- Posts: 798
- Thank you received: 0
Work with probe
change to DEBUG = 0
[DISPLAY]
DISPLAY = axis
MBED_TAB_NAME=Probe Screen
change to EMBED_TAB_NAME = Probe Screen
EMBED_TAB_COMMAND=halcmd loadusr -Wn gladevcp gladevcp -c gladevcp -u python/probe_screen.py -x {XID} probe_icons/probe_screen.glade
Please Log in to join the conversation.
I do not know how it happened, probably a copy paste
I have another question, out of curiosity.
because if I delete files .axisrc go out and return from LinuxCNC does it anyway?
Thanks
Please Log in to join the conversation.
Please Log in to join the conversation.
The first problem was the code not waiting for gcode and ocode segments to finish. I determined that the function linuxcnc.wait_complete() sometimes will return on the first probe signal and not when the ocode or gcode segment completed. This led to a variety of problems. I worked around the problem by replacing the wait_complete() with:
while self.stat.interp_state <> linuxcnc.INTERP_IDLE :
time.sleep(.1)
self.stat.poll()
This fixed all of the strangeness, but I ran into another, related issue.
Before issuing a command the code switches to MDI mode using the following lines:
self.command.mode( linuxcnc.MODE_MDI )
self.command.wait_complete()
Once again I confirmed that intermittently the wait_complete() does not wait and the command gets issued without being in MDI mode, getting an error. I put a debug statement after the wait to print an error if we get there without being in MDI mode. I can't figure out a way to work around this in the probe code. Unless someone else has some insight I'll delve into the linuxcnc code and try to figure out why it is not working. It seems odd that Linuxcnc would work at all if such a common function such as this is broken so I must be missing something.
BTW: With the first fix everything works as long as you manually set the mode to MDI before probing.
Please Log in to join the conversation.
csbrady wrote: It seems odd that Linuxcnc would work at all if such a common function such as this is broken so I must be missing something.
I wonder if it is somehow specific to Glade? Axis uses the sequence as-written and is probably the best-tested UI we have.
Please Log in to join the conversation.
Debug file information:
Can not find -sec RS274NGC -var PARAMETER_FILE -num 1
Can not find -sec DISPLAY -var INTRO_GRAPHIC -num 1
Can not find -sec DISPLAY -var INTRO_TIME -num 1
.
Traceback (most recent call last):
File "/usr/bin/axis", line 2916, in <module>
max_feed_override = float(inifile.find("DISPLAY", "MAX_FEED_OVERRIDE"))
TypeError: float() argument must be a string or a number
4796
PID TTY STAT TIME COMMAND
4845
PID TTY STAT TIME COMMAND
Stopping realtime threads
Unloading hal components
Please Log in to join the conversation.
gizmo6023 wrote: max_feed_override = float(inifile.find("DISPLAY", "MAX_FEED_OVERRIDE"))
TypeError: float() argument must be a string or a number
What is the entry in MAX_FEED_OVERRIDE in the INI? (perhaps it is blank?)
Please Log in to join the conversation.
andypugh wrote:
gizmo6023 wrote: max_feed_override = float(inifile.find("DISPLAY", "MAX_FEED_OVERRIDE"))
TypeError: float() argument must be a string or a number
What is the entry in MAX_FEED_OVERRIDE in the INI? (perhaps it is blank?)
this is the line from my .ini
MAX_FEED_OVERRIDE = 1.2
Please Log in to join the conversation.
Do you have a text editor that allows you to turn on visibility of invisible characters?
I wonder if there is something wierd on that line.
You could try deleting it and re-typing it.
Please Log in to join the conversation.
andypugh wrote: Odd.
Do you have a text editor that allows you to turn on visibility of invisible characters?
I wonder if there is something wierd on that line.
You could try deleting it and re-typing it.
I do not think my editor is that sophisticated but but not sure, Linux is a foreign language to me at the moment, only been using it for a little over a month while attempting to get my machine running.
I deleted and re entered the value in the .ini and get the same error message.
Please Log in to join the conversation.
