Removing a "glued on" encoder disc on a BLDC motor

[chris_leyson]

Today I managed to kill a Hall sensor output on a small BLDC motor, one output is stuck at 4.5V. Should have insulated the bare wire ends, that will teach me  It's probably an easy fix, just remove the rear PCB and replace the broken sensor. Trouble is the encoder hub is glued onto the rear shaft and I can't get the PCB out. Heating the encoder hub to soften the glue is not going be an option because of too much thermal mass.
The encoder sensor is a Avago part and the 250 line encoder disc might also be an Avago part but there is no grub screw to mount it onto the shaft, it's glued on, how dumb is that. It probably takes more assembly effort messing around with glue than simply tightening up a grub screw 
I'm starting to wonder if this was a deliberate design choice to make the motor non-repairable or maybe it was a cost down to save the cost of a grub screw who knows but it's poor design. BTW motor is Hurst DMA0204024B101 aka Microchip AC300022. Made in America, nice motor but shame about the electronics interface.

[chris_leyson]

Forgot to post a picture. There is no way that encoder disc is coming off and that is before I've even had a chance to find out what Hall sensors they are using. There is no diode protection built in, not that I can see so more cost down. It's a good motor but a shame about the poor electronics design.

[amyk]

Solvent might work... depends on the material.

[SilverSolder]

Are you sure it is glued -  could it just be a press fit?

E.g. they could have heated up the aluminum wheel to expand it and then press it on the steel shaft.

[chris_leyson]

I initially thought the encoder was glued on because there is a soft silicone like adhesive on the top and bottom of the encoder disc, I assume this is holding the disc onto the hub. I think a press fit is quite likely as there is a machined recess in the hub and also Hurst Manufacturing got back to me and said I need a special tool to remove the encoder. Given that the shaft diameter is 7.6mm I don't think I'm going to find a replacement any time soon so the encoder is staying put. At least I've still got two Hall sensors working as well as the encoder so I've got information to get rotor position.