Both #10 and #622 have split-frame current collection whereby the wheels on each side are connected electrically to the frame on that side. This means that the two ends of the axles are in turn isolated electrically. In the case of #10, I used split axles available from the EM Gauge Society, while for #622, I made my own from stainless steel tube and rod. Both types of axle require a layer of epoxy to isolate an inner rod from an outer tube. In the case of #622, I’m pretty certain this layer has been the failure point.
Up until now, I’ve believed that this failure was likely due to water or oil trapped in the end of the tube. However, the experiments between holiday busy times over the past couple of weeks have exposed several other alternatives.
My initial speculation was that holding onto a round 1/16″ (1.6 mm) inner axle is a lot to ask of an epoxy join. I felt that it would be better if the join were mechanical as well as chemical, and so I tried filing a flat on the rod and cutting a hole in the tube, and filling both with the same epoxy as the rest of the join. In the table below, I’ve called that configuration a “D” and a normal round rod “Plain.”
I used the same hex-head wheel apparatus as I used for testing wheel-axle joins. To keep all the trials similar, I cut the axle off flush with the “wheel” after the epoxy had set. As with #622, I used a cut-off disc in a Dremel tool to cut the axle to length. This actually melted the epoxy, but it seemed to harden back again once the cutting was complete. Going slower caused less melting.
After that initial set, I replaced the rods, but not the tubes and tried again. However, on the second set of experiments, the JB Weld epoxy contained too much hardener, and did not set correctly.
The last set of experiments enjoyed the optimal conditions: a good mix of epoxy, and a joint that had never been melted.
|1. Cut axle||D||169||450||397||339|
|1. Cut axle||Plain||560||14||440||338|
|2. Poor mix||D||142||136||203||160|
|2. Poor mix||Plain||238||198||160||199|
|3. Good mix||D||334||317||522||391|
|3. Good mix||Plain||526||555||630||570|
The results almost speak for themselves. In total, I found three good ways to make a split axle fail.
- Very carefully cutting the axle does not have a large affect on the joint, but fully melting it can cause what looks like a good join to be no join at all.
- A poor epoxy mix results in a joint that is about half as strong as a good mix.
- Trying to improve the joint by filling shaped voids with epoxy weakens rather than strengthens the joint.