Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Here's the place for discussion related to CAM/CNC and the development of the Path module.
User avatar
Kunda1
Posts: 8702
Joined: Thu Jan 05, 2017 9:03 pm

Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby Kunda1 » Mon Jan 06, 2020 4:34 pm

Discussion thread for:
Feature #issue #2916: Add knife align up function to locate knife at beginning cut precisely
JFYI, original thread in spanish: https://forum.freecadweb.org/viewtopic.php?f=14&t=20620
Want to contribute back to FC? Checkout:
#lowhangingfruit | Use the Source, Luke. | How to Help FreeCAD | How to report FC bugs and features
User avatar
sliptonic
Posts: 1976
Joined: Tue Oct 25, 2011 10:46 pm
Location: Columbia, Missouri
Contact:

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby sliptonic » Mon Jan 06, 2020 5:06 pm

I can't duplicate the problem with similar geometry. Can we get a simple example that demonstrates it?
screenshot_15.png
screenshot_15.png (5.32 KiB) Viewed 382 times
chrisb
Posts: 28913
Joined: Tue Mar 17, 2015 9:14 am

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby chrisb » Mon Jan 06, 2020 7:11 pm

This dates back to 0.17. I would just close the ticket if it cannot be reproduces easily. It is most probably fixed by all the fixes and changes made since then.
A Sketcher Lecture with in-depth information is available in English, auf deutsch, en français, en español.
mlampert
Posts: 1489
Joined: Fri Sep 16, 2016 9:28 pm

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby mlampert » Mon Jan 06, 2020 7:34 pm

sliptonic wrote:
Mon Jan 06, 2020 5:06 pm
I can't duplicate the problem with similar geometry. Can we get a simple example that demonstrates it?
IIRC the issue is the start segment, without a "lead in" the initial orientation of the knife, and therby the start of the cut, is indetermined.
screenshot_15.png
screenshot_15.png (7.44 KiB) Viewed 345 times
Russ4262
Posts: 530
Joined: Sat Jun 30, 2018 3:22 pm

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby Russ4262 » Mon Jan 06, 2020 7:34 pm

Afternoon,
I know little about this dragknife thing and less about the FreeCAD features available to deal with it.

Nonetheless, I am looking at your image posted with the similar geometry and would assume that some sort of lead-in path or dress-up is needed to orient the knife before lowering at the first/initial corner, and beginning its path. Looking at each of the other corners, they each have a 90 degree turn-out to orient the knife for the segment to cut. I think what the ticket and Sr. Robles are referring to is this need for an initial orientation path maneuver leading up to the first plunge and cut. If the user has not pre-oriented the knife to the correct position before the initial cut, that first corner will likely not be cut correctly.

... and my two cents here may be worth more like two mils, than cents. Just an observation that raised my curiosity. Reading and looking at Sr. Robles original post at Pruebas con ¿Navaja de arrastre? drag knife, I see his OP geometry and path image is almost identical to @Sliptonic's. The only variance is the second corner, moving clockwise, now DOES have the correct orientation maneuver that was lacking in Sr. Robles original. Based on that the issue, it would appear that bug has been fixed since then.

In a follow-up post in the original Spanish thread, Sr. Robles also asked about the addition of a lead-in maneuver to be added that would orient the knife correctly leading into the first cut. Any progress on this additional feature? Just asking, on behalf of Sr. Robles. If that has not mentioned since, it is likely a moot point to be dismissed.
sliptonic wrote:
Mon Jan 06, 2020 5:06 pm
... screenshot_15.png
Thanks,
Russ
mlampert
Posts: 1489
Joined: Fri Sep 16, 2016 9:28 pm

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby mlampert » Mon Jan 06, 2020 7:37 pm

mh, looking at these paths, aren't they wrong anyway? I cannot imagine how this will cut out a cross - for that the corners need overlap, right?
Russ4262
Posts: 530
Joined: Sat Jun 30, 2018 3:22 pm

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby Russ4262 » Mon Jan 06, 2020 7:50 pm

mlampert wrote:
Mon Jan 06, 2020 7:37 pm
mh, looking at these paths, aren't they wrong anyway? I cannot imagine how this will cut out a cross - for that the corners need overlap, right?
No overlap needed.

First cut drags knife down. Cutter stops past corner, leaving knife point horizontal with next cut toward right. Cutter rolls up and right, pivoting on knife point that is horizontal to next cut. Cutter proceeds to right, cutting horizontal line.

The key is to remember the knife point is offset from center of cutter(spindle). The spindle passes an intersection/corner, stopping when point is parallel and inline with next directional cut. The spindle uses the knife point as the pivot reference point, and rotates into place accordingly - inline with next cut path.

Russ
mlampert
Posts: 1489
Joined: Fri Sep 16, 2016 9:28 pm

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby mlampert » Mon Jan 06, 2020 7:57 pm

Russ4262 wrote:
Mon Jan 06, 2020 7:50 pm
mlampert wrote:
Mon Jan 06, 2020 7:37 pm
mh, looking at these paths, aren't they wrong anyway? I cannot imagine how this will cut out a cross - for that the corners need overlap, right?
No overlap needed.

First cut drags knife down. Cutter stops past corner, leaving knife point horizontal with next cut toward right. Cutter rolls up and right, pivoting on knife point that is horizontal to next cut. Cutter proceeds to right, cutting horizontal line.

The key is to remember the knife point is offset from center of cutter(spindle). The spindle passes an intersection/corner, stopping when point is parallel and inline with next directional cut. The spindle uses the knife point as the pivot reference point, and rotates into place accordingly - inline with next cut path.

Russ
Got'cha, you are absolutely correct!
User avatar
sliptonic
Posts: 1976
Joined: Tue Oct 25, 2011 10:46 pm
Location: Columbia, Missouri
Contact:

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby sliptonic » Mon Jan 06, 2020 9:50 pm

mlampert wrote:
Mon Jan 06, 2020 7:34 pm
sliptonic wrote:
Mon Jan 06, 2020 5:06 pm
I can't duplicate the problem with similar geometry. Can we get a simple example that demonstrates it?
IIRC the issue is the start segment, without a "lead in" the initial orientation of the knife, and therby the start of the cut, is indetermined.
That's what the ticket refers to but the original poster in the spanish forum seems to be talking about something else. Note the first corner after the start. It should have a pivot but doesn't.
User avatar
sliptonic
Posts: 1976
Joined: Tue Oct 25, 2011 10:46 pm
Location: Columbia, Missouri
Contact:

Re: Feature #2916: Add knife align up function to locate knife at beginning cut precisely

Postby sliptonic » Mon Jan 06, 2020 9:54 pm

Russ4262 wrote:
Mon Jan 06, 2020 7:34 pm

Nonetheless, I am looking at your image posted with the similar geometry and would assume that some sort of lead-in path or dress-up is needed to orient the knife before lowering at the first/initial corner, and beginning its path. Looking at each of the other corners, they each have a 90 degree turn-out to orient the knife for the segment to cut. I think what the ticket and Sr. Robles are referring to is this need for an initial orientation path maneuver leading up to the first plunge and cut. If the user has not pre-oriented the knife to the correct position before the initial cut, that first corner will likely not be cut correctly.
That is definitely what the original ticket was about. I never pursued the solution because I assumed a generalized leadin/out dressup might be sufficient.
... and my two cents here may be worth more like two mils, than cents. Just an observation that raised my curiosity. Reading and looking at Sr. Robles original post at Pruebas con ¿Navaja de arrastre? drag knife, I see his OP geometry and path image is almost identical to @Sliptonic's. The only variance is the second corner, moving clockwise, now DOES have the correct orientation maneuver that was lacking in Sr. Robles original. Based on that the issue, it would appear that bug has been fixed since then.
yep, that's what I queued in on too but I assumed it was a particular combination of geometry and settings that was passing by my test for whether a pivot needs to be added.