Accuracy: Not sure if this is a MakerCAM issue

an open source, web based CAM package that works!
WillAdams
Posts: 8489
Joined: Mon Apr 09, 2012 6:11 pm
Location: Pennsylvania --- south of the Turnpike, East of US-15
Contact:

Re: Accuracy: Not sure if this is a MakerCAM issue

Post by WillAdams » Fri Jun 27, 2014 10:07 am

I don't see how this could be software --- turn it around.

Define 2 pairs of objects, each milled clockwise and counterclockwise --- one a profile, the other a pocket, each using the same path where it defines the object (either create them by hand, by using follow path, or make them different by the bit diameter). Mill all of them and measure.

I believe that this will show that the machine is deflecting differently depending on whether one is doing climb or conventional milling and the bit is being pulled into the workpiece.

I guess we should code up such a file
Shapeoko 3XL #0006 w/Makita RT0701 Router w/0.125″ and ¼″ Elaire precision collets
Nomad 883 Pro #596 (bamboo)

ksaiyo
Posts: 14
Joined: Sat Jun 07, 2014 11:21 pm

Re: Accuracy: Not sure if this is a MakerCAM issue

Post by ksaiyo » Tue Jul 01, 2014 1:21 am

Just bringing attention to the fact that I made a edit to a previous post and attached new files - quoted below.
Here is another Ad-hoc test I have created for circles. I have included both the SVG's and .nc files in the folder
https://drive.google.com/folderview?id= ... sp=sharing
The circle_test_before.svg is for reference and has all the dimentions, circle_test_after.svg is the file ready for import into MakerCAM [text removed and paths combined; note it appears all interior geometry disappears but it is still there]. The .nc file is set for a 1/8 inch end mill at 15 inches per minute in both feed and plunge rate with a 0.05" plunge depth.
I was curious if the size of the interior circle made a difference when it came to accuracy and precision, as well as the direction of the profile path, so I made many different sizes of circles for comparison to see if and how the data diverges from the nominal specs.

ksaiyo
Posts: 14
Joined: Sat Jun 07, 2014 11:21 pm

Re: Accuracy: Not sure if this is a MakerCAM issue

Post by ksaiyo » Tue Jul 01, 2014 3:12 am

Okay, I think I've hit on a little success!

http://imgur.com/QMGMgye

Before I was milling MDF, I thought that since this was a isotropic material it would be and excellent candidate for calibration (or at least verification that my calibration numbers were good.)

I decided to move to expanded PVC instead. Wow, what a difference. After adjusting $0 and $1 a little, this machine comfirmed that it is able to be dialed! I ran my circle test. All (well most) dimentions, even the largest circle, was within 0.005, most were within .002! The only outlier was the pie shape where I measure form the inside corner (the center approximately) to the perimeter which was bout .008 at best. I inspected it for loose v-wheels and such and I'm pretty darn sure that it is solid as can be.

This confirms that the hardness of my material makes a dramatic effect on my overall precision (most error in the machine appears to be very repeatable). So when I attempt aluminum in the future, I will definitely be moving slow and taking small overlap (Yes, I know that you can go too slow and take too small of chip as well.) So even if the spindle can take all the abuse I can throw at it, the stiffness of the Shapeoko 2 is still a major concern. The proverb "Slow and steady wins the race" proves itself yet again.

Post Reply