#linuxcnc Logs

Sep 27 2024

01:20 AM Deejay: moin
03:33 AM lcnc-relay: <Travis Farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/vbOZEsqZgWrAyYqKNiOSmjsI/FlexGUI.png
03:33 AM lcnc-relay: <Travis Farmer> this candidate for a logo incorporates several images of CNC activities, on a plot background. 🙂
03:34 AM lcnc-relay: <Travis Farmer> Morning
03:55 AM -!- #linuxcnc mode set to +v by ChanServ
04:35 AM Tom_L: background-position: center;
04:35 AM Tom_L: background-origin: content;
04:35 AM Tom_L: background-clip: padding;
04:35 AM Tom_L: background-repeat: no-repeat;
04:35 AM Tom_L: for the logger
05:42 AM JT-Cave: https://radar.weather.gov/station/kpah/standard
05:51 AM JT-Cave: Tom_L, are we ready for flex 1.0.0?
05:58 AM JT-Cave: looks like the front rotors will arrive todaya
06:36 AM Tom_L: i'm adding a couple more probe routines
06:37 AM Tom_L: 1.0 is up to you... should be pretty close
06:38 AM Tom_L: chance it might rain?
06:41 AM Tom_L: i do want to check the prob enable button a bit when i get back
06:47 AM JT-Cave: ok
06:48 AM Tom_L: are the other mdi commands supposed to be disabled when probe enable is active?
06:49 AM JT-Cave: yes
06:50 AM Tom_L: you might toggle it a few times and watch what happens.. i'll check when i get back but it may not always happen like that
06:52 AM Tom_L: i'd been using mdi to move the probe back and i think i was able to do that with the probe still active
06:52 AM Tom_L: i just need to double check on that
06:54 AM JT-Cave: just working on a simple sim to replace the very complicated one stepconf makes
06:59 AM Tom_L: adding 'measure between and across'
07:02 AM Tom_L: i think we've covered most of the 'lacking this' complaints
07:04 AM Tom_L: might change the debug messages a bit.. i dunno
07:24 AM * rdtsc-wk really needs to get a-workin' on the mill build... just overwhelmed and so little time.
08:05 AM JT-Cave: https://www.youtube.com/watch?v=4CJG69V8KsU
08:20 AM lcnc-relay: <Travis Farmer> what did you think of the logo candidate i posted early this morning?
08:23 AM JT-Cave: I didn't see it
08:24 AM lcnc-relay: <Travis Farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/bEzrCmapZbMcavqybHJKxQYR/FlexGUI.png
08:30 AM rdtsc-wk: curious what qtdragon did on theirs, https://www.linuxcnc.org/docs/devel/html/gui/images/qtdragon_setup_page.png
08:31 AM lcnc-relay: <Travis Farmer> oh, so it loads a html setup the same name as the gcode file?
08:32 AM rdtsc-wk: qtdragon does... no Tom, not requesting it :)
08:32 AM lcnc-relay: <Travis Farmer> i just print my setup files and bring them out
08:37 AM lcnc-relay: <Travis Farmer> QT has a text browser, but not one that handles images
08:37 AM lcnc-relay: <Travis Farmer> that i know of..
08:40 AM lcnc-relay: <Travis Farmer> https://www.pythonguis.com/examples/python-web-browser/
09:05 AM -!- #linuxcnc mode set to +v by ChanServ
09:10 AM JT-Cave: the QTextEdit should be able to display a html
09:13 AM JT-Cave: travis it's kind of hard to read for me and confusing with the boundry box in the back ground
09:22 AM lcnc-relay: <Travis Farmer> ok, fair enough. i was just playing around
10:58 AM lcnc-relay: <Travis Farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/anehbfDjvHLbYFNxYMScWBLS/461500058_10161070626189543_6167212739840393036_n.jpg
10:58 AM lcnc-relay: <Travis Farmer> just the girls checking for acorns, last night 🙂
11:06 AM JT-Cave: mine quit eating from the dog bowl under the carport, I assume it's because there's a lot of acorns on the ground
11:17 AM lcnc-relay: <skunkworks8841> we have a bumper crop of hicory nuts this year. The branches were very droopy...
11:32 AM Tom_L: saw a squirrel this morning with a cluster of pecan nuts bigger than his head
11:57 AM roycroft: it's hoarding season
11:59 AM JT-Cave: it's nap time
12:05 PM JT-Cave: Tom_L, still got something wonky with jog buttons after the mdi probe
12:06 PM Tom_L: pretty close to nap yes
12:06 PM Tom_L: i know, just haven't had a chance to test that
12:06 PM Tom_L: jog buttons mess with my pendant too
12:07 PM Tom_L: jog, probe & probe enable seem to not get along always
12:08 PM JT-Cave: working on testing that but first a power nap
12:08 PM JT-Cave: oh and I've did a lot of stuff in the examples...
12:08 PM Tom_L: just resized the new icons
12:08 PM Tom_L: like what?
12:09 PM Tom_L: if you pick a good one, i'll try and add a probe tab if you want
12:09 PM Tom_L: that's a ways off for me atm
12:10 PM Tom_L: gonna try and fix these distance ones first
12:16 PM Tom_L: did you do something with overlay on the plot screen?
12:53 PM Tom_L: JT-Shop, rapid_override_lb doesn't update with rapid_override_sl
12:55 PM Tom_L: unless it needs an ini setting i don't see
12:57 PM Tom_L: ok nap time
12:58 PM Tom_L: ^^ the _lb stays on 150%
01:00 PM lcnc-relay: <Travis Farmer> JT: https://paste.debian.net/1330608/
01:00 PM lcnc-relay: my error or yours?
01:02 PM lcnc-relay: <Travis Farmer> Doh... my error... NVM
01:12 PM lcnc-relay: <Travis Farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/QzZpFNBOvtsWtIAZgEpSwMxp/Screenshot_from_2024-09-27_14-02-13.png
01:12 PM lcnc-relay: <Travis Farmer> spindle LCD readout needs work. spindle was running at 24000rpm. when at 0rpm it reads 0.0, so it is connected
01:17 PM robobub_ is now known as robobub
01:18 PM -!- #linuxcnc mode set to +v by ChanServ
01:34 PM JT-Cave: travis make sure the digitCount property is large enough to display your number including the . I think
01:34 PM JT-Cave: and get a fresh deb...
01:35 PM * JT-Cave wanders out to work on brakes
01:39 PM lcnc-relay: <Travis Farmer> oh sure, you respond once i get back inside 😉 will check on digitCount, and get a fresh deb next time i am out
01:40 PM john__ is now known as JT-Auto-Shop
01:41 PM lcnc-relay: <Travis Farmer> digitCount was 5, set to 10 just to be safe
01:42 PM JT-Auto-Shop: IIRC 5 will give you 3.1 so 1000 would show as 000.0
01:44 PM lcnc-relay: <Travis Farmer> so i need 7
02:20 PM * Tom_itx feels 12 commits is a bit far behind. so much for napping
02:21 PM Tom_itx is now known as Tom_L
02:23 PM Tom_L: override works now
02:23 PM Tom_L: thanks
02:28 PM rdtsc_away is now known as rdtsc
02:43 PM * JT-Shop takes five with Dave
02:44 PM Tom_L: one sub done 3 to go
02:45 PM JT-Shop: I have a minimal probe example started in the features directory
02:47 PM JT-Shop: Showers and possibly a thunderstorm. Some of the storms could produce heavy rainfall. High near 66. Breezy, with a north northwest wind around 21 mph, with gusts as high as 38 mph. Chance of precipitation is 100%.
02:49 PM JT-Shop: https://radar.weather.gov/station/kpah/standard
02:49 PM Tom_L: using my probe subs or your own?
02:49 PM JT-Shop: mine, just one simple one with lots of comments
02:50 PM Tom_L: i have a couple with comment every line i think
02:50 PM JT-Shop: not making a full blown probe example but rather how to add probing
02:50 PM JT-Shop: as simple as possible
02:50 PM Tom_L: i'm probably gonna update the debug messages a bit
02:51 PM Tom_L: wanted to get them all working
02:51 PM JT-Shop: looking at the probe routines chris uses both g90 and g91 and that makes sense
02:51 PM Tom_L: you just about have to
02:52 PM JT-Shop: after the probe touches during a g91 probe routine then use the actual contact spot in the next line which negates any over travel
02:52 PM Tom_L: i need to fix that on some. i have a retract distance
02:52 PM Tom_L: you don't wanna bring the probe straight up after it touches
02:52 PM JT-Shop: crap, forgot to torque the caliper mounting bolts...
03:02 PM JT-Shop: got the TIR on the LF rotor down to 0.004"
03:04 PM Tom_L: nice
03:38 PM JT-Auto-Shop: RF 0.001" TIR on the first try...
03:55 PM Tom_L: you practiced on the first side...
03:55 PM JT-Shop: it's just luck
03:57 PM JT-Shop: fronts are done... taking a bit of a break before jacking up the back
04:00 PM Tom_L: think i got material probe done in X with shortened 2nd pass distance
04:00 PM JT-Shop: cool
04:01 PM JT-Shop: it's pretty sweet how we can have spin box hal pins... I'm loving it
04:01 PM Tom_L: the bar probe is really the only one i had done that needs that fixed
04:01 PM Tom_L: you gotta go the full distance the first time
04:01 PM Tom_L: for now, i back off .1"
04:02 PM Tom_L: i considered making that a parameter
04:19 PM lcnc-relay: <Travis Farmer> you two are rock stars! just thought i would mention that, as it gets said so little.
04:20 PM JT-Shop: :)
04:22 PM lcnc-relay: <Travis Farmer> my cat on the other hand, is licking the security camera screen to get the squirrels...
04:38 PM Tom_L: ok i think i have 2 now
04:39 PM Tom_L: test on iron later on
04:39 PM Tom_L: sim looks ok
04:39 PM Tom_L: inlaws here.. should go hide.. errr at least say hi
04:40 PM Connor: Can someone talk to me about backlash comp in lcnc. It's been a while since I've played with it. I know about doubling the accel values..
04:40 PM Connor: Is the BACKLASH value Positive or Negative or does that depend on something?
04:41 PM Connor: Can you use a touch probe to help figure out the backlash? Or that just a bad idea and best to use a DTI?
05:21 PM lcnc-relay: <Travis Farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/fjeTveZrxjjzQLUBgBGkwPlp/20240927_181049.jpg
06:39 PM rdtsc: Dunno Connor; a known-good DTI is what I'd use. Maybe make a simple gcode program which moves X forward and back 0.1 units or whatever, see what the DTI says, and experiment with backlash value.
06:40 PM rdtsc: https://forum.linuxcnc.org/forum/10-advanced-configuration/1567-backlash-compensation
07:11 PM rdtsc: As an experiment, added 'BACKLASH = 0.1' to 'JOINT_3' in my .ini, then loaded a .ngc with a big, slow circle in it. Can watch joint_3 motor slow down, come to a stop, and jerk back the other way. If there was backlash, that would've nailed it.
07:13 PM Tom_L: backlash is put in at the first of a move as what i think is a rapid move
07:14 PM Tom_L: certainly not the given feedrate
07:15 PM rdtsc: in practice, it seems very wise to first qualify the machine in terms of axis accel limit, and stay well under that for these aggressive moves. because nothing would be worse than losing a few steps only sometimes.
07:17 PM rdtsc: (why is it, this part is good, but this one is off by 2%?!?!?!)
08:03 PM lcnc-relay: <skunkworks8841> You need acceleration headroom with backlash comp..
08:18 PM Tom_L: these all work now: http://tom-itx.no-ip.biz:443/~webpage/cnc/JT-SHOP/flexgui/touch-probe2.png
08:18 PM Tom_L: and metric: http://tom-itx.no-ip.biz:443/~webpage/cnc/JT-SHOP/flexgui/touch-probe2-metric.png
08:27 PM rdtsc: Tom's knocking 'em out of the park :)
08:52 PM lcnc-relay: <knoxmachining> I have it. With ClearPATH servos. Just wasn’t sure about positive or negative numbers and if I would be chasing my tail trying to use the touch probe vs a dial indicator.
08:53 PM Tom_L: last word indicator would be your best bet
08:56 PM roycroft: my shed is roofed!
08:56 PM roycroft: i still need to finish making the door, but it's ready to move stuff into it now
08:59 PM lcnc-relay: <knoxmachining> I have one of those, along with 1/2 thou DTI and a 1/2 thou Dial indicator and 1/2 tenths dti too. 🙂
09:00 PM lcnc-relay: <knoxmachining> Also have a digital mitutoyo dial indicator too.
09:02 PM lcnc-relay: <knoxmachining> This is on a router. So. Not sure if should test at the saddle or the spindle.
09:03 PM lcnc-relay: <knoxmachining> Feel like I get too much spring at the spindle.
09:38 PM roycroft: new clickspring video - it's been a while
09:57 PM CloudEvil: :)
10:05 PM -!- #linuxcnc mode set to +v by ChanServ
10:16 PM roycroft: uri tuchman also has a new video
10:16 PM roycroft: he's building a rose-ish engine thing