#linuxcnc Logs

May 15 2024

12:57 AM Deejay: moin
04:05 AM Tom_L: morning
04:54 AM -!- #linuxcnc mode set to +v by ChanServ
04:55 AM JT-Cave: morning
04:56 AM JT-Cave: https://github.com/jethornton/flexgui/blob/master/flexgui/src/libflexgui/actions.py#L254
04:59 AM lcnc-relay: <travis_farmer> Morning
06:49 AM Tom_L: JT-Cave, tooledit responds a bit different than the rest... it stays open
06:49 AM Tom_L: but that may be ok too
06:50 AM JT-Cave: I just changed tool edit from .run to .Popen
06:50 AM Tom_L: i didn't check but it may do that with axis too
06:50 AM Tom_L: i saw that
06:50 AM Tom_L: i think it should be ok like that
06:52 AM Tom_L: building a rpi deb
06:52 AM JT-Cave: yup you can open the tool editor in axis and close axis and the editor stays open
06:53 AM Tom_L: pr #27
06:54 AM Tom_L: still need to move toolchange buttons to homed before active
06:54 AM JT-Cave: working on that and tool touch off now
06:55 AM Tom_L: figured you were
06:55 AM Tom_L: several buttons are broke like that now
07:00 AM JT-Cave: yup
07:09 AM Tom_L: i was gonna try to change that to Popen last night but didn't know the syntax to pass parameters to it
07:09 AM Tom_L: then you also removed the args on the others
07:13 AM Tom_L: do they default to 'none'?
07:16 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> https://jauriarts.org/_matrix/media/v3/download/matrix.org/TuWveAWhVPuuaNjHwHbpHqPE/ima_384a892.png
07:17 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Hello guys, i have a question: i am using mesa 7i95t, connected to delta b3 servos (controlled by step/dir) and feeding encoder response back, using index from encoder for homing. However I am using stepgen position-fb to motor-pos-fb
07:17 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Delta servos are tuned to have basically 0 following error:
07:18 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> https://jauriarts.org/_matrix/media/v3/download/matrix.org/lByxPavRLHnDDtAyHnAuaiqQ/ima_ca4238c.jpeg
07:18 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> However the difference between commanded position and encoder feedback is some 0.2X mm
07:19 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Is there some kind of delay, which would cause this difference?
07:20 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> And also would it be possible to use encoder as feedback position? (Its commented out in my config)
07:20 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> +motor
07:21 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> My config is here: https://github.com/zmrdko/mesa_7i95t_config/blob/main/hallib/01_main.hal
07:22 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Question1:Is there some kind of delay, which would cause this difference?
07:22 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Question2: would it be possible to use motor encoder as feedback position? (Its commented out in my config)
07:23 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Question3: i have eliminated pid calculations from my config. Do I need them for something?
07:33 AM JT-Cave: Tom_L, yes the default args for those are None
07:33 AM Tom_L: k
07:33 AM Tom_L: later...
07:34 AM JT-Cave: ok
07:35 AM JT-Cave: pid calcs for mesa steppers improves the step accuracy or something like that in any case it's always used with mesa step boards
07:35 AM * JT-Cave starts his chicken day
07:41 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> JT-Cave: Because it somehow works without them. I am aware of the values p=1000 and ff=-1. So I will add the pid calcs back
07:43 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Thanks JT-Cave. What about the difference between encoder readings and commanded position?
08:50 AM lcnc-relay: <skunkworks8841> The PID for stepper is mainly for latency issues. (it handles spikes in latency way better than the stepgen position mode.
08:51 AM lcnc-relay: <skunkworks8841> You can switch the feedback to the encoder.. Should work. There might be some tuning you need to do.
08:52 AM lcnc-relay: <skunkworks8841> remember steppers are not like servos though.. If the stepper falls behind enought that the pid is wanting it go go faster - you will get a stall.. (but mainly - no different than pushing the stepper too hard)
08:59 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> skunkworks8841: So PID is only relevant for velocity mode?
09:00 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> I’m using:
09:00 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> setp MESA (CARD0).stepgen.00.step_type 0
09:00 AM lcnc-relay: setp MESA (CARD0).stepgen.00.control-type 0
09:00 AM lcnc-relay: <skunkworks8841> right.. The stepgens would have to be in velocity mode
09:00 AM lcnc-relay: <skunkworks8841> and all the pins hooked up correctly.
09:07 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Is it better choice to use velocity mode for step/dir controlled servo then?
09:10 AM lcnc-relay: <skunkworks8841> with mesa? for sure. Peter could probably explain it better
09:12 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> No need for further explanation. If you say so I believe you
09:12 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> So add PID and change to velocity mode
09:14 AM lcnc-relay: <skunkworks8841> yes
09:15 AM lcnc-relay: <skunkworks8841> (and hook the correct pins together) you would have to look at examples - or I think the configurators create pid stepgen configs.
09:16 AM JT-Cave: The way the Mesa hardware step generators work is that every servo thread
09:16 AM JT-Cave: invocation, the step generator hardware is given a new velocity. Without feedback
09:16 AM JT-Cave: (via the PID controller) the hardware position would slowly drift because of clock
09:16 AM JT-Cave: speed and timing differences between LinuxCNC and the step generator hardware.
09:16 AM JT-Cave: The PID controller gets feedback from the actual (fractional) step position and corrects for
09:16 AM JT-Cave: these small differences. It can also be used to correct for read/update delays (by setting
09:16 AM JT-Cave: FF2 to the time from position read to velocity write)
09:19 AM * JT-Cave heads to town
09:22 AM Rhine_Labs: Morning
09:41 AM Tom_L: JT-Cave, mdi still needs homing
09:42 AM Tom_L: tool change & touchoff look ok
09:44 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> Thanks for thorough explanation JT-Cave
10:52 AM -!- #linuxcnc mode set to +v by ChanServ
11:10 AM JT-Cave: zmrdko yw
11:10 AM JT-Cave: Tom_L, mdi fixed... nap time now
11:11 AM Tom_L: ok
11:12 AM JT-Cave: I created a parent.home_required list and if all axes are homed it enables the items in that list and if any axis is not homed then the items are disabled
11:12 AM JT-Cave: much cleaner way to deal with that
11:13 AM JT-Cave: more pythonic and adheres to PEP 666
11:13 AM Tom_L: :)
11:13 AM JT-Cave: like I give a crap about stupid PEP
11:14 AM Tom_L: puts some pep in yer step
11:16 AM Tom_L: don't think you can do much about hal pins that have mdi commands
11:25 AM Tom_L: getting harder to break things
11:30 AM lcnc-relay: <skunkworks8841> hold my beer...
11:35 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> JT-Cave: So when I will use pid and velocity mode and feedback from encoder, i guess I have to do proper PID tuning in linuxcnc, right?
11:35 AM lcnc-relay: <skunkworks8841> yes.. I think you will need to
11:41 AM lcnc-relay: <zmrdko (@zmrdko:matrix.org)> skunkworks8841: Thanks
12:23 PM Rab: solarwind, why don't you machine the brass stock and save the swarf for recasting? Seems like melting down the stock first is an unnecessary step.
12:24 PM lcnc-relay: <skunkworks8841> Starlink faired the solar storms just fine.. No issues.
12:24 PM lcnc-relay: <skunkworks8841> SO yay!
12:29 PM Unterhaus_ is now known as Unterhausen
12:30 PM Unterhausen: fine woodworking magazine's idea about how to cut a 22.5 degree angle is to have a 22.5 degree wedge
12:30 PM Unterhausen: it's 22.5 degree angles all the way down
12:33 PM lcnc-relay: <travis_farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/byYzVgwfwtkqiJgSDgMxSOpq/Screenshot_2024-05-15_at_13-25-52_Overview_Home_Assistant.png
12:33 PM lcnc-relay: <travis_farmer> i don't think i will work in my shop right now... that is the temperature in F. (don't have the A/C in yet)
12:45 PM CaptHindsight: travis_farmer: have you tried installing a big ass fan?
12:52 PM Rab: 80F's downright balmy, my shop pushes 100F in the summer...I installed a little window unit which gets it within 80-90.
12:53 PM Rab: Bad for rust, I've started wrapping tooling in the special preventative paper.
12:56 PM Rab: Almost prefer a box fan in the window, because it sucked out cutting oil smoke. Now I'm incentivized not to ventilate the shop.
01:13 PM JT-Shop: I use a dehumidifier in all my shops
01:13 PM JT-Shop: and keep the humidity to 50%
01:16 PM Rab: I have a dehumidifer setting on the window unit which I use quite a bit. Unfortunately the room is a converted carport which is open to the house attic, so natural convection cycles all the air if conditions are right.
01:16 PM JT-Shop: yeah that would be hard to keep the humidity down if it's open to the air
01:22 PM Rab: Interesting light reading, if it hasn't been posted already: https://www.ntsb.gov/investigations/Documents/DCA24MM031_PreliminaryReport%203.pdf
01:35 PM CaptHindsight: The page you're looking for doesn't exist.
01:35 PM Rab: No big surprises, but one takeaway is that the crew reacted quickly and in a seemingly logical manner. I don't think there was anything they could have done differently after the initial blackout.
01:35 PM CaptHindsight: nevermind, cut and paste worked
01:40 PM Rab: Looks like the electrical reconfiguration in port exposed a lurking equipment failure, selecting the wrong equipment for the right reasons.
01:40 PM Rab: Surprised this is so short and readable.
01:41 PM XXCoder: yeah. sometimes something happened have very complex roots but not so much for this
01:42 PM XXCoder: like that oil distillery explosion. problems was across couple days, and had many failures and problems adding up
01:44 PM XXCoder: https://youtu.be/3RFDKpwdbEA this one. report was well over 100 pages
01:48 PM XXCoder: even more complex https://www.youtube.com/watch?v=goSEyGNfiPM report is over 300 pages
02:11 PM XXCoder: sigh https://www.cbsnews.com/news/pelican-island-bridge-causeway-galveston-texas-barge-collision/
02:12 PM Scopeuk: it's amazing how similar air accident investigation reports and nautical accident investigation reports are. I guess it's not overly surprising but I was expecting more detour
02:13 PM Scopeuk: this is also only a preliminary report, the full one will be massive, probably have chunks of the data recorder data and make recomendations
02:13 PM Scopeuk: that will likely take a year or so however
02:14 PM XXCoder: probably
03:01 PM Tom_L: CaptHindsight, https://bigassfans.com/
03:04 PM Tom_L: everybody needs a big ass fan
03:04 PM lcnc-relay: <travis_farmer> https://jauriarts.org/_matrix/media/v3/download/jauriarts.org/toaHqeEbOcbLchLtEhueHtLM/20230508_104141.jpg
03:05 PM lcnc-relay: <travis_farmer> CaptHindsight: in reference to me installing a BAF... there is no room for one in that shop...
03:05 PM Tom_L: sweat is the human cooler
03:06 PM lcnc-relay: <travis_farmer> it is if there is a breeze 😉
03:06 PM lcnc-relay: <travis_farmer> now up to 84F in old shop...
03:09 PM Rab: I was looking at a BAF Haiku for my home addition, but the current lineup seems to have a tightly-integrated electronic motor controller and lighting package which requires a phone app (and maybe a cloud connection, not positive).
03:11 PM Rab: Most fans have a remote module you can clip out and bypass with hard-wiring, but the Haiku has a 3-phase brushless motor so the controller is critical.
03:11 PM lcnc-relay: <travis_farmer> for a fan???
03:12 PM lcnc-relay: <travis_farmer> ahh, i see
03:12 PM Tom_L: they're fancy fans
03:13 PM lcnc-relay: <travis_farmer> the BAF me and dad have for work has a single switch, with Hi and Lo 😉
03:13 PM Rab: https://www.tuvie.com/wp-content/uploads/haiku-advanced-ceiling-fan-by-big-ass-fans4.jpg
03:13 PM Rab: https://i.ytimg.com/vi/uiqd6Eum7Qw/maxresdefault.jpg
03:13 PM Tom_L: just build yourself a wind tunnel
03:14 PM lcnc-relay: <travis_farmer> all will be fine if i ever build the new shop 🙂
03:14 PM Rab: "Predictive learning microprocessor" what
03:14 PM Tom_L: rab, looks similar to a F&P motor
03:15 PM Rab: Fisher & Paykel?
03:15 PM Tom_L: yeah
03:16 PM Tom_L: https://www.fisherpaykeltechnologies.com/technology/direct-drive-motors
03:16 PM Rab: I think the older Haiku models were simpler, pre RGB lighting, but they no longer appear to be sold.
03:17 PM CaptHindsight: I worked with some BAF's and some even bigger ass fans. All were made rather cheaply and had issues working at low temps (-20C)
03:18 PM CaptHindsight: these were ~2m dia
03:20 PM JT-Cave: BAF's are expensive
03:20 PM * JT-Cave waits patiently for Tom_L to break something in Flex
03:20 PM XXCoder: what baf requires cloud service to work or?
03:20 PM Tom_L: JT-Cave, while you're stewing over it, fix the plotter :D
03:21 PM JT-Cave: ok
03:21 PM JT-Cave: but wait it's not broken...
03:21 PM Tom_L: did you finish the tool touchoff?
03:21 PM JT-Cave: yup
03:22 PM XXCoder: for small space i'd use a ducted fan
03:22 PM Tom_L: is that based off the machining box for Z in the ini?
03:23 PM Rab: XXCoder, the Haiku residental line: https://store.bigassfans.com/en_us/haiku
03:23 PM JT-Cave: gotta love comments like # monkey patched function from StatCanon class
03:23 PM JT-Cave: in code
03:23 PM Tom_L: i put .5 in the tool touchoff spin box and get 1.9999 in the tool offset
03:24 PM Rab: I didn't look through the manual to see what happens if no wireless/bt is available, I guess I should do that before trash-talking.
03:24 PM JT-Cave: hmmm
03:24 PM XXCoder: 1200 usd dang
03:24 PM Tom_L: that's why i wondered if it was based off the machining box
03:24 PM XXCoder: is it wired with gold
03:24 PM Tom_L: for XYZ
03:24 PM JT-Cave: prob right as .5 should be in the dro and whatever in the tool offset to make it .5
03:25 PM JT-Cave: nothing is based of the limits
03:25 PM JT-Cave: off
03:25 PM -!- #linuxcnc mode set to +v by ChanServ
03:25 PM Tom_dev: Error
03:25 PM Tom_dev: E word with no G70, G76, M66, M67 or M68 to use it
03:26 PM Tom_dev: wtf is an E word?
03:26 PM Rab: XXCoder, that's another issue. I've had bad luck with cheap fans and good luck with expensive fans, but fancy electronics adds a big most likely point of failure.
03:27 PM XXCoder: interesting. yeah I do like good fans
03:27 PM JT-Cave: how did you get that error?
03:27 PM Tom_L: pushing buttons
03:27 PM Tom_L: :)
03:28 PM Tom_L: i'll see if i can get it again
03:28 PM JT-Cave: http://linuxcnc.org/docs/stable/html/gcode/g-code.html#gcode:g70
03:28 PM JT-Cave: E - The ending distance of the profile, defaults to 0.
03:30 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/JT-SHOP/flexgui/tool_touchoff.png
03:30 PM Tom_L: those settings and push Z touchoff
03:36 PM Tom_L: i'm not using any of those G/M codes
03:36 PM JT-Cave: tool touch off?
03:37 PM Tom_L: yes
03:37 PM Tom_L: can't seem to duplicate it yet
03:37 PM Tom_L: ok
03:38 PM Tom_L: set the spin box to .5
03:38 PM Tom_L: touch off
03:38 PM Tom_L: set the spin box to 0
03:38 PM Tom_L: touch off
03:38 PM Tom_L: error
03:38 PM Tom_L: z
03:38 PM JT-Cave: hmm no error here, let try tom.ini
03:40 PM JT-Cave: https://pasteboard.co/ARNvh74uA9TE.png
03:42 PM JT-Cave: https://github.com/jethornton/flexgui/blob/master/flexgui/src/libflexgui/commands.py#L245
03:44 PM Tom_L: the top spinbox on the tool page of xyzh is the offset right?
03:44 PM JT-Cave: I just pushed a debug print but let me put that on the status bar
03:47 PM Tom_L: if you do a touchoff a 2nd time, the number is added to or subtracted from the current tool offset value
03:51 PM Tom_L: if i set the tooltable T1 Z to 0, save the file, reload it in flex, the tooloffset Z doesn't change
03:51 PM Tom_L: and
03:51 PM Tom_L: mdi G0 Z0, the DRO is Z2
03:51 PM Tom_L: the tool was previously set at Z2
03:52 PM Tom_L: not sure what's causing the DRO Z to stay at 2
03:53 PM Tom_L: xyzh.ui
03:53 PM Tom_L: not that it matters
03:53 PM Tom_L: but
03:54 PM Tom_L: after closing flex, then close the tool table it asks if i want to save it even though i saved it on the tooledit menu
03:57 PM Tom_L: it's not updating from the tool table after an external edit even though it was saved and re'read
03:58 PM JT-Shop: I may have to do something in code to update
03:58 PM Tom_L: i can't quite pinpoint what's going on but something is
04:01 PM JT-Shop: I'll have to do some testing in the morning
04:01 PM Tom_L: ok
04:02 PM Tom_L: i really should put something together to test on the mill
04:03 PM * Tom_L sweeps up the broken parts and hands them to JT-Shop
04:03 PM JT-Shop: so edit tool table and the tool info is not updated after saving and closing?
04:03 PM Tom_L: that's what it looks like
04:04 PM Tom_L: well, i didn't close it but did reread from flex
04:04 PM Tom_L: after saving in tooledit and rereading there
04:04 PM Tom_L: the file was saved
04:10 PM Tom_L: i also verified the tool table file was saved by having it open in geany. it asked to reload it since it changed
04:10 PM * JT-Shop needs to get those heavy ceramic cookers out of my truck
04:10 PM Tom_L: how many did they give you?
04:10 PM JT-Shop: I have 3 now
04:11 PM Tom_L: still making fixtures for them?
04:11 PM JT-Shop: starting over, gave up trying to make the one they made work
04:11 PM Tom_L: not that clamp you made?
04:12 PM JT-Shop: that clamp is a design of a manager lol
04:12 PM Tom_L: ahh
04:12 PM JT-Shop: I tried several attempts to make it work then finally gave up
04:13 PM JT-Shop: it would have worked with a fork lift but not with a hoist
04:14 PM Tom_L: ok if i edit the T1 Z in geany save it and reload in flex it updates
04:14 PM Tom_L: lemme try the editor now
04:14 PM JT-Shop: ok
04:15 PM Tom_L: it updated that time
04:15 PM Tom_L: with the tooledit -> save file, flex->reread
04:16 PM Tom_L: and just did a touchoff Z0 in flex and the tooledit showed the file changed
04:17 PM JT-Shop: so everything is working
04:17 PM Tom_L: did it again with Z2 touchoff and the table saved -2
04:18 PM Tom_L: and again with a Z-1 touchoff and the file saved 1
04:18 PM JT-Shop: the value in the tool table is what it takes to make the tool where you set the touchoff
04:18 PM Tom_L: yeah, (i think so)
04:19 PM Tom_L: it's the distance from Z+0 to the tool tip isn't it?
04:20 PM Tom_L: needs to be tested on some iron
04:21 PM JT-Shop: when you set a tool offset it's the value that makes the current position be that value
04:21 PM Tom_L: i alwasy use the pendant to jog for the tool offsets so i'm not used to using this
04:22 PM Tom_L: setting it to 0 is where the problem comes in
04:27 PM Tom_L: more specifically if the spinbox shows -0.00 it errors
04:27 PM Tom_L: 0.00 doesn't
04:27 PM JT-Shop: interesting
04:29 PM Tom_L: that E word error...
04:29 PM JT-Shop: that's really odd as I don't have any E word
04:29 PM JT-Shop: did you pull the latest?
04:30 PM Tom_L: not yet
04:30 PM Tom_L: not the print
04:30 PM JT-Shop: it shows the command in the status bar for testing
04:31 PM Tom_L: now i am
04:31 PM Tom_L: on either ui?
04:31 PM JT-Shop: I started to do a print but then you have to run from a terminal so I changed it to the status bar
04:32 PM JT-Shop: any ui
04:32 PM Tom_L: what will cause the printout?
04:32 PM Tom_L: i don't see it
04:32 PM JT-Shop: tool touch off
04:32 PM Tom_L: ok
04:32 PM -!- #linuxcnc mode set to +v by ChanServ
04:33 PM Tom_L: -0.00 throws an error and no print
04:33 PM Tom_L: the "Error" on the status bar probably erases it
04:34 PM JT-Shop: hmm didn't think of that
04:36 PM Tom_L: put it on the right side of the menu bar?
04:37 PM JT-Shop: maybe make a label to show the command
04:37 PM Tom_L: just lemme know
04:38 PM JT-Shop: time to make chicken crack
04:38 PM Tom_L: :)
04:39 PM Tom_L: interesting when the errors hide and don't stand out and smack you in the face
04:43 PM Tom_L: i'd consider that -0.00 a qt bug
04:54 PM JT-Shop: made a test_lb in xyz.ui
04:54 PM Tom_L: i'll add it on mine too
04:54 PM JT-Shop: ok
04:59 PM -!- #linuxcnc mode set to +v by ChanServ
05:03 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/JT-SHOP/flexgui/touchoff0.00.png
05:03 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/JT-SHOP/flexgui/touchoff-0.00.png
05:03 PM Tom_L: where'd the -2.77xxxx come from?
05:05 PM Tom_L: just a sec..
05:08 PM Tom_L: ok i redid those making the label wider
05:09 PM JT-Shop: e maybe an exponent
05:09 PM Tom_L: yeah
05:09 PM Tom_L: where the hell is that coming from though?
05:09 PM Tom_L: -2.77xxx
05:10 PM JT-Shop: https://github.com/jethornton/flexgui/blob/master/flexgui/src/libflexgui/commands.py#L251
05:11 PM Tom_L: that -0.00 is a bug imo
05:11 PM JT-Shop: must be
05:11 PM JT-Shop: I'll putz with that in the morning
05:12 PM Tom_L: test offset for '-0.00' if it is offset = 0.00
05:12 PM Tom_L: then update the spinbox too
05:16 PM JT-Shop: how do you get -0
05:17 PM Tom_L: run the spinbox negative a bit then back up
05:17 PM JT-Shop: I can't get -0
05:17 PM Tom_L: i'm special
05:17 PM JT-Shop: up
05:17 PM JT-Shop: yup
05:18 PM Tom_L: is there a spinbox setting to prevent it?
05:18 PM Tom_L: i can do that on the coordinate touchoff too
05:19 PM Tom_L: wait.. no it doesn't
05:19 PM Tom_L: i bet if you change the precision of the increment it will go away
05:20 PM Tom_L: it's .1 on the tool and .001 on the fixture
05:21 PM Tom_L: where do you set that?
05:21 PM Tom_L: found it
05:22 PM Tom_L: .01 fixed it
05:22 PM Tom_L: precision
05:23 PM JT-Shop: dynamic property precision?
05:25 PM Tom_L: no just on the doublespinbox settings
05:25 PM Tom_L: blue area
05:25 PM JT-Shop: decimals?
05:25 PM Tom_L: singlestep
05:26 PM JT-Shop: default is 1.000000000000000000000
05:26 PM Tom_L: well then i changed mine
05:26 PM Tom_L: the work offsets was .001
05:27 PM Tom_L: the tool was .1
05:27 PM Tom_L: but
05:27 PM Tom_L: .01 fixed it
05:27 PM Tom_L: i bet that's why you didn't get it on xyz
05:28 PM Tom_L: actually it should be set to .001 at least if you're using it to touch off
05:29 PM JT-Shop: odd I got it to show -0.0000000 after pressing the down button several times then the up button so it's a bug in pyqt I think
05:29 PM Tom_L: yup
05:29 PM Tom_L: just set the precision finer
05:30 PM Tom_L: yeah the default is 1.0000
05:30 PM Tom_L: just dropped a new one to see
05:31 PM Tom_L: i'm setting both mine to .001
05:31 PM JT-Shop: I'm about to call it a day and fire up the barbi for some steak
05:31 PM Tom_L: sounds like a plan
05:31 PM Tom_L: still an odd one
05:31 PM JT-Shop: very odd, but I'm glad I can reproduce it so I can see wtf is up lol
05:32 PM Tom_L: i wouldn't even call it a rounding error
06:03 PM Tom_L: touchoff_system_cb doesn't update G5x_index_lb
06:03 PM Tom_L: should it?
06:06 PM Tom_L: i didn't find it in the docs either
07:06 PM lcnc-relay: <satiowadahc#0> Inverse feed seems ridiculous to me. These guys cam generate everything so you get ridiculous values like F2109807
07:11 PM rdtsc_away is now known as rdtsc
07:29 PM rdtsc is now known as rdtsc_away
07:31 PM -!- #linuxcnc mode set to +v by ChanServ
07:41 PM rdtsc_away is now known as rdtsc
08:31 PM solarwind: What's y'alls favourite dinosaur?
08:31 PM solarwind: I think the nigersaurus is pretty cool
08:38 PM roycroft: when i was a kid, dino was my favorite dinosaur
08:38 PM roycroft: but now i think it has to be barney
08:40 PM rdtsc is now known as rdtsc_away
08:40 PM rdtsc_away is now known as rdtsc
08:52 PM rdtsc is now known as rdtsc_away
10:13 PM solarwind: XXCoder post _something_ cool. ANYTHING
10:18 PM Tom_L: penguins can fly http://tom-itx.no-ip.biz:81/~webpage/temp/penguin.jpg
10:18 PM XXCoder: whoa shocking
10:18 PM XXCoder: I wonder if they use linux
10:19 PM lcnc-relay: <satiowadahc#0> Jt/Tom for -0.00 check for number less than -0.01. In most programming languages is called eps or epsilon
10:20 PM Tom_L: i think we can avoid it by setting the precision
10:20 PM Tom_L: because there are negative numbers allowed here
10:20 PM Tom_L: not just uint
10:23 PM lcnc-relay: <satiowadahc#0> Dead band is still a thing, so if it's between 0.0000 and -0.003 and your precision is 2 digits you'll still see -0.00
10:23 PM Tom_L: i think it's a bug in pyqt
10:24 PM Tom_L: on a spin box if you roll it negative and back to 0.00 with .1 precision you get -0.00 as a result
10:24 PM Tom_L: .01 precision you don't
10:26 PM Tom_L: precision/increment