Discussion:
joint mode jogging in master
(too old to reply)
Phillip Carter
2017-04-15 10:29:22 UTC
Permalink
Raw Message
I have just noticed in sim/axis/gantry that joint 3 does not jog with
the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
--
Cheers, Phill
Gene Heskett
2017-04-15 23:17:23 UTC
Permalink
Raw Message
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog with
the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw and
sawed off a couple hunks on 1.5" thick alu to make a new set of bearing
pushers to fit the larger OD of the correct bearing. Fired up
linuxcnc -l, (this after an early morning update of LinuxCNC-2.8-pre to
LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z are still swapped at the
keyboard yet before being homed. Hit ctl+home, and it homed to the
correct places, and in the correct directions and homeing order. Tapped
the uparrow, then downarrow, and z moves left-right. Tapped the left
and right arrows, and x moves in and out. Switch to MDI, and its now
scrambled as well, as in G0X moves Z and G0Z moves X. So it appears
Phillips commit made the keyboard scramble apply to MDI too, which means
it cannot run code I ran on it just 2 days ago.

Help! I do not have another functioning lathe now.

Can you back out Phillips commit and retrigger the buildbot? It at least
ran correctly after being homed, and I could use it, once it had been
homed.

Thanks guys.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-04-15 23:35:57 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog
with the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw and
sawed off a couple hunks on 1.5" thick alu to make a new set of
bearing pushers to fit the larger OD of the correct bearing. Fired up
linuxcnc -l, (this after an early morning update of LinuxCNC-2.8-pre
to LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z are still swapped at
the keyboard yet before being homed. Hit ctl+home, and it homed to the
correct places, and in the correct directions and homeing order.
Tapped the uparrow, then downarrow, and z moves left-right. Tapped
the left and right arrows, and x moves in and out. Switch to MDI, and
its now scrambled as well, as in G0X moves Z and G0Z moves X. So it
appears Phillips commit made the keyboard scramble apply to MDI too,
which means it cannot run code I ran on it just 2 days ago.
Help! I do not have another functioning lathe now.
Can you back out Phillips commit and retrigger the buildbot? It at
least ran correctly after being homed, and I could use it, once it had
been homed.
Thanks guys.
Cheers, Gene Heskett
Touring what could have been hell, I just installed that version on both
of my mills, and things appear to be working as expected on all 7 axis's
they have. So this seems to be a lathe only problem. That should help
narrow the search.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-04-16 05:23:24 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Gene Heskett
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog
with the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw and
sawed off a couple hunks on 1.5" thick alu to make a new set of
bearing pushers to fit the larger OD of the correct bearing. Fired
up linuxcnc -l, (this after an early morning update of
LinuxCNC-2.8-pre to LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z are
still swapped at the keyboard yet before being homed. Hit ctl+home,
and it homed to the correct places, and in the correct directions
and homeing order. Tapped the uparrow, then downarrow, and z moves
left-right. Tapped the left and right arrows, and x moves in and
out. Switch to MDI, and its now scrambled as well, as in G0X moves
Z and G0Z moves X. So it appears Phillips commit made the keyboard
scramble apply to MDI too, which means it cannot run code I ran on
it just 2 days ago.
Help! I do not have another functioning lathe now.
Can you back out Phillips commit and retrigger the buildbot? It at
least ran correctly after being homed, and I could use it, once it
had been homed.
Thanks guys.
Cheers, Gene Heskett
Touring what could have been hell, I just installed that version on
both of my mills, and things appear to be working as expected on all 7
axis's they have. So this seems to be a lathe only problem. That
should help narrow the search.
Cheers, Gene Heskett
Poking and prodding with the lathe, I turned on debugging to 0x7fffffff
and ran it, and I note in the terminal a couple things that look odd,
but I don't have a working lathe to compare it to, so I'll paste them
here.

task: machine: 'The Little Monster' version 'unknown'
emcTaskOnce: Python plugin configured
emcTaskOnce: extract(task_instance): KeyError: ('task',)

emcTaskOnce: no Python Task() instance available, using default
iocontrol-based task methods
emcTrajSetJoints(2) returned 0
emcTrajSetAxes(3, 5)

The config does have a LATHE=1 in the display section.

And only XZ are specified anyplace.
***@lathe:~/linuxcnc/configs/my_LinuxCNC_machine2$ grep -i xz
my_LinuxCNC_machine2.ini
GEOMETRY = XZ
JOG_AXES = XZ
COORDINATES = XZ
KINEMATICS = trivkins coordinates=XZ

What does the above tell us? Line by line please.

Thanks all.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-04-16 05:41:08 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Gene Heskett
Post by Gene Heskett
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog
with the leftbracket/rightbracket keyboard keys but does with
the minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw
and sawed off a couple hunks on 1.5" thick alu to make a new set
of bearing pushers to fit the larger OD of the correct bearing.
Fired up linuxcnc -l, (this after an early morning update of
LinuxCNC-2.8-pre to LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z
are still swapped at the keyboard yet before being homed. Hit
ctl+home, and it homed to the correct places, and in the correct
directions and homeing order. Tapped the uparrow, then downarrow,
and z moves left-right. Tapped the left and right arrows, and x
moves in and out. Switch to MDI, and its now scrambled as well,
as in G0X moves Z and G0Z moves X. So it appears Phillips commit
made the keyboard scramble apply to MDI too, which means it cannot
run code I ran on it just 2 days ago.
Help! I do not have another functioning lathe now.
Can you back out Phillips commit and retrigger the buildbot? It at
least ran correctly after being homed, and I could use it, once it
had been homed.
Thanks guys.
Cheers, Gene Heskett
Touring what could have been hell, I just installed that version on
both of my mills, and things appear to be working as expected on all
7 axis's they have. So this seems to be a lathe only problem. That
should help narrow the search.
Cheers, Gene Heskett
Poking and prodding with the lathe, I turned on debugging to
0x7fffffff and ran it, and I note in the terminal a couple things that
look odd, but I don't have a working lathe to compare it to, so I'll
paste them here.
task: machine: 'The Little Monster' version 'unknown'
emcTaskOnce: Python plugin configured
emcTaskOnce: extract(task_instance): KeyError: ('task',)
emcTaskOnce: no Python Task() instance available, using default
iocontrol-based task methods
emcTrajSetJoints(2) returned 0
emcTrajSetAxes(3, 5)
The config does have a LATHE=1 in the display section.
And only XZ are specified anyplace.
my_LinuxCNC_machine2.ini
GEOMETRY = XZ
JOG_AXES = XZ
Now, after going thru the docs for the ini file on a line by line,
section by section, I feel like an idijit.
There, it plainly says the left-right arrows are assigned to the first
character of the above JOG_AXES string. So I now have
JOG_AXES = ZX, only place in the whole file where its "out of order",
but it looks like its working. Sheesh. Go ahead, tell me to RTFM AGAIN, I
deserve it. Thats w/o motor power, so I can't home it from here, but I'm
thinking it will be ok. It loaded the same gcode it was running 2 days
back without a single complaint, drawing exactly the same backplot.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Phillip Carter
2017-04-16 05:45:53 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Gene Heskett
Post by Gene Heskett
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog
with the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw and
sawed off a couple hunks on 1.5" thick alu to make a new set of
bearing pushers to fit the larger OD of the correct bearing. Fired
up linuxcnc -l, (this after an early morning update of
LinuxCNC-2.8-pre to LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z are
still swapped at the keyboard yet before being homed. Hit ctl+home,
and it homed to the correct places, and in the correct directions
and homeing order. Tapped the uparrow, then downarrow, and z moves
left-right. Tapped the left and right arrows, and x moves in and
out. Switch to MDI, and its now scrambled as well, as in G0X moves
Z and G0Z moves X. So it appears Phillips commit made the keyboard
scramble apply to MDI too, which means it cannot run code I ran on
it just 2 days ago.
Help! I do not have another functioning lathe now.
Can you back out Phillips commit and retrigger the buildbot? It at
least ran correctly after being homed, and I could use it, once it
had been homed.
Thanks guys.
Cheers, Gene Heskett
Touring what could have been hell, I just installed that version on
both of my mills, and things appear to be working as expected on all 7
axis's they have. So this seems to be a lathe only problem. That
should help narrow the search.
Cheers, Gene Heskett
Poking and prodding with the lathe, I turned on debugging to 0x7fffffff
and ran it, and I note in the terminal a couple things that look odd,
but I don't have a working lathe to compare it to, so I'll paste them
here.
task: machine: 'The Little Monster' version 'unknown'
emcTaskOnce: Python plugin configured
emcTaskOnce: extract(task_instance): KeyError: ('task',)
emcTaskOnce: no Python Task() instance available, using default
iocontrol-based task methods
emcTrajSetJoints(2) returned 0
emcTrajSetAxes(3, 5)
The config does have a LATHE=1 in the display section.
And only XZ are specified anyplace.
my_LinuxCNC_machine2.ini
GEOMETRY = XZ
JOG_AXES = XZ
COORDINATES = XZ
KINEMATICS = trivkins coordinates=XZ
What does the above tell us? Line by line please.
Thanks all.
Cheers, Gene Heskett
For a lathe JOG_AXES=ZX would be best and is the default if not specified.
See extract from
http://linuxcnc.org/docs/devel/html/config/ini-config.html#_display_section

*

/JOG_AXES =/- The order in which jog keys are assigned to axis
letters. The left and right arrows are assigned to the first axis
letter, up and down to the second, page up/page down to the third,
and left and right bracket to the fourth. If unspecified, the
default is determined from the [TRAJ]COORDINATES, [DISPLAY]LATHE and
[DISPLAY]FOAM values.
--
Cheers, Phill
Gene Heskett
2017-04-16 05:53:39 UTC
Permalink
Raw Message
Post by Phillip Carter
Post by Gene Heskett
Post by Gene Heskett
Post by Gene Heskett
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog
with the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw
and sawed off a couple hunks on 1.5" thick alu to make a new set
of bearing pushers to fit the larger OD of the correct bearing.
Fired up linuxcnc -l, (this after an early morning update of
LinuxCNC-2.8-pre to LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z
are still swapped at the keyboard yet before being homed. Hit
ctl+home, and it homed to the correct places, and in the correct
directions and homeing order. Tapped the uparrow, then downarrow,
and z moves left-right. Tapped the left and right arrows, and x
moves in and out. Switch to MDI, and its now scrambled as well,
as in G0X moves Z and G0Z moves X. So it appears Phillips commit
made the keyboard scramble apply to MDI too, which means it cannot
run code I ran on it just 2 days ago.
Help! I do not have another functioning lathe now.
Can you back out Phillips commit and retrigger the buildbot? It at
least ran correctly after being homed, and I could use it, once it
had been homed.
Thanks guys.
Cheers, Gene Heskett
Touring what could have been hell, I just installed that version on
both of my mills, and things appear to be working as expected on
all 7 axis's they have. So this seems to be a lathe only problem.
That should help narrow the search.
Cheers, Gene Heskett
Poking and prodding with the lathe, I turned on debugging to
0x7fffffff and ran it, and I note in the terminal a couple things
that look odd, but I don't have a working lathe to compare it to, so
I'll paste them here.
task: machine: 'The Little Monster' version 'unknown'
emcTaskOnce: Python plugin configured
emcTaskOnce: extract(task_instance): KeyError: ('task',)
emcTaskOnce: no Python Task() instance available, using default
iocontrol-based task methods
emcTrajSetJoints(2) returned 0
emcTrajSetAxes(3, 5)
The config does have a LATHE=1 in the display section.
And only XZ are specified anyplace.
my_LinuxCNC_machine2.ini
GEOMETRY = XZ
JOG_AXES = XZ
COORDINATES = XZ
KINEMATICS = trivkins coordinates=XZ
What does the above tell us? Line by line please.
Thanks all.
Cheers, Gene Heskett
For a lathe JOG_AXES=ZX would be best and is the default if not
specified. See extract from
http://linuxcnc.org/docs/devel/html/config/ini-config.html#_display_se
ction
I just found that about half an hour ago, on at least the 3rd or 4th read
before it sank in.

I believe it may be fixed now. I'll find out for sure come daytime
again. Thanks Phillip, for not calling me an idiot who can't RTFM.
Post by Phillip Carter
*
/JOG_AXES =/- The order in which jog keys are assigned to axis
letters. The left and right arrows are assigned to the first axis
letter, up and down to the second, page up/page down to the third,
and left and right bracket to the fourth. If unspecified, the
default is determined from the [TRAJ]COORDINATES, [DISPLAY]LATHE
and [DISPLAY]FOAM values.
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-04-16 19:07:48 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Phillip Carter
For a lathe JOG_AXES=ZX would be best and is the default if not
specified. See extract from
http://linuxcnc.org/docs/devel/html/config/ini-config.html#_display_
se ction
I just found that about half an hour ago, on at least the 3rd or 4th
read before it sank in.
I believe it may be fixed now. I'll find out for sure come daytime
again. Thanks Phillip, for not calling me an idiot who can't RTFM.
Post by Phillip Carter
*
/JOG_AXES =/- The order in which jog keys are assigned to axis
letters. The left and right arrows are assigned to the first
axis letter, up and down to the second, page up/page down to the
third, and left and right bracket to the fourth. If unspecified, the
default is determined from the [TRAJ]COORDINATES, [DISPLAY]LATHE and
[DISPLAY]FOAM values.
Cheers, Gene Heskett
The fix was temporary, I had been watching it while fineing up the size
of the bolt hole thru a block of alu, thinking it ought to be big enough
for the bolt (it was) but something happen inside the motor doghouse
just as I slid the feed override to 0.

I keep forgetting to re-engage the backgear, so the pid spins the motor
wide open when the spindle doesn't turn, so I had put a 100 millisecond
timer, retriggerable, netted to shut the motor drive off again by
disabling the machine if an encoder pulse has not been rx'd within the
timeout. And thats where I am. So I've got to wrestle the wheeled
toolbox its sitting on, back out of its cubby hole, and uncover the
doghouse to see whats happened. I've had so much alu swarf flying
around I will not be surprised to find a curly bit of it someplace its
not supposed to be.

But Easter Sunday dinner awaits, by baby felt like doing some of the
cooking today.

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Gene Heskett
2017-04-17 01:19:56 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Gene Heskett
Post by Phillip Carter
For a lathe JOG_AXES=ZX would be best and is the default if not
specified. See extract from
http://linuxcnc.org/docs/devel/html/config/ini-config.html#_displa
y_ se ction
I just found that about half an hour ago, on at least the 3rd or 4th
read before it sank in.
I believe it may be fixed now. I'll find out for sure come daytime
again. Thanks Phillip, for not calling me an idiot who can't RTFM.
Post by Phillip Carter
*
/JOG_AXES =/- The order in which jog keys are assigned to axis
letters. The left and right arrows are assigned to the first
axis letter, up and down to the second, page up/page down to the
third, and left and right bracket to the fourth. If unspecified,
the default is determined from the [TRAJ]COORDINATES,
[DISPLAY]LATHE and [DISPLAY]FOAM values.
Cheers, Gene Heskett
The fix was temporary, I had been watching it while fineing up the
size of the bolt hole thru a block of alu, thinking it ought to be big
enough for the bolt (it was) but something happen inside the motor
doghouse just as I slid the feed override to 0.
I keep forgetting to re-engage the backgear, so the pid spins the
motor wide open when the spindle doesn't turn, so I had put a 100
millisecond timer, retriggerable, netted to shut the motor drive off
again by disabling the machine if an encoder pulse has not been rx'd
within the timeout. And thats where I am. So I've got to wrestle the
wheeled toolbox its sitting on, back out of its cubby hole, and
uncover the doghouse to see whats happened. I've had so much alu
swarf flying around I will not be surprised to find a curly bit of it
someplace its not supposed to be.
But Easter Sunday dinner awaits, by baby felt like doing some of the
cooking today.
Cheers, Gene Heskett
Which came out well and I pigged out. That in turn put me down for the
count for a couple hours, probably too much sugar. But I recovered,
drug the toolbox out, opened up the back panel where all the spindle
electronics live, and spent 20 minutes blowing the swarf out of things.
Covered it back up, fixed the x axis drag chain a bit better, and
finished one end of a bearing pusher before calling it a night. Added
some code to take care of some of TLM's built-in taper, guessing
at .003mm/mm for external turning, bigger toward the chuck, must have
been a good guess, my calipers can't find any taper at all in what I
just did.

However, when boring I know it will take more than that as it bores small
at the chuck, and at least 15 thou an inch bigger as it moves toward the
entry of the hole. That, I've given up trying to find the why. Internal
taper and external are off in the same direction, but internal is
several times worse. I keep forgetting it was a $350 Speedway off the
Chicago Tool Touring truck parked at the local VFD for a one day sale
back in the late 90's before I retired. A Hardinge it isn't. Someday I
might get a better "little" lathe. If I last long enough. :)

Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Phillip Carter
2017-04-15 23:42:55 UTC
Permalink
Raw Message
Post by Gene Heskett
Post by Phillip Carter
I have just noticed in sim/axis/gantry that joint 3 does not jog with
the leftbracket/rightbracket keyboard keys but does with the
minus/equals keyboard keys. Is this the correct behaviour?
I also, for lack of anything better to do, rigged up my chop saw and
sawed off a couple hunks on 1.5" thick alu to make a new set of bearing
pushers to fit the larger OD of the correct bearing. Fired up
linuxcnc -l, (this after an early morning update of LinuxCNC-2.8-pre to
LINUXCNC - 2.8.0-pre1-2988-ge9f9469) x and z are still swapped at the
keyboard yet before being homed. Hit ctl+home, and it homed to the
correct places, and in the correct directions and homeing order. Tapped
the uparrow, then downarrow, and z moves left-right. Tapped the left
and right arrows, and x moves in and out. Switch to MDI, and its now
scrambled as well, as in G0X moves Z and G0Z moves X. So it appears
Phillips commit made the keyboard scramble apply to MDI too, which means
it cannot run code I ran on it just 2 days ago.
Help! I do not have another functioning lathe now.
Can you back out Phillips commit and retrigger the buildbot? It at least
ran correctly after being homed, and I could use it, once it had been
homed.
Thanks guys.
Cheers, Gene Heskett
I have just checked again with both sim/axis/lathe and
sim/axis/historical_lathe.

Keyboard jogging before and after homing works as it should:

left/right arrows move Z and up/down arrows move X

mdi moves are also moving correctly
--
Cheers, Phill
Loading...