What or how else could we control…

If you haven’t checked out the conversation over on Rene Gourley’s blog, regarding the interface and hardware we use to control our model trains, you really owe it to yourself to head on over there to catch up. The latest post is here:


Actually running the trains is such a fundamental part of our hobby but in reading through the posts on Rene’s blog, I realize that my thoughts on the subject are so very superficial. I had some thoughts on the design of the interface but the discussion is running much deeper, to the hardware level. In trying to better understand and “catch up” to the speed of the thread and the calibre of talent running through the discussion I find myself thinking about those aspects I’d like to control and how I’d like to control them.

(That conversation is brilliant. From about here on out, I can’t promise the same calibre of thought but plunge onward anyway.)

Given my interest in the dance steps involved in moving rail cars, from the customer and commodity’s perspective, I couldn’t help but wonder if this thought fit into the control thread:

A loaded freight car performs very differently than an empty one. For open cars we can replicate the appearance of a load with a loose material to represent coal, gravel, or whatever we carry in our model hopper cars. What we can’t scale and represent, as easily, is the mass of that load and the effect loaded cars have on our trains.

Using the example of covered hopper cars being loaded at the MaxYield Co-op I wrote about in One hundred cars and fifteen hours I thought about trying to find ways to distinguish between a loaded car and an empty one.

  • On the surface, as we feed cars through the mill, we can simply label them “Loaded” and be done with it. In terms of the game of operating the model railroad, we’re done;
  • Since the model is sealed, we can’t easily add weight inside the car to represent it when it’s loaded. This weight shouldn’t be permanent so we can empty the car;
  • What if we had a brake installed in the car. As we feed the car through the mill, we could use this brake to add drag to the car’s wheels and make it harder to shift. It’s not really adding weight but the “heavier” car would drag more and be different to move around compared to its “empty” state;
  • Mass, rolling resistance, and similar characteristics do not scale in our models so we’d need to consider the effect of these heavier cars in the train. Obviously, we’ve debated this one to death in conversations on the ideal weight for a model train car and what I’m proposing is a car that is only occasionally like a problem car.

My interest here in representing this car goes a bit past just having heavier models to adding something tangible to do while pretending to wait for the car to be loaded at our miniature mill. More than just setting a stopwatch timer once the car is in place.

Just an incomplete thought on a lazy Friday. It’s beautiful outside and my mind is wandering just about everywhere.






  1. Thanks for the kind words, Chris.

    If I understand your idea, I think it is eminently doable. The only hiccough would be the positive identification of load.

    We have several ways to identity which cars are in a train, but I think they all require the car to pass through a point after the locomotive. Hardware does exist to enable objects to detect if they are adjacent to one another, but I think this would imply smarts in every car. From a cost standpoint, that’s achievable on a small layout like mine, but prohibitive on a larger layout.

    However, if we can solve the detection problem then it would be easy to make the cars behave as if they are empty or loaded. All you would do is make the engine require more throttle for a given acceleration. Inversely, if you open the throttle to notch 1, say, the system would look up how many loads and empties it believes are behind the engine and adjust the acceleration accordingly.

    1. Oops.

      The unfortunate thing is that without detecting coupling, we couldn’t make this a detailed behaviour. The engine would still behave like it had ten loads after it dropped them off until it passed a detection point.

      As I say. It is doable.

      1. I think you’re touching on something I hadn’t considered: what if we could change a variable on the decoder to represent the effect of the train on the way the engine performs?

        Say we had a unit train of grain cars to load. We feed the cars through the loading point one at a time. While we’re waiting for our miniature crews to load the car we tap a “+1” button on the throttle. On click this sets a timer to measure the length of time to load the car and when that time limit is set it modifies the decoders throttle and brake settings to reflect the extra, albeit imaginary, weight. Sort of like a variable momentum driven by the circumstances.

        I hadn’t thought of this angle. It would get closer to what I was wondering about. Thanks for chiming in.


      2. So, I need to get a better handle on the roles of: the command station, throttle, and decoder.

        What I’d need is for every car added or removed a linear weight is added. This weight represents mass but in terms of the command sent to the train the weight is of a more mathematical one. Modifying the span of the typical speed step.

    2. So, we use a variety of software to drive switch lists.

      So, that means we have the cars that are in the train or will be added to our train in a table. All we’d need to do is marry the throttle to that switch list. Attributes that would naturally come across with each car in this relationship are the state of the car and how that should change the train – or at least how the throttle and brake responds to the change in mass.

      Fascinating perspective. Thank you.


  2. I read somewhere recently about drastically increasing the weight of model train cars to be more realistic, and also adding washers to railcar wheels to increase the resistance. I’m OK with the former – although it can be hard for some types of cars – but I don’t think increasing the rolling resistance is a good idea. Real rail cars are very heavy but once they start rolling, they don’t stop!

    If you have a way to add a temporary brake to a car, that would work… perhaps in today’s decoder rich world, someday soon rail cars will have decoders. You can already buy sound equipped containers…

    1. I’ve read that too, about over-weighing cars. I see the merit but it’s a permanent solution and my thoughts were toward something variable – such that the same car could be loaded or unloaded during an operating session.

      We have decoders for freight cars for uncoupling (e.g. Kadeee in HO and LGB in G) as well as car sound (e.g. Soundtraxx mechanical reefer sounds) so that’s where I was going with the car with a brake to introduce some drag. The problem is the effect on the car’s ability to roll.


      1. In N scale Microtrains couplers, by the design of their mount inside the coupler box, accordion in and out. Car-by-car this effect is multiplied along the train. Microtrains used to suggest adding a wiper to a wheelset in a caboose that would introduce some drag to soften the effect of this movement.

        When I first started thinking of this I pictured a DCC-driven brake that was proportionate and applied by the operator. Initially this was literally to represent actually setting the brake. Then I started to wonder a similar application to represent loading the car and representing this effect.

        As for the brake I pictured a simple decoder driving a servo in the car that would cause something to rub against the wheel or axle.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s