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.

2019-08-06 - Track Cleaning

Category Randall

How do we clean the track on this layout?

Turns out it’s a bit of a complicated subject. Take any model train magazine or forum and you’ll get hundreds of different replies, everyone claiming to know better.

I, for one, don’t know any better. I only know what I’ve seen done, and what I tried, with either success or failure. Let’s do a list. Sit down, that’s going to be a long story.

Isopropyl Alcohol

When the GGMRC owned the place, the cleaning process involved isopropyl alcohol and running a train with a tank car, a roller car, and a pad a few times, once per week. Despite the obvious evidence against using alcohol, some people firmly believe that’s the way it has to be done, and they won’t change their mind.

The alcohol method worked great for the older DC-based model railroads. And today’s DCC is the same, so why change? Well it does turn out there’s a huge yet invisible difference. Since DCC uses a square-wave signal through the rails, it tends to generate higher frequencies and creates micro-arcing between the wheels and the track. That in turn generates more oxidation and we get “dirty wheels” faster on DCC layouts. Unfortunately alcohol does not inhibit that, it exacerbates it. The alcohol proponents will then shift and claim one needs to use costly 90% grade alcohol instead of the “cheap” 70% one. The old fallacy of believing the price matters. They reason it has 20% less water so less oxidation, right? Well that’s… totally missing the point. It’s still a polar solvent.

OK we need something else.

CRC 2-26

Years ago I was helping someone restore an old Marklin layout in San Francisco, which had not been used for a decade or more. Original track, original trains. The poor engines couldn’t run for more than two inches without stopping. And in my toolbox I always carry a can of CRC 2-26 because that stuff is so useful and versatile. One thing to keep in mind is that the aerosol is highly flammable, but the deposited product is not as far as I know (I haven’t read the MDS in a while). So I got the idea of scrubbing the rails with a pad with CRC 2-26 and it worked wonders (I started with a paper towel and quickly realized it was a terrible idea… using a cloth that doesn’t shred at each joint!) His layout was running in no time. My idea was to use the stuff by scrubbing to remove the existing oxidation layer (just don’t let sit there… actual elbow grease is needed the first time) and then it makes a better electrical connection.

Then I got home and did the same on a small test layout and quick I ran into issues. Why, it worked great on the other layout! Yes but he had a flat layout. My trains had a good contact but suddenly they couldn’t make that 1% grade I had… oops.

Morale: Do not under-estimate the name “lubricant” on CRC 2-26. It’s really really slippery.

Issues seem to first develop in grades, as well as long steam engines with lots of wheels in curved, and the combination of both is the end of it.


“Oh I know, Wahl… that stuff is great”.

I will file that in the same category as CRC 2-26. We used that a lot at Dupont when reviving the layout, and I still use it at Randall for example to revive a yard that hasn’t been used in 4 years. Same thing, it works great when involving elbow grease to actively scrub the rail. Just placing it on a roller car and expecting to roll once and do its work is not enough. The point is to use the oil to scrub away the oxidation layer.

But then it is slippery since Wahl is also a lubricant.

And as with CRC 2-26, the oil leaves a sticky residue which after weeks attracts the dust. So next time you run your engines, they suddenly accumulate a layer of gunk on their wheels out of nowhere and after a few lap they stop running wheel.

Turpentine a.k.a. Mineral Spirits

In the same way that Wahl and CRC 2-26 have the same end result, mineral spirits and alcohol are also very similar.

My father used to clean his layout using mineral spirits. I used to love the smell. I could feel the neurons dying in groups. It’s one of these when the bottle says “use in a well ventilated area” they don’t mean just open the door. Make it a tornado in there so that you don’t get to breathe the stuff. Really.

Also mineral spirits are excellent pain thinners, so don’t drop any on the scenery.

Anyhow, we tried that at Randall and at the same time I was trying it at home. Got a bottle of it at Lowe’s, put it on a roller car, rinse and repeat. Then after a while we realized we were in big trouble. Worse than alcohol. How’s that possible?

Well it turns out the stuff I had bought wasn’t the mineral spirits I was used to. It was a “green eco friendly no-odor” version. And instead of looking like a clear liquid, it looked like a white liquid. And it was leaving a layer on the track, which accumulated at each “cleaning”, making it worse. Plus a ton of gunk on the wheels.

I then spent 2 days crawling everywhere on the layout cleaning the track with an eraser and then rolling a car with alcohol multiples times to get rid of the stuff. And getting blamed for it at the same time, with no help being offered whatsoever. That was not fun. At all.

Morale: When the product literature claims a product is a better version of the old one, they mean that literally as “it is not the same stuff”.


After being burned by the not-quite mineral spirits episode, I was told to try ATF -- Automatic Transmission Fluid -- by the same party who both congratulated me for trying the mineral spirits and then blamed me for it, so I was quite reticent at first.

All the forums conclude: use it very sparingly (one drop per rail), let the trains spread it, oh and it ruins plastics or paint or something. That sounds encouraging… not.

Eventually I decided to try it at Randall and got some great results indeed. I settled on using a cotton swab to spread one drop per rail over two inches of track at roughly opposites sides of the layout (one in Stockton Yard and the other one by the mountain before Sultan). That worked great. Renewing the process every couple months seemed to be a good schedule.

I only found two drawbacks.

First overapplication is easy. “Oh just a tiny bit more”, or “I’ve done it last week, let’s do it again, can’t hurt, right?”. Well yes it can. Second communication. It’s always fun when the party who recommended to use ATF then keep running roller cars with Wahl when it had been clearly communicated not to do so. Combine with the first one and suddenly someone indicates their Daylights GS4 can’t make it up the grade at the long curve after Stockton. So there I am again crawling in small areas scrubbing the rails to remove excess of whatever is on it, getting no help in the process, and being told communication was not enough and I should try to make it clearer. Which. I. Did.

So right now this is my policy at Randall: ATF in two locations, monthly.

I was told it would be helpful if I made a maintenance worksheet to guide volunteers. So I did that. I made so that they can fill it and sign it off so that we know when it was done last time and avoid overdoing it. And of course that worksheet gets superbly ignored because “I did not place it in a convenient location.” Yeah it’s so hard to move a piece of paper. It’s rather depressing.


Several folks in the MRH mag, the forums, and other places, now daim this to be the best thing ever. The idea is to use a graphite bar (or a pencil) and lightly apply graphite on the rail, but not too much -- it should not be visible. Graphite is a good current conductor so that makes sense. It’s also a lubricant, so avoiding overapplication is a requirement.

There are two main reasons I have not tried this approach at Randall. First I have multiple times suggested that from the very beginning, only to be totally and ridiculously ignored. It really was a lot of wasted argumentation. At some point I just gave up.

Second, given the attitudes I encountered with ATF, I can only imagine if someone passes a roller car with Wahl or runs some ATF after graphite is applied. My guess is that it will wipe the graphite layer out instantly so the exercise will be moot. We won’t even know if the system works as it will never be given a chance to demonstrate itself. And in the same category I wonder what happens when someone brings external engines and cars that have run on an oil-based layout. I’d expect these trains to also accumulate and transform the graphite layer in gunk, rendering the process inadequate.

I can see how that’s appropriate for folks who run their private layouts using their private rolling stock. They can keep their whole ecosystem closed and in unison. I don’t think it works for a shared environment like we have at Randall.

I don’t try to move mountains. I’m too pragmatic for that.

 Generated on 2020-07-05 by Rig4j 0.1-Exp-61675ac