Toward an F7
27 JULY 2010
It's been on my wish list since December 2007, when I got my first set. That old 103 chassis was just begging to be shortened to make a small North American cab-body diesel. But that tempting gap between the gearbox and the front truck would only close just so far—and not far enough, as it happens. The minimum truck spacing was 40 scale feet; F7 trucks are spaced 30 feet apart.
It was just as well; the 103s are temperamental little beasties, even when unmodified. The better approach would be to fashion American passenger coaches from stock powered 103s, and make a dummy F7 to push around. Back in 2008, Michael Denny of Canada made an F40PH that worked this way. (Sadly he lost interest in T Gauge owing to performance issues.) Unfortunately this pretty much precludes running anything but passenger trains.
When the second-generation mechanisms finally arrived in the Spring of 2010, it helped rejuvenate the interest of many T Gaugers who'd gone idle—myself included. The new mechs offer improved performance and reliability. Bashing possibilities are improved as well, since the drivetrain is self-contained, allowing one to pop on a whole new shell without having to worry about the gearbox and electrical contact strips, as with the 103s.
And yet the opportunity for a small, self-powered diesel remains elusive. In some ways the new mechanism is harder to modify instead of easier. The truck spacing on the 9000 is shorter than that of the 103, but only by a mere three millimeters, and owing to the 9000's extended gearbox, the motor prevents the non-powered truck from moving any closer than it already is.
In spite of all these obstacles, I remained unconvinced that the second-generation mechanism locked us into a long wheelbase. The more I studied it, the more it seemed as though there was an opportunity waiting to be discovered. And ultimately I found it... just not where one might have expected.
While it may seem counterintuitive, the limitation on truck spacing is actually all due to the electrical connector on the motor. It's designed such that the motor can make contact with a pair of metal strips after it's simply dropped into the chassis. This makes sense: there are no wires to solder or otherwise connect manually. Assembly is simple, reliable and supremely economical.
But the size of the connector imposes a physical limitation on the mechanism's design by forcing the motor to sit down low between the trucks; otherwise, the connector would sit too high, and wouldn't fit within the shell. And so, with the motor dropped down below the frame, the trucks cannot be positioned any closer together than they are.
So, what would happen if the motor could be moved up higher? Well, the front truck could then slip all the way under it. In fact, the truck spacing could be made to match that of an F7! Wouldn't raising the motor up higher require removing the contacts? Yup. But from the standpoint of making modifications, it doesn't matter: as a modeler, I'm free to solder wires or do whatever I need in order to deliver power to the motor; I'm not constrained by the economics of mass-production. As it turns out, a little bit of "magic" happened along the way that suggests my modification could easily be mass-produced—but now I'm getting ahead of myself.
The acid test to determine if this mod was going to be feasible was if the crown gear could be repositioned: it had to be raised up in order to remain aligned with the raised motor, and thus it also had to move back so as to remain properly meshed with the first compound gear. As it happens, this turned out to be infinitely easier than I'd expected. The slots for the retaining tabs are wider than the gear shaft, which permitted the gear to shift up and back; it was then a simple matter of modifying the retaining tabs to hold the gear in its new position (shown above, with the new position marked in green).
With the first and biggest challenge neatly addressed, the remaining modifications were relatively straightforward. I started by cutting a Kiha chassis in two places: behind the motor bearing slot and behind the motor depression (above, marked in green). I used a jeweler's saw to make the cuts. I was surprised by how easily the engineering plastic could be worked—usually this type of material is difficult to cut cleanly. Then, using a sharp knife, I sliced off the crown gear enclosure from the bottom of the rear chassis part. The result is shown below.
The shorter chassis part had to be modified to accept the motor. I made a U-shaped cut with the jeweler's saw to clear out all of the plastic from the column that supported the electrical contact strips (below, first image); then I "fine-tuned" the opening with a sharp knife until the motor would drop in (below, second and fourth images). Speaking of the motor, I removed the whole contact assembly by slicing through the gold-plated tabs right at the edge of the motor case with a diamond cutoff disc in a Dremel tool; the plastic part and the remaining tabs just snapped off easily (below, third image).
The gearbox lid also needed some more modifications. First, I clipped off the motor retaining tab with a flush cutter. Then I cut a relief opening for the top of the relocated crown gear. This was likely the trickiest process of all: it required mounting the part in a vice, inserting a jeweler's saw blade though the existing opening, and then connecting the saw blade to the handle (an exercise that would be infinitely easier if humans had three hands). Once this bit of trickery was done, I cut the slot (below, outlined in green), then released the saw blade.
The next major step was to join the two chassis parts. I spent considerably more time pondering this process than actually doing it. The problem is that engineering plastic is virtually impossible to bond with adhesives. I very nearly made a connector plate out of thick brass, which would have provided a nice low-center-of-gravity weight boost, but thought better of it as I'd have to drill multiple tiny holes in it. So, I opted for plain PC board instead, which is rigid yet easily worked and drilled.
After slicing out the new PC board "fuel tank," I drilled four #62 holes to clear 000-120 screws; then I drilled #71 holes in the chassis parts, and allowed the screws to self-tap into the plastic (above). Although I would have preferred tapping the joiner plate for the screws, I don't own a 000-120 tap; and besides, in a turn of good fortune, the engineering plastic seems to hold the screws very well.
With the chassis now whole again, it was time to begin assembling the mechanism. The motor would be a challenge, as it needed to be secured in position somehow. I really, really wanted to avoid the obvious: gluing it in place. While it would be a quick fix, it would also be risky, with a chance of CA getting inside the motor and ruining it. Plus, I was certain I'd need to dismantle things—perhaps several times—over the course of finishing the mechanism.
I wanted a "clean" solution, and I found it in the leftover piece of chassis (fortunately I have this habit of not throwing away scraps). I extracted the cylindrical post from the leftover, leaving a tiny tab of plastic on one end (above), and drilled a #71 hole in the chassis behind the motor for a 000-120 screw. The tab locks the back of the motor in position, while leaving the remains of the electrical contacts exposed (below).
This quickly brought me to a defining moment in the modification process: testing. I was now able to power the motor and see if all of this work was worth it—or if it was all for nothing. Readers can likely guess the outcome, since I wouldn't have brought them this far in such a long and detailed saga only to have the story suddenly end on a sad note (even though I've done this in the past). As I applied some juice to the motor, the mechanism instantly sprang to life and hummed away as if nothing had been done to it.
There's nothing like success to build confidence and enthusiasm. It made the final challenge—re-attaching the electrical strips—go quickly and easily, with an added bonus. My original expectation was to solder wires to the remains of the motor contacts, and then connect the wires to the contact strips. But instead I discovered that the tabs on the ends of the contact strips meant to power the LED headlights could be modified to press against the back of the motor, eliminating the need to solder any wires (below). The only other step I had to take was to apply a bit of tape to the motor case to prevent the truck mounting springs from shorting against it.
In addition to saving me a lot of trouble making tiny soldering joints, this little unexpected bonus spoke volumes to me: it demonstrated that, assuming motors could be obtained or made with truncated electrical contacts, a short power chassis could easily be manufactured at the same cost as the current design. Incidentally, I re-attached the contact strips to the gearbox by slicing off what was left of the melted mounting tabs, drilling #71 holes where the tabs were, and threading in 000-120 screws, which I clipped off underneath the gearbox lid (below).
The only thing left to do was attach the trucks and test-run the unit on track. It ran so well that I immediately placed it on my little suitcase layout, and it became the first train to run on it—a maiden run on a maiden run. I soon saw the need for a minor tweak: the unit is so tiny and lightweight that the non-powered truck spent most of its time off the rails and being dragged around. I replaced the wheels with a magnetic set from an old 103, and that fixed it.
Naturally I'm now faced with the challenge of making the shell. But by comparison to the mechanism, this will be a no-brainer. My plan is to make a prototype shell from styrene, then possibly translate that into an etched brass version. We shall see; in the meantime, please enjoy this video of my new "baby" in action. She's sure tiny!