringing in fruit from distant places was reasonable for the canneries along the Vasona branch. It was time to actually build some stock cars for the fresh fruit trade.

The Cars

My SP S-40-3 cars were built from plastic kits that Ted Culotta sold via his Speedwich Media business. The parts were originally intended for highly-detailed ready-to—run plastic freight cars sold by the Red Caboose around 2013. Ted, as a passionate freight car modeler, managed to convince Red Caboose to let him sell unassembled parts from the kit separately - I’m guessing for folks who wanted to modify the kits as well as folks who wanted the pleasure of building the kits themselves. When Ted offered these in the late 2010s, I bought three on the assumption they’d be fun to build.

Like a lot of projects, the kits sat on my shelf for a long time - I never quite had a strong reason to build them, nor the inclination to spend some long weekends building the kits.

Luckily, re-reading the Treat Ranch pear accounting when prepping for a presentation on cannery history reminded me that these cars could be useful on the layout.

There’s nothing too special about building these kits - the models were mostly built according to the instructions, painted them with rattle-cans of red primer, and decaled for the SP and T&NO. (The uncommon SP subsidiaries on the Treat Ranch accounting would have been fun to model, but the SP merged all those railroads into the T&NO in the late 1920’s, so the odd reporting marks would have been unlikely to have survived into the 1930s.)

I did bit of weathering on the cars using an airbrush. Specifically, I’d had friends talking about the inexpensive airbrushes with a small pocket-sized compressor. (Here is the airbrush I got.) These airbrushes aren’t suitable for serious painting, but they do have enough airflow to make them handy for airbrushing and other detail work. I'll admit it's also freeing to put the candy-bar-sized compressor in my pocket as I search around the layout looking for things to weather - something that's hard to do with a full-sized air compressor.

The Result

The three stock cars are now on the Vasona Branch, bringing peaches and apricots to the local canneries. I’m looking forward to crews asking why they’re bringing cattle to a cannery. They'll get an earful about how railroads make some very strange choices for assigning cars during busy seasons.

Keeping Up Appearances, Panel-wise

How much effort does a panel of meters and switches deserve?

When we build a model railroad for operations, we’re not just building it for our own amusement, but also as an object to be shown and operated by others. If we’re going to invite folks over to spend several hours reproducing prototype roles, we’d like to make the layout welcoming and easy to operate. Some of us with nice inside spaces (not me) might make sure there’s a comfortable area for crew to hang out in with some comfy sofas and model railroad magazines scattered about. Others (like me) might focus on usability: are the controls for switches in obvious locations, and are the locations marked neatly and in a matching style on the fascia.

There’s also the less-obvious improvements. We might declutter the layout and hide storage with curtains below the layout, or make sure the fascia is neatly painted. All of this can make a big difference in our first impressions of a layout. I visited a home layout last year in Minnesota that was beautiful - well lit, clean, in a carpeted basement. Future kits were stashed away neatly on shelves as if in a hobby shop, layout markings followed good graphic design traditions, and general cleanliness would make crawling under the layout for repairs would be a pleasant experience. I described it later as “exactly as childhood me thought all those beautiful layouts in the magazines looked.” After touring a lot of layouts, I can safely say not all of them are that clean or welcoming.

Cleanliness also affects how others might operate the layout. One of the local modelers thought that folks might be rougher on his freight cars because he had a layout that lacked scenery and a pleasant space. “Folks will be extra-careful with Jack Burgess’s cars because of the beautiful fascia and carpeting, but then they’ll be less careful with my boxcars when they can see the 2x4s and the pink insulation.” I’m not sure that’s true, but I could imagine the level of care the layout owner applies to the space might affect the level of caution that operators apply.

Perils of the C-Clamped Panel

When I was cleaning the layout up for an operating session at the last PCR Layout Design and Operations meet, I got thinking about this. A few years back, I realized I needed a place for some layout controls and monitoring. The first was to find a place for the fast clock controller with its switches: I made a Plexiglas panel, cut a hole in the masonite fascia, and dropped the panel in. A couple years later, I realized it was helpful to know how much power was being drawn from the repurposed AC adapters I was using for layout lighting and switch machine power. That added a second panel for a bunch of digital meters to monitor power use. After a couple of problems with DCC power, I got some track meters to check voltage and amperage. The location on the fascia wasn’t big enough for the new meters, so I cut a scrap of masonite to hold the meters, and c-clamped the assembly in a convenient spot - good enough for the debugging I wanted to do.

When I started cleaning up the layout for the op session, I realized that the temporary meter panel stuck out. It was easy to bump, and looked slapdash. There also wasn’t room where the existing panels were located for a more permanent and professional site. Moving the existing electronics wasn’t easy. I’d attached every panel with machine screws into the masonite in random locations - moving panels around required cutting new openings in the masonite, drilling new holes, and covering any openings from the previous locations. Changing the existing panel locations wasn’t easy.

Now, the way I’d attached the different electrical panels isn’t a big deal; most visitors aren’t looking at the panels. However, the ad-hoc way I’d done it did look awkward to my eye, and detraced from the overall look of the layout. To make it look better - and to make me more proud of my work - it deserved a rethink. I wanted something that would be more amenable to change if I wanted to change what meters, lights, and switches I had, or change their locations. I wanted the different panels to be more uniform and similar so that the set of panels looked intentional and related. Because I sometimes needed to pull out the panels to access all the electrical terminal strips just behind, I needed an approach where panels could easily be removed and replaced.

A Sound Solution

One bit of inspiration came from electronic music. One of the big trends for folks interested in sound synthesis is a system called “Eurorack” where similarly sized small electronics modules can be placed into a case and connected as needed with patch cords. The idea of connecting individual oscillators, modulators, noise sources, filters, and effect processors dates from the dawn of computer music; Eurorack’s standards define how devices from different manufacturers connect electrically and mechanically.

Typical Euroarck setup. From Wikipedia article on Eurorack. Credit: Matthew G Daniel, wikipedia.org, Creative Commons "Attribution-ShareAlike license.

Mechanically, Eurorack relies on some mounting channels from professional electrical equipment. Individual module panels are fixed height (about 5.25 inches, or 3U in electrical rack units) with widths that are a multiple of 0.2 inches. Panels always mount with M3 screws into screw holes or sliding nuts in the mounting channels. (There’s also an electrical standard for supplying individual modules, and for the audio signals at patch points; those were irrelevant to me. The idea of a common 12 volt supply and common ribbon cable to connect power might be useful for model railroad accessories, though.) Multiple electrical manufacturers sell the mounting channels which the Eurorack folks repurpose and resell.

Eurorack mounting rail / channel

Eurorack seemed like a decent starting point for me. It allowed for neat panels that could be adjusted or rearranged as needed. Matching the Eurorack size constraints avoided needing to rethink appropriate sizes for individual panels, and gave me measurements for locations of mounting screw holes and tolerances for widths. This wasn’t the cheapest way I could get panels done - screws into masonite would have been fine - but this solution would look good, allow the panels to change over time, and showed visitors and operators that I cared about appearances.

The channels use aluminum strip and screws to set the precise spacing; I put this together and double-checked the height between the channels. Once this was done, I mounted the channels to wood blocks glued to the Masonite fascia.

Finally, I fabricated the different panels that mounted in the new panel frame. I’d previously used thin Plexiglas for panels. These panels were rigid and held switches meters firmly, and could easily be painted and marked. However, the plastic was hard to drill and mill because of its brittleness. I instead decided to use 1/16” aluminum on these panels. The aluminum worked well because it’s rigid and easy to decorate, Some local folks reminded me that some of the sign making franchises such as FastSigns can print designs on aluminum, but I decided to do everything from scratch. I did markings by printing designs on transparency material and then mounting these with Spray-mount to the aluminum.

I did hit some challenges moving to aluminum.

First, the tolerances expected for the Eurorack channels are tight, so I had to be precise when cutting and finishing panels to size and drilling holes in exactly the right places. Panels that were cut a bit too large wouldn’t fit between other mounted panels and needed to be filed down. Eurorack channels can use captive sliding nuts or continuous rods with even screw spacing - the continuous threaded strips I got seemed better because they couldn’t slide behind another panel. However, the threaded strips turned out to make tolerances even tighter because of the requirement that the holes in the panel exactly match the spacing of holes in the threaded strips. Some commercial Eurorack modules use rounded slots instead of screw holes to allow some play when attaching in such cases. The precision required meant I needed to be very careful making the panels and drilling holes for mounting screws. I’d started out assuming I could use a center punch and a drill, but realized that it was too easy for a hole to wander a bit too much. Instead, I had to drill the mounting holes with a milling machine, a center drill to start the hole, and a final drill to size. I still ended up having to file holes larger if the hole positions weren’t quite right.

A second problem was with the lettering approach. Using transparent plastic allowed me to set up lettering exactly as I needed. However, I couldn’t drill or cut the panels with the transparency material on, or it would get pulled off, get metal fragments stuck between plastic and aluminum, or get damaged. I needed to transfer holes for panel switches precisely to do the drilling, then needed to glue the transparency material down in the same spot for the lettering to correctly line up with switches.

My final problem was cutting the large openings for meters. Cutting in thick aluminum isn’t particularly easy. I initially tried milling the openings, but had a hard time holding the aluminum sheet securely on my Sherline mill. I tried a jeweler’s saw next; it worked fine for openings close to the edge of the panel, but couldn’t cut openings in the middle of a 6” wide panel. I finally fell back to using a nibbler tool which was slow and painful but eventually got the job done. I’m proud that I didn’t give myself blisters this time - the nibbler tool is particularly successful at that.

The panel for control and monitoring is done now; it’s got controls for the fast clock, accessory voltage and current, DCC fault lights, and DCC meters. I still need to add meters for the third booster on the layout - for now, I had a blanking plate over the space where this may be placed.

Making the layout look good is important - both for my pride of craftsmanship, for ease of getting around the layout, and for its effect on the folks visiting the layout. Functionally, I now see all the meters which help me spot and troubleshoot problems on the layout. Redoing these panels wasn’t strictly needed for the layout, but it looks much better, can handle future change, and represented a fun project for me.

PCR Layout Design and Operations meet, January 31 - February 1, 2026 in Santa Clara

My favorite model railroad event, the PCR (Pacific Coast Region) Layout Design and Operations meet, is returning to the Santa Clara depot in 2026. In a reversal from the usual schedule, the clinics will happen at the Depot on Sunday, February 1 2026. Invites to operating sessions at local layouts will be the day on Saturday, January 31, 2026. Like always, the meet's a mix of talks about design and operation, layout visits, and opportunities to try model railroad operations on layouts in the Bay Area. As in previous years, you can attend virtually if you're not in the area or want to enjoy the presentations from your home.

Get more information at www.bayldops.com. Buy your tickets in advance at EventBrite. Virtual tickets give you access to watch the presentations and ask questions via a Zoom video conference. In-person tickets also include the Zoom link in case of last-minute travel problems or illness.

Like past years, we'll also be offering free layout design and operations consulting, in Santa Clara or virtually. Check the website to sign up for a session.

Bruce Morden is again running the layout design challenge. This year's challenge is to design a model railroad based on San Jose's Lentzen Ave. roundhouse and engine terminal. This is a great chance to think about how you'd design a layout based only on an engine terminal. What would you do when you're only moving locomotives around? How would you handle staging? What would an operating session be like?

You'll also be able to think about the era to model in the challenge. San Jose's Lentzen Ave. roundhouse handled South Pacific Coast narrow gauge and Southern Pacific standard gauge locomotives simultaneously before 1906. The site supported San Jose's heavy commute and freight operations throughout the 20th century. In 2004, the site turned into Caltrain's Centralized Equipment and Maintenance Facility (CEMOF), handling maintenance and repair of the commute fleet and the new electric trainsets. Any of those eras could make a wonderful small model railroad.

As in past years, we'll be inviting all attendees to an operating session on a local model railroad. For folks who have never done operations, this is a great way to gain some experience in model railroad operations, meet some of the local operations community, and find out how to join other operating sessions. For experienced model railroaders, this is a wonderful chance to operate on model railroads you've always wanted to visit.

I love the Layout Design and Operations meet because it pulls together a fun group of folks: interested in modeling specific locations, railroad history, imitating the real railroads' operating practices, and just interested in understanding what the real railroads were about. It's also a great meet if you're curious about any of these topics. The invites to operate on local model railroads got me interested in model railroad operations, and helped me understand the differences between running trains on my own versus working with a dozen other people to get trains moving on a large layout. If you're considering a new layout, or thinking about operations on an existing layout, you can sign up for time to talk with others about what you're building and what options you might consider.

We'll see you in Santa Clara next month!

3D Printing Again

I love that feeling when a project finally satisfies some itch I’ve had for years.

I’ve talked about 3d printing a lot on this blog over the years. I’d been an early adopter of using 3d printing in model railroading. Around 2009, I bought one of the early FDM (“squirt out strand of plastic”) printers, the “Makerbot Cupcake”, named for the fact that it could print an object the size of a cupcake. I loved the sense of being able to sketch a design, then turn it into a real object, but found the resulting objects were coarse - good enough for say a coat-hook or a pocket for holding a DCC throttle, but less good for scale parts.

In 2014, a company called Form Labs started selling a stereolithography-based printer. Their printer used a laser to harden a layer of photosensitive resin in a glass-bottom tank. The printer would then lift up the partially-completed piece above the tank bottom and repeat by curing another layer of resin, slowly pulling the completed part out of the resin-filled tank. Other companies had been selling printers with a similar mechanism for commercial purposes in the past, but all were much too pricey for the typical hobbyist. Formlabs appeared just as the original patents on stereolithography were expiring, allowing them to make a similar printer at prices a regular person might be able to afford.

The Form One printer was amazing; it could produce models that were close to injection-molded quality. I found it great for making detail parts, and even managed to print some HO railroad cars - flat cars, the Hart gondolas, the Harriman passenger cars, and a bunch of other stuff. The Form One printer was truly magical; I loved being able to think of a project, and suddenly have dozens or hundreds of a part. It wasn’t the fastest; most of the freight cars took several hours to create, and even detail parts could take an hour or two.

The Form One was also really cute. A little bigger than one of the original Macintoshes, it just looked really cool with its retro orange plexiglass box keeping UV out of the resin tank and curved aluminum case looking a bit more like a 1970’s Braun kitchen appliance than a 3d printer. The folks designing it also had done a beautiful job with industrial design - easy to open and pull out the build platform, easy to send the model from the computer to the printer, and a single button for start and pause.

Beyond my personal project, the Form One also seemed like a great way to share the freight cars that otherwise weren’t available commercially. For a while, I was selling the Hart gondolas, and managed to sell a bit less than a hundred of them, with occasional folks still asking about getting a model. Manufacturing my own model railroad items was an eye-opening project; I loved making these real, but I got some hard lessons on fixed costs for a new product (pilot model, boxes) and challenges for quality when I was making one or two freight cars a day. The technology also had its temperamental moments. The Form One could have print problems because of dust in the optics or resin sticking to the tank bottom. Having only a single printer also made it hard to diagnose problems - I couldn’t just test on a different printer to narrow down causes. A couple hiccups hit just as I was trying to sell a second model, resulting in a bunch of freight car bodies that were too rough to sell where I couldn’t diagnose the cause of the problem.

Ever since then in 2017, I’ve occasionally gone back to 3d printing, but various life issues - work pressures, elder care, printer misbehaving - kept me from making much progress with it. I’ve occasionally hauled the printer out and run some test prints on it, but the effort needed to adjust it and getting it working reliably was too much.

Back to 3d Printing

Form One and Anycubic Photon sitting next to each other.

Meanwhile, another set of companies have been making FDM printers in the intervening ten years. Two Chinese companies - Anycubic and Elegoo - have been selling cheaper printers. They’ve been experimenting with different ways to harden the resin (LCD screens rather than lasers), different sizes, different methods for getting the resin to detach from the tank. I’ve heard stories from friends trying these printers, generally with decent results. "When life calms down," I kept telling myself, "I really need to to check these out."

Luckily, life’s calmed down on several fronts in recent months, and I’ve got more free time. With the threat of tariffs raising prices, it seemed like a good time to try one of these printers out. Because the Form One had always been a bit too small to print 40 foot cars in easy ways, I decided to go with one of the larger printers - the Anycubic Photon Mono M7.

Verdict? Wow.

A dozen Battleship gondolas, all printed in a couple days.

Battleship gondolas in detail. These are at least as good as the Form One prints when that printer was at its best.

It's been pretty magical and fun. I started off printing various simple parts I'd designed in the past, but quickly threw something challenging at it: one of the Battleship gondolas. I'd done the model for these early steel freight cars back in 2017, but never could get them to print reliably as the Form One aged - print quality was poor, and at best I could print a single car overnight. (To be precise, the Battleships took 5 hours on the Form One.) With the new Anycubic Photon, I found I could print three cars at a time, and they'd complete an a bit more than an hour. This is what I’d intended 3d printing to be like - being able to print fast enough to build up a stock to sell. Within a couple days, I had a dozen car bodies ready. I found the quality of the cars as good or better than the Form One at its best, and the speed and yield was good enough so that I could do mass production if I really wanted to.

Anycubic vs. Form One

So far I've been really happy with the Anycubic Photon, but it's been hard not to compare it to the Form One. The speed, print quality, and volume printed are all great. One really nice feature of the Anycubic Photon is that I can use the whole print surface. On the Form One, each layer of the part was detached from the tank bottom by hinging the tank and swinging it down and away. This was a decent method, but meant that peeling forces on the part closer to the hinge were lower than the forces of the tank pulling away from the far side of the tank. As a result, particularly big or delicate models really needed to print close to the hinge to avoid portions of the part breaking off and sticking to the tank. I found myself only ever using about half the available print surface to avoid failed prints. On the Anycubic Photon, a springy clear sheet is used for the tank bottom; to detach the part, the build platform raises vertically, causing the sheet to stretch up and peel off. This different peeling process means I'm able to print larger objects and print across the whole tank without problems.

There's also been some drawbacks to the Anycubic Photon. The photosensitive resins used by both printers aren't particularly nice chemicals, but the resin provided by Formlabs had a lot less odor. I'd been able to operate the printer inside the house - particularly handy when the printer needed a clean, warm, dust-free space to avoid messing up the optics for the laser. Anycubic's resins have a much more overpowering odor, probably not helped by the printer heating the resin up to 35'C for better curing. Though the odor isn't a good proxy for safe vs. not safe, the stronger fumes definitely make me cautious about potentially dangerous fumes. It definitely doesn't feel safe to be in the same room as the printer when it's operating. FormLabs argues that some vendors' resins contain more toxic base chemicals which might explain the differences. Ventilation systems are available to route the fumes outside; I'm looking forward to trying one of these. The Anycubic printer's LED screen used for curing also is less affected by dust, so it may be reasonable to just banish the printer to the garage.

I've also found it fun to spot the differences in industrial design. The Form One was much better for retrieving parts. The hinged lid could easily be opened and closed one-handed - helpful when the other hand was trying to pull out a build platform with a just-made print still dripping resin. The Anycubic Photon has an acrylic cover that fits over the tank and build platform, but removing the cover requires two hands and a place to put the shell temporarily. The Photon's build platform also gets fully submurged during operation, causing resin to pool on top of the buld platform. This resin spills or drips whenever the build platform and part are removed from the printer. The Form One's build platform was never fully submurged, so it didn't collect resin in the same way.

For cleanup, my original Form One relied on a manual approach - a clean and dirty tank of isopropyl alcohol, and a little basket to agitate the printed parts in each tank to clean off resin. The Anycubic Photon (like later Form Ones) has a separate cleaning tank device that automates the process: all I need to do is put the printed parts in the tank and press a button for a few minutes of agitation in ispropyl alcohol. It's really nice not to have to spend several minutes manually agitating the part to get the extra resin off.

What about the future?

I'm glad I've bought the Anycubic Photon, and I'm having fun getting back into 3d printing. For now, I'm mostly seeing what I can do with the 3d models I'd previously printed on the Form One, and thinking about what projects I need most urgently for the Vasona Branch layout. My two immediate projects are some simple Ford Model A cars for a cannery parking lot scene, and lots of cannery women figures for some of the nearby streets.

The Anycubic Photon also has me thinking about whether to sell 3d printed models again. I had a lot of fun selling the Hart gondolas - figuring out how to manufacture, box, and sell them. I also like sharing these models: helping to tell the story about how freight cars evolved, helping early 20th century modelers have more realistic freight cars to put on their layout, and encouraging others to try mass-producing their own freight cars. I haven't decided whether I'll go back to manufacturing. For now, I'm enjoying not having the pressure of making and selling the cars. But with the Anycubic Photon, I can see that I could mass-produce freight cars again if I wanted.