The Aristocraft track switches are good for the price, but IMHO no way are they perfect. I've previously mentioned the original modifications wherein I cut the point rails about three inches from the end, soldered tabs under the part closest to the frog, then drilled and tapped the underside of the points themselves. The points, then, swivel on a tab soldered to the bottom of the rail nearest to the frog. When the switch changes direction, the switch motor is moving only two three-inch long pieces of rail instead of trying to pull two un-hinged rails that are about 7" long.
And in some cases modifications were made to the frogs and guard rails to try to encourage smoother operation.
However, the performance of the switch motors proved to be less than desirable. Originally they worked fine (especially inside) but exposure to the elements resulted in enough dirt and oxidation being introduced into the motor compartment that they frequently failed to work. The little nylon gears also required an unreasonable amount of maintenance.
So, being inspired by the basic Lionel O gauge track switch . .
I thought about cannibalizing a few that I had acquired in a garage sale. Their are two electromagnets, as you can see above, formed by a pair of windings that are on a nylon spool that has a divider in the center. The plunger in the center is just slightly longer than the length of one of the spools. I dismembered one of the switches, pried the U shaped cover from the spools and unwound the copper wire by re-spooling it on a piece of rod held in the electric drill.
The spooled wire is on the top shaft, half of one of the original switch motor spools is on the lower shaft. I've wound the left side, keeping a heavier concentration of windings toward the outboard end of the spool. A piece of a wine bottle cork is used as a spacer to keep the windings on the left end of the spool. After the left end was wound, I did the same to the right end.
This gave me a two-coil electromagnet on one half of the old Lionel spool. I had four wires -- two from each half of the coil. The two in the center of the spool were soldered together. The wire on each outboard end was soldered to a diode (each facing the other), then the leads to the diodes were soldered together allowing me to control the switch by just running two wires from the Aristo control box, just as they had done with their motor.
At the top of the picture you can see one of the diodes peeking out from some heat shrink tubing. I notched the ties so that the coil could be epoxied into position. The wires run under the ties to screw posts. The actuating rod comes out of the right end of the coils. It's just a piece of 1/16" brass rod. It was telescoped into a piece of plastic tubing, which in turn was slid into a hole bored into the steel armature that rides inside of the coils. That way the rod is electrically insulated from everything except the power that's applied to the switch point.
A view from the top of the switch. I originally was going to use 6-32 brass screws for the terminals, but opted for taking the knurled nuts from the Lionel switches. They're a little less obvious. See below.
Coils were given several "generous" coats of paint, red initially then flat black to make the machine less obvious.
The coil is about 3/16 above the top of the tie but still well below the railhead. There is no "latching" mechanism so in theory the points could shift under a train, but since my typical consist is four or five cars that hasn't been a problem yet. Also, they work fine as a floating trailing point switch -- no annoying derailments because the wheels just push the points aside. Pieces of 1" PVC pipe were cut in half lengthwise and glued under the switch motor in an effort to keep it away from stray pieces of ballast, water, etc.
I had an Ozark switch stand so fitted it to one of the switches. The tell-tale is connected to the points by a piece of spring steel wire and flips back and forth as the electromagnets pull the points back and forth.
These like to have about 16 volts to throw properly. I didn't make a science project out of the windings -- some coils have more turns; others less. Some of the coils emit a slight hum when energized; others don't. But when I push the button, the switch throws. And that's what matters!
Note: Aristocraft is one of the firms that has gone out of business. The switch motors have worked pretty good, although I do need to remove and rebuild one that seems to have frozen up. It was on a siding that was never used. It will be repaired and kept as a spare. The Ozark switch stands are nice although a bit fragile for outside use. I did add a capacitor to the power circuit so the coils get a bit of a 'blast' of electricity when first energized.