Thursday, 23 September 2010

First Speed Test

Last night I had a free moment and took the z-axis for a little test drive. I threw together a quick and dirty interrupt service routine for the Arduino to see how things would work out.

The good news: I'm able to take the motor up to 50,000 microsteps per second (5,000 full steps/s, or 25 revolutions/s, or 1,500 rpm), and that's with a very rough acceleration algorithm. At a threaded rod pitch of 1.75mm, this yields 44mm/s or 2.6m/min. Acceleration was comfortable at 120,000 microsteps per second per second, though I didn't try to push this further due to lack of time. All up, this is a (so far) pretty good result. It comes close to what I would hope to achieve with an Arduino. Things might change under greater load.

First z-axis test run. It won't compete with Hollywood,
but it's my first YouTube upload.

The not so good news...

First, there is something wrong with my nut mount. I think the clamp isn't tight enough so it moves a bit. This also shows up as a slight modulation in the actual sledge movement velocity. I will have to investigate this on the weekend. I believe the culprit is the depth of the groove in which the nut sits. If it's too deep, the clamp won't have enough bite.

Second, the threaded rod seems to have a mild "bend". I'm not sure if this is just the rod, or if it is a sign of mis-alignment of the nut and the two bearings. Again, the weekend will tell me more.

Third, there are several distinct levels of resonant frequencies. Probably speeds I will want to avoid once the system is in action. Note to self: Avoid resonant frequencies/speeds.

To close the post with some good news: The stepper and the entire construction feel quite sturdy. It basically doesn't notice me at all when I try to apply reasonable force against the movement (more force than I would apply using a power drill, anyhow). Very subjective, but encouraging.

No comments: