Current turning capabilities status

You could use multiple offsets with the same tool number - You may need the fanuc tool patch to do this though...

Chris M

I had a feeling that this was already solved, or at least covered in some form.
After cmorley wrote "fanuc tool path", google gave me:

sourceforge.net/p/emc/feature-requests/103/

This is dated more than 1 year ago, so probably the patch is now better/more stable. Were can I find (or know if exists) an updated version for the patch?

Unfortunately we do not have the knowledge how linuxcnc is structured / updated / contributed, and for example if cmorley did not posted, probably we would never knew about it. I hope not to sound ungrateful (because I am really not).

We already have all the control panel working and servos working. It is really nice to use the machine with the wheel jog !
Regarding the rotation of the coordinate system as Andy pointed out, it works fine (at least it looks that way).

We are now close to the first testing. For now it seams that the tool call is out priority and to check if the tool compensation (G41 G42) is really working as we use it extensively.

Basically we are working on this for 8 months.
All hardware is set up except the VFD that we had to acquire one more powerful than the one that we first started to test with. Main motor is 20kw @ 200V. As it was hard to get one compatible voltage 200V (we are in Europe). Since modern VFD can limit the output voltage, in theory, a 40Kw @ 400V would be ok to delivery the same amperage that the motor may require. The one that we could bought (ebay) was for 100kw @ 400V . Lets hope that my theory for limiting the output voltage is true, and that it will trully limit the output voltage not to damage the motor. At least I will not see "Over current" in the VFD

Today I tested how the CSS works and the tool compensation modes. Both are important for my use.
All tests were done on simulation.

The CSS appears to be fine, but the compensation probably have some issue that I do not understand.
I tested the simple code below:

Nothing strange about it. The tool 101 is well defined with a radius of 0.2mm (that on the tool table is supposedly to be in diameter. The value on the tool table is 0.2 and seams to be ok). The problem is that as can be seen on the picture attached, that the movement from point X35 Z-5 to point X35 Z-10, the machine is in diameter X35.4 instead of X35 (I know that 35.4 = 35 + 0.2 + 0.2 that could be related with the tool radius).
If the movement is along the Z axis, the commanded position in X should match the tool position even if the compensation is on. In lathes the DRO should display the imaginary tool tip. It looks that the DRO is showing the tool tip center like in mills, that is inappropriate for lathes.

Can someone please inform if G41 G42 is really working as stated on several linuxcnc documentation for lathes? I am not sure if I am missing some config.

A bit about the same issue:
The info on the following link is wrong. The X+ arrow is swapped. Linuxcnc toolpath is correct, and behaves differently from the following link.
www.linuxcnc.org/docs/2.4/html/lathe_lathe-user.html#r1

The same error is here:
wiki.linuxcnc.org/cgi-bin/wiki.pl?Lathe_Advanced_Features

and the correct tooltip direction can be seen on the file attached.

Thank you for your feedback. G18 on my code will force to be in XZ plane.
I will cover (test) this issue during today so that I can reply properly.

So I tested the best I could:

a) The backtool script change: It works good (X and Z are properly displayed), BUT the tool representation disappears. This is something that I can live with, but of course it would be better to have the tool displayed. Any ideas in where in the code (source code) this is handled?

b) the tool compensation is also not affected by the geometry rotation. The issue here is the tool tip direction that in linuxcnc is different from all others controllers that I worked with (Fanuc, Okuma, Mitsubishi, Syntec). 3 and 2 (and possible others are swapped).

Using like this, it displays correctly.

This is a edit to my original post, where I figured out that changing the tooltip direction would correct the issue (although I still think that the tip direction should be corrected and be equal to all other machines).