This is an engineering-notes blog. I plan on posting regular updates on my day-to-day happenings in a very casual format.
Playing with the OxBox Focus Controller
The spec for the OxFocus controller requires it to operate without functional limit sensors. This seems like a good idea, but in practice, the motor displaces the pin on the anti-backlash bearing if it runs into the physical limit. I tried adjusting the motor current down so the motor would stall before physically moving the pin, but the acceleration/deceleration is so high that lower motor current sometimes causes the motor to stall even when it is not physically restricted.
I couldn't find a motor current setting that allowed the stage to move reliably (no stalls during normal operation), yet stall easily (no displaced bearing pin) at the physical limit.
1.0m Mirror Cover Controller
A new Superfin-based controller has been built to control the 1.0m mirror cover and Hartmann screen. After quite a bit of work making the creation of new Superfin-based uClinux projects a simple process, the groundwork is paying off. Following a short procedure I put together, which mainly consists of copying an existing project, a new controller framework exists! The real work is actually in coding the logic that's specific to the controller.
However, during the copying and modifying, I wound myself up in Subversion hades, and had to manually copy the new files out, remove the offending directory, re-copy the project directory, then copy the new files back in. Subversion errors are not fun to decipher, so always remember to use svn cp and svn rm when you're moving svn-controlled files instead of regular system cp and rm!