Lathe Macros

I have it under the sim directory of the 2.7.0. I am not sure if I put it there myself but I dont think so.

I have put an updated file in the first post.

The difference is a more logical sequence of events on toolchange and that there is now a button to load a tool.
(Also, this is my current working version of each file).
I have remembered that the ability to do in-box unit conversions isn't part of this config, it is something I did to _all_ HAL widgets in the widget handler:
thread.gmane.org/gmane.linux.distributions.emc.devel/13550

Thanks again andy. I'll give it a shot tomorrow.

Drew

mariusl wrote: I have it under the sim directory of the 2.7.0. I am not sure if I put it there myself but I dont think so.

Ok it looks like I put it there myself but I see no reason why it cannot be placed in the distro. How do I go about voting to get it in the sim directory officially?

Hi andy,

It looks like there's a bug in all the .ngc files where variable #2 is used in the "G96 D1500 S#2 ; Constant Surface Speed" lines before it's set. The fix I tried was to move the setting of the #1 thru #8 vars to the beginning of the subroutines. That seems to work.

Drew

drogge wrote: It looks like there's a bug in all the .ngc files where variable #2 is used in the "G96 D1500 S#2 ; Constant Surface Speed" lines before it's set. The fix I tried was to move the setting of the #1 thru #8 vars to the beginning of the subroutines. That seems to work.

One change I made was to move them all to after the "setup" codes. This was to prevent tool crashes when the positions are read in before the tool offset was applied.

It looks like I need to have a more careful think about this.

I also may decide to send the numbers from the GUI to the subroutines as parameters, so that the NGC code does not have to use the "experimental" ability to read HAL pins directly.

andypugh wrote: It looks like I need to have a more careful think about this.

I also may decide to send the numbers from the GUI to the subroutines as parameters, so that the NGC code does not have to use the "experimental" ability to read HAL pins directly.

Since only the variables 1 thru 8 have been moved above the setup code then the position code shouldn't be affected. It's not much different than passing things in as parameters.

andypugh wrote: I have put an updated file in the first post.

For me, the zip file comes up with an empty "lathemacro.hal" file which is the POSTGUI_HALFILE according to the .ini file, and from what I know that file should contain the _hal[gladevcp.turn.z-f] ... definitions? linuxcnc complains.
Karl

kjacobs wrote:

andypugh wrote: I have put an updated file in the first post.

For me, the zip file comes up with an empty "lathemacro.hal" file which is the POSTGUI_HALFILE according to the .ini file, and from what I know that file should contain the _hal[gladevcp.turn.z-f] ... definitions? linuxcnc complains.
Karl

Mine is zero length too but it doesn't seem to matter. The problem you're having may be that your linuxcnc is too old. I thing the _hal stuff showed up in 2.6. I'm running 2.6.5.

Drew

drogge wrote:

kjacobs wrote:

andypugh wrote: I have put an updated file in the first post.

Mine is zero length too but it doesn't seem to matter. The problem you're having may be that your linuxcnc is too old. I thing the _hal stuff showed up in 2.6. I'm running 2.6.5.

Drew

I am running 2.7.0~pre on the machinekit branch, The problem may be in that branch, then. The gladevcp panel and everything else comes up fine.
Karl