Couple of questions about the Path Workbench

[burtbick]

Hi,

Just started working with FreeCAD a few weeks ago and it has improved a lot since I looked at it 2 years ago. I've been building a Mostly Printed CNC machine and have that working well and actually cutting.

I'm experimenting with some different CAM packages and ultimately am looking forward to the Path Workbench getting to a point where I can use it for most of what I want to do.

Right now I'm using version .16 here's the details:
OS: Linux Mint 17.1 Rebecca
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.16.6703 (Git)
Build type: None
Branch: releases/FreeCAD-0-16
Hash: 2ce5c8d2e3020d05005ed71f710e09e9aa561f40
Python version: 2.7.6
Qt version: 4.8.6
Coin version: 4.0.0a
OCC version: 6.8.0.oce-0.17

Where could I grab the latest in progress version with some of the new changes to the Parts Workbench. Have most of the changes now merged into the main branch so I could potentially just pull a nightly build to experiment, or is there another branch that would be better to pull for the Parts Workbench?

Is there an option to direct which side of the line the cutting is done on? In my initial test with v.16 I created a simple model with a slot that was 16mm x 40mm. The physical result was larger by the size of my milling bit, 3.19mm. So I'm assuming that the toolpath was such that it followed the line with 1/2 of the milling bit taking a bit outside of the line.

I had previously done a test using EstlCam with the same slot and it milled out perfectly.

Is there a setting that I missed in v.16 to compensate for this, or has this been addressed in a later version?

Also, what's the best way to set the 0 point of the model. Typically I set X, Y to the lower left corner with Z0 being the top of the piece to be milled. This is how I did it in EstlCam and HeeksCNC. What I wound up doing in FreeCAD was making a model in the Parts Workbench with a cube and moving the cube to below the plane so Z0 was the top of the cube in the lower left corner. Then I created a sketch on the top of the cube and drew in the slot and created the pocket from that. But I expect that there is a better way to do that, and I couldn't work out how to accomplish the same thing if I used the Part Design Workbench and created everything with sketches. Again, probably just being really new at all this.

In any case great job with the Path Workbench, and I'm looking forward to trying the latest changes to see all the improvements.

Burt

[Linden]

As you are running Linux mint I think you should be able to get the beading edge daily builds from the PPA this should install the daily build along side your current version.

Open a terminal and enter:
sudo add-apt-repository ppa:freecad-maintainers/freecad-daily
sudo apt-get update
sudo apt-get install freecad-daily

To start 0pen a terminal and enter:
./freecad-daily

any time you would like to update and install latest version
sudo apt-get update
sudo apt-get install freecad-daily

Have a look you will be pleasantly surprised at the progress that has been made you can choose on the line right or left as well as different algorithms for pocket and profile,

There is a long way to go but it has come this far very quickly. Hope this will get you up and going. I am using the path workbench to generate g code for my tree 325 with an old dynapath 20 controller. The code produced from freecad still needs checking and proofing prior to running but I can do more and more with less hand editing so I am happy.

[burtbick]

Thanks,

Installing now and looking forward to seeing the progress so far. If I can shake loose some time I may even dig into the code to see if there are areas where I can help. My background is SW and HW development, last few years mainly Embedded Linux ARM based systems for various companies.

The 3D printing / CNC activities are primarily hobby oriented at this point in time.

Thanks again,
Burt

[sliptonic]

Thanks,
If I can shake loose some time I may even dig into the code to see if there are areas where I can help. My background is SW and HW development, last few years mainly Embedded Linux ARM based systems for various companies.

Awesome! If you want to work on new features, please post here early and often so we can coordinate efforts. If you want to work on bugs, there's a separate Path project on the tracker: http://www.freecadweb.org/tracker

I'm almost always on the #freecad IRC channel if you have questions or want to chat about features and direction for Path.

[burtbick]

OK, I installed but here's the version that runs:

OS: Linux Mint 17.1 Rebecca
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.17.7794 (Git)
Build type: None
Branch: master
Hash: 4acca01d876f503edefdb99ba5590bde83101f3f
Python version: 2.7.6
Qt version: 4.8.6
Coin version: 4.0.0a
OCC version: 6.8.0.oce-0.17

And trying to set up my simple test I'm not seeing an option on the Pocket operation to specify inside/outside/on the line. And I have some other confusion on trying to get things set up properly.

So I'm wondering if I truly have the daily build since it indicates that this has a release date of 2016/06/20 14:49:04 and today is 2016/06/23..

Or maybe I'm missing where the option is for specifying inside/outside/on the line?

Also I'm still trying to work out the proper way to set this up as far as 0,0,0 position is concerned.

When you set up to mill/rout where are you specifying your 0,0,0 position.

For the other CAM packages I've been testing with it has involved positioning the model such that 0,0,0 is the lower left corner of the "stock" with the Z0 being at the top of the model. But I'm having some issues finding the best way to do that at the moment for the Path Workbench.

I attempted to place the cube that I start with such that it is below the plane with the top being at Z0 in FreeCAD, that worked when I tried it in v.16 but the actual origin still appears to be at the bottom of the cube so the Path Workbench says that is where 0 is even though it actually should be -26.14mm in this case. So I'm missing something basic here.

Or are you setting 0,0,0 differently for the start point?

If I can figure out how to set the proper 0,0,0 point and find the option for where the milling takes place relative to the "line" then I can do some more testing.

Thanks,
Burt

[sliptonic]

And trying to set up my simple test I'm not seeing an option on the Pocket operation to specify inside/outside/on the line. And I have some other confusion on trying to get things set up properly.

So I'm wondering if I truly have the daily build since it indicates that this has a release date of 2016/06/20 14:49:04 and today is 2016/06/23..

Or maybe I'm missing where the option is for specifying inside/outside/on the line?

I think it's correct. Pocket doesn't have an online/left/right option. Pocket always fills a region with toolpath. The region is defined by a face. Profile goes around a face and so has options for side/on line.

Also I'm still trying to work out the proper way to set this up as far as 0,0,0 position is concerned.

When you set up to mill/rout where are you specifying your 0,0,0 position.

For the other CAM packages I've been testing with it has involved positioning the model such that 0,0,0 is the lower left corner of the "stock" with the Z0 being at the top of the model. But I'm having some issues finding the best way to do that at the moment for the Path Workbench.

At the moment, gcode is generated relative to the global coordinate system. You need to position your part so 0,0,0 is where you want it. I usually make a Draft clone of my part and base my operations off the clone. Then I can either re-position the clone or edit/refine the model without screwing up the other.

I attempted to place the cube that I start with such that it is below the plane with the top being at Z0 in FreeCAD, that worked when I tried it in v.16 but the actual origin still appears to be at the bottom of the cube so the Path Workbench says that is where 0 is even though it actually should be -26.14mm in this case. So I'm missing something basic here.

The operations are guessing at start depth/end depth. So if the cube is positioned with Z=0 at the top, it should make start depth = 0 Final depth = -26.14.

If I can get some time this week, I'll record a new run-through video showing how I do things.

[burtbick]

Thanks that helps clear things up. I'll try some tests with the Draft clone approach, it sounds like what I was looking for.

I'm hoping that the changes since v.16 release will resolve one thing that I was seeing.

I did a test pocket with a slot 5mm deep. It milled nicely and was 5 mm deep, but it was larger by the width of my milling tool. It looked like it wound up milling on the line so an extra half of the tool width was cut all the way around the slot. Another test slot that I had done from the same file using EstlCam milled perfectly and didn't make the slot larger than it should have been.

There is something that I'm seeing in my current test. I was able to make an offset that appears to generate the correct gcode for depths, and on my slot the tool path looks like it should be correct to properly mill out the slot, but when I look at the gcode in the simulator it is just milling out a rectangle based on the horizontal sides of the slot.

I have tried selecting all of the sides of the slot and generating gcode from that and also tried it with just selecting the bottom face of the slot. In each case the resulting gcode only appears to be milling out a rectangle and not the arcs on the end of the slot.

Any suggestions on what I might me doing wrong that could cause that? In v.16 when I tried milling the slot it fully milled, with just the issue mentioned above with the larger size being milled out.

[EDIT} One other thing that I noticed. The Pocket operation for the slot is only cutting the rectangle, and a slight angle at the ends of the rectangle. It is not clearing all the material in the slot, but the toolpath that is drawn looks like it might do the entire area. Suggestions on how to track down what I might be doing wrong? {End of EDIT]

Thanks again, off to testing with the Draft clone approach.

Burt

[burtbick]

Think I figured out one thing. The default stepover was too large so that was why it wasn't taking out all the material. So that is corrected, but it still doesn't appear to be clearing the arcs at each end of the slot. So that is still a puzzle. The simulator in ChiliPeppr has been pretty accurate, but in this case it shows the arcs at the end so they must be defined in the gcode, but the simulation doesn't do anything with them.

I'll try a test cut to see if that might be just the simulation having indigestion for some reason.

[sliptonic]

Think I figured out one thing. The default stepover was too large so that was why it wasn't taking out all the material. So that is corrected, but it still doesn't appear to be clearing the arcs at each end of the slot. So that is still a puzzle. The simulator in ChiliPeppr has been pretty accurate, but in this case it shows the arcs at the end so they must be defined in the gcode, but the simulation doesn't do anything with them.

I'll try a test cut to see if that might be just the simulation having indigestion for some reason.

If you can share a project file, I can take a look too.

[burtbick]

On the arcs at the ends of the slot it appears that due to the size of the slot the simulator in ChiliPeppr didn't show it milling the arcs, but it did in practice. But there is an issue.

The slot is 8mm wide and 18mm from end to end, and 5mm deep. The depth is exactly 5mm but where is should be 8mm it is 5.9mm and where it should be 18mm it is 9.24mm. Those measurements are approximate due to the compression of the foam when measuring, but close to what it is actually milling. So the good news is that the latest didn't physically mill larger, but it is off. Probably something that I don't have set right.

The machine Units are set to Metric.in Path Workbench and GRBL is showing that it is set to Millimeters as well. My end mill size is set to 3.19 (doesn't say mm but doesn't let me specify mm either. So I'm assuming that it definitely is Millimeters for the Diameter

For the Pocket operation the Algorithm is libarea, clearance height is set to 5mm, Final Depth is -5mm, Finish Depth was 0 so I wonder if that might have something to do with it? Will try with -5mm as well. Safe Height 3mm start depth 0mm and step down of 1.

Climb for cut mode, Keep Tool Down false, Material Allowance 0, Start at center, Step Over 10, ZigZag false, Start point 0,0,0 Use Start Point false, And using tool #1.

Taking a closer look the tool path that is shown for the slot doesn't go close enough to the sides of the slot, which probably explains the problem but not the cause. If I increase Material Allowance from 0 to 1 it restricts the milled area more, so what would be causing it to not generate a tool path closer to the line?