CNC Lathe and 4th-axis milling...

Here's the place for discussion related to CAM/CNC and the development of the Path module.
Forum rules
Be nice to others! Respect the FreeCAD code of conduct!
User avatar
Wsk8
Posts: 182
Joined: Fri Dec 07, 2018 6:24 pm

Re: CNC Lathe and 4th-axis milling...

Post by Wsk8 »

@Russ4262
Thanks for your contribution!

FYI
- Waterline mostly crashes for me
- I also get weird lines for dropcut in x-direction
- Could you give me a short summary on how to do 4th axis (the picture i quoted previously)? It seems i'm not smart enough to get it working myself :oops: .

mfg
Attachments
fc1.PNG
fc1.PNG (125.11 KiB) Viewed 1425 times
User avatar
roerich_64
Veteran
Posts: 1465
Joined: Thu May 21, 2015 7:00 pm
Location: Ostfriesland

Re: CNC Lathe and 4th-axis milling...

Post by roerich_64 »

Hi Russel,

Thank you for your words, but i have to say thank you ;)
I am thinking the NON-linear paths are not a result of the python script; I am thinking OCL. Perhaps the significant digits, the number of decimal places, the nature of the OCL scan process itself, or another issue all together like the mesh, created from the solid, that is used for the OCL scan.
Yes, i am with you. Rounding error around sin / cos - routines in vector bill could be...
(vector bill: is that called like that in english? In german: Vektorrechnung -> https://www1.vobs.at/maturawiki/index.p ... b=Vektoren

Yesterday i have done more tests to find out could the 'nonlinear issue' come from the scan over the 'Model-Body'.
So i copy the Model-Body to a new file and test it with the FEM-WB:
3D-Surface_XI.jpg
3D-Surface_XI.jpg (421.6 KiB) Viewed 1409 times
OCL_Surface_Test-FEM.FCStd
(189.01 KiB) Downloaded 45 times
Could be, but then we should have all lines 'not linear' ;)

Now i am thinking around rounding error when the Y-Vector of the 'solid - face' comes closer to the Z-Vector...
And it could be that the other vector combinations have also this issue...
So i am with you that is an OCL issue...

To find more out in this direction i have generated a second test file:
OCL_Surface_Test_II.FCStd
(33.28 KiB) Downloaded 64 times
3D-Surface_Test_II_1.jpg
3D-Surface_Test_II_1.jpg (143.89 KiB) Viewed 1409 times
It is a rotating object to generate one line to have those stair railing bars: https://www.treppenshop-dresden.de/Railing-Rods-Beech
3D-Surface_Test_II_2.jpg
3D-Surface_Test_II_2.jpg (630.74 KiB) Viewed 1409 times
The Result is: the generated gCode have no arcs.

Code: Select all

(begin operation: Surface)
(Path: Surface)
(Surface)
(Beginning of layer 0)
G0 Z54.0203
G0 X0.0000 Y-0.5311
G1 Z70.0000 F800.00
G1 X59.2593 Y-0.5311 Z70.0000 F800.00
G1 X69.1358 Y-0.5311 Z69.3270 F800.00
G1 X79.0123 Y-0.5311 Z66.6247 F800.00
G1 X88.8889 Y-0.5311 Z61.4842 F800.00
G1 X98.7654 Y-0.5311 Z56.5373 F800.00
G1 X118.5185 Y-0.5311 Z60.0203 F800.00
G1 X167.9012 Y-0.5311 Z68.7278 F800.00
G1 X177.7778 Y-0.5311 Z70.4693 F800.00
G1 X187.6543 Y-0.5311 Z72.2108 F800.00
G1 X197.5309 Y-0.5311 Z73.9149 F800.00
G1 X207.4074 Y-0.5311 Z75.5144 F800.00
G1 X217.2840 Y-0.5311 Z77.0321 F800.00
G1 X227.1605 Y-0.5311 Z78.4654 F800.00
G1 X237.0370 Y-0.5311 Z79.8174 F800.00
G1 X246.9136 Y-0.5311 Z81.0887 F800.00
G1 X256.7901 Y-0.5311 Z82.2788 F800.00
G1 X266.6667 Y-0.5311 Z83.3904 F800.00
G1 X276.5432 Y-0.5311 Z84.4181 F800.00
G1 X286.4198 Y-0.5311 Z85.3647 F800.00
G1 X296.2963 Y-0.5311 Z86.2318 F800.00
G1 X306.1728 Y-0.5311 Z87.0199 F800.00
G1 X316.0494 Y-0.5311 Z87.7290 F800.00
G1 X325.9259 Y-0.5311 Z88.3570 F800.00
G1 X335.8025 Y-0.5311 Z88.9032 F800.00
G1 X345.6790 Y-0.5311 Z89.3707 F800.00
G1 X355.5556 Y-0.5311 Z89.7600 F800.00
G1 X365.4321 Y-0.5311 Z90.0708 F800.00
G1 X375.3086 Y-0.5311 Z90.3017 F800.00
G1 X385.1852 Y-0.5311 Z90.4504 F800.00
G1 X395.0617 Y-0.5311 Z90.5207 F800.00
G1 X404.9383 Y-0.5311 Z90.5230 F800.00
G1 X414.8148 Y-0.5311 Z90.4517 F800.00
G1 X424.6914 Y-0.5311 Z90.2998 F800.00
G1 X434.5679 Y-0.5311 Z90.0688 F800.00
G1 X444.4444 Y-0.5311 Z89.7599 F800.00
G1 X454.3210 Y-0.5311 Z89.3726 F800.00
G1 X464.1975 Y-0.5311 Z88.9050 F800.00
G1 X474.0741 Y-0.5311 Z88.3556 F800.00
G1 X483.9506 Y-0.5311 Z87.7269 F800.00
G1 X493.8272 Y-0.5311 Z87.0196 F800.00
G1 X503.7037 Y-0.5311 Z86.2332 F800.00
G1 X513.5802 Y-0.5311 Z85.3672 F800.00
G1 X523.4568 Y-0.5311 Z84.4174 F800.00
G1 X533.3333 Y-0.5311 Z83.3880 F800.00
G1 X543.2099 Y-0.5311 Z82.2788 F800.00
G1 X553.0864 Y-0.5311 Z81.0893 F800.00
G1 X562.9630 Y-0.5311 Z79.8194 F800.00
G1 X572.8395 Y-0.5311 Z78.4662 F800.00
G1 X582.7160 Y-0.5311 Z77.0306 F800.00
G1 X592.5926 Y-0.5311 Z75.5134 F800.00
G1 X602.4691 Y-0.5311 Z73.9145 F800.00
G1 X612.3457 Y-0.5311 Z72.2108 F800.00
G1 X622.2222 Y-0.5311 Z70.4693 F800.00
G1 X632.0988 Y-0.5311 Z68.7278 F800.00
G1 X681.4815 Y-0.5311 Z60.0203 F800.00
G1 X691.3580 Y-0.5311 Z58.2788 F800.00
G1 X701.2346 Y-0.5311 Z56.5373 F800.00
G1 X711.1111 Y-0.5311 Z61.4838 F800.00
G1 X720.9877 Y-0.5311 Z66.6315 F800.00
G1 X730.8642 Y-0.5311 Z69.3191 F800.00
G1 X740.7407 Y-0.5311 Z70.0000 F800.00
G1 X800.0000 Y-0.5311 Z70.0000 F800.00
G1 X800.0000 Y-0.4689 Z70.0000 F800.00
G1 X740.7407 Y-0.4689 Z70.0000 F800.00
G1 X730.8642 Y-0.4689 Z69.3201 F800.00
G1 X720.9877 Y-0.4689 Z66.6330 F800.00
G1 X711.1111 Y-0.4689 Z61.4853 F800.00
G1 X701.2346 Y-0.4689 Z56.5387 F800.00
G1 X691.3580 Y-0.4689 Z58.2802 F800.00
G1 X681.4815 Y-0.4689 Z60.0217 F800.00
G1 X632.0988 Y-0.4689 Z68.7292 F800.00
G1 X622.2222 Y-0.4689 Z70.4707 F800.00
G1 X612.3457 Y-0.4689 Z72.2122 F800.00
G1 X602.4691 Y-0.4689 Z73.9156 F800.00
G1 X592.5926 Y-0.4689 Z75.5142 F800.00
G1 X582.7160 Y-0.4689 Z77.0312 F800.00
G1 X572.8395 Y-0.4689 Z78.4670 F800.00
G1 X562.9630 Y-0.4689 Z79.8198 F800.00
G1 X553.0864 Y-0.4689 Z81.0895 F800.00
G1 X543.2099 Y-0.4689 Z82.2788 F800.00
G1 X533.3333 Y-0.4689 Z83.3881 F800.00
G1 X523.4568 Y-0.4689 Z84.4180 F800.00
G1 X513.5802 Y-0.4689 Z85.3671 F800.00
G1 X503.7037 Y-0.4689 Z86.2332 F800.00
G1 X493.8272 Y-0.4689 Z87.0193 F800.00
G1 X483.9506 Y-0.4689 Z87.7269 F800.00
G1 X474.0741 Y-0.4689 Z88.3559 F800.00
G1 X464.1975 Y-0.4689 Z88.9050 F800.00
G1 X454.3210 Y-0.4689 Z89.3725 F800.00
G1 X444.4444 Y-0.4689 Z89.7597 F800.00
G1 X434.5679 Y-0.4689 Z90.0688 F800.00
G1 X424.6914 Y-0.4689 Z90.3000 F800.00
G1 X414.8148 Y-0.4689 Z90.4517 F800.00
G1 X404.9383 Y-0.4689 Z90.5230 F800.00
G1 X395.0617 Y-0.4689 Z90.5207 F800.00
G1 X385.1852 Y-0.4689 Z90.4504 F800.00
G1 X375.3086 Y-0.4689 Z90.3016 F800.00
G1 X365.4321 Y-0.4689 Z90.0709 F800.00
G1 X355.5556 Y-0.4689 Z89.7602 F800.00
G1 X345.6790 Y-0.4689 Z89.3706 F800.00
G1 X335.8025 Y-0.4689 Z88.9030 F800.00
G1 X325.9259 Y-0.4689 Z88.3568 F800.00
G1 X316.0494 Y-0.4689 Z87.7291 F800.00
G1 X306.1728 Y-0.4689 Z87.0202 F800.00
G1 X296.2963 Y-0.4689 Z86.2318 F800.00
G1 X286.4198 Y-0.4689 Z85.3645 F800.00
G1 X276.5432 Y-0.4689 Z84.4181 F800.00
G1 X266.6667 Y-0.4689 Z83.3906 F800.00
G1 X256.7901 Y-0.4689 Z82.2797 F800.00
G1 X246.9136 Y-0.4689 Z81.0888 F800.00
G1 X237.0370 Y-0.4689 Z79.8177 F800.00
G1 X227.1605 Y-0.4689 Z78.4658 F800.00
G1 X217.2840 Y-0.4689 Z77.0330 F800.00
G1 X207.4074 Y-0.4689 Z75.5157 F800.00
G1 X197.5309 Y-0.4689 Z73.9160 F800.00
G1 X187.6543 Y-0.4689 Z72.2122 F800.00
G1 X177.7778 Y-0.4689 Z70.4707 F800.00
G1 X118.5185 Y-0.4689 Z60.0217 F800.00
G1 X108.6420 Y-0.4689 Z58.2802 F800.00
G1 X98.7654 Y-0.4689 Z56.5387 F800.00
G1 X88.8889 Y-0.4689 Z61.4850 F800.00
G1 X79.0123 Y-0.4689 Z66.6257 F800.00
G1 X69.1358 Y-0.4689 Z69.3285 F800.00
G1 X59.2593 Y-0.4689 Z70.0000 F800.00
G1 X0.0000 Y-0.4689 Z70.0000 F800.00
(End of layer 0)
G0 Z56.0203
(finish operation: Surface)

At last:
as @Sliptonic says: go, Russel, go :) ;)
I am also very happy that more axis comes to FreeCAD :)
https://forum.freecadweb.org/viewtopic.php?f=24&t=33408

BR
Walter
Last edited by roerich_64 on Fri Mar 29, 2019 1:34 pm, edited 1 time in total.
Die Liebe wird siegen, denn sie ist unzerstörbar :) ;)
chrisb
Veteran
Posts: 53943
Joined: Tue Mar 17, 2015 9:14 am

Re: CNC Lathe and 4th-axis milling...

Post by chrisb »

We have seen these parkinson-style paths before, you may dig in the archives of Path forum for it.
A Sketcher Lecture with in-depth information is available in English, auf Deutsch, en français, en español.
User avatar
roerich_64
Veteran
Posts: 1465
Joined: Thu May 21, 2015 7:00 pm
Location: Ostfriesland

Re: CNC Lathe and 4th-axis milling...

Post by roerich_64 »

chrisb wrote: Fri Mar 29, 2019 1:29 pm We have seen these parkinson-style paths before, you may dig in the archives of Path forum for it.
Thank you Chris, i have to read ;)

BR
Walter
Die Liebe wird siegen, denn sie ist unzerstörbar :) ;)
herbk
Veteran
Posts: 2657
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: CNC Lathe and 4th-axis milling...

Post by herbk »

If i look at the gcode Walter postet i get in mind that the path jumps from meshpoint to meshpoint. This points are not at the same Z level, for that the tool creates this "waves".
Gruß Herbert
User avatar
roerich_64
Veteran
Posts: 1465
Joined: Thu May 21, 2015 7:00 pm
Location: Ostfriesland

Re: CNC Lathe and 4th-axis milling...

Post by roerich_64 »

herbk wrote: Fri Mar 29, 2019 5:48 pm If i look at the gcode Walter postet i get in mind that the path jumps from meshpoint to meshpoint. This points are not at the same Z level, for that the tool creates this "waves".
I am back from digging ;)

Herbert, look at:
3D-Surface_Test_II_3.jpg
3D-Surface_Test_II_3.jpg (216.14 KiB) Viewed 1355 times
OCL_Surface_Test-FEM-2.FCStd
(125.75 KiB) Downloaded 38 times
When i read this http://www.anderswallin.net/category/cn ... utter-cam/
from Anders Wallin and have a look to the pictures of him, then i am coming to the point, that the internal mesh in FreeCAD is to rough.

Can we manupulate the FreeCAD internal mesh for testing?

BR
Walter

Edit:
chrisb wrote: Fri Mar 29, 2019 1:29 pm We have seen these parkinson-style paths before, you may dig in the archives of Path forum for it.
Have you a link?
Die Liebe wird siegen, denn sie ist unzerstörbar :) ;)
User avatar
roerich_64
Veteran
Posts: 1465
Joined: Thu May 21, 2015 7:00 pm
Location: Ostfriesland

Re: CNC Lathe and 4th-axis milling...

Post by roerich_64 »

Hmm, when i see this:
3D-Surface_XII.jpg
3D-Surface_XII.jpg (335.96 KiB) Viewed 1343 times
then i came to the 'mesh issue' not to 'rounding issue'...

Thanks to Russ to correct the 'high issue' :)
Workes great ;)

Good Night
Walter
Die Liebe wird siegen, denn sie ist unzerstörbar :) ;)
chrisb
Veteran
Posts: 53943
Joined: Tue Mar 17, 2015 9:14 am

Re: CNC Lathe and 4th-axis milling...

Post by chrisb »

roerich_64 wrote: Fri Mar 29, 2019 7:33 pm
chrisb wrote: Fri Mar 29, 2019 1:29 pm We have seen these parkinson-style paths before, you may dig in the archives of Path forum for it.
Have you a link?
No I havent't. If I had it at hand I would have posted it in the first place.
A Sketcher Lecture with in-depth information is available in English, auf Deutsch, en français, en español.
herbk
Veteran
Posts: 2657
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: CNC Lathe and 4th-axis milling...

Post by herbk »

Hi,
roerich_64 wrote: Fri Mar 29, 2019 9:26 pm Hmm, when i see this:
then i came to the 'mesh issue' not to 'rounding issue'...
In my mind path creation from a mesh never can have result with parallel pathes, because the base for the calculation are the mesh nodes wich are not placed at one level. If you use a smaller triangulated mesh the deviance will be smaller to, but will be still there.

Again a question i asked earlier: Isn't it more easy (and more exact) to "cut" the part to slices and use the outline of the slices for the path creation?


@ Walter:
How many lines of gcode you have for the result of your last pic? ;)
Gruß Herbert
User avatar
roerich_64
Veteran
Posts: 1465
Joined: Thu May 21, 2015 7:00 pm
Location: Ostfriesland

Re: CNC Lathe and 4th-axis milling...

Post by roerich_64 »

Hi,
another nice shinig and warm spring day :)
FreeCAD is calculating the gCode for Herbert... and calculating... and calculating... :)
herbk wrote: Sat Mar 30, 2019 7:01 am Hi,
roerich_64 wrote: Fri Mar 29, 2019 9:26 pm Hmm, when i see this:
then i came to the 'mesh issue' not to 'rounding issue'...
In my mind path creation from a mesh never can have result with parallel pathes, because the base for the calculation are the mesh nodes wich are not placed at one level. If you use a smaller triangulated mesh the deviance will be smaller to, but will be still there.

Again a question i asked earlier: Isn't it more easy (and more exact) to "cut" the part to slices and use the outline of the slices for the path creation?


@ Walter:
How many lines of gcode you have for the result of your last pic? ;)
Do not throw the flint into the grain so fast, Herbert ;)
When i see this video from Anders Wallin: https://www.youtube.com/watch?time_cont ... eci8wSmeGM

at 0:00:13 we see:
3D-Surface_XIII.jpg
3D-Surface_XIII.jpg (129.71 KiB) Viewed 1283 times
then i am very hopefully, we can have nice and smooth gCode-paths out of meshes ;)

Edit:
The point is the interface between FreeCAD and OCL.
When we can give OCL a mesh with small triangles to calculate, then we should have smooth pathes...
Die Liebe wird siegen, denn sie ist unzerstörbar :) ;)
Post Reply