Saturday, February 23, 2019

Installing the Interlocking Machine at West San Jose

I'd shown the interlocking machine for the WP crossing a couple posts back. Now that I've shared it, it's given me a good kick-in-the-butt to actually install it on the layout. After a couple of weekend work sessions, I'm proud to say it's installed now. See the video above to see it in action, or read on and learn some of the details about installing it.

I've argued I need the interlocking because crews will otherwise forget the crossing is there. In case you doubt it, here's a photo from an early op session showing someone dropping cars directly across the crossing.

And finally, some details about the scene:

The Derails

The West San Jose tower was a mechanical interlocking - that is, the levers on the interlocking machine moved piping which would cause switches, derails, and signals to change position and state. Seeing the Barriger photo of the crossing from the 1930's, I really wanted to model the control devices - the derails that would push cars off the tracks, and the movable rods that would move a derail out of the way so a train could pass.

I'd thought about doing something physical to block the tracks, but the alternatives I thought about, such adding a switch point in the track or dropping a derailing device on the rail, seemed both fragile and troublesome. I chose to do something less prototypical, embedding an LED between ties. The LED glows red if the derail is set, encouraging crews not to roll through. The lights don't force crews to deal with the interlocking, but they're at least an encouragement.

I'd also considered laying out the actual rodding to control the signals and derails. That idea also ended pretty quickly; I realized most solutions wouldn't be able to stand up to the aggressive track cleaning needed for a garage layout. I'd considered using something substantial such as piano wire for the rodding, but immediately had thoughts of poking a bit of rod straight through one finger. (Perhaps I could bend right angles at each end of a section of pipe so the ends are firmly in the roadbed?) For now, there will be no piping.

The Signals

The real West San Jose crossing would have been protected by tall semaphore signals - one set of signals close to the crossing, and another about a mile back. Semaphore signals wouldn't survive last long at this point on the model railroad. The area around Auzerais Street requires a lot of reaching in to couple and uncouple cars, and the low upper deck means operators need to reach right in. Rather than watch semaphores get destroyed each session, I decided to use dwarf signals at the crossing. Like all modelers, I've usually got some interesting stuff in the scrap box for a project. The scrap box held some dwarf signal castings I'd probably bought at the Trains-Nothing-But-Trains closing sale back in 1983, but I only had two of those left.

Instead, I fell back on my favorite crutch - the 3d printer. With about an hour of work, I'd sketched up a signal and had it printing on the 3d printer. I needed to refine the design widen the holes so they'd fit my chosen LEDs, but still had usable signals within a day of changing my plan. The design may seem a bit simple, but it's got all the same detail that my 1970's era white metal signal had... and I don't need to run to the hobby store to get more.

The Switches

The interlocking machine controls a pair of track switches for the interchange track between the WP and SP. (Full disclosure: there was no such track here. When the SP and WP interchanged cars, they did so at a small yard along South Fourth Street. Switching interest won out over accurate trackage.) Like all switches on my layout, I use Tortoise switch machines to control them, both so crews don't need to reach into the scene to throw switches and so I've got electrical contacts to avoid dead frogs. The Tortoises work by reversing polarity, so the switches on the fascia are DPDT switches wired as reversing switches. That won't work with the Modratec contacts - it provides SPDT contacts for each lever.

Instead, I replaced the Tortoises with the the MP5 switch motors I'd used on the Market Street modular layout. The switch machines can be controlled via SPDT contacts. They're also easier to install - the position of the throw wire can be adjusted after the switch machine is screwed onto the layout. The MP5s do use tiny screws for mounting, but I've worked around this by mounting them to thin plywood with #2 screws at the workbench, then using larger screws to attach the plywood to the benchwork.

The Lights

The area around the Western Pacific crossing hasn't gotten a lot of attention; apart from a coat of paint soon after the track was laid, there's been little work on the area for the last... oh, ten years. I did build a model of Western Pacific's tower years ago.

Putting in the interlocking also forced me to do a few other jobs - I added dirt to hide the bare homasote, glued down a fence leading to the Del Monte cannery. I also ended up improving the lighting. When I started on the Vasona Branch, I used under-the-counter fluorescent fixtures. They worked ok, but there wasn't always enough space for the twenty inch long fixtures. I also was always a little hesitant about threading 120 volt wire through the layout just in case the wrong wire got chafed or cut. When I'd checked out LED strip lighting years ago, I found the lights weren't really bright enough for layout lighting, and the printed circuit carriers weren't easy to mount.

Last year, I'd spotted some cool LED units in Fry's electronics components aisle. These were 12 volt LED modules, with white LEDs on a plastic carrier. At $1.50 a unit, they were too pricey for an entire layout. Searching on eBay, however, I found the same modules were often used for hollow sign lighting, and that I could buy strips of a hundred of these lights for almost nothing. These are still available on Ebay (like this - search for "LED module 5050" (5050 is the part number for the bright white LEDs) and there's some that exactly match mine, and a lot of other similar fixtures. I like the waterproof ones; they've got sealed packages. I use 12 volt power supplies for laptops to power them - they're cheap ($10), come with a cord and plug, and don't cover up outlets like wall warts.

Next step: get some operators to actually test out the interlocking!


  1. Really cool to see this project come together. Nicely done!

  2. Why go to the expense of building an interlocking that 1) doesn't match your layout
    and 2)leaves out all the detail of the interlocking, proper signals-Semaphore, not dwarf on the main, derails, not leds, and the siding in the plant. not to mention the added switch to the drill track, not shown on your diagram. Otherwise great detail on the rest of the layout.

  3. Like everything else for the model railroad, I do it because it's a fun project and contributes enough to the overall vision and theme of the layout. There's always going to be inaccuracies, excessively-compressed or rethought locations, and stuff that's not perfect when first built, but having a working and reasonably finished layout is much better than having a perfect layout.

    The interlocking is appropriate for the location. The main differences are the missing distant signal levers and the fact that I didn't model a Saxby and Farmer machine. The signals and LEDs for derails were a workwhile tradeoff; I could build them in reasonable time and they're unlikely to get destroyed during operating sessions like some working semaphores might. Matching the actual track diagram wasn't going to happen; the track layout was decided fifteen years ago, long before I thought of the interlocking project. The layout's already been selectively compressed with the end of siding switch at an unrealistic distance from the crossing (not to mention Shell Oil at the wrong location on the other side of the tracks), but I live with those inaccuracies because it made things fit. (For a similar cheat, see Seth Neumann's switches between the NUMMI plant and Warm Springs Yard where three or four miles of track was compressed to about six inches between the switches and signals.)

    If I ever get frustrated with any part of the scene, I'll take another pass on it like I've done at other parts on the layout. Until then, I'll enjoy what got built!

  4. what could have been the diamond on the layout looks like it is just an after thought. anyway, the detail in the rest of the articles is suburb, you can please some of the people all the time, all the people some of the time, but you can't please all the people all the time. Keep up the good work

  5. Well, to be fair, the WP crossing is an afterthought in absolutely every aspect:

    * It's a crossing of a rarely used SP branch line with a glorified industrial branch of the Western Pacific. It's not some busy tower like a NYC-Pennsylvania mainline crossing approaching Chicago, nor is it even a shadow of well-known Southern Pacific crossings like Niles Tower or the College Park tower approaching San Jose.

    * It's a branch that was so under-used that the tower stopped being manned in the late 1920's, and was torn down in the late 1930's. If I was being prototypical, I'd just have a "push to cross" button for the WP as the SP trains crossed as if the WP wasn't there at all - pretty much the same way the SP thought about the WP overall.

    * It's a crossing that was hidden in the middle of a block, stuck behind a set of canneries and other industries. This isn't a much-loved photogenic location, visited by thousands of rail fans. Check out the Barriger photo in the first article on the crossing to see the unassuming look of the location.

    * It's a crossing on a half-a-two car garage layout focused on switching and canneries. I've repeatedly chosen operations and switching over museum-quality models and a
    "don't touch" scene, and discounted mainline operations to focus on switching boxcars. Placing the canneries and other industries won out when I was designing the layout, and the WP crossing was wedged in as a "nice to have" afterthought because it isn't the focus of the layout.

    * There's no trains on the WP track, so there's never going to be trains crossing here. It's not a busy crossing. The play value of the crossing is only to slow down the switching crews, and making them think a bit about getting them thinking about the appropriate places to place cars, and correct etiquette around a railroad crossing.

    * It's not even a particularly important part of the layout. If I had more space and had the choice of improving the WP crossing or making the Del Monte Plant #3 scene more realistic... well, I hope you like the smell of cooking tomatoes and peaches!

  6. * If you don't ask a question, you never get an answer.

    * The reason I initially contacted you is because you did build a mechanical lever frame interlocking. Whether or not prototypical, on a seldom used branch, or a busy mainline doesn't matter, it does work.

    * I too am interested in designing a prototypical mechanical lever frame interlocking. The problem is there isn't much good (american), but lots of united kingdom info available. The problem is that there tracks through an interlocking plant are direction sensitive, (up and down directions), route sensitive, not speed sensitive.

    * I also found that old MR article for building one. The reason they needed to build a working model was that they had little real info to go on, and there wooden model would prove if it worked, Just like the real thing.

    * You do have a picture of the frame. I don't think many of these even existed here from that time period.

    1. I'm extremely happy with the Modratec kit; I can't imagine building one on my own just because of the sheer number of design details to get right. Even if the Modratec instructions are focused on British-style operations, the actual interlocking design didn't seem to be constrained to British style; when designing the software, you get to specify all the levers that must be in a specific position, regardless of their purpose. They do color their levers in the software and hint at British style track equipment, but that doesn't affect the finished kit. I'd mocked up the San Jose Market Street interlocking as an exercise, and it seemed to be able to capture all the craziness of three routes from the Los Angeles direction. It was also able to force unused switches on a parallel route to point away from an active track.

  7. * If I ever get to the point of actually building an interlocking, I will share what I have learned in the process.

    John Tranes