#flex-dev Logs

Jun 09 2025

04:42 AM Tom_L: morning
05:17 AM JT-Cave: morning
06:30 AM Tom_L: you may want to create a separate prefx from probe_ with reversed logic so not to interfere with other probe_ widgets
06:56 AM JT-Cave: not sure I follow
07:20 AM JT-Cave: 1: on startup, estop, power on, home, -> settings use ^ v arrows to change a setting, it changes the first press then locks up
07:21 AM JT-Cave: I don't get that behavior, I clicked into search feed tp3 and used the up down arrows and no freeze
07:22 AM JT-Cave: I do see that the spin box should be disabled at startup but it's not
07:24 AM JT-Cave: I see the prefix is pbe_ hmm
07:26 AM JT-Cave: possibly they are disabled because they have the dynamic property set_var
07:26 AM JT-Cave: when I grep for pbe I get no results
07:29 AM JT-Cave: grep in scr
07:31 AM JT-Cave: the only place I can find pbe_ being used is rpi_mill_probe_ops.ui and touch-probe3.ui
07:33 AM Tom_L: i changed one of my settings to probe_ to test
07:34 AM Tom_L: it must have been pbe_ early on and got changed
07:35 AM JT-Cave: searching git I only find commits from you that have "pbe_"
07:35 AM Tom_L: i wonder if i renamed them so they weren't disabled
07:35 AM Tom_L: since the logic seemed backward at the time
07:35 AM Tom_L: that's a strong possibility
07:37 AM Tom_L: on 1: if i do that sequence with probe_ widget doublespinbox and push the ^ arrow to change it, it will do one change then quit
07:37 AM Tom_L: (disable)
07:38 AM Tom_L: i would make a different prefix un_probe_something with the logic reversed so it doesn't interfere with other probe_ widgets
07:38 AM Tom_L: the buttons work great
07:39 AM Tom_L: this didn't surface until i started 'using' the probe routines instead of just 'testing' them
07:43 AM Tom_L: we can leave the settings with pbe_ prefix and they seem to work 90% of the time
07:43 AM Tom_L: i haven't quite pinpointed the other 10% yet
07:44 AM Tom_L: ops use the .var file as well
07:44 AM Tom_L: but that shouldn't make any difference
07:46 AM Tom_L: did you put a timeout on set_var?
07:46 AM JT-Cave: a spinbox that sets a var "can" have any object name... if the prefix is probe_ then it is disabled until probing is enabled
07:46 AM Tom_L: so it didn't change as quickly
07:46 AM Tom_L: right
07:46 AM JT-Cave: let me look
07:47 AM Tom_L: i can do some more testing too
07:47 AM Tom_L: i'm pretty sure set_var had a timeer on it to slow it down a bit
07:48 AM Tom_L: which that time works well
07:49 AM JT-Cave: yup a 1/2 second timer for update
07:52 AM Tom_L: don't change that, that's just about perfect
07:54 AM Tom_L: i think the main issue was after changing settings with pbe_ vars i had to toggle the probe enable button to get the probe buttons back
07:55 AM Tom_L: and using probe_ on the vars is backward logic
07:58 AM Tom_L: the other issue with rigid tapping i can put a warning to make sure feed_override and spindle_override are at 100%
07:59 AM Tom_L: i'd changed them quite a bit when testing the ops and forgot about rigid tapping
08:00 AM Tom_L: basically i'd turned the spindle off to watch the motion
08:00 AM JT-Cave: probe buttons? do you mean the mdi buttons that start a probe routine?
08:00 AM Tom_L: lemme fire up the pc
08:00 AM Tom_L: i think they're mdi yes
08:01 AM JT-Cave: I'm running touch-probe3
08:01 AM Tom_L: whatever activates the probe routine buttons
08:02 AM Tom_L: yes those are mdi
08:02 AM JT-Cave: I see that now...
08:03 AM Tom_L: ok, do an edge probe sim using the trip button
08:03 AM Tom_L: that works fine
08:03 AM JT-Cave: I think it's a mode issue
08:04 AM Tom_L: go to the settings page and change the XY Search dist
08:04 AM Tom_L: go back to the probe page and the buttons are disabled
08:04 AM Tom_L: even though probe enable is green
08:04 AM JT-Cave: I think that is a mode issue after changing the setting
08:04 AM Tom_L: could be
08:05 AM JT-Cave: I'll change touch-probe3 to add a mode label so I can watch it
08:05 AM Tom_L: it doesn't hurt anything to change the settings with probe enable button active or inactive really
08:05 AM Tom_L: mode is at the top right
08:06 AM Tom_L: you shouldn't need to add one
08:07 AM Tom_L: after a probe routine it is in manual
08:07 AM Tom_L: after changing a setting it is in mdi
08:08 AM Tom_L: watch mode before and after each change
08:10 AM Tom_L: but toggling the probe enable button to get the probe buttons back doesn't change the mode
08:13 AM JT-Cave: I see the mode label now lol and like I thought it's stuck in mdi
08:14 AM Tom_L: we had a dreamlifter fly overhead yesterday
08:14 AM Tom_L: damn those things are big
08:25 AM Tom_L: thinking back now i'm pretty sure i changed the prefix on the settings from probe_ to pbe_ because of the enable logic
08:26 AM Tom_L: so they would be excluded from being disabled
08:28 AM JT-Cave: hmm with probe enabled and changing the value in a random named spin box the traj mode changes when the value is updated and I think that's the issue
08:28 AM Tom_L: gotta go here in 10
08:28 AM JT-Cave: it changes from teleop to coord
08:28 AM Tom_L: hmm
08:28 AM JT-Cave: yup time to start my day as well
08:29 AM JT-Cave: and a spin box with a prefix of probe_ is enabled after homing which seems wrong
08:30 AM Tom_L: didn't notice that
08:30 AM JT-Cave: I just modified xyz to test
08:30 AM Tom_L: the buttons aren't
08:30 AM JT-Cave: can't tell in touch probe 3 when something is disabled or not
08:31 AM Tom_L: yeah, i've been testing with it alot longer
08:31 AM JT-Cave: I have probe disabled and changed the spinbox with the prefix of probe_
08:32 AM Tom_L: yep, that shouldn't happen
08:32 AM Tom_L: later...
08:32 AM JT-Cave: once traj mode is coord it's ok so the change is the issue
08:32 AM JT-Cave: yup
09:35 AM lcnc-relay: <roguish> goood morning....
09:37 AM rdtsc-w: morning from the midwest
09:46 AM lcnc-relay: <roguish> JT-Cave: hey, do you know how the gcode preview works? why it doesn't really work with marcros ???
09:46 AM lcnc-relay: <roguish> or MDI's
11:31 AM JT-Cave: nom clue
11:31 AM JT-Cave: no
11:31 AM JT-Cave: what is a macro?
11:52 AM lcnc-relay: <roguish> macro, when a program calls another program.....
11:52 AM lcnc-relay: <roguish> just checking
11:53 AM lcnc-relay: <roguish> I like to see the probing path, but it's not often visible,
11:56 AM JT-Shop: oh a subroutine
12:00 PM JT-Shop: a probe move has an unknown outcome so it can't be previewed AFAIK
12:02 PM lcnc-relay: <roguish> not necessarily ttue. when it's written with definite distances, it can be shown. but I think it has to be fully written out so it goes thru the interpreter and the path planner.
12:03 PM JT-Shop: I've not tested that
12:03 PM JT-Shop: anyway if Axis shows it Flex will as well as they both use the same plotter
12:04 PM lcnc-relay: <roguish> i've written some out, and run them just as normal program, they show.
12:04 PM JT-Shop: ok
12:04 PM lcnc-relay: <roguish> i'll mess around with it a bit and see.
12:04 PM JT-Shop: nap time here...
12:04 PM lcnc-relay: <roguish> go for it
12:22 PM Tom_L: my sim shows the probe moves
12:37 PM Tom_L: JT-Shop, i will say if you're running a subroutine and have a program loaded, the active line marker (yellow) will follow the loaded gcode lines even though they're not being executed
12:37 PM Tom_L: i failed to mention that one earlier
12:41 PM Tom_L: roguish, running my ops routines also show up in the plot and they're all subroutines calle
12:41 PM Tom_L: d
12:56 PM Tom_L: yeah, preview would be near impossible
01:36 PM JT-Cave: the highlighted line is the current line being read by the interpreter I think
02:29 PM Tom_L: i'm not entirely sure what teleop or coord are
02:30 PM JT-Shop: seems like teleop is a single axis move and coord is more than one axis moving at the same time
02:30 PM JT-Shop: still trying to figure out exactly what is causing the issue
02:30 PM Tom_L: ok
02:30 PM Tom_L: new to me for sure
02:31 PM JT-Shop: new for me as well just fingered that out a bit ago
02:35 PM Tom_L: AI: In LinuxCNC, coord mode is for moving the machine using Cartesian coordinates and kinematics, while teleop mode is used for jogging the machine in world coordinates, with the joints moving as a coordinated system.
02:35 PM Tom_L: Coord mode uses the main trajectory planner for motion based on queued commands, while teleop mode primarily uses the axis GUI for jogging
02:52 PM lcnc-relay: <roguish> Tom_L: running your touch_probe3 now. it shows the movement lines, but does not preview the probe routine.
02:53 PM lcnc-relay: <roguish> I'd like to see a full preview of the probe routine.
02:55 PM JT-Shop: how would you preview a mdi command button?
02:55 PM lcnc-relay: <roguish> no clue
02:56 PM JT-Shop: me neither
02:56 PM JT-Shop: all the probe buttons are mdi command buttons...
02:59 PM lcnc-relay: <roguish> Tom_L: JT-Shop hey, I am not bitching or complaining. ok. just asking questions....
03:00 PM JT-Shop: and I'm just trying to answer
03:00 PM lcnc-relay: <roguish> ok, cool, thanks
03:02 PM JT-Shop: I really need to move my tool boxes to the other shop... I don't work on cars in the garage anymore
03:06 PM JT-Shop: however... that might be an interesting concept to have the button just load the file and use the normal run controls...
03:08 PM lcnc-relay: <roguish> might work. i'm looking back at the old axis sim examples. particularly the ngcgui ones.
03:09 PM lcnc-relay: <roguish> would that be much work for you to figure out how to just load a program with a button???
03:10 PM lcnc-relay: <roguish> JT-Shop: did you see the Goodyear blimps are now 100 years old????
03:10 PM JT-Shop: I think that would be pretty easy so long as the program is in one of the paths in the ini file
03:10 PM lcnc-relay: <roguish> didn't you drive one of those ???
03:10 PM JT-Shop: no, I drove the Saints blimp
03:11 PM lcnc-relay: <roguish> adding paths isn't too difficult. just gotta be careful with the format
03:13 PM lcnc-relay: <roguish> I have a probe routine or 2, that are similar to regular programs. the speeds and distances come from user parameters added in the .par file.
03:13 PM lcnc-relay: <roguish> when they're loaded, they're previewed.
03:18 PM JT-Shop: yup when you load a file it gets previewed
04:15 PM Tom_L: i think it would be pretty hard to preview an undetermined path
04:16 PM Tom_L: certainly able to as it is moving
04:16 PM JT-Shop: damn I found the blower relay plug on walmart and I have to create an account so they can bombard me with crap
04:17 PM JT-Shop: ah fleabay has the same plug
04:17 PM Tom_L: there you go
04:17 PM Tom_L: amazon likely does too
04:18 PM xxcoder: surpising walmart listens to your unsub request
04:18 PM xxcoder: I ordered rice cooker years ago, unsubbed and I neve got email since
04:18 PM xxcoder: ugh I still need to find lid for it. I want rice someday
04:19 PM JT-Shop: my wife has her grandmothers rice cooker and it looks that old
04:19 PM xxcoder: nice. I love mine, it is just add right amounts and hit one button.
04:19 PM xxcoder: it pops to warm mode automatically
04:21 PM Tom_L: i wonder if i take the sub wrappers off one of these files if it would run normally
04:22 PM JT-Shop: I imagine it would
04:23 PM JT-Shop: did you see the thought I had about just loading files with a button?
04:25 PM Tom_L: yeah but not quite sure i follow completely
04:25 PM Tom_L: it loaded and did a preview but wasn't where the actual path ran
04:25 PM Tom_L: probe to side
04:25 PM Tom_L: very basic one
04:27 PM Tom_L: http://tom-itx.no-ip.biz:443/~webpage/cnc/JT-SHOP/flexgui/probe_preview.png
04:27 PM Tom_L: the purple/red is actual path
04:28 PM Tom_L: so it took the parameter max distance to probe and displayed that but was tripped before it reached it
04:29 PM Tom_L: all i changed was removing the sub endsub lines
04:30 PM Tom_L: i saw one subroutine that all the parameters were called from the command line
04:30 PM Tom_L: passed as system parameters
04:32 PM Tom_L: if you had the button load the file you would still have to have the parameters set somewhere and passed to it
04:33 PM Tom_L: you could in effect do that with all my subroutines as long as the parameter tab had the proper vars set up ahead of time
04:34 PM JT-Shop: those would be 1-30 only iirc
04:34 PM Tom_L: those what?
04:34 PM JT-Shop: any file can read parameters in the var file
04:34 PM Tom_L: right
04:34 PM JT-Shop: parameters passed on the command line
04:34 PM Tom_L: oh yeah i agree with that
04:35 PM Tom_L: it might be a bit trickier to keep track of the files if they were in the regular nc_files path
04:35 PM JT-Shop: they can be in path that's in the ini
04:35 PM Tom_L: that's true
04:35 PM Tom_L: i guess that's what i do as well now that i think about it
04:36 PM Tom_L: so the only thing really gained is somewhat of a preview
04:36 PM Tom_L: not an accurate one
04:37 PM Tom_L: it would preview all the max parameter values
04:37 PM JT-Shop: exactly
04:38 PM Tom_L: that may be helpful for some i suppose
04:39 PM Tom_L: the bolthole pattern ones make calls to other subs though so it may not be so accurate for the ops ones
04:40 PM Tom_L: i set the 2nd op up that way so other ops could be easily added to a pattern
04:41 PM Tom_L: not sure what else you would need but the option is there
04:42 PM JT-Shop: my old gcode generator iirc you could spot, drill, ream, counter bore the same pattern
04:42 PM Tom_L: spot drill, drill, bore, chamfer, countersink, ream
04:42 PM Tom_L: that about covers hole ops afik
04:42 PM JT-Shop: yup
04:43 PM JT-Shop: I had to leave the emc channel...
04:43 PM Tom_L: you could in effect use the drill op for reamer
04:43 PM Tom_L: oh?
04:43 PM Tom_L: what did i miss?
04:43 PM JT-Shop: nothing but politics bullshit
04:43 PM Tom_L: nothing?
04:43 PM Tom_L: oh
04:43 PM Tom_L: i could crack the whip again
04:44 PM JT-Shop: they should goto #politicsandothercrap channel
04:46 PM JT-Shop: I need to figure out how to remove the sun visor from the ridgeline and see why it's loose
04:46 PM JT-Shop: it's just out of the b2b warranty
04:47 PM Tom_L: that should do it for a day or two
04:47 PM JT-Shop: what's that?
04:47 PM Tom_L: i scoulded the children
04:47 PM JT-Shop: lol
04:48 PM Tom_L: some will push the boundaries as far as possible
04:49 PM Tom_L: lemme know if/when you figure out that telop/teleporter thing
04:50 PM JT-Shop: so we can teleport them out of the channel?
04:50 PM Tom_L: i can do that now
04:50 PM Tom_L: it's never been a channel noted for that though
04:52 PM Tom_L: go see the doc about my bp tomorrow. i'm sure that shake i just had isn't going in my favor....
04:53 PM JT-Shop: I've been taking my bp every morning and it's doing good
04:53 PM JT-Shop: seems my left and right are closer to the same now
04:53 PM Tom_L: i've gotten off the dash diet but it's been regulated fairly good
04:53 PM JT-Shop: the right was 20 higher
04:54 PM Tom_L: usually the bottom is 80-85 which is what they're looking at
04:55 PM Tom_L: tip: don't check it right after having a shake
04:55 PM xxcoder: what is "shake"?
04:55 PM Tom_L: ice cream
04:55 PM JT-Shop: does a shake increase your bp?
04:55 PM JT-Shop: milkshake
04:55 PM JT-Shop: I prefer a malt
04:55 PM xxcoder: ah yes. yeah quick temp decrease can have fun effects
04:56 PM Tom_L: something did but i just came in the door and wanted a worst case numbe set
04:56 PM Tom_L: i got one :)
04:56 PM xxcoder: hmm I want something ice cream now
04:57 PM xxcoder: but it might melt completely before I can get back home, and its only few min drive
04:57 PM Tom_L: mine was gone before i hit the door
05:00 PM xxcoder: my car is deathly hot during hot summer days too
05:00 PM JT-Shop: I wonder if that deer eating dinner under the carport will eat the apples
05:11 PM lcnc-relay: <roguish> Tom_L: interesting. you set the probing speeds, etc. in the .par file
05:11 PM lcnc-relay: <roguish> I've assigned them to hal pins.
05:11 PM lcnc-relay: <roguish> can do either.
05:12 PM JT-Shop: they are set in the GUI and stored in the .var file
05:12 PM JT-Shop: and read from the .var file
05:14 PM lcnc-relay: <roguish> I just made a short probe routine that reads those .var numbers when it's opened. the code is previewed.
05:16 PM JT-Shop: any program you load will be in the preview and probe moves will show the maximum move
05:17 PM lcnc-relay: <roguish> yup
05:17 PM lcnc-relay: <roguish> so if a button can load a program, we're set to go...
05:33 PM JT-Shop: Flex just keeps getting more and more flexible
05:34 PM Tom_L: roguish, the number parameters were your idea
05:34 PM Tom_L: set up a dynamic property for set_var and assign a number to it
05:36 PM Tom_L: you just have to manually make sure the number parameter exhists in the .var file
05:37 PM JT-Shop: I've been pondering a different method of updating the .var file that could add missing variable as well
05:38 PM Tom_L: that's what's nice about the hole ops is you can change the parameters very easy and never write a bit of gcode
05:38 PM JT-Shop: yup
05:38 PM Tom_L: you can if you like but i assign mine in blocks and use them as needed
05:39 PM JT-Shop: I pretty much do the same
05:39 PM Tom_L: i'l go ahead and add a few extra and update the readme to reflect what they do
05:39 PM JT-Shop: cool
05:40 PM Tom_L: they add up fast
05:40 PM JT-Shop: yup
05:41 PM JT-Shop: looks like a good morning to ride tomorrow
05:42 PM Tom_L: yeah been warm here today
05:42 PM Tom_L: mine range from 2000 to 2110 with 2120 reserved
05:43 PM lcnc-relay: <roguish> Tom_L: i've used 4900 to 4920 before
05:43 PM Tom_L: i never thought i'd use that many
05:44 PM lcnc-relay: <roguish> just gotta avoid the ones already used
05:44 PM JT-Shop: 31-5000 - G-code user parameters. These parameters are global in the G code file, and available
05:44 PM JT-Shop: for general use.
05:44 PM lcnc-relay: <roguish> yes
05:45 PM JT-Shop: if you use set_var I "think" it checks for duplicates
05:46 PM lcnc-relay: <roguish> i should be getting a good ride in tomorrow morning...... going to a friend's place in Crocket (by the river) to help with the Haas he has. I insisted on a ride. he used to be a track sprinter. rode for Jamainca in the PanAm games once.
05:48 PM Tom_L: the vo'tech here has a whole classroom full of haas simulators
05:48 PM Tom_L: and a few out in the shop
05:49 PM Tom_L: i think they get em free in trade for demoing them to prospective clients
05:50 PM lcnc-relay: <roguish> this one is ancient. it was owned by a guy I used to work with. he didn't want it anymore and gave it away. my other friend took it off his hands......
05:51 PM Tom_L: the vo'tech is a big part of NIAR
06:10 PM * JT-Shop calls it a day
06:15 PM * Tom_L 2nds that