Thursday 13 February 2014

Concept Development - Autofocus 2



Maybe it's because this part is tricky to work out, or maybe it's because I was doing this at 8am having not slept all night, but starting to build the rails for the autofocus system took longer than I was expecting. To ensure the lens barrel moves smoothly, I am arranging two steel rails at the top and bottom as close to the optics as I can get (without piercing through key components around the sensor, such as the shutter in blue above).  Having the rails as close to the optics as possible and having them firmly fixed to the body at their top and base are both high priorities. To begin, I extruded the rails where I wanted them to see where they would fall on the uni-body.





Oh dear, they seem to fall in a gap, filled by the shell of the main shutter button at the back of the camera. With the rails unable to be located anywhere else (due to the sensor and shutter being rectangular - longer horizontally then vertically, and therefore only leaving enough space top and bottom) and the shutter button also unmovable, I had to get creative with the 3mm of space between the unibody and the back of the circuit board.




I firstly put a large spare across the button hole, and modified the button so that it could be snap fitted into place from the outside rather than dropped in from the inside. I also created a hole allowing for the button to penetrate through and be able to press it's electronic switch underneath (circuit mounted facing out).




Thinking I had come up with a clever solution, I examined the design in cross section only to find that it of course, didn't work at all. With the button having a subtle shallow to it's top surface, the inside wall thickness was resting against my new spar. This meant that the button could not be pressed, as it simply had no where to go. A re-thinking was in order, and after blindly hacking at the geometry, I ended up with a raised profile spar running across the surface of the uni-body, with space for the button to be pressed, and all fitting within the 3mm gap. In fact, with such tight space, I raised the spar to 3mm and decided it would form part of the support holding the board up.




Not only this, but the bosses into which the rails are secured penetrate through the bottom board and support the sensor board at the correct elevation. This helps kill 3 birds with one stone: a firm rail location point, a spar to keep the main circuit board raised, and bosses to keep the sensor board raised. Below we can see the final arrangement at the back.