#linuxcnc Logs

Feb 24 2024

01:09 AM Deejay: moin
02:54 AM mrec: does anyone know how NC files are transferred to Mitsubishi M60 (or similar) controllers via parallel port?
02:55 AM mrec: is it just dumping the g-code to the parallel port? I was able to resurrect the PC that was infront of that machine but the transfer program seems to require some hardware dongle to start it up (it's pretty old stuff)
02:56 AM mrec: my plan is first try to get it run somehow, use it for some time and lateron maybe change it to linuxcnc
03:02 AM jpa-: are you sure it is parallel port and not 25 pin serial port?
03:11 AM mrec: it went through the parallel port for sure
03:12 AM mrec: I will check the pc settings later, the mainboard was dead I just replaced it
04:16 AM Tom_L: https://www.scribd.com/document/187997363/M60-RS-232-Procedure
04:17 AM Tom_L: alot of those old ones used odd speeds like 4800 baud
04:19 AM Tom_L: and i had to introduce alot of waits between bytes on some
04:21 AM Tom_L: realterm is excellent for those types of things
04:22 AM Tom_L: https://sourceforge.net/projects/realterm/
04:22 AM Tom_L: https://alternativeto.net/software/realterm/?platform=linux
04:23 AM Tom_L: ^^ alternatives since realterm is windows
04:25 AM Tom_L: that one looks like 9600 E 7 2
04:25 AM Tom_L: even parity, 7 bits, 2 stop bits
04:28 AM Tom_L: on db9 to db25, 2 & 3 are straight thru, otherwise they're crossed
04:28 AM Tom_L: 25 to 25 are X as are 9 to 9
04:29 AM Tom_L: unless you're absolutely sure it's a parallel port and not a db25 serial
04:30 AM Tom_L: all the ones i've come across were serial
04:44 AM JT-Cave: morning
05:24 AM DerChristian: You wanted me to give you a quick headsup if I got any errors with mesact, now I got one https://imgur.com/zl0T6cf.png
05:30 AM DerChristian: JT, ss promised I wanted to give you a little update on my progress after you helped me out a couple of days ago. Mesaact indeed seemed to have fixed something pnfconfig did wrong. After configuring with mesact I was at least able to start the spindle. After around in search for the problem I found an mirror, and it was me... the whole damn time. I was so deep in to wiring the cabinet, that I didn't read
05:30 AM DerChristian: the manual of the stepper driver correctly and put plus 5V on the Enable input of the stepper drivers, not realising that it will disable the steppers
06:03 AM JT-Cave: DerChristian, thanks for the bug report, it's been fixed now
06:04 AM DerChristian: lovely, thank you very much
06:05 AM JT-Cave: I've always thought it was odd to have a disable pin on stepper drivers...
06:06 AM * JT-Cave gets back to working on Flex
06:55 AM CaptHindsight: steppers are used sometimes just to move something for a set number of steps and there is no concern about their present position or if they have moved since they were last driven
06:56 AM CaptHindsight: they just enable the drive and step and then disable the drive
07:46 AM hermann: hi, after updating to LINUXCNC - 2.10.0-pre0-2816 i get some errors opening g codes using small hole feature for plasmac, File "/usr/bin/qtplasmac_gcode", line 495, in check_math
07:46 AM hermann:     tmp1 = data.split(axis)[1]
07:46 AM hermann: IndexError: list index out of range
08:01 AM JT-Cave: the plasmac guy is only on the forum and he's prob asleep right now being from down under IIRC
08:02 AM JT-Cave: unless there is some feature in master you "must" have it's not wise to use it
08:03 AM hermann: how can revert the update? i ran apt-get upgrade
08:04 AM JT-Cave: in the synaptic package manager you can uninstall 2.10 and install 2.9
08:07 AM JT-Cave: hmm apt upgrade should not have installed a newer version...
08:08 AM JT-Cave: what os are you using?
08:08 AM hermann: plasma@linuxcnc:~$ uname -a
08:08 AM hermann: Linux linuxcnc 4.19.0-25-rt-amd64 #1 SMP PREEMPT RT Debian 4.19.289-2 (2023-08-08) x86_64 GNU/Linux
08:09 AM hermann: i did do a apt-get upgrade
08:09 AM JT-Cave: I don't see that any debian package has 2.10
08:09 AM JT-Cave: https://packages.debian.org/search?keywords=linuxcnc-uspace
08:10 AM JT-Cave: what does lsb_release -d return?
08:11 AM hermann: Description:   Debian GNU/Linux 10 (buster)
08:11 AM hermann: plasma@linuxcnc:~$ cat /etc/apt/sources.list.d/linuxcnc.list
08:11 AM hermann: deb http://buildbot.linuxcnc.org/ buster master-rtpreempt
08:11 AM hermann: deb-src http://buildbot.linuxcnc.org/ buster master-rtpreempt
08:11 AM JT-Cave: linuxcnc is not even in debian 10
08:11 AM JT-Cave: ah your using buildbot
08:12 AM hermann: yes
08:12 AM Tom_L: what's 4.19 preempt-rt from?
08:12 AM JT-Cave: change master to 2.9 something then sudo apt update
08:12 AM JT-Cave: preempt-rt is the kernel
08:12 AM hermann: how?
08:13 AM JT-Cave: you can edit that in the synaptic package manager
08:13 AM hermann: Master branch (development - 2.9)
08:13 AM hermann: Buster (arch: amd64, armhf)
08:13 AM hermann: uspace: realtime with RT-Preempt, and simulation
08:13 AM hermann: deb     http://buildbot.linuxcnc.org/ buster master-rtpreempt
08:13 AM hermann: deb-src http://buildbot.linuxcnc.org/ buster master-rtpreempt
08:13 AM hermann: should be Master branch (development - 2.9)?
08:13 AM hermann: but its installed 2.10
08:13 AM hermann: from http://buildbot.linuxcnc.org/
08:14 AM JT-Cave: master is 2.10 now
08:14 AM JT-Cave: looks like this page is not up to date http://buildbot.linuxcnc.org/
08:15 AM JT-Cave: deb http://buildbot.linuxcnc.org/ buster 2.9-rtpreempt
08:15 AM JT-Cave: deb-src http://buildbot.linuxcnc.org/ buster 2.9-rtpreempt
08:15 AM JT-Cave: try that
08:15 AM * JT-Cave starts his chicken day
08:17 AM hermann: after update and uprade: The following packages have been kept back:
08:17 AM hermann: linux-image-rt-amd64
08:17 AM hermann: should i do a sudo apt autoremove?
08:40 AM hermann: its working now, seems like its not a good idea to be using the dev branch,
08:40 AM Tom_L: not unless you need a feature it has
08:41 AM Tom_L: JT-Cave, PR #86 mesact up
08:51 AM mrec: Tom_L: many thanks; the system only has a parallel port and db9; I will doublecheck on Monday
08:53 AM Tom_L: mrec, one reason i think it's serial is purely due to the cable length for parports
08:53 AM Tom_L: unless the pc is sitting right beside the cnc
08:55 AM Tom_L: ours were connected to a pc in the cad room and some were quite a distance from that
08:55 AM Tom_L: a couple i added 'black box' extenders to increase the distance
08:56 AM mrec: the PC is right infront of the CNC
08:56 AM Tom_L: in your case yes
08:58 AM Tom_L: https://www.blackbox.com/en-ca/store/product/detail/Async-RS232-Extender-over-CATx-DB9-with-Control-Signals-to-Terminal-Block/ME890A-R2
08:58 AM Tom_L: kinda surprised you can still find those
09:10 AM Tom_L: certainly didn't pay that much for them
09:10 AM Tom_L: all it is is a max chip inside
09:10 AM Tom_L: https://www.ti.com/lit/ds/symlink/max232.pdf?ts=1708773635232&ref_url=https%253A%252F%252Fwww.google.com%252F
09:19 AM JT-Cave: Tom_L, if you get a chance see if I'm missing anything https://paste.debian.net/1308460/
09:25 AM mrec: Tom_L: I will check the cable next week I did not see any adapter in between
09:25 AM mrec: the machine itself seems to have a db27 connector
09:26 AM mrec: who knows what's behind that
09:26 AM mrec: seems like the machine was controlled via a software from Taiwan called cnc tools which required a hardware dongle - which is not there anymore
09:27 AM mrec: everything's very outdated
09:28 AM Tom_L: yeah they loved dongles back then
09:29 AM Tom_L: the only reason we used those black box was to extend the length. i doubt you have anything like that
09:30 AM Tom_L: one way you could tell parport from serial was the gender but i don't remember which was which now
09:31 AM mrec: I have the mainboard here it's parallel
09:32 AM mrec: I took the PC with me becaues it didn't boot I told the owner that I'll make it work..
09:32 AM mrec: if I can get the machine to work I'll buy the machine very likely, especially we can have the space for free as long as we cooperate with the owner with some projects so 1-2 years rent for free for sure
09:50 AM Tom_L: mrec, keep in mind alot of times it costs more to move one than they're worth
09:50 AM Tom_L: unless you can move it yourself
10:22 AM mrec: I don't want to move it all
11:07 AM JT-Cave: Tom_L, when you get a chance see if anything is missing https://paste.debian.net/1308460/
11:09 AM Tom_L: headin out but i'll check when i get back
11:09 AM JT-Shop: ok
11:12 AM JT-Shop: nap time here
12:11 PM solarwind: Can linuxcnc be run headless?
12:12 PM solarwind: I'm contemplating overhauling it into a more modular system
12:30 PM Unterhausen: It used to be able to be run headless, but I think that part is currently broken
12:30 PM Unterhausen: people talk about it occasionally. It seems like most get really frustrated and stop posting about it
12:30 PM Unterhausen: maybe because they get it to work, idk
12:37 PM Tom_L: i thought somebody managed to get that to work recently
12:40 PM chris131313: Nospark you around?
12:48 PM Tom_L: JT-Shop, where you putting things like run, pause, resume, step?
12:48 PM Tom_L: also file open, reload, show hal, clear mdi, reload tool table ?
12:50 PM Tom_L: manual, mdi, auto...
01:59 PM NoSpark: hey guys, any idea how to add the other cards to pncconf?
02:00 PM NoSpark: there used to be a mesa firmware package, but I can't remember if that still works?
02:40 PM Unterhaus_ is now known as Unterhausen
02:49 PM Leeloo_ is now known as Leeloo
03:08 PM XXCoder: sigh lately make it extreme isnt really great channel to watch
03:14 PM XXCoder: dumped it. got tired of it
03:23 PM Unterhausen: I think I figured out a mosfet is bad, but I really don't want to desolder a surface mount mosfet
04:11 PM Tom_L: JT-Shop, i'm not quite sure i follow what you did there
04:23 PM JT-Cave: Tom_L, run step pause and resume and stop are command buttons
04:23 PM JT-Cave: open reload etc are menu items as well as tool bar items
04:24 PM JT-Cave: the paste is just the enable disable for different states
04:24 PM chris131313: hey folks quick question here. no led are illuminated on my 6i25. never used it before just did a lspci -v and got 0.0 pci bridge: texas instruments xi02001 pci express to pcie and a 2:0 DPIO module device 2718:5125
04:25 PM chris131313: is it safe to say it sees the card
04:25 PM chris131313: shouldnt i have atleast one led or light on?
04:26 PM JT-Cave: I don't see any leds on my 6i25 but it's hard to see...
04:27 PM JT-Cave: install mesact and it can check the board
04:27 PM chris131313: JT-Cave: can i just sudo apt get mesact?
04:27 PM chris131313: or use package managre
04:27 PM chris131313: manager**
04:28 PM Tom_L: https://github.com/jethornton/mesact
04:29 PM Tom_L: JT-Cave, where are the debs?
04:30 PM chris131313: thanks Tom_L i will start reading. thanks
04:30 PM Tom_L: https://gnipsel.com/linuxcnc/mesact/install.html
04:31 PM Tom_L: the deb links are there
04:31 PM chris131313: thanks
04:31 PM JT-Cave: debs for mesact?
04:32 PM JT-Cave: https://github.com/jethornton/mesact/releases
04:33 PM JT-Cave: so the enable/disable logic for example estop off almost everything disabled estop on enable power
04:34 PM Tom_L: so you're grouping them for different senario?
04:38 PM JT-Cave: just trying to make sure I have the logic correct when to disable and when to enable a control like what happens in axis
04:39 PM JT-Cave: actually I see a mistake I think, you need both a file loaded and homed to run a g code file or do mdi
04:41 PM JT-Cave: https://paste.debian.net/1308491/
04:42 PM JT-Cave: I think axis has the enable/disable down pat so trying to do the same
04:42 PM JT-Cave: my deer seem to be done with dinner so out to the machine shop
04:51 PM chris131313: I just installed mesact under"mesa setup i selected drop down under board. i tried the 6i25 and the 5i25 . i left the ip address black and clicked "find IP board"it looked at 10.10.10.10 and 192.168.1.121 and found nothing. seems odd because the lights flashfor a second when i start the machin
04:52 PM chris131313: Machine= PC
04:52 PM JT-Shop: 5/6i25's are not ip boards
04:53 PM chris131313: ok. so i just clicked "verify board" and is asked for password. then said 5i25
04:53 PM chris131313: Thanks JT-Shop
04:53 PM JT-Shop: both boards are 5i25
04:53 PM JT-Shop: you're welcome
04:54 PM JT-Shop: glad I can help
06:11 PM JT-Shop: say goodnight Gracie
06:28 PM JT-Cave: Tom_L, I fixed mesact to not enable find ip board if a pci(e) board is selected so a build is in order when you have time
06:37 PM XXCoder: sigh. watching someone work on very large 3d printer. he uses usual reprapy way of using limit switches
06:37 PM XXCoder: ie it moves into and smashes into limit bar
06:37 PM jdh: smash into them?
06:37 PM XXCoder: something go wrong, you lose the switch
06:38 PM XXCoder: yep
06:39 PM XXCoder: he doesnt say, but I hope he uses NC switches in least
06:39 PM XXCoder: No is no-no
06:44 PM XXCoder: https://youtu.be/Tbrdz3eWDQM btw, long and lot of it is talky
06:44 PM XXCoder: I'll never do it, because hotend is $220,000 lol
06:45 PM XXCoder: when something is worth half of house, I wont use it
06:46 PM * XXCoder says that, but he used to run 5 million dollar cnc cell. lol
06:49 PM CaptHindsight: solarwind: yes it can run headless, machinekit tried to fork it to be more modular and easier to run a network GUI, take a peak and learn from their mistakes
06:49 PM chris131313: hey all i am to the point of trying to open linux cnc but keep getting an error on lines 9 and 10 of the hal. loadrt hm2_eth board_ip="192.168.1.121" config="num_encoders=1 num_pwmgens=0 num_stepgens=4"
06:49 PM chris131313: setp hm2_7i76e.0.watchdog.timeout_ns 5000000
06:50 PM XXCoder: hey CaptHindsight hows opensls going
06:50 PM chris131313: i have tried to change it to hm2_5i25 .0 but it doesnt make the error clear
06:51 PM chris131313: im guessing its something in line 9 with the board ip. anyone got any ideas
06:51 PM CaptHindsight: XXCoder: I haven't looked in a while, been tied up with other things
06:51 PM XXCoder: no problem
07:20 PM solarwind: CaptHindsight ok cool thanks. Ideally it should be modular like a microkernel service and communicate via IPC or over the network
07:22 PM CaptHindsight: how many years free do you have?
07:23 PM solarwind: Is it that badly written?
07:23 PM solarwind: Conceptually it's not complicated at all
07:23 PM CaptHindsight: XXCoder: looks like openSLS is coming along and working, not sure when he will open source "everything"
07:24 PM XXCoder: interesting. thanks :)
07:24 PM CaptHindsight: solarwind: it's all the little stuff, features etc
07:25 PM CaptHindsight: but sure starting up a new project wit the broad strokes won't be too much work
07:26 PM CaptHindsight: getting all the drivers in RT for all the hardware will be a full time job
07:26 PM solarwind: Well I don't care about _all_ of it, just the parts that pertain to me
07:27 PM solarwind: If I do a good job of the architecture and make it easy for other people to contribute, well that's open source at work
07:29 PM CaptHindsight: change course and go into Bio, lots of tools needed there for neuro, cancer, etc etc
07:30 PM CaptHindsight: still the wild west
07:30 PM solarwind: Yeah I'm aware of how shitty it is
07:30 PM solarwind: it's also full of shitty business assholes
07:31 PM CaptHindsight: build the infrastructure to work around them
07:31 PM solarwind: I wouldn't know where to start
07:31 PM solarwind: You need connections
07:31 PM CaptHindsight: they appear to be more selfish than intelligent
07:31 PM solarwind: biologists are idiots, they're not like physicists/chemists
07:32 PM CaptHindsight: has to start somewhere
07:32 PM CaptHindsight: you're young and have the time :)
07:35 PM CaptHindsight: kickstart the open source DNA programming movement
07:36 PM solarwind: I don't like humans enough to try and save them
07:36 PM solarwind: That is, I won't go into a space where nobody will listen to me
07:37 PM solarwind: They've already chosen their path and shown us how fucking retarded they are
07:37 PM CaptHindsight: yeah, that space needs to be created
07:37 PM CaptHindsight: it's been well controlled
07:38 PM solarwind: Aaron Swartz tried to do the right thing
07:38 PM solarwind: and he ended up killing himself
07:38 PM solarwind: that space doesn't care about anyone, they just want money
07:40 PM CaptHindsight: hah "Swartz died by suicide on January 11, 2013.[111] After his death, federal prosecutors dropped the charges."
07:40 PM CaptHindsight: after
07:40 PM solarwind: And nobody tried to fight for hinm
07:41 PM solarwind: The prosecutors should be executed
07:41 PM Tom_L: chris131313, what ip is the board set for?
07:43 PM CaptHindsight: solarwind: how about Klipper?
07:43 PM solarwind: I'm _just_ starting to look into Klipper
07:43 PM solarwind: I have an Flsun V400 that I'm about to assemble tonight
07:44 PM solarwind: I don't know anything about Klipper though and the flsun v400 is designed to work with it out of the box
07:47 PM CaptHindsight: maybe someone is already infiltrating the authorities the same way that fascists have the past few decades, only for good vs evil
07:49 PM CaptHindsight: playing the long game
07:52 PM Tom_L: JT-Cave, PR#87 mesact up
08:02 PM solarwind: Hmmmmmm I'm going to build a centrifuge
08:02 PM Tom_L: for chips?
08:02 PM solarwind: for containers about the size of a grease gun cartridge
08:03 PM Tom_L: :)
08:03 PM Tom_L: we had big ones to recycle the coolant from chips
08:03 PM solarwind: Need to get some air out of fluids without having to use a vacuum tube because it would boil off some of the more volatile compounds of the fluid
08:04 PM Tom_L: don't pop the cardboard casing
08:04 PM Tom_L: or are they all plastic now?
08:05 PM roycroft: i finished my internal wood threading device today - all that's left is to make a cutter, which i'll do tomorrow
08:05 PM roycroft: i don't know if i want to wait a week to soak a sample bore in oil before i test the cutter, so i might make a pretty rough go at it tomorrow
08:07 PM XXCoder: nice
08:07 PM XXCoder: cant wait to see results
08:09 PM Tom_L: kid spent the day at work cleaning out the coolant tank on his machine
08:09 PM Tom_L: only 1200 gallons
08:09 PM XXCoder: wow sounds like nicer shop
08:09 PM XXCoder: shop I worked at before dont really clean any coolant
08:10 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/cincinnati/Cincinnati_5.jpg
08:10 PM Tom_L: the yellow grate is the tank
08:11 PM XXCoder: fun
08:11 PM Tom_L: i think he said it was like 10' deep
08:13 PM Tom_L: eventually you gotta clean them or you'll just be pumping chips and clogging the tubes
08:13 PM XXCoder: indeed
08:19 PM roycroft: and when the tubes get clogged homer something something
08:33 PM CaptHindsight: ever work with a shop where "when in doubt, just throw it out" is the philosophy?
08:34 PM Tom_L: his last place had a very high scrap rate
08:34 PM CaptHindsight: can you fix it in under 15 minutes? NO, throw it out.
08:35 PM Tom_L: they did tell him to run the spindle bearings as is until it blew up
08:35 PM CaptHindsight: is it near the door and will it get in the way? Yes, throw it out.
08:35 PM Tom_L: they didn't run them in properly on a 70k spindle
08:36 PM Tom_L: not rpm $$
08:36 PM CaptHindsight: as long as it doesn't give the worker a single more cent pay
08:36 PM Tom_L: 25-30 hp
08:36 PM CaptHindsight: just throw it out
08:37 PM Tom_L: there are plenty of those to go around
08:38 PM CaptHindsight: should I spend 1-4 hours fixing something to save thousands? Not if you can just throw it out.
08:39 PM Tom_L: he spent 2 weeks flattening the work plate on one of those to prevent scrapping parts
08:39 PM roycroft: conspicuous consumption seems to be a right wing tactic for compensating for a small member
08:39 PM Tom_L: probed the low spots and cut to that
08:41 PM solarwind: https://www.youtube.com/watch?v=a_o0v9mPkhU [ External Gear Pump - Cavitation ]
08:41 PM solarwind: I see it but I don't see how it can hurt the metal over time
08:42 PM CaptHindsight: lack of lube in the short term voids
08:42 PM Tom_L: you don't see the 'wear' on the leading edge?
08:43 PM solarwind: OH
08:43 PM Tom_L: on the top gear
08:43 PM Tom_L: more visible there anyway
08:43 PM solarwind: So is that the primary mechanism for wear? No lubricating film at the voids?
08:44 PM Tom_L: or those could be bubbles... not sure
08:44 PM Tom_L: or both
08:44 PM CaptHindsight: my theory from watching 10 seconds of it
08:45 PM CaptHindsight: https://www.youtube.com/watch?v=C3-BKvswQkE slowerer video
08:45 PM Tom_L: watch that on a big screen high sometime :)
08:46 PM XXCoder: love how bubble reacts
09:01 PM CaptHindsight: https://www.youtube.com/watch?v=4q79VH2QGqw the void
09:01 PM CaptHindsight: Cavitation Bubbles Behind a Sphere
09:02 PM roycroft: bubble bubble, toil and trouble
09:02 PM CaptHindsight: out to lunch!
09:03 PM CaptHindsight: faster oil would help
09:05 PM XXCoder: fun video