Advanced Search

I Solved it! Holy cow! it was crazy simple.  So here's the lowdown of what needed to happen.  The ngc in the remap had to be lowercase.  That's it.  I had no idea that you couldn't use an upper case ngc name when remapping.  Is this a bug or it's supposed to be that way?  Goodness that would have saved me about a year of challenge. 

I believe Andy has a more elegant solution using pure python as: index_lathe_tool_with_wear but I can't write in python so I was in too deep.  
So here's my road to success with T.   Paths and directories are very important and you need to pay attention to where your python and subroutine code lives (mine is home/jake/linuxcnc/python, and ./) respectively.
Since I had already put together a M6 remap it got pretty tricky with what to do.  I kept thinking that remapping with the fanucy lathe example was creating a circular loop but it wasn't. 
My turret has a closed loop stepper with fault output an air cylinder to lock/unlock the turret, an index, and a inductive sensor to see if the hirth joint is locked.  With that my m6 remap ngc just goes through and rotates the stepper motor as joint 2/axis A and I scaled everything so that one tool pocket is equal to one unit on A. ex: T0=A0 T1=A1.0 T4=A4.0  It has an exit if the stepper is faulted for whatever reason, or the turret hirth fails to lock.  The downside to using this method and not a carosel component is that g30 causes the A axis to want to go home so use G28 instead. In the end I leave the remapped M6 alone.
Next up was making sure the remap.py in the python directory had the remap.py with has an added"get_t_number" from the fanucy_lathe example config. 
Also make sure you've got stdglue.py and toplevel.py in that same directory.  
Modify your ini file to remap T using the prolog=get_t_number and then this is where I made the fatal mistake I named an ngc Ttotool.ngc
(I always got this weird error "file not open" with that name)  the name ended up ttotool.ngc and it worked perfectly.
Next make sure you copy the toolchange.ngc that's called out in the lathe_fanucy example remap T but rename it with the exact subroutine you plan on having in your subroutine directory.  MAKE SURE IT IS ALL LOWERCASE. 
Remapped T should theoretically work after you complete these steps.  If you get issues with python crashes, just go through the remap linuxcnc.org/docs/devel/html/remap/remap.html and reimport the python directories and try again editinging with the lathe_fanucy python changes.  
Good luck. 

After following the advice from tommylight and downloading and installing the files in that order did in fact work , the names aren't the same but i will post the files that worked for me .

After downloading i wasn't able to install by the GUI Package Installer , so i used the cmd :

the same for the rest of the files , after that is just to continuing to follow the tutorial from this link and it will be done.
Thanks to tommylight and rodw for the continuos help.

Lcvette wrote:

you are not in full screen mode, hit F11 to allow the display to correctly fit the screen.
 

I was going to do a reinstallation but first I wanted to try full screen and that was indeed it, now the man auto and mdi buttons are displayed correctly.
Forgive my ignorance but I am very new to all this.

Thank you so much.

The hm2_eth manual page has this information

man hm2_eth

Notes:

1. ethtool must be present (looks like it may be missing in the latest
LinuxCNC distribution so needs to be added)

2. This only applies to computers with Intel Ethernet hardware

 

spumco wrote:

Ismacr63 wrote:

I don't know why those modes aren't shown in my version. I just downloaded it, followed the installation instructions and it looked like this. 


How can I add them?
 
Sounds like something got messed up in your PB installation.

What Linux distribution are you using, and what version of LCNC?  This might have a bearing - I've had problems in the past with certain combinations/versions of Debian/LCNC/qtpyvcp/PB.  Seems like frequently some dependency or other invisible 'backstage' issue led to weird problems with PB.

I know it'll be a bit painful, but my suggestion is to make a copy of your config and then uninstall and reinstall PB using the instructions here:
forum.linuxcnc.org/qtpyvcp/53871-probe-b...on-start-here#310377

I'm using LMD6 and LCNC 2.9.3

I'm going to try to do a reinstall.

I'm using axis GUI if that matters.

Assume I have homed X, Y, and Z

Later I want to re-home say the X axis, if I do this via the Axis GUI it will pop up a message asking if I'm sure that's what I want to do and if I click yes, it now re-homes that axis.
.........
Now doing the same using HAL commands, after power-up, I can home say the X axis with halui.joint.0.home, and that works fine.
Later if I want to re-home, and I issue another halui.joint.0.home, I get a "must be in joint Mode" error message
I suppose I could send a halui.joint.0.unhome, followed by a halui.joint.0.home and this works.

Is there a more direct way to re-home using HAL after the initial home?
 

I'm have a .py script that sends keystrokes to LinuxCNC.

Basically you define the the keystroke in one array for example:
sendArray { "G1", "X",  "space"}  etc.

Then in a matching array you specify if that item should be sent as text or as a keystroke, so for the above example  "G1" and "X" would be specified as 2, 2 means just send as text, "space" would be specified as type 1 and "space" would be sent as if you pressed the space bar key.

I have this working for all desired keys except the backspace key.

I'm not sure what the correct text is for the backspace key, I have tried:
backspace
Backspace
Back Space
back space
None of these are correct

I struggled with the Enter key earlier until I tried "Return" and that works, so just the backspace key is unknown at this point.

Where can I find the correct definition that invokes the backspace key?

I'm using this with Axis if that matters.

Edit: solved the correct sequence is "BackSpace"

I tried to reverse the z-axis and x-axis motors but the problem remains the same. this evening I'll try to create a new minimum configuration to test, using the 5th free exit of the mesa instead of this one. If it doesn't work, as soon as I can I'll dismantle all the components of the axle to check them. I can't think of anything else...

Everything you need is here
forum.linuxcnc.org/10-advanced-configura...ning-detailed-how-to

You might also add:

pcie.aspm=0 r8168.eee_enable=0 pcie_aspm=off idle=poll

To your kernel command line as this has helped on other systems

OK I would leave them at 20000 since is has no significant effect on performance
and eliminates a possible source of errors.

Could still be mechanical, a problem with the drive or encoder, etc.

Can you swap drives easily?

No. error finishing read continues.

The result I posted in above last post is with IRQ in CPU1.

I have attached two screen shot with latency numbers and tmax values (linuxcnc and latency test run together). Please refer the file name for CPU number.

I noticed that while it was running (i.e before read error occured), the tmax values were similar, bur with IRQ to CPU0, network-latency spikes appear and causes read error.

So, IRQ affinity is a must.

are there any one on one tutor or services available, that can help some one over come there learning curve, can someone point me in the right direction pls I really  need help. I have mesa CT installed and I am using 7i797T 

Lets see if this gets to you, it is your topic so you get emails by default on every reply.

It's not possible to switch to manual from gcode but you could set an analog output in gcode to indicate to a python component that you want to switch to manual mode when the gcode finishes or aborts (ie the interpreter switches to IDLE):