Hi All,
Most of my work has been with one-of-a-kind polyurethane foam shapes, so I'm encountering new challenges when I work with hard maple - using my 2002 C40. The problem is that I'm trying to optimize the production of a bowl shape in hard maple. I've gradually "tightened up" every parameter I know about and am surprised at how low I need to turn the feed override in order to avoid seeing little "oval dimples" in the surface, which I hope show up in the attached image.
Right now, I'm using an ArtCam model, created by spinning a smooth curve of about 1.5" radius through a plan curve of about 5" in radius. The pixel count is 300/inch. I've smoothed it maybe 500-600 passes. I've created various toolpaths for 3/4" ballnose (regular length), stepover .030" , rpm 15000, feed 300 ipm. I reduced my usual .001 tolerance to .0001 for latest toolpaths.
When I run a raster toolpath (either straight or curved), I'm surprised that even when I reduce the tangency factor down to G09F1, I also need to cut the feed back considerably lower than I would like, in order to get rid of the dimples and see only the cusps which I expect at that stepover. I was under the impression that down at G09F1, the machine would limit speed through the turns to whatever ipm is necessary to "stay on track." On the contrary, I seem to have to turn the override down to about 50% or lower to get rid of the dimples and see only the cusps. [Then I would address the stepover and what cusp is optimal.]
I hope I'm missing something which could help me cut repetitive maple pieces quickly enough to allow a little profit. [And, no - the shape can't be turned on a lathe.] I'd appreciate any suggestions about how to avoid excessive sanding of difficult shapes without spending too much machine time at low speeds.
Thanks. Norm Kinzie
need smoother finish
Moderators: Mike Iubelt, Jason Susnjara, Larry Epplin, Clint Buechlein, Mike Iubelt, Jason Susnjara, Larry Epplin, Clint Buechlein
-
- New Member
- Posts: 2
- Joined: Tue, Nov 30 2010, 2:12PM
- Company Name: Landfoam
-
- Guru Member
- Posts: 451
- Joined: Thu, Dec 08 2005, 6:10PM
- Company Name: Williamson Finewoodwork
- Country: UNITED STATES
- Location: Capac MI
- Contact:
Re: need smoother finish
I would consider that normal, as the bit goes in one direction it cut with the grain and when it returns it is cutting against the grain. Even on a lathe you would have a similar problem, With tear out on opposing quadrants. I think the bit is jumping slightly as it crosses the grain from early wood to late wood.
Have you tried to cut this with the profile modeler?
What style bit are you using? I have been using 4 flute roughing cutters at .02 step over and 750 IPM but we still put them on the lathe for a finish sanding.
Give me a call 810 310 1111 Will Williamson
Have you tried to cut this with the profile modeler?
What style bit are you using? I have been using 4 flute roughing cutters at .02 step over and 750 IPM but we still put them on the lathe for a finish sanding.
Give me a call 810 310 1111 Will Williamson
Will
http://www.willmade.com
KEG/Intel Core i 7 CPU K875 @ 2.93 GHZ/12G Ram
Dual boot XP PRO/Windows 7 Nvidia Quatro 600 1Gig Ram
http://www.willmade.com
KEG/Intel Core i 7 CPU K875 @ 2.93 GHZ/12G Ram
Dual boot XP PRO/Windows 7 Nvidia Quatro 600 1Gig Ram
-
- Thermwood Team
- Posts: 361
- Joined: Wed, May 17 2006, 11:25AM
- Location: Thermwood
Re: need smoother finish
Norman,
If you could, please post the nc code or email it to us at program@thermwood.com. We can look at the code to see if these ripples are indeed in the code or if it is something else going on at the machine. Thanks.
If you could, please post the nc code or email it to us at program@thermwood.com. We can look at the code to see if these ripples are indeed in the code or if it is something else going on at the machine. Thanks.
-
- New Member
- Posts: 2
- Joined: Tue, Nov 30 2010, 2:12PM
- Company Name: Landfoam
Re: need smoother finish
Good Morning Daniel,
Thanks for your reply. I sent a sample of the code, but it bounced back. Please tell me if second try did not reach you.
I had a helpful conversation with Will Williamson and tested same toolpath with 18 lb/cf foam and G09F1 - found the same marks. He also suggested I check grounding. I'm not sure how to test it other than to recall we drove a long ground rod down close to machine. [I seem to recall being told to disconnect the redundant connection to the building ground.]
Norm Kinzie
802 728 7098
Thanks for your reply. I sent a sample of the code, but it bounced back. Please tell me if second try did not reach you.
I had a helpful conversation with Will Williamson and tested same toolpath with 18 lb/cf foam and G09F1 - found the same marks. He also suggested I check grounding. I'm not sure how to test it other than to recall we drove a long ground rod down close to machine. [I seem to recall being told to disconnect the redundant connection to the building ground.]
Norm Kinzie
802 728 7098
Re: need smoother finish
Norm,
G09F# is the tangency factor which controls how fast the machine tries to flow through segmented geom. If this number is to low, you could potentially get a slight hesitation of motion between each segment of code. Typically a good value for running segmented style code is G09F8 or G09F12. Another G code that can be used in conjuction to help smooth motion of your tool would be the Acceleration factor code (G800). Typcially a value of G803 is good starting point. The Acceleration factor can be set up to G809, but will slow the cnc machine motions considerably. I would like for you to try the following in the header of your program and keep us up to date on your progress:
G09F12
G803
If you still do not get the results you desire, I would like for you to email us your machine number (located on serial tag, back of base on cnc machine) and we can make further suggestions as what can possibly be adjusted. The next step would be working with the Gains by utilizing AFL to write machine specific settings in your part program.
email: program@thermwood.com
G09F# is the tangency factor which controls how fast the machine tries to flow through segmented geom. If this number is to low, you could potentially get a slight hesitation of motion between each segment of code. Typically a good value for running segmented style code is G09F8 or G09F12. Another G code that can be used in conjuction to help smooth motion of your tool would be the Acceleration factor code (G800). Typcially a value of G803 is good starting point. The Acceleration factor can be set up to G809, but will slow the cnc machine motions considerably. I would like for you to try the following in the header of your program and keep us up to date on your progress:
G09F12
G803
If you still do not get the results you desire, I would like for you to email us your machine number (located on serial tag, back of base on cnc machine) and we can make further suggestions as what can possibly be adjusted. The next step would be working with the Gains by utilizing AFL to write machine specific settings in your part program.
email: program@thermwood.com