Blog & News

About the Model Railroad

RTAC Software


The Randall Museum in San Francisco hosts a large HO-scale model model railroad. Created by the Golden Gate Model Railroad Club starting in 1961, the layout was donated to the Museum in 2015. Since then I have started automatizing trains running on the layout. I am also the model railroad maintainer. This blog describes various updates on the Randall project and I maintain a separate blog for all my electronics not directly related to Randall.

2019-05-18 - Work Update: Napa and Sultan

Category Randall

Last week I spend most of the Sunday cleaning the Napa yard and its balloon track. The goal is to eventually be able to get a train out of the Napa yard and on the mainline. There’s still a number of challenges.

This Saturday, I did install a Frog Juicer Auto-reverser on the Napa balloon track. I simply put it in series with the reversing toggle, which means power can still be cut simply by using the “A power” 3-state toggle (yard, N, valley). Usage is thus simplified, one just needs to select the A or B push-buttons to align the entry turnout and the yard ladder.

I ran a train from Napa to Sultan to understand the various issues. I was running the TGV in MU with two opposite “head” units. That was sort of required as, although I spent a lot of time cleaning that track last week, it’s still kinda rough. It needs to see more usage before it gets smooth. A single engine will typically lose power at the most inconvenient places.

Let’s look at the track schematic on how to go from the yard to the mainline. The “normal forward” direction on the mainline (when operators run in a continuous loop) on this schema is from right to left:

Mainline is in blue. A train normally arrives from block 170 and goes straight through the T140 turnout, also known as “Sultan” (note: I got the turnout T140 schematic wrong, the blue mainline track is straight through the turnout and the black lead track should be the diverging branch).

To bring a train from the Napa yard to the mainline: go straight through block B160, and simply throw T140 in reverse.

  • Known Issue 1: Need remote control of Sultan’s switch.
  • Known Issue 2: Intermittent power loss just on B150 just before the Sultan turnout.

Here’s why issue #1 exists: The turnout controls are on the Valley Panel, on the opposite side of the layout room, with a mountain in between. When operators are next to the turnouts, they are far from the panel. And if they move to the panel, they are far from the track. Lots can happen between when a turnout is thrown and the operator moves to the other side to go see a train. This is also in the public area of the layout, which means access may not be trivial and operators will get distracted on a busy day.

I spent a bit of time understanding that power loss. I realized each time I tested this spot without train or with a single engine, it was fine. I only saw it happen to me and others with a somewhat long train.

The power loss happens just before reaching the leg of the turnout on T140 and sometimes it works, sometimes it doesn’t. I’m suspecting a bad solder in a lead wire, or a broken/oxidized rail joint. The problem is that there’s a tunnel right in that curve, making it harder to look at the track.

Some operators insist on using the Sultan turnout to get in an out of the Napa and Richmond yards. I’ve been asked to fix it repeatedly as it would be “convenient” to use these yards for Saturday ops. But is that the only solution, or just the one everyone focuses on? And honestly I’m only half sold on the value of it (it definitely has value but as explained above on a busy public Saturday ops day, it’s probably not the best place to be, too many things to go wrong). Still, are there alternatives?

Looking at the track schematics above, obvious alternatives would be to use T150 or T151 to go to and from the mainline to the yards.

  • Ideally a train could go past T151, stop, throw T151 and reverse. That would be the most obvious one.
    • Except... one of the two turnouts on this crossover does not throw. I’d have to crawl under the layout and figure why, likely yet another broken wire.
    • On the map, I wrote it down as a “T/F” type, meaning one side is a new Tortoise (thus updated in the 2000s) and the other side is one of the old Fulgurex (likely from the 1970s’). This matters because if I need to add electronics to control it, something like an NCE Switch-8 can directly drive a Tortoise but not a Fulgurex, so I’d need an extra relay. No big deal, just need to account for it.
    • If I can make T151 work, we have a good replacement for Sultan. Once T151 is thrown correctly, an operator can use T162 to choose between Napa and Richmond, and this one works.
  • Another choice is to use T150 with a train going forward from mainline onto the B150 block.
    • It must then fit fully in B150 before fouling the T140 turnout, stop, throw B150 back to normal and reverse.
    • Which means the length matters.
    • It also means the lead engine would be stopped in the tunnel just before T140 so if anything happens (loss of power, derailment), it becomes yet another task to get it out.

An important consideration is that due to the physical design of the layout room, all these operations involving turnout T140 “Sultan” or the leads to the Napa or Richmond happen in the a public and congested place. The turnout control panels are on the opposite side of the room, with a mountain in between, offering thus no visibility on the operations.

Even if we have remote operation of the turnouts, it means the operators need to be next to the location. I can partially solve the lack of visibility by adding a small pinhole camera and set up a tablet for remote view of that turnout.

However one thing I cannot solve is that this all needs coordination between operators, to ensure another train is not going down the mainline while one is entering or leaving the yard. Lack of operator communication and attention to what’s ahead of their train is still a typical issue that I observe every Saturday due to the “passive” nature of a run-around-the-loop operation scheme, coupled with the lack of visibility. Operators just naively assume there’s nothing possibly wrong in front of their train when they lose sight of it, and just expect it to come back fine on the other side of the layout. I’ve done so myself so I’m exempt here. The rigor is just not there, and it is bound to be error prone.

Any derailment becomes a public facing issue. On a busy Saturday museum day, this is yet another challenge.

 Generated on 2021-11-29 by Rig4j 0.1-Exp-666f4a7