The goal is a roughly two-inch diameter wooden ring that will surround a pinewood derby wheel. The ring needs to provide about 2 ounces of weight, to simulate the load on a wheel during the race.
For my first attempt, I tried using a simple hole -- a 2" diameter for the outside, and a 1 1/4" diameter for the inside. It turns out my hole saw, and I assume others commonly available, have a lot of runout, and are unsuitable for attempting to do work at any real precision. (They make a perfectly fine hole for a door knob, and that's their main function, I think.) I was unhappy with the ring that resulted, as it had on outside diameter of about 1 7/8" or less, and an inside diameter that I felt was too large.
I also discovered that a ring of these dimensions that's made of wood is way too light -- about 0.3 ounces, rather than the desired 2 ounces. So I needed to find some way to add some significant weight to the ring.
I ended up using my router and a home-made router "table" to fashion the ring.
The router "table" is really just a piece of 1/2" MDF that attaches to the router base. There are also two adjustable fences, also of 1/2" MDF. (This set-up is mainly for routing out weight pockets in the bottom of derby cars.)
I put a nail through a piece of 3/4" x 3/4" scrap, and then drove the point of the nail into the middle of a piece of stock for the ring. The scrap is then clamped across the top of the two fences, and acts as an overhead arm that lets me locate the pivot at any desired distance from the router bit. (I was concerned that this basic nail-as-pivot arrangement would develop some play during use, but this turned out not to be the case.)
One thing I didn't pay enough attention to initially was managing the arm's travel in the Y dimension (left to right movement in the picture above). It's obvious with a little thought that the pivot-to-bit distance depends on both the X- and Y-location of the pivot, so it's important to mark where the arm crosses each fence. It's also important that the fences be parallel to each other, or the marks won't really mean anything when you change the X-location. After the picture above was taken, I added some F-clamps to the arm itself, to hold it absolutely steady.
In hindsight, setting the bit-to-pivot distance would have been made easier if I'd first centered the pivot on the router bit, without the stock in place, and marked that zero point on both fences. The pivot-to-bit distance could then be established by clamping the arm at a measured distance along both fences. Having not done this, I had a set each distance by trial and error.
I first cut the outside diameter with a 1/4" bit, raising the bit 1/8" with each pass. On the fourth pass the bit cuts all the way through the 1/2" stock, and you have a disc where previously you had a rectangle.
I then switched to a 1/8" bit, also raising 1/8" on each pass, but this time only cutting a 3/8" deep channel in the ring. I was quite pleased with how this came out, really much better than I would have thought possible (at least for me).
Finally, I switched back to the 1/4" bit and cut through the disc again to form the inside diameter of the ring. Setting the radius for this cut involved a lot of trial and error, since it's desired to fit
the plastic car wheel pretty closely. (As the very last bit of material is removed, the ring becomes disconnected from the pivot, resulting in a small defect in the inner wall. I was later able to
correct this with a rotary tool sanding drum.)
I then cut a strip of lead flashing 3/8" by 12", and inlaid it into the routed-out channel in the ring by wrapping around twice. (I find that my cheap Harbor Freight paper cutter works great for making strips of lead flashing.) The ring now weighed about 1.4 ounces; 1.5 when the derby wheel is added.
To make up the last half ounce or so, I considered applying a ring of lead flashing, seen here. This would have worked (total of about 2.2 ounces), but the lead ring looked terrible. (I'm reluctant to use things like a rotary tool because of the lead dust and chips that would be generated. Unfortunately, tin snips and the like leave a pretty rough edge when trying to cut circular lines.)
Instead, I decided to try widening the channel in the ring, and adding another strip of lead. With the original pivot no longer available, I added a third fence to my router set up, and used the arm-and-nail pivot as a point fence in the middle of the ring. (I also moved the side fences in so they just touched the outside of the ring.)
With these fences all in place, I was able to use the 1/8" bit again to
widen the channel. Unfortunately, there was some tear-out in the ring's inner wall, probably because there ended up being just a little more play in the set-up than there should have been.
However, most of the wall is OK, and I was able to add the third strip of lead. The ring now weighs 1.8 ounces or so.
Here's the jig with a wheel in place, using three pieces of rubber band as shims. Note that, with only three points of contact on the inner wall, the damaged sections have only cosmetic effect.
After painting, I hope to pursue Stan's suggestion of adding a revolution counter of some kind, probably by cannibalizing the wheel encoder from an old non-optical mouse.
