Using the stock 0.125 cutter with a 5mm acrylic sheet, I used the default parameters to cut out a 1" x 1" square with a 0.5" diam hole in the center. Measuring with calipers, the part came out as a 0.990" square with a 0.507 diameter hole. When I measure the diameter of the cutter, it is undersized, but the resulting path is oversized. I make custom cutter settings setting the bit diameter to 10 thou larger and the new result was 0.998" square and a 0.498 hole. Does one need to change the apparent cutter diameter for different types of material? Or is there a different setting I should be using to compensate to improve the dimensional accuracy? Thanks for any feedback!
http://docs.carbide3d.com/article/47-measuring-runout
See http://www.shapeoko.com/wiki/index.php/Calibration_and_Squaring_the_Machine#Diamond_Circle_Square_Test for more on that.
Thank you William. It appears that my machine is square, but my inaccuracy is from runout. What is the suggested way to correct for runout? And would I expect the runout to change with material type?
I’m a big believer in test cuts — unfortunately, I can’t recall having cut two different materials w/ the same bit, so can’t answer that, beyond suggesting that you always make test cuts.
2 Likes
Thank you William. It appears that my machine is square, but my inaccuracy is from runout. What is the suggested way to correct for runout? And would I expect the runout to change with material type?
Runout, the wobble in the router that affects the end mill, varies somewhat with the material being cut. More torque is necessary to keep the RPM at the requested level for denser/harder materials and larger end mills (more cutting surface). The higher torque requirement can affect the runout. Unless one is really pushing their router hard, this isn’t too large a factor.
Runout is an issue… but so it “stick out” and deflection. The “stick out” is how far the end mill extended past the collet (and collet nut). One wants the stick out to be as small as possible. The more stick out, the more deflection (and deflection increases much more rapidly than linear (with “stick out”). Deflection affects all end mills in all materials but becomes a major issue for 0.125" end mills and smaller. I’ve seen people running a job with the deflection that was larger than the end mill diameter!
Deflection can be compensated for in the feeds and speeds (F&S) - material, spindle/router, end mill diameter and composition, and stockout). G-Wizard, a feeds and speeds package, is a good source of such data. The performance of the router can even be added to the calculation such that an optimal F&S is obtained.
Some of the inaccuracy you’re seeing is due to runout, some due to deflection, and some due to the inaccuracies of the positioning of the CNC machine. If your F&S is too “high”, deflection can become a major factor.
I’m a big believer in test cuts
I agree with @WillAdams - test cuts are always a good thing. I recommend chevrons; a dependable and trustworthy set of calipers is required. YMMV.
mark
1 Like