Home Page
Prototype
Layout
Equipment
Articles
Library
About
Site Map

Peter's Model Railroading | The Equipment | Cabin Cars | PRR N6b
Construction: Data Gathering

When I started this web site many years ago, I didn't want it to be another site that just shows the models I bought in a store and plopped on my layout. I wanted this site to be a tutorial or how-to site. That is why I always take a ton of photos while I am doing something related to the hobby, in the hopes that between the photos and the accompanying texts, you can figure out what I did and perhaps copy or adapt that approach to your own modeling, regardless of chosen scale. So, since I have decided to build this PRR cabin car from scratch, I wanted to show you how I approach a project like this from the very beginning until I have a fully painted, decorated, and running model. Since this is a hobby and I have limited time to devote to the hobby, this project is going to take some time, so be patient with me. This page starts the initial phase, which is the basic research phase.

What to Build?

The first step in determining what to build is to narrow-down the model type and the era you are modeling. This will help you figure out which actual prototype model you want to build. The Pennsylvania Railroad had many different kinds of cabin cars. Since I model the summer of 1924, I am looking for any car that was built before that date. Various sources indicate that the N6b was built from 1914 through 1923. So, any version of this car would be suitable.

Studying prototype references (books, magazine articles, online photos, etc.), I learned that there were two versions of this car. One had the cupola in the center of the car, and the other had it offset to one end. The built-from-new cars had the centered one. The ones that started life as the two-axle predecessors, were rebuilt ones which had the offset one; they simply added a new section to one end of the car, to make the car the legal minimum size of 24 feet). For this build, I am going to build an offset one.

Source Material

The next phase of the data gathering process is to find any and all scale drawings, prototype photos, and any other text data that might help in determining the basics, such as dimensions, configurations, and maybe even models that were specific to a particular geographic region only. Your ultimate goal is to find a prototype photo (if possible) of a car that you want to model as close as possible. This will then also give you an idea of the road number, and the other lettering of the car that you want to apply to your model. Sometimes the railroad may have made changes to a particular car, so if you compare several photos of the same kind of car, you may notice several differences. This may get confusing, and you may wind up with a model that is a hybrid of several different versions. So, I prefer to do as much research as possible, but then find a good quality photo of one particular car, and then just set out to build a model from that one photo. For this particular car, the Pennsylvania Railroad Cabin Cars by Robert L. Johnson, published in 2021 is the best resource. The diagrams included in this book are a perfect fit; not an exact match, but they provide enough details to "fill in the blanks".

(external link: Pennsylvania Railroad Cabin Cars)

Exterior

For this car, the exterior of the main body is made out of individual vertical boards. Prototype drawings show the main body being 24 feet and 7-3/4" long. That comes out to 295.75 inches. I then found the best prototype photo I have of the car (found on the web somewhere), copied the top portion of it into the graphics program I use, and placed a series of 10 dots, one on each board, going across the car (it is easy to lose count when they all look alike, hence this approach). This yielded exactly 9 groups of 10 boards, or 90 boards. Dividing 90 into 295.75" yields right at about 3-1/4" wide boards were used.

I have decided to use OpenSCAD to draw the car's design as accurately as I can. The primary reason being that this forces me to find all of the actual measurements and make final decisions as to what the model is to look like. My first actual construction effort will be to scratchbuild one of these cars from wood and styrene. However, since OpenSCAD can support being exported to a format that allows for 3D-printing, by investing my time in creating this CAD design, when I, in the future, buy a good-quality resin printer, I might be able to print multiple models of this car from this design. This diagram shows one of the side walls of the body only. It measures a scale 24'7-3/4" long and 7' tall.

From the prototype drawings, I concluded that the window openings in the wall are 22-1/2" wide and 28" tall. They start at 46-7/8" in from the outside edge of the body, and are 36" from the bottom of the exterior boards. I could then also draw the other side wall. The outside surfaces of these two walls are 9'1-3/4" apart (109.75"). The exterior boards are shown to be 7/8" thick; in my CAD drawings I am making them 1" thick, for simplification, which only impacts any interior measurements. Note that the prototype drawings show a marker lamp attachment panel/door in the upper corner of each side wall. I don't know when these went away, but I have yet to find any prototype photos that show those and no model railroading models were ever made with those (regardless of scale), so I am ignoring those as well.

Next, I drew the ends of the body to fit in between the two side walls. Getting that curvature correct for the tops of the walls was several hours of work! For now, I am drawing these walls as solid, flat surfaces. If I ever use this drawing for 3D-printing, then I'll have to take the time to produce an individual-boards version of these, but that is a problem for future-Peter to tackle.

There are two windows in the ends of the body, and they are identical in size and vertical location to the ones in the side walls. These are 6-3/4" away from the their respective corners of the body. For now I am just drawing an opening in the wall, but eventually the design will be flushed out with the actual window sills, muntins, and framing.

Next are the doors. The exterior body goes down to the bottom of the frame of the car. However, there is an internal floor that sits on top of the frame. So, the end platforms not only have steps that help the conductor to get up into the car, but also to get to the vertical position of the floor. All that is to say, that the door sits raised up from the bottom of the ends. From prototype drawings, I concluded that the door measures 6' tall and is 25" wide. Its bottom is at about 15" from the bottom of the exterior boards.

The purple roofs have been added to the diagram. The color is just to distinguish it from the rest of the car in this diagram. Since I am building an offset-cupola version, there is a short roof and a long one. There is an overhang of 28.125" over the end platforms. This then makes the short roof be 8'9-1/8" (105.125") long, and the long roof 14'9-1/8" (177.125") long. The spacing in between, for the cupola, is 6 feet long.

(drawing is in the process of being revised) The long roof takes up the remaining non-cupola space, which winds up being 153.25" long.

The interior floor sits up at 6-1/2" from the bottom of the exterior walls, as best as I could determine from the prototype drawings. I then accounted for a floor to be 1" thick. So, the bottom of the interior side walls starts at 7-1/2" from the bottom of the exterior walls, as drawn here in orange. The interior of the car also uses the same board slats, but they are oriented horizontally. The Kappler Mills wood that I'll be using is 1/16" real-world thick, which translates to 4" scale inches in S. Gluing the interior wall to the exterior wall will yield an 8-inch thick wall, which is pretty close to what the prototype's total wall thickness is (they used 6x6 framing and then 7/8" exterior and interior boards, which comes out to close to 8 inches).

(drawing is in the process of being revised) The roofs actually extend out from the main body by another 2'4-1/8" according to the prototype drawings (the platform below it is actually 3 feet wide). So, I extended the two roof sections by 28-1/8" to get them to their final lengths (109" and 329", respectively).

(drawing is in the process of being revised)

(drawing is in the process of being revised) Now that we know where the bottom of the door is, we can draw the interior floor to match (shown in red).

(drawing is in the process of being revised) With the floor established, the interior walls can now be modeled and added. From prototype drawings and restoration project photos found on the Web, it appears the walls were constructed of a framework of 4" boards, with the exterior slats mounted to that framework, and with interior slats mounted to the inside of the framework. As described above, the exterior slats ran vertically, to aid in rain water resistance. On the interior, however, the slats ran horizontally. Together, that makes for a very strong and stable wall. So, theoretically, the walls should be 6" thick, but my experience has shown that wall thickness doesn't really translate well into miniature models. So, I made the design of the interior walls be 3-1/4" thick, so, combined with the 3/4"-thick exterior slats, the walls in the diagram are 4" thick, total. This is important, as the windows are inset into the body, so that must be accounted for in the actual construction later on. The diagram shows, in orange, the two interior side walls added. Note how the bottom of the interior walls are shorter, which matches up with where the floor sits (hidden from the design here to better show just the interior walls).