EMD F7 in SCALE
————————————————
In order to model something you need to know something about that something. So my first step after deciding to model an F7 was to learn as much as I could about it. I'm not exactly ignorant of what goes on inside a locomotive, but I did need specifics that would assist me in making good design choices. For example, I needed weights and dimensions in order to start making some sketches and translate a 50 foot long, 230,000 pound locomotive into a reasonably accurate scale equivalent. The
Ixquick Internet search engine got quite the workout, saving me the indignities of having to paw through endless books and take endless notes during my information quest.
One of those searches netted me an image of an old EMD print for the F3, whose exterior dimensions are identical to the F7. I enlarged this image so I could reproduce it on two letter-size pieces of paper that could be taped together (I didn't have a large format printer available to me at the time). With that done, I could convert the full size dimensions to their scale equivalents—programmable pocket calculator to the rescue!
- F7A Dimensions Worksheet Cab End
- F7A Dimensions Worksheet Vestibule End
With basic dimensional information at hand I now knew roughly how big this beast was going to be—a bit more than 6-1/2 feet over the coupler pulling faces, and exactly 4 feet from one truck kingpin to the other—and could turn my attention to just how I would build it. Here is where I made some very specific choices.
Most gas-engined model locomotives are more like riding lawnmowers than locomotives when viewed from a performance perspective. That wasn't at all what I wanted. I wanted a locomotive whose behavior would emulate that of the prototype in all respects. This meant that the engineer's controls and the propulsion system could not be the typical riding scale setup. Here are some of the features that I felt would be required to emulate the full sized F-unit:
- Power and speed solely controlled by a notched throttle with idle and eight power positions.
- Separate reverser with neutral, working independently of the throttle.
- Air brakes on both loco and control (riding) car.
- Automatic propulsion transition with manual override.
- Indefinite relationship between prime mover RPM and ground speed.
The above items would all be factored into the design, and would in turn, dictate how some things get executed. I'll go over all that in detail later on.
I had already made the preliminary decision to power it with a Briggs & Stratton V-twin engine, whose general design and performance characteristics are commensurate with what is required to power a large-scale model locomotive. However, before I could commit to the V-twin I needed more information about it, especially its physical size and wet weight. A couple of phone calls resulted in me acquiring some DXF CAD files of the V-twin.
- Briggs & Stratton V-Twin DXF Drawing
The V-twin is mostly built to the metric system, which meant I found myself converting many dimensions to English sizes, resulting in more work for the programmable calculator.
Speaking of DXF prints, dimensions and such, and before I get too far along, I should mention something about drafting, as a lot of drawings (several hundred, in fact) were produced as this project proceeded. I use a relatively inexpensive two-dimensional CAD package called
DeltaCAD to do my mechanical drafting.
¹ I have never liked AutoCad and as Microsoft's MS-DOS morphed (or degenerated, depending on your perspective) into Windows, AutoCad seemed to get clunkier, slower and more difficult to use. Everything I do in mechanical drafting is two-dimensional, so I don't need the capabilities (and expense) of three-dimensional CAD. DeltaCAD has worked out very well for me and I highly recommend it—versions exist for both the Mac and Windows. The below image was taken from a wheelset drawing I did in DeltaCAD, using the built-in export-to-image function.
- Example DeltaCAD Drawing
Getting back to the F-unit's prime mover, the V-twin presents several engineering challenges, the most significant being that it is a 90 degree design and hence is pretty wide across the cylinder heads, making for tight quarters inside the unit (if only Briggs had used a 60 degree Vee—which I'm sure they didn't do because a two-throw crankshaft and a separate counterbalance shaft would have been required to produce a smooth-running engine). As the V-twin is air cooled, I had to ensure that adequate engine room ventilation was available, as well as carefully plan the machinery layout to keep things that get hot away from things that don't like heat.
The position of the oil filter on the V-twin is awkward, increasing the effective engine width to 16.75 inches—an oil filter sticking out of the side of the loco would be quite unsightly.
The filter's obtrusiveness would have been a show-stopper if it weren't for the fact that its mounting bracket is removable and Briggs makes a remote filter mounting kit available. So one of the things I would have to do was find a new home for the oil filter, keeping in mind that it had to be readily accessible for routine servicing, and that it would be one of those hot things to be kept away from things that don't like heat.
Another consideration with the V-twin is its overall height of approximately 17-1/2 inches. In scale, the F7's height over the railhead is 22-1/2 inches, measured at the horns. It was clear without even firing up a calculator that the V-twin's mounting in the loco would have to be relatively low to maintain interior clearance. Lowering the prime mover would mean that everything attached to it that had to be coincidental with the crankshaft centerline would be lowered as well, creating potential clearance issues with other parts of the loco's structure. As an aside, the dry weight of a 16 horsepower V-twin is approximately 90 pounds, and that weight is slightly asymmetric relative to the engine's crankshaft centerline.
Mentioning "structure" brings up yet another design point. The real F7 did not have a separate frame as is the case with modern locomotives, instead using a truss structure in the carbody to act as the primary load-bearing element of the unit. The body panels were riveted and bolted to this truss, adding some rigidity. Removable hatches in the roof that were attached with a gazillion bolts gave access to the engine room for removal and installation of the machinery (yep, the prime mover—all 30,000-plus pounds of it—had to be hoisted through the roof). Three such hatches were present: the main hatch, which also supported the radiator cooling fans; the dynamic brake hatch, which supported the dynamic brake grid cooling fan (if dynamic brakes were fitted); and the steam generator hatch located at the number two (vestibule) end of an A-unit.
From a modeling perspective, the F7's body design isn't very practical. If I had used it I would have been forced to install and remove the prime mover through the roof like the original, which given that the width across the cylinder heads approaches that of the body, would have resulted in an appearance grossly different than the prototype's. Servicing the machinery would have been very cumbersome. The only practical design is to use a rigid frame and relegate the body to a primarily cosmetic role. As it would be necessary to mount the prime mover and attached machinery relatively low in the unit, a drop-center frame would be necessary. I further decided to mount the body to the frame so that it would open like the body on a drag racing funny car, which would give access for refueling, routine maintenance and light repairs. The body would be completely removable for heavy repairs, just as is done with modern locomotives.
However, before I could design the frame, I needed to design the propulsion system, especially the "power assembly," which is what I call the amalgam of the prime mover and power transmission pieces that are attached to it. That will be described in my next post.
————————————————————————————————————
¹
Regrettably, DeltaCAD has been discontinued. The final version was 10.0, released in 2020.