Saturday, May 29, 2010

Precision Machining

I am increasingly pleased with my ability to make stuff very accurately.  I received an order for some ScopeRollers for the Celestron CI-700 tripod a few days ago.  Much of the work on machining a part is the setup: putting the cutting tool in place; installing the drill chuck in the tailstock; positioning the tailstock; zeroing the handwheels.  Therefore, if you are going to make three parts, making nine identical parts isn't even twice the work. I wasn't putting any extraordinary effort into this, but because I have become fairly adept at this, I turned out nine perfectly squared cylinders that ranged from 3.599" to 3.602" in length; that's a .08% variance in length.

6 comments:

  1. Neat!!

    Here in Wisconsin, however, tolerances are only measured in "10ths".

    That means 10,000ths of an inch.

    But if you're good with trig, I can get you an interview as an apprentice machinist!

    Seriously--that's good stuff, though.

    ReplyDelete
  2. I suppose it depends how accurately stuff needs to be made. In this case, .1" would be sufficient on length. The diameter is where it matters--and even here, .005" is adequate.

    I've thought that if software engineering went away, I could probably get a beginner job as a machinist. Of course, teaching me to program a CNC controller wouldn't be any great struggle.

    ReplyDelete
  3. careful there, Clayton. "Pride cometh before the fall," and all that.

    As a woodworker I can only imagine those variances. 0.003"? The relative humidity can affect wood more than that on a 3.6" length. Sheesh!

    ReplyDelete
  4. teaching me to program a CNC controller wouldn't be any great struggle

    No...but then, having a CNC running Unix with native C and able to print/fax/copy while doing 5-axis cutting might be unique...

    Keep it in mind. Good machinists make decent money.

    ReplyDelete
  5. "The relative humidity can affect wood more than that on a 3.6" length. Sheesh!"

    Fortunately, acetal is not much affected by temperature changes. Turning the diameter of these cylinders isn't quite as precise, because you have to very precisely center the cylinder in the chuck, and even then, you get a bit of variation from end to end. For eight of the nine cylinders, I ended up with a variation of 2.743" to 2.478". One of them I screwed up and ended up with 2.740" to 2.743".

    ReplyDelete
  6. Year or so back bought an old Sears benchtop lathe, about 18" bed, at a flea market. And it's been kind of fun, and a little amazing, being able to turn stuff I need or want.

    ReplyDelete