January 2011 Archives

Version 1.2 - CNC upgrade update

|
| Comments (0)
A couple minor updates. Made use of the new power supply and moved the unit, PSU and computer into the corner of the shop so I don't keep tripping on wires.

My shop - CNC

Upgraded the bridge axis. Added a bearing to take lateral load off the flex hose connecting the threaded rod to the stepper. The bearing mount is 4 sheets of 3/16" plywood - cut by the CNC - how cool is that! The bearing is sandwiched in between. The stepper is further offset by the old plate to reduce tension from any wobble. I can now move this axis at up to 7 inches per minute - previous rates were around 6 in/mn. The stepper can only turn at 7/min in free air (no threaded rod attached), so this is pretty good.

Stepper mount version 2

Another touch is to mount the power transistor / driver board so the cables don't flop around so much.

Driver mount version 1

Finally - a small reminder that I do not have all the bells and whistles installed. I really need limit switches. A recent job ran the table to the end and started to pull the flex tube off the stepper shaft. Yikes!

Table limit hit!

That's all for now!

Math homework - first CNC project complete!

|
| Comments (0)
So my son was watching doodle youtube video clips in math class (his teacher was trying to teach something) and as an assignment was asked to doodle in math class - how cool is that. 


So my son takes it a step further and explores the over / under principles described in the video above, creates a doodle and then using Blender, creates a graphical 3D render of the image.

curvy_final_01.JPG

Then I look at my CNC machine and have the urber nerdy idea to actually render it real 3D using wood. The first step is a Z-pass filter to convert the vertical profile into a grey scale. White is the bottom, black at the top.

curvy_z_pass_01.PNG

Next I used EMC 2.4 / Axis (my milling software) to raster scan the image and generate g-code to mill out the wood. At first I tried a 2 x 2 inch square, but the details were lost.

P1200041.JPG

Then I went for a 4 x 4 inch square. At 2 hours per pass, 7 passes in total, it took just over 14 hours of milling - but here is his math assignment rendered in beautiful cedar!

4 x 4 cedar math doodle


More lessons learned - slippage is bad

|
| Comments (0)
After many, many, many hours or milling I spotted another fail this morning.

X axis fail.

The X axis had stopped working! What went wrong? I started looking at the electronics wondering if I had lost power, ground a signal, something. That was all good, then I looked closer at the stepper and noticed that the motor was turning, however the coupler had slipped off and the tie wraps were jammed against the frame stopping the threaded rod from turning.

Stepper coupler fail

The fix has been to grind a key into the stepper shaft and use tie wraps on the stepper shaft.

Fail! - or win? This is why we test

|
| Comments (0)
TorusTestFail
And this is why we test things. Although I'm not 100% sure what broke or failed, the end result was the Z-Axis not raising. After many a number of hours of successful milling, while on the 3rd pass, the Z-axis stopped raising. Each row would start down from were it left off. The final swerve was most likely a result of being too deep for the rapid no-milling return movement.