The Randall Museum in San Francisco hosts a fairly large HO-scale model train layout. Initially created by the Golden Gate Model Railroad Club started 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 de-facto layout maintainer. This blog describes various updates on the project. Links: description of the layout, description of the automation software.

2018-03-30 - Doodlebug

Category Randall

Here’s an interesting experiment for the branchline automation: using a Doodlebug.

This one comes courtesy of Martin Perry, who bought the Doodlebug and coach back in the 2000s. In 2005, he relettered both cars and donated them to the GGMRC for the Branchline operation, so it would be quite fitting to have it run today on the branchline automation. The engine is part of Bachmann’s Spectrum line and ten years later still works like a charm.

Since I have no record of when maintenance was done on it, I should probably open it and lubricate it before using it for daily automation. Right now it’s on “probation” to make sure it works and does not derail on the branchline dual-gauge turnouts… these have been doing fine so far but I’ve had some not-so great experience running a previous 2-6-0 engine right there a couple years ago.

2018-03-28 - Error Conditions

Category Randall

Now that we have some basic automation running and, most importantly, some feedback on how it is running, it’s time to account for error conditions. The post from 2018-02-12 did mention that earlier but it was more academic with any actual basis. Now I do have basis:

  • On the Branchline, a repeated issue is the RDC failing to stop and reverse on the reverse block.
  • Now that the RDC is moved to the mainline, I did experience the same issue at least once. However that’s not clear cut as in this instance I was also trying to update/restart RTAC remotely. It happened in a second instance for the RDC.
  • At least once the same issue happened with the Amtrak train.

That picture is, ironically, exactly what I want to avoid and which happened:

In trains parlance, this is called a runaway.

Click here to continue reading...

2018-02-28 - Atlas Master Amtrak GE Dash 8-32BWH

Category Randall

So it’s the beginning of January 2018, I am being told the museum opens in one month and I have two engines that are constantly derailing. Not a good way to demonstrate that automated train layout that I’ve been working on for the last 2 years (or at least trying to, when I had access).

Logical plan: have backups. Nothing new, I had mentioned this multiple times in the past. In an effort to move on towards that goal, I got hold of some of the previous GGMRC-owned engines appropriate for the task, namely three P42 and three GE Dash 8-32BWH. The P42s seem in dubious shape, having run quite a lot.

The three GEs seemed better. One was still DC and two had their original Atlas decoders and seemed to run adequately. These are from the Atlas Master line, and came equipped with dual DC/DCC decoders.

After some lubrication (and learning the hard way how to take the shells off) followed by basic speed matching, two GE Dash 8-32BWH are used in the automation program and we get through the Randall Museum’s Reopening Ceremony fairly well, including two follow up Saturday.

The third Saturday is the Science Fair exhibit at the museum. Very large attendance. Record on the RTAC tablets show that the branchline RDC ran 120 times (!) and the Amtrak train ran 45 times.

And by the end of the day one of the engines stalls on some turnouts.

I take both engines home, remove the shell and realize that one of the motor is… smoking :-/

Click here to continue reading...

2018-02-24 - Mainline Issues

Category Randall

Running under automation is only one aspect for the train layout on the Randall Museum.

The other aspect is supporting people who want to run on Saturdays. There are a few challenges to solve here, such as complying with museum requirements and defining our own screening and approval process. That could be tricky since ex-members might want to run as if they were back in their old club, using their old ways.

One point I have been trying to make is that running on this layout is more tricky than it looks. The panels are confusing, labeled yet hard to understand. Most people have the natural bias of overestimating their own proficiency and that is clearly visible when operating this layout. That is why I want to insist in everyone, including ex-members, in having a training before running again yet at the same time there’s little traction for it from said people -- the typical “how hard can it be”.

But before getting there, one issue to address is the lack of maintenance on the mainline.

5913 approaches turnout T322. The point was stuck and throwing it shorted the mainline.

Click here to continue reading...

2018-02-18 - Automation Toggles

Category Randall

Now that the barrier is up, the automation toggles are hard to reach.

The original design goal was that operators can quickly look whether the toggles are physically up or down, and being able to easily flip with up or down. So having them in a central location made a lot of sense. During the renovation, we used to work in the middle area and have direct access to the mountain panel. Now we don't.

Passenger Automation

Branchline Automation

What would make sense now is to have them in the valley panel area.

Another choice is to have both, e.g 3 state toggles (off, rest, on), repeated on both panels. Push up for on, down for off. The computer keeps the state. By then at that point, we might as well not have physical toggles and move the control to the tablet. The “downside” is that it makes the tablet indispensable for the automation, which is a big no-no.

2018-02-12 - Automation Feedback from the Randall Opening Day

Category Randall

Issues noted during the Randall Opening Day:

  • Passenger:
    • One derailment from 204 going down on B320 between the signal bridge and the turnout.
    • Two derailments from 209 going up right when leaving the station. This is again a very very subtle left turn.
  • Branchline:
    • About 4~5 times, the RDC stopped at a station and never left again. One time it missed the stop at the reverse station.
    • 2 times we couldn’t figure where it was.

To deal with the Passenger train issue leaving the station: this happens where the train accelerates and there’s a very very slightly left curve. ⇒ Delay the acceleration till the train is in the straight line after, which leads to a right curve.

Click here to continue reading...

2018-02-01 - Rapido F40PH

Category Randall

The main passenger automation runs using two new Rapido F40PH engines. They have superb detail and superb sound.

Since we use them for the automation in a back-and-forth route, our initial concern was that some dirty track where the engine stops and reverse would have them lose power. Thus we got two ESU current keeper which I installed.

Many weeks pass and I finally complete the automation of the route. At first everything is fine and then we realize quickly that one engine derails consistently in one of the curves, just before the canyon bridge by the entrance.

Oddly enough we have seen hundreds of engines go through that route, and even though the layout track is not exactly new, it does seem OK.

Click here to continue reading...

2018-01-13 - Brancheline Power

Category Randall

The power for the Branchline includes a Lenz Digital Plus LK100:

This is a DCC Reverse Loop polarity reverser.

Yes, the branchline is a reverse loop.

An engine going from B321 (mainline) to branchline will come “in the back” in the tunnel in the reverse direction.

2018-01-12 - Sonora Signal Mast

Category Randall

The signal mast on Sonora is a double track signal bridge.

What we want, for trains going up hill:

  • T330 Normal aka “mainline” (block 320). Mainline green, Siding red.
  • T330 Reversed aka “siding” (block 321). Mainline red, Siding green.

First I need to identify wires under the layout, there are likely 6 of them, 3 for each signal.

Apparently most of these use a common anode (e.g.

[Update ⇒ 5, one is a common anode.]

Click here to continue reading...

2018-01-05 - Activation Buttons

Category Randall

Issue: short push on activation buttons are not detected fast enough by the AIU-01 polling.

Solution: use a basic RC to make the signal seem longer.

Experiment at home:

  • Use a 1000 µF between GND and the input.
  • Use a 110 Ω R in serie from input to button (to avoid shorting the capacitor).
  • This results in a hold time of about 5-7 seconds when fully charged.
  • When powering up, this also activates the input while the capacitor charges.

Quick drawing:

 Generated on 2018-05-21 by Rig4j 0.1-Exp-40204e8