Wrong profile path (strange loops)

Here's the place for discussion related to CAM/CNC and the development of the Path module.
herbk
Posts: 1567
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Postby herbk » Sat Oct 15, 2016 5:43 am

Good Morning,

@sliptonic
On OpenSuse i can use only OCC for "path profile" in this time, if i use libarea. got a error message:

Traceback (most recent call last):
File "/usr/lib64/FreeCAD/Mod/Path/PathScripts/PathProfile.py", line 333, in execute
output += self._buildPathLibarea(obj, edgelist)
File "/usr/lib64/FreeCAD/Mod/Path/PathScripts/PathProfile.py", line 258, in _buildPathLibarea
lead_out_line_len)
File "/usr/lib64/FreeCAD/Mod/Path/PathScripts/PathKurveUtils.py", line 348, in profile2
set_good_start_point(curve, False)
<type 'exceptions.NameError'>: global name 'set_good_start_point' is not defined

If you take a look like i have to made the hole in the middle, there you can see that i have a'n other prob.wich was fixed sometimes before: If i cutt out the center hole before i run "path profile" for the outline, i got no outline path...

But i'm in mind, that both, invalid running of libarea und not making a path profile on a part with a hole, are problems which are evoked by the changes for OpenSuse...

@mlampert
by converting your magnet wheel to a "Solid" (Part menu) and build the path around that
I also tried that, but got also some loops, but on different places.


For the loops i'm with sliptonic and chrisb: it seams to be a a rounding bug...
Last edited by herbk on Sat Oct 15, 2016 11:47 am, edited 1 time in total.
Gruß Herbert
mlampert
Posts: 1363
Joined: Fri Sep 16, 2016 9:28 pm

Re: Wrong profile path (strange loops)

Postby mlampert » Sat Oct 15, 2016 5:44 am

Obviously I'm not very good at letting go .....

The offending piece are your t-bones (which I admire very much -nicely done). If you make your t-bones slightly larger (radius = 1.6mm) you won't have any issues without having to play with extra offset and such. It also works on the Cut directly, no need to make a Solid out of it.

(your post just came in, I also had to comment that line for set_good_start_point out - I'm on Debian)
herbk
Posts: 1567
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: Wrong profile path (strange loops)

Postby herbk » Sat Oct 15, 2016 6:03 am

If you make your t-bones slightly larger (radius = 1.6mm) you won't have any issues without having to play with extra offset and such.
OK, i'll try... But: in this case its not a problem to change the radius a little bit, but what if it is...?
I also had to comment that line for set_good_start_point out
?? I don't set a start point...?
Gruß Herbert
mlampert
Posts: 1363
Joined: Fri Sep 16, 2016 9:28 pm

Re: Wrong profile path (strange loops)

Postby mlampert » Sat Oct 15, 2016 6:13 am

Maybe I got it wrong - I thought the radius of your t-bone cylinders 1.5mm and your tool radius is also 1.5mm. I tried 1.6 for the t-bones with a 3mm diameter tool and it worked fine. You can experiment with smaller radii if the 0.1mm is too much for your needs. If you the actual fusion instead of a copy for the cut experimentation is easier/faster.
herbk
Posts: 1567
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: Wrong profile path (strange loops)

Postby herbk » Sat Oct 15, 2016 6:40 am

i'm on my Mint
now.
here i got no ouline path, no different using OCC or libarea.

But with Mint (ubuntu, Debian) im not familiar and i'm sure i have not the ppa's i need...

OS: Linux Mint 17.3 Rosa
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.17.8676 (Git)
Build type: None
Branch: master
Hash: b8163a03dc80b64b38744ed1289e8a7bdab0d378
Python version: 2.7.6
Qt version: 4.8.6
Coin version: 4.0.0a
OCC version: 6.8.0.oce-0.17

FreeCAD don't show libarea, It is: 0.0.20160313.120+gitf1986ac+201603142132
Gruß Herbert
herbk
Posts: 1567
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Postby herbk » Sat Oct 15, 2016 7:14 am

Hi mlampert,

back on OpenSuse... with radius of a little bit biger than the tool radius it works here to, means: I got a outside path without loops. I have tried with radius 1.6, 1,8 and 2.
My be it's necessary to set the radius a bit biger aus the tool... I will have a eye on it.

Edit:
Also i got i outside path with the cuttet of center hole if the "dogbone radius" is biger as the tool radius.
Last edited by herbk on Sat Oct 15, 2016 11:39 am, edited 1 time in total.
Gruß Herbert
chrisb
Posts: 19484
Joined: Tue Mar 17, 2015 9:14 am

Re:

Postby chrisb » Sat Oct 15, 2016 10:07 am

herbk wrote: My be it's necessary to set the radius a bit biger aus the tool... I will have a eye on it.
Ah, I hadn't seen that. So it is probably nothing but the well known coplanar faces problem. And here it is even more evident. You cannot expect that going right of the left side is the same as the path left from the right side. Probably nothing to do in Path WB.
herbk
Posts: 1567
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: Wrong profile path (strange loops)

Postby herbk » Sat Oct 15, 2016 12:27 pm

Part is ready now..
loops3.jpg
loops3.jpg (387.48 KiB) Viewed 825 times
loops4.jpg
loops4.jpg (370.71 KiB) Viewed 825 times
:D :D
Gruß Herbert
chrisb
Posts: 19484
Joined: Tue Mar 17, 2015 9:14 am

Re: Wrong profile path (strange loops)

Postby chrisb » Sat Oct 15, 2016 12:37 pm

Thanks for sharing (with the good old Mitutoyo ...). Perhaps we should collect the path showcases somewhere to demonstrate the capabilities the Path WB has reached. A topic with links to existing discussions or complete showcases?
Just curious: it is not really magnetic the material you used :? :) ?
herbk
Posts: 1567
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: Wrong profile path (strange loops)

Postby herbk » Sat Oct 15, 2016 2:28 pm

Hi Chris,
i have to say "thank you" to the guys they made it possible :!:

We can use e thread i have startet some time before fo it... viewtopic.php?f=15&t=15495

"Good old" is the right name for the Mitutoyo... Ihave it at least 30 years and it still works perfect.

The material i used is GFK, the pockts are the place for the magnets, - each 40x10x5mm. They have to be placed very exactly araond the disk, so CAD CAM is the right way to produce it.
Gruß Herbert