pushing a gear on a fp

[Curt Henderson]

Originally Posted by AdamHeard

All sounds like good advice to me.

I've pressed on a lot of gears in that shaft size in recent times. Another thing to notice is that sometimes the shafts are oversize and the gear is way under. Sanding a running motor with emory paper (cover holes to prevent dust from entering motor) is a decent way to fix this. Otherwise, it's almost impossible to press on straight.

Back in the late 90's I used to work in the PowerWheels Plant as an Automation Engineer that assembled these same motor gearbox assemblies and I can support your troubles in pressing a gear onto the shafts of these motors. I was derictly invloved in the design of the machine that pressed on these gears so I can give you some do's and don'ts. First, you must support the end of the shaft so the motor bearings do not take the axial load of pressing the gear onto the shaft and it goes for the removal of a gear as well. The bearings in these motors are very intolerent to axial loads. Second, I would not recommend sanding the shaft dia smaller for two reasons. First, the ribs on the shaft diameter were intended to deform the gear material during the pressing of the gear onto the shaft to provide interference between the two parts to prevent slipping under loading conditions. Secondly, during your sanding process, you mentioned blocking the motor holes to prevent debris from getting in the motor, by blocking the holes you are blocking the air flow used to cool the motor internels, which is necessary even for short durations, to extend the life of the motor.

The process we used to press the gear onto the shaft was to lay the motor on its side in a vee block, support the non gear shaft end with a fixed anvil, support the gear with another vee block so its centerline is matched to the motor's and then press the gear onto the shaft with a quick pressing motion. This may sound easy, but we struggled with this for quite some time due to the variables in the motor assembly and the gear manufacturing tolerances, but we were trying to do this in a 2.5-3.0 second time frame per assembly.

Hopefully this helps!