
My intention was not to make an ultra-accurate, prize-winning model; it was to experiment. In particular, I had an idea of how to imbue the tiny little thing with detail I could not possibly fabricate physically. My thought was to make a near-full-body decal that would present the grilles, seams, rivets and other relatively subtle surface details on the sides that would ordinarily be no thicker than paint in T scale if modeled accurately. The only details I'd have to fabricate would be the fans on the top.

I was quite surprised by how thick I could make the shell. The math was straightforward: the real thing is ten and a half feet wide, which is 0.28 inches in T. The mechanism is 0.20 inches wide, leaving a difference of .080, meaning the shell could be made from .040 thick sheet styrene. This turned out to be quite an advantage: it allowed me to easily file and sand all of the many curved surfaces.

As I predicted, the nose was a bit of work, especially since it had to fit around one end of the mechanism. I had to push the windshields forward several scale inches to make room for the end of the motor. But at the model's tiny finished size, I doubted this discrepancy would be noticeable.

I'd hoped that I might be able to complete the shell without having to resort to filler putty, but this was only a fantasy. I had to shape the nose twice to get it right, as it was misshapen the first time around: the headlight wound up way too low, and it looked like a Baldwin babyface. So I slathered on the Squadron putty and had at it again.

While the putty was drying, I made the cap-top fans and other details. Then I shaped the nose once more. Along the way, I learned a trick to compensate for the fact that the putty is too soft and porous to hold small, intricate shapes: I soaked the surface of the putty with CA, which sealed and strengthened it.

Finishing the little beastie was made harder by virtue of my choice of road name: CNJ, in the classic tangerine and navy scheme. I could have made things much easier on myself and gone with something like all-black PC. But I chose CNJ for a specific reason, and it wasn't because it's my favorite road: the yellow-orange stripe down the side makes the grilles clearly visible.

Using photographs I found on the Internet, I made four decals: the two sides (both made from the same photo, with the other side just flipped—with the lettering corrected), the lower nose, and the windshields. Printed on white decal material, the images were startlingly sharp for their size: the lettering was all still readable, and the windshield wipers were clearly visible.

The toughest part of the finish work was the nose, which couldn't be covered by a decal, and had to be solid yellow-orange. Unfortunately, I neglected to prime the dark green putty first, and so I wound up applying coat after coat of paint (which turned out to be a good thing—it helped smooth the still-rough surface of the putty). Making matters worse, I didn't have the right shade of blue paint on hand. In the end it was not a pretty sight, and I almost declared the shell ruined by virtue of a botched paint job. But, remembering that this was a proof-of-concept project, I decided to keep it, and disguise the flaws with a healthy dose of weathering.

For a proof-of-concept model, I think it came out rather well, even though it's technically not even finished—I've yet to make the fuel tanks and the front coupler. One little letdown I encountered was that there wasn't room for an LED to make the headlight function. It could conceivably be done if the shell around the nose was thinner; out of necessity, this one is all solid styrene.

Although I have my doubts, I do hope that someone at Eishindo is following my blog. This model would be so easy to manufacture—the mechanism modification would not be hard to replicate, and the shell would be a no-brainer. And since Eishindo ink-jets their shells, they could produce no end of colorful paint schemes.

Here's a video of F3 #56 in action. Please excuse the fact that I erroneously refer to it as an F7.
I don't have the right words for your work but it's awesome!
ReplyDeleteAs usual, bloody fantastic. What are your plans, if any, for boxcars and such? On one hand I can't imagine it'd be economically viable to use EMU sets for a whole train, but on the other I'm not sure whether the single unit would have enough guts based on your previous posts?
ReplyDeleteWill, over time I hope to make some freight cars. But I'm also thinking about making an F7 ABBA lashup, which should address any issues of pulling power.
ReplyDeleteDavid-
ReplyDeleteVery cool. For the next iteration, and assuming you stick with the F3 artwork, the F3 rooftop details should be even easier to make than the F7 ones you included. Tall boxy fans, and rectangular grills instead of the round dynamic brake fan. And I'm not sure the CNJ F3's had steam generators, so you might be able to leave those details off as well.
@skibbs4 Actually, the rooftop details are correct for later phases of the F3 (including #56), except for the steam generator. I do have a drawing of an F3 with a steam generator, but I have not found a photo of one yet. So, admittedly this model is a foob.
ReplyDeleteThis is legendary.
ReplyDeleteBut quick question: Why couldn't you adapt your process you used for your crossing gate for a single headlight?
@Nathan Installing any LED would have meant grinding out the interior of the shell to make room for an LED plus resistor plus wires (it wasn't a matter of just installing an LED). And I didn't want to solder the LED, etc., to the chassis, so I would have had to devise and make room for connectors, too. As this was a proof-of-concept to see if I could make the shell, the headlight was not a major concern. Next time...
ReplyDelete