RampEntry Dressup

Here's the place for discussion related to CAM/CNC and the development of the Path module.
chrisb
Posts: 19512
Joined: Tue Mar 17, 2015 9:14 am

Re: RampEntry Dressup

Postby chrisb » Thu May 25, 2017 8:57 am

herbk wrote:
chrisb wrote:We already have a "Cutting edge angle" in the tool description. Is this what you mean or do we need another attribute in the tool table?
Hi Chris,
in my mind "Cutting edge length" is the dimension for that what mmxe is talking about with " that allows the flutes full engagement".
If you want to add a attribute to the tooltable, then a checkbox for "tool allows vertikal touch in".
You are right with your explanation of cutting edge length, but I was talking about the cutting edge angle
For using the full tool size for cutting, you need also a machine which is powerfull and strong enough to handle it. i think most of us don't own a machine like that, - and for that reason we work with little "step down steps"...
The principles stay the same, no matter how big the machine is, and of course we should head for a universal solution. In the topic concerning helical holes there is a discussion about ramping as well, perhaps we should open a new topic for both or continue the discussion only in one of them?
herbk
Posts: 1570
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: RampEntry Dressup

Postby herbk » Thu May 25, 2017 9:25 am

Hi Chris,
herbk wrote:in my mind "Cutting edge angle" describes the shape of the tool. E.g. tools for milling a prism or a V-groove.
i wrote also...
The principles stay the same, no matter how big the machine is, and of course we should head for a universal solution.
of course... i just tryd to explane why we, at least i, are not using the "whole tool".
perhaps we should open a new topic for both or continue the discussion only in one of them?
you are right again.. this thread would be the better place
Gruß Herbert
roivai
Posts: 117
Joined: Thu Feb 02, 2017 5:29 pm
Location: Oulu, Finland

Re: RampEntry Dressup

Postby roivai » Fri May 26, 2017 5:49 am

mmxe wrote: Hi,
I'm a mechanical engineer and I work sometimes in a very busy machine shop. I think you're over-thinking this ramp operation. Ramp entries are great, but the first consideration is what ramp angle can your cutter handle. This is specified by the cutter manufacturer, and you respect it or you damage your cutter.
So, from this standpoint, by trying to compute this angle in software based on the requested Z level plunge depth, you take away all the certainty required by the machinist to respect the tool capabilities and specs, and you force him/her to pull out a calculator to make sure your software isn't going to ruin the cutting tool.
Thanks for your insightful comment. I agree that the specified ramp angle should be respected at all times.
The angle is not calculated by default, but given as a parameter. The issue is about what to do when the given cutting path is too short to to reach the desired Z with the given angle. My initial quick solution was to tighten the angle, which is, like you said, a wrong one in many cases. The better solution should be to go back and forth until the Z is reached and I already implemented that to my "method 1" and will try to implement it to the others as well.

The discussion about effective material removal and adaptive clearing etc. is a big topic and deserves its own thread and guys with big brains. :)
herbk
Posts: 1570
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: RampEntry Dressup

Postby herbk » Fri May 26, 2017 8:31 am

Hi roivai,

like mmxe explaned and chris's picture shows here (i think you already read it) https://forum.freecadweb.org/viewtopic. ... 60#p175459 a to steepen ramp kills some tools.

But if we want to take respect of all that "tool factors" we have to rework the tooltable. Let's think about what should be includet there...
Gruß Herbert
mlampert
Posts: 1366
Joined: Fri Sep 16, 2016 9:28 pm

Re: RampEntry Dressup

Postby mlampert » Sun Jun 04, 2017 6:47 pm

FWIW, I had some training last week and they used vcarve-pro. The ramp entry there goes down in quite a tight zig-zag. I think the rule was to make elongation a bit more than the tool diameter. Just another data point.
polymer
Posts: 228
Joined: Fri Sep 12, 2014 8:49 am

Re: RampEntry Dressup

Postby polymer » Mon Jun 05, 2017 5:39 am

The Angle depends from the Tool so the best way would be to set a value while creating the Programm in freecad.
Every Tool supplier gives you a max entry angle.

For on page 198:
http://shop.lmt-tools.de/media/import/l ... lungen.pdf
roivai
Posts: 117
Joined: Thu Feb 02, 2017 5:29 pm
Location: Oulu, Finland

Re: RampEntry Dressup

Postby roivai » Sun Jun 11, 2017 7:52 am

Sorry for not doing much for this lately.

Now I modified the Method3 so that it always respects the given angle, so only Method2 will not do that. I think that method should be removed eventually, I don't think it is very useful.

sliptonic encouraged me to create PR out of this, so I did that. https://github.com/FreeCAD/FreeCAD/pull/814

Edit: It is now merged. Please, try to break it! I'm sure you'll succeed. :D
mlampert wrote:
Sun Jun 04, 2017 6:47 pm
FWIW, I had some training last week and they used vcarve-pro. The ramp entry there goes down in quite a tight zig-zag. I think the rule was to make elongation a bit more than the tool diameter. Just another data point.
Hmm, maybe that is to reduce the area where the ramp affects the surface quality? I would guess that going with minimum number of direction changes would be fastest and least stressful for the machine.. But if that is desired, I could add a parameter to restrict the maximum length of the ramping path.
mlampert
Posts: 1366
Joined: Fri Sep 16, 2016 9:28 pm

Re: RampEntry Dressup

Postby mlampert » Mon Jun 12, 2017 5:17 am

roivai wrote:
Sun Jun 11, 2017 7:52 am
Edit: It is now merged. Please, try to break it! I'm sure you'll succeed. :D
wohooo! Thanks, gotta go and check it out! :)
roivai wrote:
mlampert wrote:
Sun Jun 04, 2017 6:47 pm
FWIW, I had some training last week and they used vcarve-pro. The ramp entry there goes down in quite a tight zig-zag. I think the rule was to make elongation a bit more than the tool diameter. Just another data point.
Hmm, maybe that is to reduce the area where the ramp affects the surface quality? I would guess that going with minimum number of direction changes would be fastest and least stressful for the machine.. But if that is desired, I could add a parameter to restrict the maximum length of the ramping path.
I have no idea why they do it this way and if it's better for anything, or if it was just the simplest way to implement it.
Anyway, we got our own now! :)
User avatar
kkremitzki
Posts: 1789
Joined: Thu Mar 03, 2016 9:52 pm
Location: Texas

Re: RampEntry Dressup

Postby kkremitzki » Mon Jun 12, 2017 6:45 am

It looks like this may have introduced a bug. If you activate the Test WB and run all the tests you'll see an error:

Code: Select all

Workbench failure

cannot import name PathDressupRampEntry

Code: Select all

Traceback (most recent call last):
  File "/usr/lib/python2.7/unittest/case.py", line 329, in run
    testMethod()
  File "/home/kurt/Desktop/freecad/builds/master/Mod/Test/Workbench.py", line 46, in testActivate
    self.fail("Loading of workbench '{0}' failed: {1}".format(i, e))
  File "/usr/lib/python2.7/unittest/case.py", line 410, in fail
    raise self.failureException(msg)
AssertionError: Loading of workbench 'PathWorkbench' failed: Test on activating workbench PathWorkbench failed
Like my FreeCAD work? I'd appreciate any level of support via Patreon, Liberapay, or PayPal! Read more about what I do at my blog.
roivai
Posts: 117
Joined: Thu Feb 02, 2017 5:29 pm
Location: Oulu, Finland

Re: RampEntry Dressup

Postby roivai » Mon Jun 12, 2017 7:10 am

Wow, you are right. I just built a recent master and I get that error on startup.. But not with my branch where I did the development even though it was rebased to master very recently..

Edit: OK, so the CMakeList was not updated with the new file. I will create PR asap. Sorry, I've never dealt with CMake.