#linuxcnc Logs

Jul 22 2022

#linuxcnc Calendar

01:57 AM -!- #linuxcnc mode set to +v by ChanServ
02:58 AM pere: anyone know how to test tolerance vs. speed automatically, ref <URL: https://github.com/LinuxCNC/linuxcnc/pull/1853 >?
03:11 AM jpa-: pere: one way to test it would be to run e.g. G1 X0 Y0 G1 X100 Y100 G1 X200 Y0 and have HAL minmax component record the maximum Y coordinate reached
03:11 AM pere: jpa-: any idea how to script it?
03:13 AM pere: I see from tests/motion/g0/ a way to run the G1 blocks, but have no idea how to measure the time it take to complete it.
03:13 AM jpa-: at its simplest, it could be just a shell script with halcmd setting up the components & connections (i think motmod & minmax would be enough?) and then mdi command to send the G-code
03:16 AM jpa-: for time measurement you could use HAL time component and connect it to is-running or in position signal
03:17 AM jpa-: hmm, except time only measures seconds
03:17 AM jpa-: maybe some kind of counter could be used for measuring servo thread cycles
03:19 AM jpa-: tests/motion/g0 samples the whole path to a file, which is reasonable also; then you can post process it in any programming language
03:21 AM pere: halui.program.is-running seem to be set when a command is running, but sadly not when homing.
03:30 AM jpa-: motion.in-position might show that
04:00 AM pere: jpa-: look like it could work. Got rid of race condition in <URL: https://github.com/LinuxCNC/linuxcnc/pull/1859/files > as far as I can tell.
04:16 AM JT-Cave: morning
04:51 AM Tom_L: morning
04:51 AM Tom_L: jt did you get the new deb? i posted it yesterday
05:08 AM -!- #linuxcnc mode set to +v by ChanServ
05:15 AM -!- #linuxcnc mode set to +v by ChanServ
05:25 AM -!- #linuxcnc mode set to +v by ChanServ
05:33 AM JT-Cave: HexChat is setup on the new PC... one down a billion to go
07:40 AM Tom_L: getting a new pc is like taking out the trash. nice time to clean house
07:44 AM jpa-: i just move the trash to the new pc.. and when i bought a desktop PC alongside my laptop, i copied the same crap to both :)
07:44 AM jpa-: ls ~ | wc -l 966
08:50 AM tcurdt: pcw---home what the max mA the 7i92 breakout boards should use of the 5V on the IDC connector? I cannot seem to find that in the manual.
08:55 AM pcw---home: 1A is probably a safe limit
08:55 AM tcurdt: thx!
09:01 AM tcurdt: pcw---home and about the 74ACT541 buffers: those are directly connected to the FPGA, right? or should there be a resistor between the FPGA output (IOxx) and the buffer input?
09:24 AM JT-Cave: crap I forgot to copy the armhf deb this morning...
09:27 AM JT-Cave: I think I'm going to try Debian 11.4 then change to unstable maybe
09:29 AM pcw---home: tcurdt: the buffers can be driven directly
09:29 AM tcurdt: thx!
09:32 AM CaptHindsight[m]: JT-Cave: the big difference in the latency test results on the AMD Ryzen systems is using isolcpus or not with preempt_rt
09:33 AM JT-Cave: I don't understand your question if it is a question
09:33 AM CaptHindsight[m]: try your current kernel with isolcpus=1,2,3,4,5,6,7,8,9,10.11
09:34 AM CaptHindsight[m]: https://forum.linuxcnc.org/18-computer/46292-current-motherboards?start=40#247993
09:35 AM CaptHindsight[m]: the latency histogram on top is without the use of isolcpus in the kernel parameters
09:35 AM CaptHindsight[m]: the one below is with isolcpus being used
09:36 AM CaptHindsight[m]: so if you were to use your system to test or run software stepping use isolcpus
09:37 AM CaptHindsight[m]: if you are using just a servo thread or day to day use, browsing or compiling skip using isolcpus
09:39 AM JT-Cave: ok
09:39 AM JT-Cave: I'm going to try Debian 11.4 to see if it's more friendly
09:40 AM CaptHindsight[m]: it's been stable for me
09:40 AM JT-Cave: good to know
09:41 AM CaptHindsight[m]: there is a backport from unstable to deb11 for 5.18.0-2-rt-amd64
09:41 AM JT-Cave: I am more used to Debian than LinuxMint any more but I used to use LinuxMint
09:42 AM JT-Cave: if Debian 11.4 knows about my mb nic I'll just use it
09:43 AM JT-Cave: bye
09:43 AM CaptHindsight[m]: JT-Cave: I haven't installed from 11.4 but there were some changes to how the docs get built so po4a was backported to deb11
09:45 AM -!- #linuxcnc mode set to +v by ChanServ
09:46 AM gywilo[m]: Okay, round two; I think I'm going to try this approach. If I understand correctly, does it mean install debian from debian iso? then install synaptic package manager, preempt-rt, linuxCNC from cmd line? Would it be best to start with debian 11.4.0 iso-BD? 'debian-edu-11.4.0-amd64-BD-1.iso'
09:46 AM * gywilo[m] uploaded an image: (30KiB) < https://libera.ems.host/_matrix/media/r0/download/jauriarts.org/qraYBiCmdnFIYtsCoCxNQcHf/unknown.png >
09:47 AM gywilo[m]: I didn't get Tom_L's full response there in the reply. was:... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/dbcb34ab4f5f595e965e598998cf4d4335847cff)
09:48 AM JT-Cave: I just fired up hexchat on the old pc for now
09:51 AM CaptHindsight[m]: gywilo: install debian from debian iso ....
09:51 AM CaptHindsight[m]: synaptic package manager will already be installed with it
09:53 AM CaptHindsight[m]: as far as which Deb 11.4 iso to use I'd say the one with the free and non-free already included
09:54 AM CaptHindsight[m]: sorry I don't have a link since I haven't tested a Deb 11.4 ISO yet
09:54 AM gywilo[m]: that's okay, I think I've found it here
09:55 AM JT-Cave: CaptHindsight[m], Debian 11.4 does recognize my NIC so great
09:55 AM CaptHindsight[m]: \0/
09:57 AM * JT-Cave heads to work now
09:59 AM CaptHindsight[m]: https://repository.qtpyvcp.com/repo/linuxcnc_debs/
10:01 AM CaptHindsight[m]: gywilo: getting your graphics up is the main hurdle, after that the LCNC install gets easier
10:04 AM gywilo[m]: and graphics drivers are part of the 'non-free' packages?
10:10 AM CaptHindsight[m]: yes, where they are
10:13 AM CaptHindsight[m]: I never use Intel so I can't help you there
10:14 AM CaptHindsight[m]: well i have once in the past 20 years
10:39 AM roycroft: iirc my biggest hurdle when i built the lcnc machine for my mill was getting the driver for the wifi card to work
10:40 AM roycroft: the second biggest hurdle was installing the wifi card - it's a tiny chip and i needed something like an m2.5 screw to hold it in place
11:02 AM roycroft: eek
11:03 AM roycroft: 38 on thursday next
11:03 AM roycroft: and i have meetings after work that day, so i'll need to be out and about
11:25 AM GTKplusplus[m]: Still messing with probe screen, just can't get it to qork
11:25 AM GTKplusplus[m]: s/qork/work/
11:25 AM GTKplusplus[m]: Removed it, just in case
11:26 AM * GTKplusplus[m] uploaded an image: (215KiB) < https://libera.ems.host/_matrix/media/r0/download/jauriarts.org/ngjpRdUtqkTcTlibBxNRewIG/IMG_20220722_182325_042.jpg >
11:26 AM GTKplusplus[m]: This is what happens now with my old tool change macro
11:26 AM GTKplusplus[m]: Note. This always worked
12:05 PM * roycroft is squirting out boxes like crazy, and is happy to see forward progress on that project
12:12 PM Tom_L: wonder what you ate to be squirting boxes...
12:14 PM roycroft: the good filament
12:15 PM roycroft: the crappy filament was getting me all stuffed up
12:16 PM Tom_L: hardly ever pays to buy on the cheap
12:16 PM roycroft: yup
12:16 PM roycroft: cheap stuff is usually too expensive
12:17 PM roycroft: the "cheap" filament was actually costing me over $100/kg for the filament alone, not to mention the time and electricity i was wasting
12:18 PM roycroft: and if i can't find a place to recycle what i have left, i'll have to add in landfill costs as well
12:18 PM Tom_L: bury it in the back yard
12:19 PM Tom_L: use it in your weed eater
12:19 PM roycroft: that creates its own issues
12:19 PM roycroft: i think it's way too brittle for that
12:19 PM roycroft: and a bit thin
12:20 PM roycroft: the weedeater uses 2.5mm nylon line
12:20 PM GTKplusplus[m]: Tried changing axis_2 in the macro to axis_z. Now it doesn't error out anymore (yay). But instead it just stays stuck when it should show a popup to confirm the manual tool change
12:20 PM roycroft: this stuff is 1.75mm
12:20 PM roycroft: ish
12:20 PM GTKplusplus[m]: Again working and tested macro, and I stopped using probe screen completely
12:20 PM Tom_L: yeah
12:20 PM roycroft: with a big emphasis on the "ish"
12:20 PM roycroft: and dammit
12:21 PM Tom_L: they quality control checked it with some micrometers the neighbor welder used as a clamp
12:21 PM roycroft: i was just out in the shop and forgot *again* to measure my stepper motor shaft diameter
12:21 PM Tom_L: .250"
12:21 PM Tom_L: 6mm
12:21 PM roycroft: i'm ready to start working on a rack and pinion design for my router and need to know that
12:21 PM roycroft: yes, one of those, probably
12:21 PM roycroft: or possibly 3/8", but that's mostly found on nema34 motors
12:22 PM Tom_L: did i happen to mention something about a nut driven ballscrew?
12:22 PM * roycroft makes a special trip to measure, because he's certain to forget again next time he goes out there for something else
12:22 PM roycroft: several times :)
12:22 PM Tom_L: i was kinda hoping somebody would try that
12:25 PM roycroft: it makes me wonder how many shares of spinning-balls-r-us, inc. you own
12:25 PM roycroft: and the answer to the motor shaft question is: none of the above
12:25 PM roycroft: it's 10mm
12:25 PM Tom_L: big one
12:26 PM roycroft: i'm glad i checked before i went timing pulley shopping
12:26 PM Tom_L: my last steppers i got were 3/8
12:26 PM roycroft: yeah - bigger than i expected
12:26 PM Tom_L: but stepped up from 1/4"
12:26 PM Tom_L: you could order them either way
12:26 PM roycroft: these are 525oz-in nema 23 motors
12:26 PM roycroft: bigger is ok
12:26 PM Tom_L: that was my thinking
12:26 PM roycroft: if i only find pulleys with smaller bores i can make the bore bigger
12:27 PM roycroft: making it smaller is messy
12:27 PM Tom_L: sdpsi has pulleys and i generally have to bore at least one
12:27 PM roycroft: involving a bushing, which is generally disagreeable to me
12:27 PM roycroft: they are one of the vendors i'll be checking with
12:27 PM Tom_L: i did a bushing on my encoder but there's no stress on it
12:28 PM roycroft: it seems that a 3:1 reduction is pretty common on systems that use rack and pinion to move the gantry
12:28 PM roycroft: so i'll want the motor pulley to be pretty small
12:28 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/Mill_Steel/Spindle/Pulley1.jpg
12:28 PM roycroft: because i don't want a ginormous one driving the pinion
12:28 PM Tom_L: brass bushing
12:29 PM Tom_L: i had to cut the bottom out of the pulley to get the clearance i needed for the encoder pulley
12:29 PM roycroft: i'm thinking i'd like flange pulleys on both the pinion and motor, to help keep any dust out
12:29 PM roycroft: but if i can't find flanged ones, i'll be ok
12:30 PM Tom_L: you really only need it on one
12:30 PM Tom_L: well, on a belt system anyway
12:30 PM roycroft: i'm not sure how much dust protection it would give me anyway
12:31 PM Tom_L: the dust will smash in there and keep things tight :)
12:31 PM roycroft: if the motor mount is machined properly and the motor shaft and pinion shaft are aligned properly, i shouldn't need flanges at all
12:31 PM roycroft: and if the belt is made well
12:31 PM Tom_L: you get something in there and it can move
12:32 PM roycroft: well first i have to go shopping and see what's available
12:32 PM Tom_L: mine pretty well rides in the middle anyway though
12:32 PM roycroft: the belt on my mill rides in the middle
12:33 PM roycroft: well, i guess it does ride a bit low
12:34 PM Tom_L: wow my router is arriving 2 days early
12:34 PM Tom_L: go amazon!
12:34 PM roycroft: https://roycroft.us/Mill-Motor-Mount/Mill-Motor-Mount-3.jpeg
12:34 PM roycroft: i should be able to adjust that though
12:34 PM Tom_L: oh i remember that
12:35 PM roycroft: it probably rides low because the motor mount plate dips down a bit
12:35 PM Tom_L: you had flex in the motor mount and wanted flanges added
12:35 PM roycroft: i haven't gotten around to bracing it yet
12:35 PM roycroft: yes
12:35 PM Tom_L: there is a round tuit for that
12:35 PM roycroft: i haven't used the mill very much since i installed the new motor
12:35 PM roycroft: i think the new hour meter shows like 3 hours
12:36 PM roycroft: i do need to work on that mill some more after my router build is done
12:36 PM roycroft: i should work on it before that, since i'll be using it to make parts for the router
12:36 PM roycroft: unless i end up hiring jt to make all my parts, and not just some of the bigger ones
12:37 PM Tom_L: that would be the easy way out
12:37 PM roycroft: which depends on my cash flow and jt's available time
12:43 PM Tom_L: hope port fwding is easy on the new router
01:16 PM roycroft: it seems that 10mm is an unusual shaft size
01:16 PM roycroft: 9mm is fairly common, as is 10.3mm and 10.4mm
01:16 PM roycroft: but this is why we have boring bars and reamers
01:26 PM TheMightyFozz: Hi folks. I'm picking Drag Chain for my router. i need about 4m(12ft) of it. the machine is running 4 Nema34 closed loop motors, A spindle with water-cooling, led lighting under the gantry and z axis, and home switches on both ends of all axis. also an extra connector for future use near the spindle. What size is recommended for such a setup? I'm guessing 1"x2" roughly.
01:30 PM roycroft: you will probably need multiple sizes
01:30 PM roycroft: one to carry the z axis components
01:30 PM roycroft: one to carry the x+z axis
01:30 PM roycroft: and one to carry the x+y+z
01:30 PM ZincBoy[CAON][m]: Two factors, size must fit the cables and must also meet the bending radius requirements.
01:31 PM ZincBoy[CAON][m]: Go with a larger radius whenever possible as it will extend the life of the cables/water lines.
01:32 PM roycroft: i may not use drag chain on my z axis
01:32 PM roycroft: i'm still getting that sorted
01:33 PM ZincBoy[CAON][m]: Probably not needed on the z with the limited travel in a router setup. I didn't bother on mine. Just on x and y.
01:33 PM TheMightyFozz: y axis is about 150mm 6". its huge
01:33 PM roycroft: my y axis will have 2000mm of travel or so
01:33 PM roycroft: x will be 700mm or so
01:33 PM TheMightyFozz: same as mine
01:33 PM roycroft: but z will be 150mm
01:34 PM TheMightyFozz: are you printing youchain? or have you had enough of printing for a while
01:34 PM roycroft: i purchased some
01:34 PM TheMightyFozz: *your chain
01:35 PM Tom_L: probably cheaper just to buy that
01:35 PM roycroft: first of all, my printer will be busy for quite a while with the boxes
01:35 PM TheMightyFozz: yes.
01:35 PM roycroft: second, injected will probably hold up better than extruded
01:35 PM TheMightyFozz: i cant wait to see them when they are done. its going to look ace
01:35 PM roycroft: third, what tom_l said
01:35 PM ZincBoy[CAON][m]: Just be aware that drag chain is not the best option unless you have motion rated cable. Suspending from the ceiling is better as it minimizes bending.
01:35 PM Tom_L: i _think_ i went with 10 x 15 on the x y axis
01:36 PM roycroft: suspending from the ceiling is not an option for me
01:36 PM TheMightyFozz: nor me. the cables will get tangled. the ceilings too low
01:37 PM TheMightyFozz: the z axis just misses the rafters.
01:37 PM Tom_L: i stuck mine on the wall
01:38 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/Mill_Steel/Wiring/X_Axis_Dragchain.jpg
01:38 PM roycroft: you'll have to wait a while for that, themightyfozz
01:39 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/Mill_Steel/Wiring/Y_Dragchain_Bottom.jpg
01:39 PM roycroft: i'll be squirting out boxes for at least another month just to get the stuff from the old hardware bins done
01:39 PM roycroft: and then i have boxes of fasteners that have never been stored/organized properly
01:40 PM roycroft: i got some of this to experiment with - it's 15x30:
01:40 PM roycroft: https://www.amazon.com/gp/product/B00AUB24VC/
01:40 PM Tom_L: i still do
01:40 PM Tom_L: yeah i used something like that on Z. forget the exact size
01:40 PM roycroft: there's no way i could print them for that price
01:41 PM roycroft: and it would take a long time if i did decide to print them myself
01:41 PM ZincBoy[CAON][m]: Just try to maximize the bending radius whenever possible. Motion rated cable will get you tighter bends but you don't want to have intermittent broken conductors a few years down the line.
01:41 PM roycroft: at the very least make sure all your cables have stranded wire, not solid
01:41 PM roycroft: the motion rated stuff will have a larger number of smaller strands
01:42 PM Tom_L: the motor cables were none too long
01:42 PM roycroft: kind of like the super-flex welding cables
01:42 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/Mill_Steel/Spindle/Drag_Chain1.jpg
01:42 PM ZincBoy[CAON][m]: lol! for sure. Solid wire is right out.
01:42 PM TheMightyFozz: I know that. using full copper for things that dont move and braided for stuff that does.
01:43 PM ZincBoy[CAON][m]: Solid copper wire has no place in a CNC machine. There is too much vibration for it to be reliable.
01:43 PM Tom_L: how big is your mill/router?
01:44 PM TheMightyFozz: it will cut an 8x4 sheet
01:44 PM TheMightyFozz: with 250mm of travel in the z axis for engraving tables etc
01:45 PM Tom_L: nice
01:45 PM TheMightyFozz: Zinc. i'm sorry should clarify. the solid wire is being used in the control box. the box is seperate to the machine
01:57 PM TheMightyFozz: wow 200 quid. almost worth doing a insane 3d print run.
02:02 PM gywilo[m]: Are you printing from a set? I was considering gridfinity.
02:02 PM gywilo[m]: https://hackaday.com/2022/04/18/gridfinity-3d-printed-super-quick-tool-storage-and-retrieval/
02:04 PM gywilo[m]: debian 11.4 successfully installed!
02:04 PM * gywilo[m] uploaded an image: (3601KiB) < https://libera.ems.host/_matrix/media/r0/download/jauriarts.org/BRxjIgOopFohIJZMRekjAFrr/IMG_20220722_115937.jpg >
02:08 PM gywilo[m]: but I'm not able to change resolution from 800x600 to anything else...?
02:25 PM unterhausen: gridfinity looks interesting, but I'm not sure about the cad situation now that I read that hackaday article
02:25 PM gywilo[m]: I tried different, lower resolution monitor with same result, only option is for 800x600
02:31 PM gywilo[m]: it looks like native f360 and STEP files are available
02:31 PM gywilo[m]: https://myhub.autodesk360.com/ue2b114c8/g/shares/SH35dfcQT936092f0e43a9d5084dee87f7ba
03:31 PM roycroft: i considered a grid system for the bottoms of my fastener drawers, but in the end decided that for what i'm doing it's not really necessary
03:31 PM roycroft: i've dimensioned my drawers so that the boxes fit fairly snugly, and i don't usually pull a large number of fasteners out at any time
03:31 PM roycroft: nor do i tend to slam the drawers shut
03:32 PM XXCoder: well you could add it in anytime anyway
03:32 PM XXCoder: just make grid smaller than gap between box to box
03:32 PM roycroft: not really, as i did not accomodate that in the box design
03:32 PM XXCoder: you did make boxes tapered
03:32 PM roycroft: yes, but the draft is only a couple degrees
03:33 PM XXCoder: thats still nough to make 1 mm or so
03:33 PM roycroft: i'd have to measure, but there's really not a lot of room between the boxes
03:33 PM roycroft: and as i said, i just don't think it's necessary
03:33 PM XXCoder: yeah, just saying that its always possible since you added raft :D
03:33 PM XXCoder: draft
03:33 PM roycroft: the only time it would help would be when i have drawers that are not completely full, and i've already designed some rods to hold the boxes in place in partially-filled drawers
03:34 PM roycroft: as long as the entire row is full i'll be fine
03:34 PM XXCoder: yeah
03:34 PM roycroft: and i can always pad a partial row with empty boxes
03:34 PM roycroft: which will be waiting to be filled anyway :)
03:34 PM XXCoder: or make special wide boxes for "random stuff" or tooling
03:34 PM roycroft: i'm actually going to reprint some of my boxes
03:35 PM roycroft: the base size is ~2"x2" square
03:35 PM roycroft: actually 1.925"x1.925"
03:35 PM roycroft: i have been printing 2x2, 2x3, 2x4, 2x6, and 4x4 boxes
03:36 PM roycroft: the 2x3 boxes are a pain to deal with because i need to have even numbers of them in any row where i use them
03:36 PM roycroft: so i've decided to eliminate that size and put the stuff that went in that size in 2x4 boxes
03:36 PM roycroft: and the 4x4 may end up being a pain in the butt as far as organizing goes
03:36 PM roycroft: i may end up designing a 2x8 box to use instead of 4x4
03:37 PM roycroft: which would have to print at an angle and would still almost overflow my build plate
03:37 PM roycroft: but i think i can print that size
03:37 PM XXCoder: ok
03:38 PM roycroft: this is one of those cases where something looks good on paper but complications turn up when it's actually put in use
03:38 PM XXCoder: yeah
03:38 PM roycroft: there will be no reprinting/replacing boxes until the old stuff is off the wall, though - that is my primary goal right now
03:39 PM * roycroft thinks he'll get his quota of 20 printed again today
03:41 PM roycroft: that gridfinity guy is kind of nuts, rather annoying, and it takes a lot of work to watch him
03:41 PM roycroft: but i did enjoy his gridfinity video a while back in spite of all of that
03:41 PM roycroft: there are some good ideas there
03:41 PM XXCoder: got link?
03:42 PM roycroft: https://www.youtube.com/watch?v=ra_9zU-mnl8
03:42 PM roycroft: it's what folks were discussing above
03:42 PM roycroft: you are a lucky guy
03:42 PM XXCoder: ah yes
03:42 PM roycroft: you cannot hear
03:43 PM XXCoder: yeah hes annoying lol
03:43 PM roycroft: you only have to endure his words
03:43 PM XXCoder: I like videos but couldnt sub'
03:43 PM roycroft: not his sounds as well
03:43 PM XXCoder: his subtitles shirt is fun video
03:43 PM gywilo[m]: this guy, Alexandre Chappel, has a similiar thing going, is a bit easier to watch
03:43 PM gywilo[m]: https://www.youtube.com/watch?v=OsLc76k4KeM
03:44 PM gywilo[m]: https://www.youtube.com/watch?v=OthwoWeLb_U
03:44 PM gywilo[m]: https://www.youtube.com/watch?v=adi_tjCwbUw
03:44 PM roycroft: yes, i've seen alexandre chappel's videos too
03:44 PM roycroft: i spent a lot of time thinking about storage boxes before i came up with what i'm using
03:45 PM XXCoder: first of gywilo links have grid similiar to what i thought of for yours roy.
03:45 PM XXCoder: nothing complex, and no changes to boxes lol
03:48 PM roycroft: my boxes are very similar to what the gridfinity guy's, with the label-holding tab and the big curved bottoms
03:48 PM roycroft: but i was printing mine last year, and the gridfinity just came out
03:48 PM roycroft: i think the design is fairly obvious
03:50 PM roycroft: speaking of printing, it's time to pull some boxes off the build plate and queue up another job
04:25 PM _unreal_: hello
04:25 PM * roycroft just finished his work tasks for the week, but has another half hour to go before the weekend
04:27 PM _unreal_: ok so I got a big black thing of starboard
04:28 PM _unreal_: will make a perfect BACK to the pendent when I design and build it
04:36 PM Tom_L: i like when you track an amazon package and it gets close it tells how many stops away
04:38 PM roycroft: but they won't tell you which law enforcement agencies they share your porch cam pictures with
04:38 PM Tom_L: i'm ok with that anyway
04:38 PM Tom_L: they use my cameras all the time anyway
04:40 PM roycroft: steve bannon has been found guilty
04:40 PM roycroft: that was a really fast trial
05:02 PM roycroft: it's the weekend!
05:19 PM * roycroft heads out to finish his bandsaw outfeed table, and perhaps start slicing long veneers
05:29 PM _unreal_: t4nk_fn, sup
05:29 PM _unreal_: just got my ESP32 D1 R32 boards
05:30 PM _unreal_: https://github.com/grblHAL/ESP32 only just got home a little while ago got other stuff to do but over this weekend I'm hoping to get that version of GRBL complied https://github.com/grblHAL/ESP32/blob/master/main/espduino-32_wemos_d1_r32_uno_map.h
05:31 PM _unreal_: preping to install ubuntu on a spare laptop
05:31 PM _unreal_: not fast or current but it will do the job.
05:52 PM ZincBoy[CAON][m]: Because they are not taps. They are endmills that need to cut in a helical path. They need to spin fast to do an M2 thread.
05:53 PM mrec: that motor can go relatively fast, I can still engage the bldc motor ontop
05:53 PM ZincBoy[CAON][m]: You can tap with your setup but even there I like to run ~1200rpm for an M2 forming tap. That is still a bit slow.
05:54 PM Tom_L: what material?
05:54 PM mrec: this is just so I can keep track of the position
05:54 PM mrec: aluminum
05:54 PM Tom_L: single point or form tap?
05:54 PM Tom_L: i lost connection so i missed...
05:54 PM mrec: I'm still working on it
05:55 PM ZincBoy[CAON][m]: I wouldn't connect the stepper while running the spindle on the BLDC. The back emf would be very high and you might get winding breakdown.
05:55 PM mrec: today a spring loaded tool will arrive, so I can put the tapping tool into it. Unfortunately I only have taps here for hand tapping (I'll get other ones next week)
05:56 PM mrec: of course the motor is removed when the spindle is used
05:56 PM ZincBoy[CAON][m]: The thread mills work well with the BLDC and do not require spindle synchronized motion. You use them by programming a helical cutting path.
05:57 PM mrec: I would have to replace the BLDC driver for that
05:57 PM ZincBoy[CAON][m]: They should be rigidly mounted.
05:57 PM ZincBoy[CAON][m]: Why?
05:57 PM mrec: I can only set up the speed via linuxcnc
05:58 PM ZincBoy[CAON][m]: Thread milling works with a dumb spindle. It doesn't need any sync.
05:58 PM mrec: I have only interfaced the potentiometer
05:58 PM ZincBoy[CAON][m]: Tapping required spindle sync.
05:58 PM mrec: and the lowest rpm is like 150
05:58 PM mrec: well ... we're talking about m2 here
05:58 PM ZincBoy[CAON][m]: For an M2 thread mill you want >10k rpm.
05:59 PM mrec: interesting they really have thread milling cutters for m2 and even lower
05:59 PM ZincBoy[CAON][m]: Yes, but you need a very precise machine with very little backlash and a very high speed spindle.
06:00 PM mrec: mine isn't that precise for sure I guess a D-nut would do but I don't have space for that on the Y axis
06:01 PM mrec: I think mine is at around <=0.1mm
06:01 PM mrec: (accuracy)
06:01 PM ZincBoy[CAON][m]: Then small thread milling is not for you. Feel free to try but you will just break them.
06:02 PM mrec: it's too slow for sure
06:02 PM mrec: I have hundreds of threads to mill
06:02 PM ZincBoy[CAON][m]: Roll form tapping works great and is not as sensitive to machine mechanics. You just need spindle synchronization.
06:03 PM mrec: that's the plan with the stepper
06:04 PM ZincBoy[CAON][m]: I have done 1000s of M2 and smaller roll form threaded holes on my mill with no issues. The main factor in getting it to work is the hole size. You have very little margin for error and I suggest gauging the holes before tapping.
06:05 PM ZincBoy[CAON][m]: With spindle sync motion, you don't need the spring loaded holders either. It doesn't hurt to use them but they are not really needed.
06:05 PM mrec: I will try the cheap taps first anyway (I noticed if I make them glow on the gas oven and cool them down they'll last much longer... in fact I never broke on of those treated ones yet)
06:05 PM mrec: butI can easily break the non treated taps
06:06 PM mrec: it should all be better now since the mill is not shaking
06:10 PM ZincBoy[CAON][m]: As a potential future upgrade, I had a mill very similar to yours in the past. I replaced the spindle drive with a Mesa 8i20 driver board. Combined with a spindle encoder I was able to do rigid tapping and had much more torque than the factory drive.
06:10 PM mrec: I have a JGL-100 on the shelf
06:10 PM mrec: 2500W
06:11 PM mrec: including spindle encoder
06:11 PM mrec: I will put it on the mini mill when I have some more time :-)
06:11 PM mrec: I guess it will need some counter weight too
06:12 PM ZincBoy[CAON][m]: That will be a good upgrade for sure!
06:13 PM ZincBoy[CAON][m]: I doubt it. If you are pulling out the original motor/spindle the net difference shouldn't be too much.
06:13 PM mrec: the spindle has >10kg
06:13 PM mrec: 15kg if I remember
06:18 PM _unreal_: Interesting. so I just started moving shit out of the way so I had a place to put that low end laptop
06:19 PM _unreal_: that I'm going to load ubuntu onto
06:19 PM _unreal_: and....
06:19 PM _unreal_: my mini laser etcher that has been laying on its side for prob a year I started to move it. and said.......... I should put this together. long story short the original laser burned out....
06:20 PM _unreal_: in no time as well
06:20 PM _unreal_: so I ditched the POS controller it came with. NEJE 36x36mm laser etcher
06:20 PM _unreal_: tiny
06:21 PM _unreal_: and got some nifty little arduino nano board that holds two motor controllers. I also had to do some funky wiring etc....
06:21 PM _unreal_: now I have a few wires I dont know what they went to... and need to sort it out again.
06:21 PM _unreal_: I just "assembled it" in like 15min... so I'm kind of mad at my self for putting it off for so long. I was dreading the task....?
06:22 PM _unreal_: once I sort the ... 3 wires
06:22 PM _unreal_: I can put it back together
06:22 PM _unreal_: like stupid level of ? WHY didnt I
06:22 PM _unreal_: in anycase once I get it back together
06:23 PM _unreal_: and check the controller tuning.. I'm upgrading the laser from a half watt to a 3 watt if mem serves
06:24 PM _unreal_: way over powered for what it will be used for
06:24 PM _unreal_: but at least I can burn tiny logos into wood coasters
06:28 PM _unreal_: so much for installing linux. I have to re-install windows 10 on it first I forgot its still got its original OS loaded.
06:36 PM roycroft: so the bandsaw outfeed table is functional
06:37 PM roycroft: i think i'm not going to slice any veneers today, though
06:37 PM roycroft: i'll wait until tomorrow, so i can glue them to the substrate right after slicing
06:37 PM * roycroft hopes he can get consistent, wafa-thin veneers
06:40 PM XXCoder: nice
06:42 PM roycroft: yes, finishing up that project will be really nice
06:43 PM roycroft: and until now i was just not set up to do large veneering like i need to
06:43 PM roycroft: i have the vacuum bag, but i did not have the rest of the stuff i needed for taht
07:13 PM -!- #linuxcnc mode set to +v by ChanServ
07:18 PM -!- #linuxcnc mode set to +v by ChanServ
07:29 PM t4nk_fn: yeah, howdy master _unreal_
07:34 PM t4nk_fn: haven't used grbl with esp yet, so you will be chopping down the trees and bushes, paving the road ahead with diamond-inlay-gold for us
07:35 PM t4nk_fn: orrrr... you just blow up the board and blame it on something else :b
07:36 PM t4nk_fn: mmm, there aren't many pins on the board to work with though
07:41 PM _unreal_: its an arduino form factor drop in
07:41 PM _unreal_: and will work with arduino uno hardware
07:42 PM t4nk_fn: ah, that will significantly increase the number of available i/o indeed
07:42 PM _unreal_: ?
07:43 PM _unreal_: oh ya...
07:43 PM _unreal_: I'm just planning on swapping it into my laser etcher
07:43 PM _unreal_: if when that happens
07:43 PM -!- #linuxcnc mode set to +v by ChanServ
07:43 PM _unreal_: regardless it will make that machine 10x faster
07:43 PM _unreal_: if I get it going
07:44 PM _unreal_: right now I'm just TRYING to get the @#$@# win10 key off of this laptop so I can wipe it out
07:48 PM t4nk_fn: ah, yeah, I guess,.. I never used arduino
07:48 PM _unreal_: I want to load linux on it but I dont want to do that till I recover the win10 key
07:49 PM _unreal_: then I can reload win10. and then load ubuntu
07:49 PM t4nk_fn: nor would I ever, ever, ever, ever want to have had and will not ever
07:58 PM t4nk_fn: lol, and I guess that goes the same for windows ;)
07:58 PM t4nk_fn: well, unfortunately I did once use that :((
08:58 PM roycroft: i should clarify that i personally find it easier to install linux first, then install windows, in a dual-boot environment
08:58 PM roycroft: and that may be due to my lack of experience with windows
08:59 PM roycroft: of which i'm neither proud nor ashamed
08:59 PM roycroft: just meh
08:59 PM XXCoder: me, I start at linux... and end there ;)
09:01 PM roycroft: i was using unix before bill gates founded microsoft and borged dr-dos to put on the ibm pc
09:01 PM XXCoder: yes, elder nerd ;)
09:01 PM roycroft: and unix was already fairly mature by the time ms-dos came out
09:02 PM roycroft: so i never saw a reason to use that stuff
09:02 PM roycroft: especially when xenix was ported to the ibm xp
09:02 PM roycroft: er
09:02 PM XXCoder: my earliest use was well i forgot year but it was 1990s, post-1995. it wasnt great (red hat linux)
09:02 PM roycroft: the at
09:02 PM XXCoder: next use was at my college years 2003 to 2010
09:02 PM roycroft: minix was the first unix-like os to run on an xp
09:03 PM XXCoder: small skip to 2013 or so, thats my perment move to linux and still using since
09:04 PM roycroft: i've worked with a lot of operating systems, but unix and unix-like oses have always been my main ones
09:07 PM t4nk_fn: how old are you, XXCoder, 12 or so? :))
09:08 PM XXCoder: someone have problem with math ;)
09:08 PM roycroft: maybe 12 in vulcan years