#hazzy Logs

Jul 11 2020

01:15 AM Watze[m] is now known as 21WAAB4B0
01:16 AM Joco[m]: question: getting lost in a pile of forum threads on HAL. Any of you guys know how to us a hal pin that is IO (i.e. has both In and Out). When I try and link it to a qtpyvcp hal button Out pint via a net command I get an error about not being able to add the Out pin of the button.
01:17 AM Joco[m]: * question: getting lost in a pile of forum threads on HAL. Any of you guys know how to us a hal pin that is IO (i.e. has both In and Out). When I try and link it to a qtpyvcp hal button Out pin via a net command I get an error about not being able to add the Out pin of the button.
01:30 AM TurBoss: you need to do net on postgui hal
01:30 AM TurBoss: hal postgui
01:31 AM TurBoss: @joco
01:33 AM Joco[m]: had this line in a hal file called in the postgui_call_list.hal
01:33 AM Joco[m]: ```
01:33 AM Joco[m]: net glass-scale-reset <= [HMOT](CARD0).7i76.0.0.enc0.reset => qtpyvcp.halbtn-resetEnc0.out
01:34 AM Joco[m]: still got error
01:35 AM * Joco[m] uploaded an image: image.png (8KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/xsZTRBUiTsJmDxUudObYpPnA >
01:36 AM Joco[m]: from halshow
01:36 AM Joco[m]: the challenge seems to be that the Direction is both In/Out
01:36 AM Joco[m]: hang on let me get it to fail and I'll post the exact error message
01:40 AM * Joco[m] sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/DsBnKdStrpwWvMvmzSMjcpLa >
01:41 AM Joco[m]: bit stumped at the moment.
01:47 AM Joco[m]: I can do a manual setp via halshow no probs
01:48 AM Joco[m]: TurBoss: so if you have some more good ideas, I'm all ears. 🙂
02:15 AM TurBoss: Joco: whats the goal?
02:15 AM TurBoss: :P
02:15 AM TurBoss: ```
02:15 AM TurBoss: net glass-scale-reset qtpyvcp.halbtn-resetEnc0.out <= [HMOT](CARD0).7i76.0.0.enc0.reset
02:16 AM Joco[m]: I have this glass scale linked to enc0 on the 7i76 in mode 2. This is one of the MPG encoders. I want to use this to map the ballscrew in an autmated manner. So when I home the axis I need to reset the encoder to zero so that measurements are from the correct reference.
02:17 AM TurBoss: ahhhh
02:17 AM Joco[m]: I could do it by hand using halshow and a setp command. That is how I have been testing. But that seems really manky
02:17 AM Joco[m]: so want to link to a btn in gui if at all possible
02:17 AM Joco[m]: for ease of use
02:17 AM Joco[m]: eventually I will get a real encoder add on board as I want to mount these scales permanently and use them to inform eoffsets
02:18 AM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/bctuCpyMMdBFtEwTsRqOKjjO >
02:18 AM TurBoss: should be in?
02:18 AM Joco[m]: yes
02:18 AM Joco[m]: no its a button
02:19 AM TurBoss: ok
02:19 AM TurBoss: do the button show in halmeter?
02:19 AM Joco[m]: yes
02:19 AM TurBoss: 🤔
02:19 AM Joco[m]: visiible in halshow
02:20 AM Joco[m]: let me start machine up so I can take some halshow snips
02:20 AM TurBoss: oki
02:22 AM * Joco[m] uploaded an image: image.png (31KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/NhGUlmWiZRgHryWZibsxnBOp >
02:23 AM * Joco[m] uploaded an image: image.png (33KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/mubWbJZFjZvyteUpMOGWkxPP >
02:24 AM Joco[m]: its like a need to have an intermediate component that has an IO link on one side and on the other separated IN and OUT pins
02:25 AM TurBoss: qtpyvcp.halbtn-resetEnc0.out is OUT only
02:25 AM Joco[m]: I know
02:25 AM Joco[m]: that's the problem
02:25 AM TurBoss: you want to add IO to halbtn?
02:26 AM Joco[m]: but what can I put in the middle to act as an agent to allow the linkage?
02:26 AM TurBoss: in the main python
02:26 AM TurBoss: of thevcp
02:26 AM TurBoss: * of the vcp
02:26 AM Joco[m]: ok
02:26 AM TurBoss: hmmm
02:27 AM TurBoss: meybe we can add a setenable I bit to hal btl
02:27 AM Joco[m]: I will hunt through all the HAL docs and see if there is an existing component that will solve this. I can NOT be the first person hitting this.
02:28 AM Joco[m]: there is where lcnc is weak. The HAL documentation is skimpy and the IP in the forums is a nightmare to wade through
02:29 AM TurBoss: clicked is a good nema?
02:29 AM Joco[m]: huh?
02:35 AM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 09-25-59.png (92KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/zStjPPhGGAdKCkKLLPmfwvfI >
02:36 AM TurBoss: it needs some logic but is working
02:36 AM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 09-26-52.png (18KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/udOEAHAjBmJNmyzoGnbxUjBs >
02:40 AM Joco[m]: your plc gadget?
02:41 AM TurBoss: just added a in pin to the hal button
02:41 AM Joco[m]: so - would that allow it to be connected to an IO pin?
02:43 AM TurBoss: yes
02:46 AM Joco[m]: ok - will see if I can figure out how to do that
02:46 AM TurBoss: I'm doing
02:47 AM TurBoss: I'mt trying to figure some logic stuff
02:47 AM Joco[m]: rgr
03:24 AM TurBoss: Joco: I can't figure how to hook the logic correctly
03:25 AM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 10-15-58.png (162KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/MFMEJRIOzhMjKgXMcMwsWxaz >
03:31 AM Joco[m]: I'll have a look later. Watching movie at the moment. 8:22pm here
03:46 AM TurBoss: okay
05:17 AM jthornton: morning
05:28 AM jthornton is now known as JT-Cave
05:36 AM TurBoss: hello
05:37 AM TurBoss: jthornton: how are your sercos devices?
05:37 AM TurBoss: just curious
05:37 AM TurBoss: wich model
05:37 AM JT-Cave: the Koreans are working on it via my laptop and TeamViewer
05:37 AM JT-Cave: allen bradley sercos device
05:38 AM TurBoss: cool
05:38 AM TurBoss: great
05:38 AM JT-Cave: http://www.plctalk.net/qanda/showthread.php?t=125522
05:39 AM JT-Cave: I think it is not a program issue as I found out the start of the problem was a night shift maintenance guy blew it out with the power on and that's when it died
05:39 AM TurBoss: oh :(/
05:40 AM JT-Cave: I'm thinking the Kinetix 2000 drives are the issue
05:49 AM Joco[m]: TurBoss: might have cracked it. Just coded an activated io pin on the hal_button and blended it into the logic of the existing press code. Hooked it up and works. The next part would be for it to honour a True value coming in. Not sure how that would be used. At the moment its just sink for incoming.
05:50 AM TurBoss: you did?
05:51 AM * Joco[m] posted a file: hal_button.py (3KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/WvOtLEsHttwKZXBccDrsHCto >
05:52 AM Joco[m]: just tested it on real machine. HAL now works and the gui button activates the reset pin on the encoder
05:52 AM Joco[m]: see attached file above
05:54 AM TurBoss: great
05:54 AM TurBoss: I'm booting dev
06:07 AM Joco[m]: TurBoss: not saying it’s great. Just it worked.
06:07 AM TurBoss: xD
06:07 AM TurBoss: just testing
06:07 AM TurBoss: did you make the button checkable?
06:12 AM TurBoss: we need a keep
06:13 AM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 13-03-32.png (16KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/rlJjmUUDGxnajYVIVAWtKqDj >
06:14 AM Joco[m]: Nope. Just a push button. Not on/off.
06:14 AM TurBoss: then that should work
06:16 AM Joco[m]: Yup. It’s working on my real rig. Just used it to setup a test of the automated mapping gcode I have from the forums. Just did 256 samples going in positive and negative directions with the scale readings all being correctly recorded to file.
06:17 AM Joco[m]: pretty awesome to watch the machine just doing all this automatically
06:18 AM TurBoss: little vid?
06:18 AM TurBoss: :P
06:18 AM Joco[m]: lol - not very interesting to watch. But will be doing another run in a few mins. Will take a short vid and post
06:18 AM TurBoss: great
06:19 AM TurBoss: so I will know whats this about lol
06:19 AM Joco[m]: lol
06:19 AM Joco[m]: ball screw mapping
06:19 AM TurBoss: that means?
06:19 AM TurBoss: lol
06:20 AM TurBoss: hehehe
06:21 AM Joco[m]: you basically measure the machine travel v's an independent measure of travel (e.g. DRO scale) and record that info for the same set of distance points going from in one direction then going back in the other. (e.g. from 0 to max travel, then from max travel back to 0).
06:21 AM Joco[m]: This is put into a specific format used by the linuxcnc screw compensation processing.
06:22 AM TurBoss: ohhhhhhhhhhhh
06:22 AM TurBoss: ahhhhh
06:22 AM Joco[m]: It allows lcnc to cope with the variations that exist in C7 ball screws
06:22 AM TurBoss: innteresting
06:22 AM Joco[m]: as well as backlash
06:22 AM TurBoss: sorry didin't know bout
06:22 AM Joco[m]: that's ok
06:23 AM Joco[m]: if you have high spec ball screws you wouldn't need it. But for C7 where there is variation in the pitch of the screw even if you have double nuts.
06:23 AM Joco[m]: It has the potential to really tighten things up.
06:23 AM Joco[m]: Well that's the plan anyway
06:24 AM TurBoss: i think i have it
06:28 AM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 13-18-29.png (77KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/HNSpUwzXcWqjaSXjnOuByvAq >
06:29 AM TurBoss: hmmm 😕 not sure...
06:29 AM TurBoss: if value changed i don't need to asign it again
06:30 AM TurBoss: the I don't need setButtonDown functio
06:30 AM TurBoss: * the I don't need setButtonDown function
06:31 AM Joco[m]: are you trying to make it work for a sticky button? i.e. checkable?
06:31 AM TurBoss: just trying to sync hal and user click
06:32 AM TurBoss: if you set it from hal the click on the button with the mouse
06:34 AM Joco[m]: i'm not sure you need to set the pin in the setbutton proc
06:34 AM Joco[m]: actually - you might
06:34 AM Joco[m]: depends how the the other functions get fired
06:35 AM Joco[m]: you should probabltyy be checking it not None as is done in the other functions
06:35 AM TurBoss: oh ok
06:35 AM Joco[m]: I'm guessing that defensive pattern is there for a reason
06:36 AM TurBoss: not sure
06:39 AM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 13-29-29.png (14KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/WxMGrPQJvIDTXcEyzrTXADVG >
06:39 AM TurBoss: what do you think?
06:40 AM TurBoss: ```
06:40 AM TurBoss: setp qtpyvcp.halbutton.io 1
06:41 AM TurBoss: I think thoose safeties are there in case the click evenge gets fired before hal gets linked... 🤔
06:42 AM TurBoss: * I think thoose safeties are there in case the click event gets fired before hal gets linked... 🤔
06:42 AM Joco[m]: could work. Need to step through it to see what is happening
06:45 AM TurBoss: Joco: are you managing this from Gcode?
06:45 AM Joco[m]: managing what? The rest?
06:46 AM Joco[m]: reset
06:46 AM TurBoss: I mean you can use
06:46 AM TurBoss: ```
06:46 AM TurBoss: DEBUG(EVAL(vcp.widget.button).setDown({1})
06:46 AM TurBoss: from gcode
06:47 AM TurBoss: not sure if correct synyax
06:47 AM TurBoss: * not sure if correct syntax
06:47 AM Joco[m]: nah - I just get it all setup correctly at start via GUI
06:47 AM TurBoss: cool
06:47 AM TurBoss: there are also motion pins
06:47 AM Joco[m]: not trying to over complicate it. Just had to solve the io v's out problem
06:48 AM TurBoss: okay
06:48 AM TurBoss: not casino
06:48 AM TurBoss: :(
06:48 AM TurBoss: :P
06:48 AM Joco[m]: the gcode I have is logging correctly
06:51 AM Joco[m]: well ... seems to be sort of working. About to have another go. Will do quick video
06:52 AM TurBoss: hurray!
06:55 AM * Joco[m] uploaded a video: video_31eb7a5.mp4 < https://matrix.org/_matrix/media/r0/download/matrix.org/tdccnyugIVyciyBtHmIhafJJ >
06:56 AM TurBoss: ohhh
06:57 AM Joco[m]: Hmmm seems to have chopped it a bit.
06:57 AM Joco[m]: Or is that just my phone.
06:57 AM TurBoss: so there are absolute encoders?
06:57 AM TurBoss: or no
06:58 AM Joco[m]: Glass linear scales. So no absolute. Hence why I need to get them to home position and reset the encoder counting.
06:58 AM TurBoss: ahhhhhhhhhhhhhhhhhhhhhhh
06:58 AM TurBoss: 👂️
06:58 AM TurBoss: got it
07:07 AM Joco[m]: https://drive.google.com/file/d/1Bc3rCVidJqjyzkLD2PYMyxPQKPpOApaK/view?usp=sharing
07:08 AM Joco[m]: hopefully better
07:08 AM Joco[m]: looks like video up there is ok after all
07:09 AM Joco[m]: nice thing about this process is I can set it up and walk away and leave it. :)
07:09 AM TurBoss: I just created a branch
07:09 AM TurBoss: with the halbutton
07:14 AM Joco[m]: cool
07:15 AM Joco[m]: running another sample. Going to do 3 or 4 and see how the numbers compare
07:15 AM Joco[m]: that will be a tad over 2000 samples
07:16 AM Joco[m]: And tomorrow I will tackle the problem of perm mounting these things
07:23 AM TurBoss: nite
07:23 AM Joco[m]: nite
08:23 AM Lcvette: whats crackin?
08:23 AM Joco[m]: been hacking halbutton
08:24 AM Joco[m]: needed an io pin on it. Think TurBoss has a branch
08:24 AM Joco[m]: got screw comp auto creation working. video back a little on chat
08:24 AM Joco[m]: now 1:14am and I really should go to bed
08:25 AM Lcvette: Nice joco
08:25 AM Joco[m]: but just processing a screw compensation file based on 2000 odd samples
08:25 AM Lcvette: So how is the accuracy now?
08:25 AM Joco[m]: about to build a new compensation file and run the logging code to see how it compares to the dro scale
08:27 AM Lcvette: But have you just wanted basic measuring yet?
08:27 AM hazzy-m: Joco: nice!!!
08:30 AM Joco[m]: <Lcvette "But have you just wanted basic m"> not sure what you mean. Sorry - brain starting to get rathey fuzzy
08:31 AM Lcvette: ie commanded move verse dro move amount
08:31 AM Joco[m]: yeah - that is what have been measuring
08:31 AM Joco[m]: in an automated wat
08:31 AM Joco[m]: way
08:32 AM Joco[m]: 256 sample in forward direction, 256 sample in reverse (negative) direction
08:33 AM * Lcvette uploaded an image: image.png (346KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/yvRlkQOzlgKTqnrfxDRCvCqE >
08:34 AM Lcvette: is that 449.800 vs 449.736?
08:38 AM Lcvette: * But have you just tested basic measuring yet?
08:38 AM Lcvette: is this with the scrw settings set at 5mm?
08:41 AM Joco[m]: step scale has been adjusted based on a reading over ~450mm of travl
08:41 AM Joco[m]: STEP_SCALE = -400.266844563042
08:42 AM Joco[m]: Otherwise if becomes stupidly out.
08:42 AM Lcvette: that was what i am looking for is what was the 5mm setting on that 450mm travel
08:43 AM Joco[m]: You could reverse engineer from the scale. 2000 steps per turn
08:43 AM Lcvette: vs the dro
08:43 AM Lcvette: k
08:44 AM Joco[m]: Actually I could do that now. Lol
08:44 AM Lcvette: cool looks like you are havign fun!
08:44 AM Lcvette: :d
08:44 AM Lcvette: * :D
08:44 AM Lcvette: \o/
08:44 AM Joco[m]: tired. Going to quickly see how this new compensation file goes then call it a note.
08:45 AM Joco[m]: You are bang on. Now I have the scale I should use that for a 480mm reading to calc the steps/mm
08:46 AM Lcvette: yeah if you have the benefit of the scale its like setting up whatever blocks you want
08:46 AM Lcvette: :D
08:46 AM Lcvette: im interested to see how far off those screws are
08:52 AM Joco[m]: running the logging code with the new compensation file
08:54 AM Joco[m]: done in one direction.
08:54 AM Joco[m]: max error looks to be about 0,02mm
08:54 AM Joco[m]: the vast builk at 0.015mm or less
08:55 AM Lcvette: nice
08:56 AM Lcvette: sounds like you got it dialed in
08:56 AM Joco[m]: every now and then I get a 0.025
08:56 AM Lcvette: you should make a deviation graph
08:56 AM Joco[m]: on the return run
08:56 AM Joco[m]: I will tomorrow
08:56 AM Joco[m]: the points are all being logged
08:56 AM Lcvette: for how much it had to be corrected to get there
08:57 AM Joco[m]: I will do this with no compensation file and the nominal 400/mm step setting
08:57 AM Joco[m]: just to see how far off things are.
08:57 AM Lcvette: YES!
08:57 AM Joco[m]: Graphin that could be interesting
08:57 AM Lcvette: that is the hard data i find most interesting
08:57 AM Lcvette: to see if it is close to its C7 rating
08:58 AM Joco[m]: but tomorrow. I'm knackered. 1:48am here
08:58 AM Lcvette: ok,, that is acceptable
08:58 AM Lcvette: lol\
08:58 AM Lcvette: get soe sleep
08:58 AM Joco[m]: lol
08:58 AM Lcvette: * get some sleep
09:00 AM Joco[m]: i reckon if I can get these scales on proper encoders and do some clever feedback to eoffsets I might be able to get things even tighter. solid compensation file and some 0.01 to 0.02mm dynamic tweaking based on realtime scale feedback. Could be good. lol - will dream that. How sad.
09:02 AM Joco[m]: right scale done. Still looks good. Scans in both directions have the vast bulk of errors sub 0.01mm with s prinkling of points with 0.015, 0.02 or 0.025mm error. Overall. orders of magnitude better than it was. Next proof in the pudding will be cutting of course. But at least the base position is much netter than it was.
10:01 AM * hazzy-m is a little sad, just had to cut down our last standing peach tree
10:02 AM TurBoss: hello
10:02 AM hazzy-m: TurBoss: hey
10:02 AM TurBoss: sup hazzy ?
02:57 PM TurBoss: hazzy: did you figured the spamminator issue?
02:57 PM TurBoss: maybe if you fix could you link it to the #kcjengr_log:jauriarts.org spam room
02:58 PM TurBoss: 😲 that room is not on IRC
02:58 PM TurBoss: Lcvette: did a lot of work on LATC VCP
02:58 PM TurBoss: also hans is helping with VerticalVCP
03:10 PM Joco[m]: Morning. I think its morning. Im still fuzzy headed. In bed with ipad trying to wake up. Lol
03:10 PM Lcvette: TurBoss: great!!
03:10 PM Lcvette: :D
03:10 PM TurBoss: morning
03:10 PM Lcvette: \o/
03:10 PM Lcvette: i love new helpers!
03:11 PM Joco[m]: Hans is hands?
03:11 PM TurBoss: he did what told me 2 days in 30 mins
03:11 PM TurBoss: * also hansd is helping with VerticalVCP
03:11 PM TurBoss: * also hands is helping with VerticalVCP
03:11 PM TurBoss: yes
03:11 PM TurBoss: hands:
03:11 PM TurBoss: sorry
03:11 PM TurBoss: :OP
03:11 PM Joco[m]: Ah. Hard core dev. 20 yrs in the game. Gonna be a big help.
03:12 PM Joco[m]: Yeah - him. Hi hands
03:13 PM Joco[m]: For when hes back. 😇👍
03:13 PM Joco[m]: Gah. Head soooo muzzzy
04:01 PM Joco[m]: Right. Ups. And measurements for Lcvette
04:02 PM Lcvette: \o/
04:13 PM Joco[m]: Lcvette: so we want scale at 400/mm (i.e. nominal) with NO backlash compensation or screw compensation. Yes?
04:19 PM hands[m]: I have found a difference in the same control between the normal probe_basic and probe_basic_vertical... Specifically the vertical version has some dynamic properties (statusitem, format, etc) AND a rule... where as the same control only has a rule on the probe_basic template..
04:20 PM Lcvette: Joco: yes
04:20 PM Lcvette: hands: the vertical is outdated
04:21 PM TurBoss: he is updating it :)
04:21 PM * hands[m] uploaded an image: unknown.png (102KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/aLdQUcUBlXXZSECnmlHfyfPZ >
04:21 PM hands[m]: vs.
04:21 PM Lcvette: vertical needs to have the newer probe basic horizontal items brought over
04:21 PM * roguish[m] uploaded an image: image.png (61KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/LBqbymxKURgYAnRpWqEcmCdR >
04:21 PM roguish[m]: is it warm out yet ????
04:21 PM * hands[m] uploaded an image: unknown.png (117KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/DQLiBwAvMlYsgEGzFkMhgxNM >
04:22 PM hands[m]: however, probe_basic has some controls that still use statusitem.
04:22 PM Lcvette: i have not been in the vertical UI in some time
04:22 PM hands[m]: are those dynamic properties deprecated and I should move those to rules when I seem them?
04:22 PM TurBoss: hands: dynamic properties noremally are remove properties from the widget
04:23 PM TurBoss: * hands: dynamic properties noremally are removed properties from the widget
04:23 PM * Joco[m] uploaded an image: ima_b1ccf63.jpeg (156KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/duICxDMKCDscVclueowkxBcx >
04:23 PM hands[m]: ok, so when I see them delete them? (just confirming)... but that sounds like what you are suggesting... 🙂
04:23 PM Lcvette: i would compare them to the current probe basic and change to the same as the current probe basic style
04:23 PM TurBoss: yes safe to delete tem
04:23 PM TurBoss: * yes safe to delete them
04:24 PM Joco[m]: Lcvette: there you go. That is 0 to 480 move in single go at 200mm/min
04:24 PM Lcvette: and be careful the naming conventions remain the same as many of the items are connecte to subroutine calls
04:24 PM hands[m]: I am about 3/4 the way through the controls that are on all the screens.. then on to the middle switchable space.
04:24 PM hands[m]: Lcvette, that is what I am doing, control by control..
04:24 PM hands[m]: but this was soo different I wanted to confirm 🙂
04:25 PM Lcvette: awesome and thank you!
04:25 PM Lcvette: Joco: that looks pretty bad
04:25 PM Lcvette: :o
04:25 PM Lcvette: hands: yes different because so outdated
04:25 PM TurBoss: hands: in the main py file just saw a little thing :)we need to rename the ProbeBasic in supper car too
04:25 PM hazzy-m: hands: yes, statusItem has been removed (just confirming the confirmation lol)
04:25 PM TurBoss: * hands: in the main py file just saw a little thing :)we need to rename the ProbeBasic in supper call too
04:26 PM Joco[m]: 0.26mm our
04:26 PM Lcvette: i figured once probe basic was finished and locked down i would update instead of continuously updating two
04:26 PM Joco[m]: * 0.26mm out
04:26 PM Lcvette: but thanks to you it will get done sooner!
04:26 PM Lcvette: :D
04:26 PM Lcvette: \o/
04:26 PM Lcvette: Thank you Hands
04:26 PM Lcvette: TurBoss: o/
04:26 PM Lcvette: :D
04:26 PM Lcvette: \o/
04:26 PM Lcvette: fresh help!!
04:26 PM TurBoss: booting dev
04:26 PM hands[m]: well I am building a machine and I want vertical for it so... yah Open Source 🙂
04:27 PM TurBoss: Lcvette: we have to take a look at thw linear atc changes I did yesterday
04:27 PM Lcvette: ok
04:27 PM Lcvette: im at dev
04:28 PM Joco[m]: <hands[m] "well I am building a machine and"> Yup that how it works. Find and itch and scratch but others benefit from the scratching. It’s really a bit group scratch season.
04:28 PM Joco[m]: * > <@_discord_418751959818960896:jauriarts.org> well I am building a machine and I want vertical for it so... yah Open Source 🙂
04:28 PM Joco[m]: Yup that how it works. Find an itch and scratch but others benefit from the scratching. It’s really a bit group scratch season.
04:28 PM Lcvette: Joco: can you run it over 300mm
04:28 PM Lcvette: that is the tolerance std
04:28 PM Lcvette: supposed to be +/-50 microns lead error
04:29 PM * TurBoss uploaded an image: Captura de pantalla de 2020-07-11 23-19-31.png (21KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/VWCMoRhMXBsBbqtkUKCwHzop >
04:30 PM Joco[m]: <Lcvette "Joco: can you run it over 300mm"> I'm just running a complete screw mapping with these "default" settings. That will give me a picture of what is happening using nominal values with 256 measurements across the full travel
04:31 PM Joco[m]: love this program.
04:31 PM Lcvette: TurBoss: i cannot launch latc
04:31 PM Lcvette: on the new master
04:32 PM Joco[m]: got it from this thread: https://forum.linuxcnc.org/38-general-linuxcnc-questions/36044-possible-error-in-logic-of-linuxcnc-homing-with-leadscrew-compensation
04:32 PM TurBoss: gotany error message?
04:32 PM * Lcvette sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/WdFkzzvFjpTGyPIXvlKwadMx >
04:33 PM TurBoss: this should fix the "no module error"
04:33 PM Lcvette: im in the new branch master
04:34 PM TurBoss: ```python
04:34 PM TurBoss: python -m pip install --upgrade .e .
04:34 PM TurBoss: in the probe_basic directory
04:34 PM TurBoss: needed because some changes
04:34 PM * Lcvette uploaded an image: Screenshot_2020-07-11_17-24-47.png (37KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/IAJbuUYuorLMHsmiHzUrHEkq >
04:34 PM TurBoss: * ```python
04:34 PM TurBoss: python -m pip install --upgrade -e .
04:34 PM TurBoss: ```
04:35 PM Lcvette: ok ran that and now when i try an open i get a blank temrinal scren
04:35 PM * Lcvette uploaded an image: Screenshot_2020-07-11_17-26-08.png (11KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/UQRKGApxsWpbuyUmtptMGnQa >
04:35 PM TurBoss: yes there
04:35 PM TurBoss: is fine
04:35 PM Lcvette: no no
04:36 PM Lcvette: this is what i get when i try and run probe basic after runnong the pip
04:37 PM * Lcvette posted a file: pb_launch_issue.mkv (368KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/WsbqPJaFksnzTxehXDjwUcVc >
04:37 PM Lcvette: turboss ^^^
04:39 PM hazzy-m: it's not running any command it looks like
04:39 PM hazzy-m: strange
04:40 PM TurBoss: that messed
04:40 PM Lcvette: /o\
04:40 PM Lcvette: lol
04:41 PM TurBoss: can you open a new terminal an run
04:41 PM TurBoss: ```
04:41 PM TurBoss: linuxcnc /home/path/to/latc/ini.ini=
04:41 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/uRzEsdsofLBmpUFopvcXgrvB >
04:42 PM TurBoss: just saw the vid
04:44 PM * Lcvette sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/xTSJjpGvnNnKjpAwGSDQFeoW >
04:44 PM TurBoss: replace /home/cris/dev/probe_basic/config/pro...
04:46 PM Lcvette: ?
04:46 PM TurBoss: you need to run the command linuxcnc then the path to the latc ini
04:47 PM TurBoss: >/home/path/to/latc/ini.ini
04:47 PM TurBoss: this was just an example
04:47 PM TurBoss: * >
04:47 PM TurBoss: /home/path/to/latc/ini.ini
04:47 PM TurBoss: this was just an example
04:48 PM Lcvette: no idea
04:48 PM Lcvette: why doesn't it work with my launcher?
04:48 PM Lcvette: ```
04:48 PM Lcvette: xfce4-terminal -e "linuxcnc /home/chris/dev/probe_basic/config/probe_basic_latc/probe_basic.ini" --hold
04:50 PM TurBoss: ```
04:50 PM TurBoss: linuxcnc /home/chris/dev/probe_basic/config/probe_basic_latc/probe_basic.ini
04:50 PM TurBoss: * ```
04:50 PM TurBoss: linuxcnc /home/chris/dev/probe_basic/config/probe_basic_latc/probe_basic_latc.ini
04:50 PM TurBoss: ```
04:51 PM TurBoss: probe_basic_latc.ini
04:52 PM TurBoss: Lcvette: ^^
04:53 PM TurBoss: any luck?
04:55 PM * Lcvette sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/RPtHhJvLNMzdLhzfGfUXZOqh >
04:55 PM Lcvette: same error
04:55 PM Lcvette: but it trys starting now
04:56 PM TurBoss: ahhhh
04:56 PM TurBoss: you need to run """qcompile ."""
04:56 PM TurBoss: under probe_basic directory run
04:56 PM TurBoss: ```
04:56 PM TurBoss: qcompile .
04:59 PM Lcvette: you broke all the formatting, but it opened
04:59 PM Lcvette: \o/ and /o\
05:00 PM TurBoss: well
05:00 PM Lcvette: TurBoss: ^^^
05:00 PM TurBoss: its fragile
05:01 PM Lcvette: broke many things
05:01 PM Lcvette: /o\
05:01 PM TurBoss: its much broke?
05:01 PM Lcvette: think gonna have to go back and revrt
05:01 PM Lcvette: very much
05:01 PM Lcvette: on all pages lots of things
05:01 PM Lcvette: i did not update my master branch yet
05:01 PM TurBoss: yes
05:01 PM Lcvette: just checkout out yours
05:01 PM TurBoss: sure
05:08 PM * TurBoss posted a file: debug.ngc (1KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/elvTMgofFTyYzbFjSVtLPtpz >
05:08 PM TurBoss: Lcvette: run this file on LATC
05:14 PM Lcvette: i tried running now luck does nothing
05:15 PM TurBoss: in the latc tab?
05:16 PM Lcvette: ?
05:16 PM * Lcvette uploaded an image: Screenshot_2020-07-11_18-07-08.png (121KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/UYwwTfYCaOrvnyHmpPXfFZNU >
05:17 PM TurBoss: swich to lineat atc tab
05:17 PM TurBoss: the click cycle run
05:17 PM TurBoss: * then click cycle run
05:19 PM * TurBoss posted a file: debug.ngc (1KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/kfJCglVojPiQJNlpRzxeHMsV >
05:20 PM Lcvette: does nothing but flash the buttons for a second
05:20 PM Lcvette: what am i looking for?
05:21 PM * TurBoss uploaded a video: linear_atc_demo.webm (473KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/ETJbKjXpYJIMCChgZKPhxYBr >
05:21 PM TurBoss: didi a vid
05:22 PM Lcvette: \o/
05:22 PM Lcvette: i see yes im showing tools
05:22 PM TurBoss: hurray!
05:22 PM Lcvette: but i was not on linearatc tab when i first ran it so was not seeing it change
05:22 PM Lcvette: hurray!!
05:22 PM Lcvette: \o/
05:22 PM TurBoss: ya
05:22 PM TurBoss: great!
05:22 PM TurBoss: so whats the breakage?
05:22 PM TurBoss: layout?
05:23 PM Lcvette: all the formatting and font and colors
05:23 PM Lcvette: alot of broken stuff
05:23 PM Lcvette: not sure why
05:24 PM Lcvette: not idea where to look to fix it
05:24 PM * TurBoss uploaded an image: Captura de pantalla de 2020-07-12 00-14-42.png (259KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/zPXBJjMeWKqILHFVZgdOuKGv >
05:24 PM TurBoss: not good?
05:25 PM TurBoss: me is confused
05:25 PM TurBoss: needs to smoke
05:25 PM Lcvette: see the big 0 in the top left box
05:25 PM Lcvette: * see the big 0 in the top right box
05:25 PM TurBoss: ah
05:26 PM Lcvette: different from the box underneath it
05:26 PM Lcvette: click the atc setup tab top left corner
05:26 PM Lcvette: all box test is formatted wrong
05:26 PM Lcvette: the probe screen buttons are all wrongs
05:27 PM Lcvette: backgrouns coloring etc
05:27 PM TurBoss: qss file is old
05:28 PM Lcvette: why is it old?
05:28 PM Lcvette: what other files are wrong?
05:29 PM * TurBoss uploaded an image: Captura de pantalla de 2020-07-12 00-19-15.png (310KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/HgWNWrFQwMJEQLsSipvgxyPP >
05:29 PM TurBoss: fixed
05:29 PM Lcvette: \o/
05:29 PM TurBoss: xD
05:29 PM Lcvette: are the subroutine folder the enwest one?
05:30 PM TurBoss: it was a mess because there were things done to regular probe_basic and I had to begin picking from from regular to latc
05:31 PM TurBoss: Lcvette: I'll push the style fix
05:32 PM Lcvette: TurBoss: that scares me
05:32 PM Lcvette: because there is alot different im sure so now im not sure what else is mixed up
05:32 PM Lcvette: /o\
05:32 PM TurBoss: the old branch has all the work done in regular pb
05:34 PM Lcvette: give it a try
05:34 PM Lcvette: yes
05:34 PM Lcvette: my master branch is current
05:36 PM TurBoss: just pushed
05:36 PM Lcvette: k
05:36 PM Lcvette: \o/
05:38 PM TurBoss: ok something less
05:38 PM TurBoss: thats because the new field entries are setings and the old qss file didn't know how to paint them
05:54 PM Joco[m]: Lcvette: got some graphs of the error over the screw with nominal settings.
05:54 PM * Joco[m] uploaded an image: x_base_neg.png (17KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/lwfKNvoVRITaESLEawgsidlP >
05:54 PM * Joco[m] uploaded an image: x_base_pos.png (18KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/JmbtjuJEgiELgzTLXoMxlITK >
05:54 PM Joco[m]: thats with 400/mm step setup
06:02 PM Lcvette: TurBoss: sorry got called to the dinner table
06:03 PM TurBoss: no problem
06:03 PM Lcvette: joco, so i guess now you can dial the steps per to flattent hat line out
06:03 PM Lcvette: good news is it looks fairly linear
06:04 PM Lcvette: then you can dial in the lead error
06:04 PM Joco[m]: yeah - might need to run things again
06:04 PM Joco[m]: here's some more pics
06:04 PM Lcvette: im trying also now to help someone communicate with their new 7i76e card and not going well
06:05 PM Joco[m]: These are the pics with a "better" steps/mm setting
06:05 PM * Joco[m] uploaded an image: x_steps_neg.png (19KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/zNJTqbDCkGypFkZTnWvTXkYe >
06:05 PM Lcvette: joco do you have static ip setup on mesa?
06:05 PM * Joco[m] uploaded an image: x_steps_pos.png (19KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/vdtTJIMTXWktDsyIlpoWGMcZ >
06:06 PM Joco[m]: <Lcvette "joco do you have static ip setup"> I'm currently running a plug-n-go kit. Mate has his set for the default that comes out of the box
06:06 PM Lcvette: sp looks like you could tweak it a hair more
06:06 PM Lcvette: turboss yours is ethernet card correct?
06:07 PM Lcvette: or no you have 6i25/7i76
06:07 PM TurBoss: pcie
06:07 PM TurBoss: yes
06:07 PM Lcvette: do you guys know how to set the static IP?
06:07 PM Joco[m]: on the card or on oc?
06:07 PM Lcvette: https://forum.linuxcnc.org/38-general-linuxcnc-questions/33969-7i76e-quick-setup
06:08 PM Lcvette: i tried following this but it errors
06:08 PM Lcvette: we set the jumper on the card for 10.10.10.0
06:08 PM Joco[m]: ok
06:08 PM Joco[m]: if you type ifconfig in terminal what comes back?
06:11 PM Lcvette: bash
06:11 PM Lcvette: ipconfig?
06:12 PM Joco[m]: sorry - yeah
06:12 PM Joco[m]: ipconfig
06:12 PM Lcvette: bash on that too
06:12 PM Lcvette: command not found
06:12 PM Joco[m]: no - its ofconfig
06:12 PM Joco[m]: * no - its ifconfig
06:13 PM Joco[m]: no ifconfig???
06:13 PM Lcvette: command not found
06:13 PM TurBoss: ```
06:13 PM TurBoss: ip a
06:13 PM * Lcvette sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/FfCvdkNtvNGbnFiEjJmgCaEl >
06:13 PM Joco[m]: fuck - that install has moved to the bloody ip commane
06:14 PM Joco[m]: ip link
06:14 PM Lcvette: eth0
06:14 PM Joco[m]: did you use eth0 in the networks setup?
06:14 PM Lcvette: yes
06:14 PM Joco[m]: cool
06:15 PM Joco[m]: so if you use ip address
06:15 PM Joco[m]: you get the 10.10... you are expecting on eth0?
06:15 PM * Lcvette sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/UhNdoakKvDwwkTZToXiKUejP >
06:15 PM * Lcvette uploaded an image: image.png (53KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/tRxKQXTVHvDEbqdmIBFCdYFw >
06:16 PM Lcvette: turboss joco ^^^
06:16 PM Joco[m]: what is you loadrt line?
06:16 PM Lcvette: that is the error message i get
06:17 PM Lcvette: ```
06:17 PM Lcvette: loadrt hostmot2
06:17 PM Lcvette: loadrt hm2_eth board_ip="10.10.10.10" config=" num_encoders=1 num_pwmgens=0 num_stepgens=5 sserial_port_0=00xxxx"
06:18 PM Joco[m]: why is the address in the networks not 10.10.10.10?
06:18 PM Lcvette: just following what was in that link
06:18 PM Lcvette: i dunno
06:19 PM Joco[m]: oh - i know
06:19 PM Lcvette: you know the issue?
06:19 PM Lcvette: :o
06:19 PM Lcvette: \o/
06:19 PM Joco[m]: no - lol - I know why addresses not the same
06:20 PM Lcvette: oh
06:20 PM TurBoss: Lcvette: is you mesa car directly connected?
06:20 PM Lcvette: yes
06:21 PM TurBoss: what does
06:21 PM TurBoss: ```bash
06:21 PM TurBoss: ip a
06:22 PM TurBoss: you have two entries in for eth0 in you network settings
06:22 PM Joco[m]: do you have a wifi dongle in the machine?
06:22 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/vjAGnursnzGcljFyuiTiyTjr >
06:23 PM Joco[m]: snap - you beat me
06:23 PM Lcvette: ok testing
06:23 PM Lcvette: sek
06:26 PM Lcvette: same error
06:26 PM Lcvette: how can i test comunication with 7i76e?
06:27 PM TurBoss: Lcvette:
06:27 PM TurBoss: ```
06:27 PM TurBoss: ping 10.10.10.10
06:27 PM Lcvette: TurBoss: what should the reponse be
06:27 PM Lcvette: copy results here
06:27 PM TurBoss: past
06:28 PM * Lcvette uploaded an image: image.png (35KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/nITfTzTULQrtVvvcBbfEHCMy >
06:28 PM TurBoss: no comm
06:28 PM Lcvette: so then what?
06:28 PM Lcvette: /o\
06:28 PM TurBoss: ```
06:28 PM TurBoss: ip a
06:28 PM TurBoss: then show here
06:28 PM * Lcvette sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/FCGXQrowcDtYhdKtkoKavykK >
06:29 PM TurBoss: for some reason eth0 is not getting an ip
06:29 PM TurBoss: where are you editing the configs?
06:29 PM TurBoss: wich file path
06:30 PM Lcvette: i forgot to run sudo /etc/init.d/networking restart
06:30 PM Lcvette: after i changed it
06:30 PM Lcvette: retesting
06:30 PM TurBoss: ping works?
06:30 PM TurBoss: ok
06:31 PM Lcvette: worked after network restart\o/
06:32 PM Lcvette: hurray!!
06:32 PM Lcvette: :D
06:32 PM TurBoss: great
06:32 PM Lcvette: party!!
06:32 PM Joco[m]: lol
06:32 PM TurBoss: but didi you remove the ofending eth0 entry?=
06:32 PM Joco[m]: you should do
06:32 PM Lcvette: ?
06:33 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/aOIcUoEiUMsUzxPfPWLhRwZB >
06:33 PM Lcvette: i did not restart the netowrk for changes to take effect
06:33 PM Lcvette: yes
06:33 PM Lcvette: removed that
06:33 PM TurBoss: great
06:33 PM Lcvette: did not retsrat network after making that change
06:33 PM Lcvette: after restart network it worked
06:34 PM Joco[m]: cool - you are away then
06:34 PM Joco[m]: I have to do all this soon when I convert my rig to ethernet
06:35 PM Joco[m]: the card is sitting there with an accusing look on it
06:35 PM Lcvette: geterdone!
06:37 PM Joco[m]: yes Keith
06:44 PM Joco[m]: Lcvette: refined the steps/mm. Think its quite a bit tighter now. Here are the error graphs.
06:44 PM * Joco[m] uploaded an image: x_steps_neg.png (21KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/akHwsyhIcoqxrWgkaSgYSXVV >
06:44 PM * Joco[m] uploaded an image: x_steps_pos.png (21KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/fuxacZDhsKSURZVmoJVJlrSw >
06:45 PM Joco[m]: now I need to process a new compensation file.
06:45 PM Lcvette: is blc off still?
06:46 PM Joco[m]: yup
06:47 PM Lcvette: so the directional error is your backlash amount maybe?
06:47 PM Joco[m]: to a point. Along with some likely variation in the screw
06:47 PM Lcvette: 0.07mm?
06:47 PM Joco[m]: but yeah circa 0.7mm
06:47 PM Lcvette: well start of travel appear to change from one graph to the next
06:47 PM Joco[m]: 0.07mm
06:48 PM Lcvette: try throwing that in and run it again and see if it matches
06:48 PM Joco[m]: the process is start at x0, rapid to x480, start samples back to 0, then revrse sample back to 480
06:49 PM Lcvette: ie brings the start to the same place the previous move ends on the scale
06:49 PM Lcvette: that shold be a great way to nail down blc
06:50 PM Joco[m]: part of the challenge is that there is measurable variation in travel per turn as well. The joys of rolled v's ground.
06:51 PM Joco[m]: but the backlash part is really moot. The compensation file overrides the backlash setting when you use a compensation file.
06:51 PM Joco[m]: the fact there is backlash is essentially baked into the compensation file.
06:55 PM Joco[m]: will crack in to gettting this new compensation file built then will try another test using that.
08:35 PM travis-ci: kcjengr/qtpyvcp#1657 (QPlainTextEdit - 9862885 : Kurt Jacobson): The build was fixed.
08:35 PM travis-ci: Diff view : https://github.com/kcjengr/qtpyvcp/compare/7fde9def96f3...98628857f2cd
08:35 PM travis-ci: Build details : https://travis-ci.org/kcjengr/qtpyvcp/builds/707282396
08:35 PM TurBoss: \o/
08:35 PM TurBoss: some spam
09:24 PM * hands[m] uploaded an image: unknown.png (184KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/ILYmRmFtUthLcsqaFlmejZDX >
09:24 PM hands[m]: a lot still broken.. but progress...
09:25 PM hazzy-m: hurray!
09:25 PM Joco[m]: progress is good.
09:25 PM Joco[m]: We like progress :-)
10:41 PM hazzy-m: Joco: This might replace our work on the QScintilla based gcode editor:
10:41 PM hazzy-m: https://github.com/kcjengr/qtpyvcp/commit/2c9cb4c4e211c356d7be99f127f3b55be5d0f0e0#diff-7e72d6a061da7173f85774db8e12de9e
10:41 PM hazzy-m: its pure QT using and based on a QPlainText edit
10:41 PM hazzy-m: to much more light weight and faster
10:41 PM hazzy-m: * so more light weight and faster
10:43 PM * hazzy-m uploaded an image: image.png (32KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/gvEeQvFtdhTwpdFpOIBsRODX >
10:43 PM Joco[m]: Lite weight sounds good.
10:43 PM hazzy-m: still needs tweaking. but it mostly works very well
10:43 PM hazzy-m: JT wrote it a while back
10:43 PM Joco[m]: I’m all for simple
10:44 PM hazzy-m: even has context menus for Run from line
10:57 PM Joco[m]: Nice.
11:35 PM Joco[m]: hazzy: would seem to be a much slicker implementation. Worth pursuing I think
11:49 PM Joco[m]: Re done my compensation settings on X axis. Getting the process pretty smooth now with some macros in libre calc to help move data about. Results looking good. There are error graphs on the commanded v's actual position. Measure going from 0 to 480m and from 480mm to 0.
11:49 PM * Joco[m] uploaded an image: x_test_neg.png (20KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/jxVYerHDFWtIWWTSqpHUxKdH >
11:49 PM * Joco[m] uploaded an image: x_test_pos.png (20KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/HWGuvMPxyVmZrhLDutBTCotF >
11:55 PM Joco[m]: It's looking a heck of light better than it was when I started out.