Close

Bearings and Axles

A project log for Goliath - A Gas Powered Quadcopter

A BIG Gas Powered Quadcopter

peter-mccloudPeter McCloud 01/29/2016 at 00:200 Comments

In the previous project log, the first of the new rotor pulleys with the new axles had been tested. The new hardware was considered to be good because nothing broke (yes, not breaking is a low bar to set). The next step was supposed to be assembling the rest of the rotor pulleys and axles and doing a more complete test.

Two things happened after that. The first as that there was a question regarding the fit of the axles. It wasn't required to press fit the axles into the rotor pulleys, which shouldn't have been the case. The second was an unidentified noise during the testing. The first hint of an answer to the second question came to me from YouTube user +visnevskiscom, who commented that during the latest video at 2:30 there was a vertical oscillation that appeared to correspond to the noise. After investigating this further, the source of the noise was found to be the rotor hitting the belt occasionally. It turns out that there was slight amount of play between the axle and the bearing to allow the rotor to wobble just slightly and just clip the belt.

This led back to the suspicion that something wasn't right about the interference fit. The axles had just been ordered from McMaster Carr and the drawing states that the axle was 0.625" +0.000/-0.001". The bearings were pulled out of a set of idlers almost two years ago, and I knew less about. Below is the product page that hasn't changed from when the bearings were ordered. (link to site)

So the inside diameter is stated to be 5/8" (0.625"). There is no tolerance set, so in theory the ID could be just slightly greater than 0.625" and the shaft could be a thousandth less and it wouldn't need an interference fit. Since there wasn't any more information from Circular Motion, I looked up the bearing part numbers and did some internet searching. It appears that the bearing is a specialty bearing and was hard to get the specs for, but after a while I did find a drawing. (link to site)

The typical applications is stated to be a 5/8" sprocket idler bearing. However the drawing states that the ID is 0.635" , 0.01" larger than what Circular Motion states the ID to be. I don't know enough about sprocket idler design to know why the bearing is 0.01" larger that the nominal ID or if doing so is a standard practice. What is clear is that the current bearings wouldn't work with standard axle shafts (and that I'll never order another bearing without seeing a drawing sheet)

After a short period of despair, thinking about the amount of time and money that went into a set of rotor pulleys that had useless bearings, a reasonable solution was found to the problem. It turns out that a bearing was available from McMaster-Carr that not only would fit the shafts correctly (0.625 +0.0000/-0.0005), the rotor pulleys just needed to be bored out just slightly larger to accommodate the new bearings AND they're cheaper.

So after ordering a new set of bearings, boring the bearing pockets to be larger, and installing the new bearings, the hardware was reassembled. This time the bearings and the axles required a press to fit them together:

Test 29 was mostly a repeat of Test 28. This time around, two of the new rotor pulleys were tested, both mated to the new axles. The latest revision of the flight rotors were also used.The test went well and there were no unidentified noises. The rotor bolts continue to hold up. The next step is now to complete the other rotor assemblies.

Discussions