#hazzy Logs

May 04 2020

12:23 AM JamesW[m]: WIN. Zero new virus cases today.
04:21 AM RacoonRider|MovT: gmorning
04:22 AM JamesW[m]: morning/evening
04:22 AM RacoonRider|MovT: hehe
04:49 AM jthornton: morning
05:14 AM * JamesW[m] posted a file: Kazam_screencast_00000.mp4 (2014KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/pyuiGwPpPpUrDOumMXjINmFG >
05:18 AM JamesW[m]: little video to show the manual tool change routine. But it also shows some issues with the VTK plot
05:18 AM JamesW[m]: rendering issues of the machine extents
05:18 AM JamesW[m]: is is running off the latest pulled head from git
05:20 AM RacoonRider|MovT: @JamesW Very cool screen!
05:21 AM RacoonRider|MovT: have you tried to increase the clipspace from 100 to eg. 10000?
05:21 AM RacoonRider|MovT: its in inch always 100 in metric 10000 but need to test if 100 inches are real from the viewpoint
05:24 AM JamesW[m]: my test system sim is metric as that is what my real world machine is
05:24 AM JamesW[m]: so this boundary is 480mmx180x300
05:24 AM JamesW[m]: How do I change the clipspace?
05:59 AM RacoonRider|MovT: if you have the latest qtpyvcp you can set it in the ini file
06:00 AM JamesW[m]: running of git clone and its up to date
06:02 AM JamesW[m]: what is the ini setting?
06:03 AM RacoonRider|MovT: err you need to set it in the vtk_backplot.py
06:03 AM * RacoonRider|MovT sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/schSckqOZAwiRfMQthUeiZpQ >
06:03 AM RacoonRider|MovT: but it seems you already have it
06:03 AM RacoonRider|MovT: when you pulled today
06:04 AM RacoonRider|MovT: maybe you need more than 10 meters
06:04 AM JamesW[m]: yup
06:04 AM RacoonRider|MovT: i also need more
06:06 AM JamesW[m]: yup - checked. Have that code
06:06 AM RacoonRider|MovT: thas the patch https://github.com/kcjengr/qtpyvcp/commit/951a1a4a82a9
06:07 AM RacoonRider|MovT: have u tried self.clipping_range_far = 100000.0
06:07 AM RacoonRider|MovT: ?
06:07 AM RacoonRider|MovT: 100 meters
06:07 AM RacoonRider|MovT: that worked for me
06:07 AM JamesW[m]: no
06:08 AM JamesW[m]: only just seen this stuff
06:08 AM JamesW[m]: also that patch does not look like the code I have
06:08 AM JamesW[m]: sorry - yes it does
06:09 AM JamesW[m]: just reading it wrong
06:09 AM RacoonRider|MovT: strange
06:10 AM JamesW[m]: just tried the 100,000
06:10 AM JamesW[m]: no change - same issue
06:11 AM RacoonRider|MovT: whats the issue exactely?
06:12 AM RacoonRider|MovT: whats the issue exactly?
06:12 AM RacoonRider|MovT: the white machine borders?
06:12 AM * JamesW[m] uploaded an image: image.png (12KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/EFtwHtVGvfivNcLrBXHiHEqH >
06:13 AM JamesW[m]: machine boundaries on bottom not rendering
06:14 AM RacoonRider|MovT: what happen when you rotate the view a few degrees?
06:14 AM JamesW[m]: extents in absolute value is x=480, y=180, z=300
06:15 AM * JamesW[m] uploaded an image: image.png (11KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/WGKklALrMlsAdugnSuAcRfBz >
06:16 AM * JamesW[m] uploaded an image: image.png (11KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/GxvVCxptYrSFGVCOzuEFHBSx >
06:18 AM RacoonRider|MovT: its a strange thing. I trie it on my probe-basic. Need to go in the basement. brb
06:18 AM RacoonRider|MovT: its a strange thing. I try it on my probe-basic. Need to go in the basement. brb
06:23 AM RacoonRider|MovT: back. have the same behaviour. it looks like the render supress the "invisible" edges
06:23 AM RacoonRider|MovT: but in your case there are not
06:24 AM RacoonRider|MovT: all my front edges will render right
06:24 AM JamesW[m]: I have the same issue on two separate machines
06:24 AM JamesW[m]: with very different h/w
06:25 AM RacoonRider|MovT: thats weird
06:25 AM RacoonRider|MovT: what happen wenn you switch off the grid?
06:25 AM RacoonRider|MovT: wenn=when
06:25 AM JamesW[m]: will see
06:26 AM JamesW[m]: fixes it
06:26 AM JamesW[m]: grid breaks it
06:27 AM RacoonRider|MovT: i have another guess. it overlay the boundary
06:27 AM RacoonRider|MovT: because the are black
06:28 AM JamesW[m]: from a usability angle I can't see the grid so will probably not use it
06:28 AM RacoonRider|MovT: we need to change the order of rendering. boundary is the last we need to ernder i guess
06:28 AM RacoonRider|MovT: yeah i just switched it off too
06:29 AM JamesW[m]: or decide if grid is actually worth it. Cose unless it is visible its not adding anything
06:29 AM RacoonRider|MovT: GRID_LINES should only render the grid not the square
06:30 AM RacoonRider|MovT: than all will be fine
06:30 AM JamesW[m]: it would be nice to have the visual reference.
06:31 AM JamesW[m]: But I find the machine and prog boundaries more useful
06:31 AM JamesW[m]: might be just me
06:31 AM JamesW[m]: anyways ... 11:30pm and I have lovely work again tomorrow. So need sleeps
06:31 AM RacoonRider|MovT: yes for mee too. i will take a look into vtk_backplot
06:31 AM JamesW[m]: sweet
06:31 AM RacoonRider|MovT: okay. sleep well!
06:32 AM JamesW[m]: later
06:32 AM RacoonRider|MovT: you are on the westcoast?
06:32 AM RacoonRider|MovT: l8er
06:32 AM JamesW[m]: nope. NZ
06:32 AM JamesW[m]: its Monday night
06:32 AM JamesW[m]: 4th May
06:33 AM RacoonRider|MovT: ahh okay 🙂 nice!
06:33 AM JamesW[m]: i'm in your future ROFL
06:33 AM RacoonRider|MovT: hehe
06:34 AM RacoonRider|MovT: i was sure i hear voices 🙂
06:36 AM JamesW[m]: arghh - how do I set away on this thing
06:38 AM JamesW[m]: ah well -- walk off now.
10:07 AM hazzy-m: JamesW: neat video!
10:47 AM roguish[m]: hazzy: how goes the external widgets? when can Chris's probe routines be made into standalone ??
10:50 AM roguish[m]: hazzy: also, did you see that rene has master up on python3 ??? soo good.
12:49 PM hazzy-m: I'm working on some other things currently, but will try to get to making a stand alone probe dialog or widget
12:50 PM hazzy-m: do you have an idea for the type of layout you would like?
12:50 PM hazzy-m: I was thinking about making it like the renishaw probe interface, that kinda walks you thru it
12:51 PM hazzy-m: Yes, Rene has been doing a lot of great work!
12:56 PM Not-944e: [02qtpyvcp] 07KurtJacobson pushed 034 commits to 03StorageDeviceManager [+0/-0/±5] 13https://git.io/JfGOu
12:56 PM Not-944e: [02qtpyvcp] 07KurtJacobson 034f8ccd8 - removable device combo - send initial signals on show event
12:56 PM Not-944e: [02qtpyvcp] 07KurtJacobson 03c8e39db - add new device info DataChannel
12:56 PM Not-944e: [02qtpyvcp] 07KurtJacobson 03dd4b62a - file system - remove duplicate nc_files location attribute
12:56 PM Not-944e: [02qtpyvcp] 07KurtJacobson 03d0b4aba - add atDeviceRoot signal
02:16 PM JamesW[m]: Morning
02:21 PM JamesW[m]: Wanted to check. Is there a persistence framework in qtpyvcp? Im about to make an initial manual touch-off screen and was keen to have the value of the reference object (usually a guage block or pin) be persistent. Had a vague hope you just set a check box on an input item and things just got taken care of for you. :-D
03:51 PM JamesW[m]: I can see a data persistance plugin and can see the sample yaml config. Just not sure how it is intended to be used.
04:28 PM Freeway[m]: @TurBoss#0000 wondering if you could help me out. I've just recently switched to linuxcnc with a 7i76e and went through the steps to install probe_basic. My problem is the tool measurement isn't working, it doesn't recognize the probe input and tries to jam the tool all the way through.
04:34 PM JamesW[m]: you have the probe-in mapped in HAL?
04:38 PM hazzy-m: JamesW: there are a couple of ways to do persistence in QtPyVCP, but they all use the plugin as the back end
04:39 PM JamesW[m]: <Freeway[m] "@TurBoss#0000 wondering if you c"> As a starting point you need to map the pin on your 7i76e that has the probe input to motion.probe-input\
04:39 PM hazzy-m: I’ll give to some ideas when done working
04:39 PM JamesW[m]: Freeway: assume you have that setup?
04:42 PM * JamesW[m] sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/REuwZXyOYKIizmwkPbNIZRIY >
04:43 PM JamesW[m]: hmmm ... silly auto pretty formating
04:47 PM Freeway[m]: @JamesW#0000 correct, except I'm using or2 as well
04:47 PM JamesW[m]: so when you manually trigger your probe you can see motion.probe-input go high?
04:48 PM JamesW[m]: assuming your probe is one of those stylus like ones that you can bump with a finger to activate it
04:58 PM JamesW[m]: Freeway: I just checked one of the probe ngc files. it's using all the standard stuff I would expect. So nothing special. You obviously need to have specified the correct setup params. But I think the first principles check you need to do is to make sure that you are actually triggering motion.probe-input when you think you should be. Then move on to the software side.
05:03 PM Freeway[m]: @JamesW#0000 ill get back to you when I get home. Thanks
05:04 PM JamesW[m]: all good.
05:04 PM JamesW[m]: I've just been through all this setup on a different gui so its all fresh in my head,
08:39 PM hazzy-m: JamesW: you around?
08:39 PM JamesW[m]: keeping an eye on things
08:40 PM JamesW[m]: dipping in and out of work calls
08:40 PM JamesW[m]: ducking, diving. Generally being shifty
08:40 PM hazzy-m: sweet. I've been doing the same, first day working at home since COVID hit
08:41 PM JamesW[m]: that makes more sense with a srong east london accent lol
08:41 PM hazzy-m: lol
08:41 PM JamesW[m]: ah - been working from home for 6 odd weeks now
08:42 PM hazzy-m: we have a small core crew we keep in the office, most people have been working at home
08:43 PM JamesW[m]: I'm not critical services so home work is only option
08:43 PM hazzy-m: but one of our warehouse guys is positive, so decided to stay home for a bit
08:43 PM JamesW[m]: good idea
08:44 PM JamesW[m]: we seem to be getting it under control here. single digit cases for a while. zero yesterday
08:44 PM JamesW[m]: will be interesting to see what today is
08:44 PM hazzy-m: Fantastic!
08:44 PM JamesW[m]: ~5 people in hospital now.
08:44 PM JamesW[m]: no one in intensive care
08:44 PM JamesW[m]: 85% of cases recovered
08:46 PM hazzy-m: 85% is very good
08:46 PM JamesW[m]: NZ is only 5m people. But we closed down hard and fast. So it's paying off
08:46 PM JamesW[m]: all our deaths are from a retirment home cluster. So all 70+ yrs old
08:46 PM JamesW[m]: with underlying health issues
08:48 PM hazzy-m: yeah, it seems like its very mild unless you have other problems
08:49 PM hazzy-m: my coworker does not even have noticeable symptoms
08:50 PM JamesW[m]: thats the nasty part of this thing. Some people will sail through no issues but they are contagious. Others will get taken out.
08:51 PM hazzy-m: yeah, he would have been spreading it without even knowing, except he went in for minor ankle surgery and the tested and noticed he was positive
08:53 PM hazzy-m: JamesW: ok, for persistent data, you can use the persistent data manager directly like so: https://github.com/kcjengr/qtpyvcp/blob/master/qtpyvcp/widgets/dialogs/open_file_dialog.py#L62
08:54 PM hazzy-m: any type that can be JSON or pickle serialized can be stored. Its just key value to very easy
08:54 PM JamesW[m]: ok - so I just add some code in
08:54 PM hazzy-m: I'm not sure exactly what your trying to do, but yes, thats one way
08:54 PM JamesW[m]: All I want is a way to have some adjustable parameters in the gui that get persisted and loaded up again later
08:55 PM hazzy-m: oh, that's easy!
08:55 PM hazzy-m: Use settings
08:55 PM hazzy-m: they are defined in the YAML file
08:55 PM JamesW[m]: Where is that? I missed it
08:56 PM JamesW[m]: and they can be changed in the gui later and the change kept?
08:56 PM hazzy-m: https://github.com/kcjengr/probe_basic/blob/master/probe_basic/probe_basic.yml#L31
08:56 PM hazzy-m: yes
08:57 PM JamesW[m]: nice
08:58 PM JamesW[m]: code to link those to gui elements? Or some other method?
08:59 PM hazzy-m: they are settings widgets that provide a user interface displaying/modifying the settings value
09:00 PM hazzy-m: sek, trying to find an example
09:00 PM JamesW[m]: ok - will pick through probe basiic to see how its setu\p
09:00 PM hazzy-m: you can also use settings in rules
09:00 PM hazzy-m: they are exposed under settings:setting-name
09:00 PM JamesW[m]: cool
09:01 PM JamesW[m]: at this point its the persistant part I need
09:02 PM hazzy-m: and you can also reference settings values in NGC subroutines
09:02 PM hazzy-m: this is how Lcvette probe routines work
09:03 PM hazzy-m: really need to work on documenting this stuff, even I forget how everything works
09:22 PM JamesW[m]: yeah - the initial doc's are good. But as soon as you get to these types of areas I find myslef hunting through code and asking questions. Not really the most eficient use of time
09:23 PM JamesW[m]: what are lcvette rountines?
09:23 PM JamesW[m]: are they the ngc probing code that probe-basic uses/
09:25 PM JamesW[m]: AHAH!!!! SettingsLineEdit!
09:25 PM JamesW[m]: know I get it. Completely missed that little gem
09:35 PM hazzy-m: its like hunting for Easter eggs
09:57 PM Freeway[m]: @JamesW#0000 just looked for motion.probe-input in the halmeter and its not there. Is that where it would be?
09:58 PM hazzy-m: look harder
09:58 PM hazzy-m: its there
09:58 PM JamesW[m]: look in halshow
09:59 PM JamesW[m]: under the pins part of the tree
10:00 PM JamesW[m]: but it should have been in halmeter
10:01 PM JamesW[m]: just easier to find in halshow as it's got a nice treey structure
10:02 PM JamesW[m]: if it's not there then you need to check that you actually set it when you configured the base config. did you use the wiazrd to define your intial .ini and .hal files?
10:03 PM JamesW[m]: http://linuxcnc.org/docs/2.8/html/config/pncconf.html
10:05 PM hazzy-m: the motion pins should always be there, no matter what config ...
10:17 PM JamesW[m]: yeah - correct
10:17 PM JamesW[m]: forgot that
10:17 PM JamesW[m]: Freeway: can you see them in halshow
10:19 PM hazzy-m: Freeway: you could also try looking at the pin value with halcmd
10:19 PM hazzy-m: halcmd show pin motion.probe-input
10:19 PM * JamesW[m] uploaded an image: image.png (50KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/AhSHYNLAALKGhqOAwDZRNnHL >
10:21 PM JamesW[m]: but what you REALLY want to see is that there is something connected to that pin :)
10:21 PM JamesW[m]: I tihinnk
10:22 PM JamesW[m]: hmmm ... hazzy there should be something connected?
10:26 PM hazzy-m: yes, the signal from the probe should be connected to the motion.probe-input
10:27 PM JamesW[m]: this pic is off a sim so not "perfect" representation in that respect
10:28 PM hazzy-m: correct, unless there is a simulcasted probe input connected
10:43 PM * JamesW[m] uploaded an image: image.png (65KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/XfREoRNqikpAtwqZeYmIczPz >
10:44 PM * JamesW[m] uploaded an image: image.png (64KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/FbMkQtEePlShHKsWVnHtBecl >
10:44 PM JamesW[m]: oops - spelling mistake
10:45 PM hazzy-m: looks great!
10:46 PM hazzy-m: doubled the wrong letter lol Took me a good while to find it
10:46 PM JamesW[m]: this is my manual touch off screen.
10:47 PM hazzy-m: so you will put the touch off tool dia in?
10:47 PM JamesW[m]: for use with a wobbler or beep electric probe or a guage block/pin in Z
10:47 PM hazzy-m: sweet!!
10:47 PM JamesW[m]: yeah that is the offset amount.
10:48 PM JamesW[m]: half the diam of a wobbler and size of the guage block
10:48 PM hazzy-m: I've been intending to do that for a while, since I'm always setting the current axis offset with the wring sign
10:48 PM JamesW[m]: I will put in a full automated probe screen on the left side. But sometimes that is just ott
10:49 PM JamesW[m]: And not everyone has those clver probes
10:49 PM hazzy-m: yes, and sometimes its good enough to just touch off with the drill bit etc.
10:49 PM JamesW[m]: yup
10:50 PM JamesW[m]: I will need to solve the numeric entry problem. Does the framework have a pop up numeric keyupad?
10:52 PM hazzy-m: not in master, but there is a virtual keyboard branch (VKB) that's mostly functional
10:52 PM hazzy-m: need to finish it up and merge
11:06 PM Freeway[m]: I was really hoping that was it, motion probe-input shows both probe and tool setter working. Neither of them work while trying to measure a tool.
11:23 PM Not-944e: [02qtpyvcp] 07KurtJacobson pushed 031 commit to 03StorageDeviceManager [+0/-0/±3] 13https://git.io/JfGVD
11:23 PM Not-944e: [02qtpyvcp] 07KurtJacobson 03b8cd23a - add default file locations
11:23 PM JamesW[m]: that makes it more complex then.
11:24 PM JamesW[m]: so if you issue a tool probe from MDI and trigger the probe by hand before it gets anywhere near something hard it stops?
11:25 PM JamesW[m]: ie: http://linuxcnc.org/docs/2.8/html/gcode/g-code.html#gcode:g38
11:26 PM JamesW[m]: testing that should show if it's something to do with the prob ngc or not.
11:26 PM JamesW[m]: or some linkage/setting re that.
11:44 PM Freeway[m]: Probing is good, is there a setting im missing in the offsets tab for touching off?