#hazzy Logs

Aug 16 2020

12:09 AM Joco[m]: yeah. The error isn't linear though which is a bit of a bugger. Cheap scales. But I hope can come up with a comprise. tweaked it so test cut is within 0.01mm but a retest on the 100mm travel shows a 0.02mm undertravel. So a little more tweaking to do.
12:09 AM Joco[m]: * yeah. The error isn't linear though which is a bit of a bugger. Cheap scales. But I hope can come up with a compromise. tweaked it so test cut is within 0.01mm but a retest on the 100mm travel shows a 0.02mm undertravel. So a little more tweaking to do.
12:54 AM Joco[m]: hazzy: Well that is promising. Did a 100mm measurement test using DTI as the control datum. And got it bang on. The. Did a test cut with a target of 79mm and mic’d in at 79.01mm
12:54 AM Joco[m]: * hazzy: Well that is promising. Did a 100mm measurement test using DTI as the control datum. And got it bang on. Then did a test cut with a target of 79mm and mic’d in at 79.01mm
01:17 AM Joco[m]: hazzy: ok - pocket test done. Finish looks good. No backlash marks visible and can't feel any. Target size was 20mm and on measuring looks to be 20mm within 0.02mm. Very happy camper.
05:19 AM JT-Cave: morning
05:34 AM Joco[m]: JT-Cave: morning
06:39 AM TurBoss: morning
09:48 AM hazzy-m: Joco: fantastic!
01:11 PM Joco[m]: Monday. Blah.
01:22 PM TurBoss: morning Joco
01:34 PM Joco[m]: Hi ya. What’s happening?
01:56 PM Not-61bf: [02qtpyvcp] 07adargel commented on pull request 03#47: ENH: Add support for virtual keyboard and numpad - 13https://git.io/JJdx2
02:41 PM TurBoss: been playing old videogames with friends
02:58 PM Joco[m]: Cool. I spent a big chunk of the weekend problem solving on the cnc mill
04:28 PM * hazzy-m uploaded an image: ima_60fe82b.jpeg (353KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/GDCYVyzEJkWaHmtHqoWyPGSu/ima_60fe82b.jpeg >
04:30 PM hazzy-m: Filled ye old stuck F350 with use diesel crankcase oil. Hopefully she'll chooch after the SO2 does it's thing
04:30 PM hazzy-m: *H2SO4
04:50 PM Joco[m]: chooch chooch
04:58 PM Joco[m]: haz
04:59 PM Joco[m]: hazzy: I took following error samples of my circular milling test using halsampler. Will put graphs up. Always interesting to see how the machine is actually behaving,
04:59 PM * Joco[m] uploaded an image: xy_test_cut-v07.png (63KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/FayvcdFUYJUAUexPTwScNrAI/xy_test_cut-v07.png >
04:59 PM * Joco[m] uploaded an image: y_test_cut-v07.png (85KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/tExGiIuHQptOQbweiSCaFCBf/y_test_cut-v07.png >
04:59 PM * Joco[m] uploaded an image: x_test_cut-v07.png (83KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/XjtCCARdIvIiOlqGKMlWKKAL/x_test_cut-v07.png >
05:04 PM TurBoss: Joco: did you know there is a hal plot? :P
05:05 PM Joco[m]: is that the holiscope thing?
05:05 PM TurBoss: no
05:05 PM Joco[m]: in that case - no
05:06 PM TurBoss: https://forum.linuxcnc.org/qtpyvcp/36838-plot-widget-for-qtpyvcp
05:07 PM Joco[m]: oooo. Is that a standard widget in the designer?
05:07 PM Joco[m]: Would be good to add to a diagnostics page
05:07 PM TurBoss: yes
05:07 PM TurBoss: it is
05:08 PM Joco[m]: cool - gonna add that to my gui diagnostic page. It take it is realtime only? You are not saving the samples to file?
05:08 PM TurBoss: save has to be implemented
05:09 PM TurBoss: other way
05:11 PM TurBoss: https://github.com/kcjengr/qtpyvcp/blob/master/qtpyvcp/widgets/hal_widgets/hal_plot.py
05:11 PM Joco[m]: ok. So I wanted the saved data. But for realtime visulasation this looks a nice addition
05:42 PM * Joco[m] sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/WIFenjAEZlraDLeCKHgOLZwC/message.txt >
05:45 PM TurBoss: :O
05:46 PM Joco[m]: ah - series 1 is enabled by default. Solving
06:05 PM Joco[m]: hmmm ... need a series active to see the hal pins but you need to have something linked to the hal pin else the gui shits its self. But you can't see the naming pattern without out being able to start linuxcnc
06:06 PM Joco[m]: at least that's what looks like is happening.
06:22 PM Joco[m]: yup - if you link in a pin via hal its happy
08:30 PM hazzy-m: Joco: interesting, I'll gave to take a look at that issue, or see it the original author can take a look
08:44 PM Joco[m]: hazzy: would be good. The failure point is the update routine. It does test for an enabled data series but of course and disabled data series has no pins. lol
08:44 PM Joco[m]: bit of a catch 22
08:48 PM Joco[m]: * hazzy: would be good. The failure point is the update routine. It does test for an enabled data series but of course a disabled data series has no pins. lol
08:49 PM Joco[m]: a simple fix would be to just wrap that stuff in a try block and log any error with the pin name that is not connected.
08:50 PM Joco[m]: of course given it is an update routine that might be a little excessive