#linuxcnc Logs

Jul 02 2023

12:48 AM solarwind: I LOVE this kind of "grinding wheel dresser" https://www.kutritemfg.com/images/dressers_zoom.jpg specifically for angle grinder wheels that get clogged with rust and such
12:59 AM solarwind: Definitely not the kind you want to use on bench grinders or surface grinders, but great for coarse wheels
12:59 AM solarwind: Diamond points/bars just smooth out angle grinder wheels too much and just make them worse
12:59 AM solarwind: The crude thing I linked above pummels the wheels and makes them nice and rough again. Huge improvement to material removal speed
01:17 AM roycroft: solarwind: i use that kind of wheel dresser on my coarse wheels
01:19 AM roycroft: i use my 2x72 belt grinder for most of my grinding needs these days, though, and the belts don't need to be dressed
01:20 AM Deejay: moin
02:57 AM pere: <URL: https://openhub.net/p/emc > show an impressive track record of the project.
04:30 AM travis_farmer[m4: Morning ☕️
04:58 AM mrec: does anyone have experience with small round bldc motors? (<20mm dia)?
05:09 AM JT-Cave: morning
05:43 AM JT-Cave: rooster just crowed
05:45 AM * travis_farmer[m4 is debating on if he should try the ADS1115.comp project again, or if he should shell out $99 for a 7i87 when he only needs 3 inputs (so far)...
05:48 AM jpa-: mrec: at that size, inrunners have very high RPM and pretty low torque
05:49 AM jpa-: travis_farmer[m4: do you need them realtime or could it be userspace component?
05:50 AM travis_farmer[m4: one channel is the 0-10V out from the VFD for sensing spindle RPM, the other two are connected to air PSI sensors
05:53 AM jpa-: so probably not very time-critical, assuming the VFD does basic speed regulation by itself
05:54 AM travis_farmer[m4: yes, the VFD does it's own speed regulation
05:55 AM jpa-: in that case you could write a basic driver for whatever ADC board in python, with ability to use USB or any connection like that
05:57 AM travis_farmer[m4: Hmmm...
06:02 AM CloudEvil: 20mm brushless motors may in principle not have much torque, but alone they can accellerate to maximum speed in well under a second.
06:02 AM CloudEvil: Often ridiculously less than a second.
06:14 AM Tom_L: morning
06:31 AM Tom_L: mrec, i have a handfull of these: http://tom-itx.no-ip.biz:81/~webpage/pdf/MC68332/maxonmotor.pdf
06:32 AM Tom_L: with gearheads on the end
06:32 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/maxbot/image001.jpg
06:33 AM Tom_L: Allegro 3971 to drive them
06:34 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/maxbot/image024.jpg
06:36 AM pere: Anyone else able to reproduce <URL: https://github.com/LinuxCNC/linuxcnc/issues/2559 >?
07:44 AM pere: is there any way to get the current tool height compensation value?
08:00 AM * JT-Cave starts his chicken day
08:01 AM jpa-: pere: i don't fully understand the bug report, but might it be just the axis executing the G-code when loading up the preview?
08:03 AM pere: jpa-: no idea what is going on, but I did not expect the lines to be printed or the nofitication to show up.
08:04 AM pere: jpa-: did you test in a simulator? did your program run?
08:05 AM jpa-: no, i did not test - but normally axis runs G-code by itself (in userspace) to get the lines it draws on display; i wouldn't be surprised if output from that ends up in stdout
08:09 AM jpa-: it's at least possible to get the tool offsets using status.tool_offset in Python api
08:25 AM pere: jpa-: python api approach might be an idea. did not occur to me.
08:28 AM pere: jpa-: are you talking about linuxcnc.stat.tool_offset ? Not sure where to find status.tool_offset.
08:29 AM jpa-: yes, that
08:29 AM jpa-: status = linuxcnc.stat(); print(status.tool_offset)
08:37 AM * travis_farmer[m4 is thinking of not re-upping his personal webserver when it is time to renew his domain... the server is useless, and is a waste of electricity... :-(
08:40 AM CloudEvil: How limited is it? I have a google cloud and am using it gree as my usage is so low.
08:40 AM CloudEvil: free
08:41 AM travis_farmer[m4: how limited is what?
08:44 AM Tom_L: google spies on you
08:45 AM JT-Cave: pcw-home, what is the default firmware on the 7i76e?
08:46 AM * JT-Cave should go clean chicken crap
08:46 AM * travis_farmer[m4 is chicken crap...
09:03 AM * JT-Cave smells bacon
09:03 AM pcw-home: 7i76e_7i76x1
09:28 AM skunkworks[m]: JT-Cave: I am taking care of some friends chickens this weekend.. 4 eggs so far lol
09:29 AM Tom_L: mmm breakfast!
09:29 AM Tom_L: our neighbor brings us eggs from their property all the time
09:29 AM Tom_L: they have some rural land
09:32 AM travis_farmer[m4: Hmm, seems to be raining... what a wet miserable day...
09:34 AM JT-Cave: pcw-home, thanks
09:34 AM JT-Cave: skunkworks[m], did you candle them?
09:39 AM * JT-Cave goes to town
09:41 AM skunkworks[m]: JT-Cave: no roosters.
10:22 AM mrec: Tom_L: I think about using one of those motors for the Z axis of my pnp machine, I still need to look around what's available
10:28 AM mrec: https://streamable.com/trdcir feeder electronics connected to the feeder bus
10:31 AM santydofus2014[m is now known as santydofus0[m]
11:24 AM acer: hello yall. just posting, I just got a N3-210-C-U VFD I'm looking to SELL. USA.. If anyone is interested let me know.
11:24 AM acer is now known as _unreal_
11:26 AM _unreal_: Its a 10HP 230v 35amp VFD
11:29 AM Tom_L: mrec, where are you?
11:31 AM mrec: Taiwan
11:33 AM Tom_L: mmm
11:33 AM Tom_L: those are very nice motors i posted
11:34 AM mrec: I will work on a ballscrew Z setup for the pick and place machine mid term, for now I have a belt based system
11:34 AM mrec: X/Y are ballscrews, but I want to improve Z
11:34 AM Tom_L: i forget the gear reduction on them off hand
11:36 AM mrec: the feeder bank is just nice all very low power
11:36 AM _unreal_: Tom_L, I just got a few more BLDC drives
11:37 AM _unreal_: beyond that been endlessly stuck dealing with my father and his health issues.
11:47 AM JT-Cave: skunkworks[m], you candle the eggs to check for blood or other defects
11:56 AM travis_farmer[m4: blood isn't a defect, it is just a chicken at that point ;-)
12:13 PM ZincBoy[CAON][m]: Use G38.3 and G38.5.
12:14 PM ZincBoy[CAON][m]: They are the same thing as G38.2 and G38.4 except you need to handle the errors yourself.
12:14 PM pere: how is it signaling the error? Is the source file nc_files/remap-subroutines/qt_auto_probe_tool.ngc incorrect?
12:16 PM JT-Cave: hmm the a coil is frozen up...
12:17 PM ZincBoy[CAON][m]: No idea on the remap subs but if you want to probe without GUI error messages, you need to use 38.3 and 38.5. That is what I use. You just check #5070 to see if you made contact and handle things appropriately. Most of the probing code around lcnc is poorly tested at best.
12:18 PM pere: ok. reporting a patch to change to G38.3 for the file, to get more eyes to look at it.
12:19 PM mrec: whow...
12:19 PM mrec: LCSC has stopped receiving orders from conflict zones and suspended shipping all product categories to all customers and end-users in Russia, the Donetsk and Luhansk regions of Ukraine, and Belarus.
12:19 PM pere: what about africa and the middle east...
12:19 PM ZincBoy[CAON][m]: Does the code handle the condition where it doesn't make contact correctly? I wouldn't replace G38.2 with G38.3 unless the surrounding code supports it. See my previous comment about poorly tested probing code.
12:20 PM pere: ZincBoy[CAON][m]: yes. See <URL: https://github.com/LinuxCNC/linuxcnc/blob/master/nc_files/remap-subroutines/qt_auto_probe_tool.ngc >.
12:24 PM ZincBoy[CAON][m]: There is no check for contact after the first G38.2 rough probe. You need to check there as well. It will also depend on what the epilog does with the -3 return code. It should halt the program as this would indicate a missing or broken tool.
12:26 PM pere: <URL: https://github.com/LinuxCNC/linuxcnc/pull/2562 > is my first draft to get the discussion going.
12:29 PM Tom_L: JT-Cave, change the filter
12:30 PM Tom_L: i had that happen when there wasn't enough air flow
12:30 PM ZincBoy[CAON][m]: Also that code will have unexpected behavior if you are using multiple work offsets. It doesn't back out the offset values so the effective tool length will change depending on the work offset z value.
12:30 PM Tom_L: JT-Cave, froze the copper outside as well
12:30 PM Tom_L: turn the heat on.. a sure way to defrost it :)
12:31 PM Tom_L: i used a torch
12:31 PM pere: ZincBoy[CAON][m]: that sound exactly like the problem I am having with my setup
12:31 PM ZincBoy[CAON][m]: Refrigerant is probably low.
12:32 PM Tom_L: could be but i'd do the filter first
12:32 PM Tom_L: which reminds me, on the 5th i need to pick up another box of filters
12:33 PM JT-Cave: I changed the filters and have the fan only running and it's melting
12:33 PM Tom_L: were the filters plugged?
12:33 PM JT-Cave: very
12:33 PM JT-Cave: they were a year old lol
12:33 PM Tom_L: maybe all you need to do then aside from defrost
12:33 PM Tom_L: mine was a nice block of ice
12:34 PM JT-Cave: already changed the filters just waiting for the ice to melt
12:34 PM * travis_farmer[m4 has discovered how unimportant his existence really is...
12:34 PM JT-Cave: I'm starting to get a little air flow so I know it's melting
12:35 PM Tom_L: yard mowed & watered, showered, eating lunch then a nap
12:40 PM ZincBoy[CAON][m]: @pere see https://pastebin.com/qABSrGzt for the relevant section of my code. It backs out the workoffset and the current tool offsets to calculate the tool length. The measured tool length is independent of any offsets so works in any conditions.
12:43 PM pere: ZincBoy[CAON][m]: thanks. will take some time to wrap my head around it, too many variables and settings I am not familiar with.
12:44 PM ZincBoy[CAON][m]: Most of them are specific to my code so it is really just an example.
12:46 PM pere: of course, but that only make it harder to understand for me, not easier. :)
12:46 PM pere: and I need to look up every parameter number as I do not know these by heart yet. :)
12:47 PM JT-Cave: fan blowing better now... give it a bit before turning on the ac again
12:47 PM Tom_L: learning to speak gcode is a prerequisite of this channel ya know...
12:47 PM JT-Cave: lol
12:48 PM Tom_L: did you look at the A coil?
12:48 PM Tom_L: mine was pretty thick on the underside near the top
12:48 PM JT-Cave: no, it's not accessible at all
12:48 PM pere: Tom_L: I guessed as much, but it take a while to get there. :)
01:05 PM JT-Cave: pcw-home, is there a replacement for the 7i76E in the works?
01:14 PM -!- #linuxcnc mode set to +v by ChanServ
01:36 PM pere: ZincBoy[CAON][m]: With <URL: https://github.com/SebKuzminsky/MazakVQC1540/commit/97070f745a1df2b46018be16761ea099aa2684ee > in place, the tool height seem to be limping along.
01:38 PM * skunkworks[m] uploaded an image: (881KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/hKSxWPdhxCfbnqbsPszuswfi/PXL_20230702_183640537.jpg >
01:40 PM skunkworks[m]: newest online DJ
01:40 PM skunkworks[m]: (we are not online)
03:19 PM JT-Cave: I like the Dennis the Menace look
03:29 PM skunkworks[m]: Lol - yes.. He needs a hair cut...
03:29 PM skunkworks[m]: and still in his favorite jammies..
03:32 PM XXCoder: dennis the menace story is funny. apparently there was also russian dennis the menace. its about black haired boy that does bad stuff.
03:32 PM XXCoder: both versions authors never knew about each other for years
03:33 PM Tom_L: skunkworks[m], you should get some sound proofing for that
03:33 PM XXCoder: https://static.wikia.nocookie.net/beano/images/9/99/Dennis_of_the_21st_Century.png/revision/latest?cb=20230224215822
03:33 PM XXCoder: russian one
03:34 PM XXCoder: https://www.plagiarismtoday.com/2010/10/18/the-odd-case-of-dennis-the-menace/
03:34 PM XXCoder: actually uk I misremembered the country
03:35 PM skunkworks[m]: This is just a test setup - the actual users have to figure out how they want to setup the streaming software..
03:36 PM skunkworks[m]: * Radio broadcast automation software
03:36 PM skunkworks[m]: so many terms I am not used too...
03:36 PM skunkworks[m]: You think linuxcnc is pretty small...
04:38 PM pcw-home: JT-Cave: Yes 7I76EU (replaces 7I76E and 7I76ED)
04:42 PM JT-Cave: cool, are they getting close to being in stock?
04:44 PM pcw-home: still going to be a while, first are 7I95T 7I94T 7I80HDT 7I97T
04:54 PM JT-Cave: thanks what's a 7i94T?
04:55 PM JT-Cave: I found the 7i94
04:56 PM JT-Cave: looks like an ethernet card for mostly sserial?
05:01 PM pcw-home: Yes
05:43 PM JT-Cave: what's the next board coming up?
09:58 PM -!- #linuxcnc mode set to +v by ChanServ
10:42 PM Tom_L: http://www.xylotex.com/XS3525V202.pdf
10:42 PM Tom_L: anybody ever use one of these with linuxcnc?