#linuxcnc Logs
Feb 17 2020
#linuxcnc Calendar
12:18 AM _unreal_: 1am heading to bed
12:23 AM ve7it: gnite
12:23 AM ve7it: found the servo board!
12:23 AM ve7it: just like the picture only dustier
02:14 AM Deejay: moin
04:49 AM BitEvil is now known as SpeedEvil
04:55 AM fogl81: Hello everybody. Can someone please help me with a liitlle hint: Where can i read the "rtapi_print_msg(RTAPI_MSG_ERR,", dmesg doesnt show anything. I am using preempt_rt.
08:46 AM pcw_home: On preemt-rt the messages are printed on the console (so start LinuxCNC in a terminal window)
10:35 AM t4nk_freenode: bought the wrong bag of concrete methinks... 'no mix fast concrete'
10:35 AM t4nk_freenode: *sigh*
10:37 AM t4nk_freenode: meant to dig a hole, fill it with water and then dump the concrete, put a pole in the hole.. and you can let go of the pole in 2 minutes... sets in 15mins
10:46 AM t4nk_freenode: I'll make a little mold tonight... mix anyhow and be quick about it, see if I can put it into the mold within a minute, see what happens
11:07 AM selroc: hi , the latest version of pycam looks very promising
11:11 AM selroc: it has evoled quite a bit
11:11 AM selroc: evolved
12:01 PM miss0r: goodevening
12:13 PM Connor: Anyone use the REF mode in their DRO? (I.E. to find the Index mark for when you power off and need to find the position again?)
12:13 PM Connor: My DRO has that feature, but has 3 modes, but no explication as to what they're for. REF, AB, LEF_AB
12:15 PM -!- #linuxcnc mode set to +v by ChanServ
01:39 PM veegee: https://www.youtube.com/watch?v=cbo5UYq_Q1E
01:39 PM veegee: LOLLLL pressure sensitive E stop that fits between your buttocks
01:40 PM pink_vampire: veegee: yeah, this is AvE
01:54 PM veegee: I don't understand what he's using to create a vacuum
01:54 PM veegee: The idea is to get to zero pressure under the workpiece right? No way a venturi thing can do that
01:56 PM veegee: what the hell, the modern ones can do 29.5 "Hg? Did NOT expect that
01:58 PM pink_vampire: he is trying to build a case for the copper hammer
01:58 PM veegee: No I linked the vacuum clamp video
01:59 PM veegee: For some reason I didn't think the venturi effect was that intense
02:00 PM veegee: Because I want to build a little wet vacuum to suck up dirty contaminated oil/coolant and stuff but any other pump will fail because the dirt will clog the piston or damage the membrane
02:00 PM veegee: I wonder if a venturi pump will work
02:05 PM CaptHindsight: not likely if purchased from HF (harbor freight)
02:06 PM CaptHindsight: https://www.coleparmer.com/p/air-powered-venturi-vacuum-pumps/6906 up to 29.5" Hg
02:08 PM roycroft: you don't need 29.5" of vacuum to hold a workpiece in place
02:08 PM roycroft: it's nice that some of the venturi vacuum generators do that, but it's not really necessary
02:08 PM roycroft: 25" is plenty to hold things in place
02:08 PM roycroft: air is pretty heavy
02:08 PM roycroft: people don't realise how heavy it is
02:09 PM CaptHindsight: I have a warehouse full of it
02:10 PM roycroft: i run my vacuum press for wood at a bit over 22"
02:11 PM roycroft: it does a brilliant job
02:12 PM roycroft: https://www.roycroft.us/Vacuum_Bag.jpeg
02:12 PM roycroft: there's a panel being glued up
02:12 PM roycroft: https://www.roycroft.us/Vacuum_Pump.jpeg
02:12 PM roycroft: and there's the vacuum pump i built to do it
02:48 PM * JT-Shop just rapid the 2.5" face mill into the vise... dunno what happened to my edit
02:50 PM CaptHindsight: have done similar, ouch
02:50 PM XXCoder: ow
03:01 PM JT-Shop: the worst part is I have no clue what caused Z4.090 to change to Z0... it's the same keys over and over when I'm sizing stock
03:01 PM miss0r: That sounds odd
03:02 PM JT-Shop: no, it sounds BANG
03:02 PM miss0r: so.. 'nasty' ? :)
03:03 PM miss0r: Would you know if it is possible to stretch a nylon tube to a larger ID when it is heated? (I need to fit a tube on a stud that is a "little" too large..)
03:03 PM JT-Shop: trashed the shell mill, hit the vise jaw, not looked at it close to see if the slot is trashed or not
03:03 PM miss0r: damn
03:03 PM * miss0r feels your pain
03:05 PM DaViruz: miss0r: that has worked fine for me with pneumatic nylon tubing at least
03:05 PM miss0r: DaViruz: How "stretchy" was it? :)
03:05 PM DaViruz: very
03:05 PM DaViruz: it's a thermoplastic, so heating it should make it very compliant
03:05 PM miss0r: hah... Alright.
03:06 PM miss0r: I wonder if it is the right choice here.. :) It should not get that hot in there as it is for measuring the oil pressure on an engine
03:06 PM miss0r: and that means theres realy no flow through the pipe
03:06 PM miss0r: so.. the engine temperature should not equal that inside the tube
03:09 PM miss0r: it lists at 80 degrees celcius stable
03:09 PM miss0r: I think that will be alright
03:10 PM miss0r: I hope it will, atleast :) I don't want a nasty leak
03:11 PM * JT-Shop put an oil pressure gauge on his VW Beetle and the tube touched something hot and melted... had to rebuild the engine to get home
03:11 PM rmu|w: Arduino vs. Evil should have used foam rubber gasket instead of O-Ring
03:13 PM miss0r: DaViruz: Would you think the 4mm tube this is made for is regarded as ID or OD? https://dk.rs-online.com/web/p/pneumatisk-vinkelfittings/2223823/
03:15 PM gregcnc: the thread is M5 so....
03:17 PM rmu|w: miss0r: https://docs.rs-online.com/0d53/0900766b8130e860.pdf
03:18 PM JT-Shop: at least it was not my brand new never been used shell mill...
03:18 PM rmu|w: seems to say OD 4mm ID 2,5mm
03:18 PM miss0r: yeah
03:18 PM miss0r: it seems so :) thanks
03:19 PM gregcnc: pneumatic fittings and standards are such a mess
03:19 PM miss0r: (which is the hose I wanted)
03:19 PM miss0r: gregcnc: hear hear
03:24 PM p0g0_: miss0r, I'd use teflon, copper, steel or brass. as noted, you will trash the engine if it fails.
03:24 PM JT-Shop: wow a new set of hard jaws is $120
03:25 PM miss0r: p0g0_: Yeah. Thanks. That was a wakeup call... I've been looking at nylon thinking about teflon all along :o
03:26 PM p0g0_: I have seen, up close and personal, a V6 throw a rod- it's not something you want to endure.
03:26 PM miss0r: but that sure af won't stretch :D
03:27 PM p0g0_: teflon is unforgiving that way
03:27 PM p0g0_: and heat won't help much
03:27 PM miss0r: I'll need to make an adaptor then :)
03:27 PM miss0r: I don't want to ruin this vintage V8 with such a treatment, no
03:28 PM miss0r: +260 - -200 celcius. does not leave alot for the imagination :)
03:28 PM CaptHindsight: I finally hit the top of a vise a few weeks ago, only took 7 years
03:28 PM CaptHindsight: twice in the same day
03:29 PM CaptHindsight: so my average is 1:3.5 years :)
03:30 PM miss0r: CaptHindsight: Thats not too bad. glad you did it twice, then I am ahead of you :D I've had my vise for 4½ years and never hit it
03:32 PM miss0r: meh. I'll have to go speak to my local machinist supply for some of that PTFE tubing. I don't want to order 50 meters of it, just to get two :D
03:32 PM miss0r: See you around, i'll call it a night.
03:33 PM p0g0_: I'd go for copper
03:33 PM p0g0_: gn8
03:35 PM Deejay: gn8
04:38 PM JT-Shop: well that screw up only cost $260 if you don't count the ruined inserts 5 @ $15 each
04:39 PM Tom_L: do tell
04:39 PM Tom_L: i take it you broke the inserts beyond rotating them
04:40 PM JT-Shop: I was sizing some material and editing the Z height and somehow it went to Z0 so it rapided into the vise
04:40 PM JT-Shop: but Z0 should have been an error without the trailing .
04:40 PM Tom_L: oooo crap
04:41 PM JT-Shop: trashed a 2.5" shell mill and the rear hard vise jaw
04:44 PM JT-Shop: $240 worth of parts to replace the knock sensors and the wires that a rat ate under the intake manifold of Rusty
04:46 PM Jymmm: Moth balls in a platic jar with a #60 hole drilled in the lid, place under hood, and refill as needed.
04:46 PM Jymmm: Dollar Tree has bags of moth balls for $1
04:47 PM CaptHindsight: oh sure you car can already make it to the Dollar Tree
04:47 PM JT-Shop: I have rat poison under the hood now, I can't stand the smell of moth balls
04:54 PM Tom_L: kids finally settled on the hondas
05:06 PM JT-Shop: which one?
05:08 PM Tom_L: civic hatchback
05:08 PM Tom_L: x3
05:09 PM Tom_L: she's gettin some crap added to hers so i haven't taken delivery of that one
05:10 PM JT-Shop: fog lights?
05:11 PM Tom_L: dimming rear mirror, center console wireless charger and something else...
05:11 PM Tom_L: it has fog lights
05:25 PM _unreal_: ok making parts for my X axis drive system
05:26 PM _unreal_: and just got off the phone with a friend. he's going to weld the parts together for my XZ interface assembly
05:33 PM _unreal_: JT-Shop, got photos yet?
05:40 PM JT-Shop: I'd rather forget it than document it
05:42 PM Tom_L: can't say i blame ya on that one
06:03 PM _unreal_: one more of the three part sections to go and I'll have my two parts for the new machine cut
06:04 PM _unreal_: then I still have to put one of them together and tap it and other crap
06:04 PM _unreal_: sigh
06:04 PM _unreal_: At least these parts will get the X axis setup
06:33 PM CaptHindsight: ok, now I'm getting somewhere with synchronizing motion with image slices
06:33 PM CaptHindsight: again
07:05 PM unterhaus_: I smashed something in my fatigue machine I was building and I kept it around for years. Finally hit the metal recycling with it
07:30 PM Bushman: hey guys
07:31 PM CaptHindsight: howdy
07:31 PM Bushman: i have a question regarding the built-in image to gcode filter
07:31 PM CaptHindsight: go ahead
07:31 PM Bushman: does the tool diameter get ignored when using V bit?
07:32 PM Bushman: i've programed a roughing pass with a 3mm tool but everything got milled away
07:32 PM CaptHindsight: sounds familiar but I rarely use the filter myself
07:32 PM Bushman: i've rouined my stack :(
07:32 PM Bushman: *stock
07:33 PM Bushman: is there anyone online right now proficient enought to have a look in the code or something?
07:33 PM Bushman: the manual does not cover this
07:33 PM Bushman: it only bluntly states "the diameter of the tool" but doesn't mention if it's used when V-carving
07:34 PM CaptHindsight: I'd dig through the forums
07:34 PM Tom_L: i haven't used image to gcode
07:35 PM CaptHindsight: I recall having to play with the settings a bit
07:35 PM Tom_L: is it programming the tool edge or center?
07:35 PM Bushman: i've used it in the past with flat endmils and it does what it says on the tin
07:35 PM Tom_L: i always program the center
07:35 PM Tom_L: what did you give it for diameter?
07:35 PM Bushman: it did cut with a tool of specified diameter... but this time i need to use the V bits
07:36 PM Bushman: Tom_L: "3"
07:36 PM Tom_L: 3 what? mm?
07:36 PM Bushman: i entered "3", not "3mm"
07:36 PM Tom_L: nm
07:36 PM Bushman: in mm units
07:36 PM CaptHindsight: https://i.imgur.com/EAF5oPj.jpg I did get it to work here with an airbrush
07:37 PM Bushman: CaptHindsight: oh! lovely
07:37 PM CaptHindsight: think I messed with the Z to get wider and thinner lines a bit
07:37 PM Tom_L: i read that too.. i'm not paying attention tonight :(
07:39 PM CaptHindsight: Bushman: that was written many years ago and hasn't been touched in a while
07:39 PM Bushman: Tom_L: according to GUI the diameter is in "units", which i selected as mm
07:39 PM Tom_L: right
07:39 PM Tom_L: but is it a v bit?
07:39 PM Bushman: yes, i have v bits with various tip diameters
07:40 PM Tom_L: so enter the tip diameter there
07:40 PM Bushman: and i wanted to rough it out with 3mm
07:40 PM CaptHindsight: so diameter changes vs Z position, similar to the airbrush
07:40 PM Bushman: Tom_L: i did
07:40 PM Tom_L: k
07:40 PM Bushman: the tip diameter is 3
07:40 PM Bushman: mm
07:40 PM Bushman: i later planned to go down to 0.1
07:40 PM Bushman: but i can't dig 2mm in brass with 0.1mm v bit
07:41 PM Tom_L: i'd listen to capn about the Z since it's a V bit
07:41 PM CaptHindsight: I'd have to play with that app again
07:42 PM Bushman: the problem is i'd use a flat endmil but that carves out the "slopes" too
07:42 PM Elmo40: CaptHindsight, you like playing...
07:42 PM Elmo40: who wouldnt like playing
07:42 PM Bushman: so in the end the v bit wouldn't be able to mill anything later
07:42 PM Elmo40: i also like tinkering ;-)
07:42 PM CaptHindsight: Elmo40: there was something counter intuitive about how it converted, at least to me
07:43 PM Bushman: i can't really use a flat endmill cause my hight map has only 2 heights, 0 and 100%
07:43 PM Bushman: so no slopes would be milled
07:43 PM Elmo40: nothing to FULLDEPTH!
07:43 PM CaptHindsight: think I had to simplify the image to 10 x 10 pixels
07:44 PM Bushman: Elmo40: o_O?
07:44 PM Tom_L: what about a ballnose?
07:44 PM Bushman: don't have any
07:44 PM CaptHindsight: Jymmm: still here?
07:44 PM Bushman: also they generate similar path as flat, only that they take the ball shape into account
07:45 PM Bushman: i though that V 60° would take the diameter into account also and generate against a trapezoid
07:45 PM Bushman: cause that's what the shape of the tool is in the end
07:46 PM CaptHindsight: it might but after you understand the configs
07:48 PM CaptHindsight: https://github.com/LinuxCNC/linuxcnc/blob/master/docs/src/gui/image-to-gcode.txt
07:48 PM Bushman: well, the GUI seems to be relatively straightforward
07:48 PM Bushman: everything else works like a charm right out of the box
07:48 PM Bushman: except the V bit flat diameter
07:49 PM CaptHindsight: Bushman: what is it doing wrong (from your perspective) with the V-bit?
07:49 PM Bushman: well...
07:49 PM Bushman: it mills off what it just tried to avoid
07:49 PM Bushman: as if the tip diameter was 0
07:50 PM Bushman: it moves up, and then down on a slope
07:50 PM Bushman: without the flat part of whatever the flat section is wide + tool dia
07:51 PM Bushman: imagine an image 1mm thick line
07:51 PM Bushman: it goes up to go over the line...
07:51 PM Bushman: then drops ater just 1mm
07:51 PM Jymmm: CaptHindsight: I think I'm here
07:51 PM Bushman: as if the tip was 0mm in diameter
07:52 PM CaptHindsight: Jymmm: are you the image-to-gcode guru?
07:52 PM Bushman: CaptHindsight: :D
07:52 PM Bushman: Jymmm: please be the image-to-gcode guru! T_T
07:55 PM Jymmm: guru?
07:55 PM CaptHindsight: Bushman: https://www.scorchworks.com/Fengrave/fengrave.html might be another option
07:55 PM Jymmm: CaptHindsight: what's the question?
07:56 PM CaptHindsight: I forget who helped solved my issues with image-to-gcode
07:56 PM Bushman: it is another option, but it does not do the raster row/columns but rather voronoi pattern
07:56 PM Jymmm: CaptHindsight: cradek wrote a img2gcode program
07:56 PM Bushman: Jymmm: does the image to gcode take tool diameter into account if you select V endmil
07:57 PM Elmo40: doubt it
07:57 PM Jymmm: http://www.linuxcnc.org/docs/2.4/html/gui_image-to-gcode.html
07:57 PM Bushman: one would exped unused variable to be grayed out, like other ones are.
07:58 PM Bushman: yes, i've read that page. no mention about tool diameter behaviour for V bits
07:59 PM Bushman: i would go with flat endmills but i don't have such tiny ones for the level of precision i'm trying to shoot for
08:00 PM Bushman: so a V bit is a must. but all the V bits i have always have flat end
08:00 PM Bushman: even if it's just 0.1
08:02 PM Elmo40: https://www.thingiverse.com/thing:721620
08:02 PM Elmo40: Touch Probe.
08:05 PM Bushman: CaptHindsight: i've also looked into F-Engrave (which i use for DXFs) but it does not take PNG, and converting it to BMP in GIMP doesn't help. says "no image loaded", so this option is not an option ATM :(
08:08 PM CaptHindsight: Bushman: what image format are you starting with?
08:10 PM Elmo40: .cdx
08:10 PM Elmo40: >_<
08:10 PM Elmo40: damn corel draw...
08:10 PM CaptHindsight: a good PNG file should convert to BMP in GIMP, I just did it
08:13 PM Bushman: CaptHindsight: PNG
08:13 PM Bushman: it does convert
08:13 PM Bushman: but fengrave complains
08:13 PM Bushman: Elmo40: speaking of touchprobes, this is one i've made: https://imgur.com/a/IUm710q
08:13 PM CaptHindsight: likely some BMP nuance
08:14 PM Bushman: CaptHindsight: most likely
08:15 PM Bushman: hmm... ok, apparently it doesn't like 16 bit bmps
08:15 PM CaptHindsight: ask in 14 hours from now
08:15 PM CaptHindsight: or post to the LCNC forums
08:17 PM Bushman: umm... the F-Engrave suffers from the same problem... it doesn't have tip diameter
08:18 PM Bushman: i only use the 90° V bit with relatively small tip for wood so it doesn't really show.
08:18 PM Bushman: but this is tiny and it's brass
08:18 PM Bushman: so this trick won't work :(
08:19 PM CaptHindsight: https://github.com/robomechs/image-to-gcode
08:20 PM CaptHindsight: this version ^^ goes into V-bit details
08:20 PM Bushman: oh! shiny!
08:21 PM Bushman: why isn't it updated in the main trunk tho? :P
08:21 PM CaptHindsight: maybe try this version
08:21 PM CaptHindsight: different version
08:21 PM Bushman: i will! it has 2 diameters, just as i need
08:21 PM CaptHindsight: a fork of the one included
08:22 PM Bushman: let's hope for the best
08:22 PM Bushman: maybe i can just swap the .py?
08:22 PM CaptHindsight: maybe, if life was easy
08:23 PM CaptHindsight: i was just battling some pyhton and it ends up that typos DO MATTER
08:23 PM Bushman: heh
08:23 PM CaptHindsight: python even
08:24 PM Bushman: what will win? 2000 lines of code or one coma?
08:24 PM Bushman: :D
08:30 PM Bushman: i will run it stand alone for a while and if it will work fine i'll integrate it
08:31 PM Bushman: still big help! thanks!
08:31 PM CaptHindsight: ask again in 14 hours if you still are wondering
08:31 PM Bushman: should be good ;)
08:32 PM CaptHindsight: you might catch some of the early risers in the US that are fast asleep by now
08:34 PM Bushman: normaly i should be asleep too, but i'm on a night shift at work. and since there's nothing to do, i'm playing with engraving
08:35 PM Bushman: i work in maintenance and so far everything is going smooth. let's hope it stays that way till morning XD
09:05 PM unterhaus_: I'm tired of hearing about how Mach can do stepping at 50khz and lcnc can't
09:05 PM unterhaus_: they have some sort of hack, right?
09:06 PM Javabean: called closed-source/proprietary technology... its probably just a disguised flux capacitor
09:06 PM unterhaus_: I think I used to understand how they did it a little better
09:07 PM Javabean: don't share that knowledge, its probably patented
09:08 PM unterhaus_: you can talk about patented tech all you want
09:11 PM unterhaus_: I think Les Newall is right, they took over a timer channel
09:19 PM unterhaus_: I remember when someone first wrote a step/dir driver for EMC. I bet they never thought it would cause so many people to be confused
09:22 PM skunkworks: From what I have heard - you can do more than 50khz stepping with mach - but even art doesn't recomend it...
09:22 PM skunkworks: *with the right computer that plays well
09:23 PM skunkworks: network disabled... no antivirus...
09:23 PM unterhaus_: yeah, takes xp though, so as soon as you get it running someone encrypts your hard drive
09:24 PM Javabean: not without networking or physical access
09:24 PM unterhaus_: I have some machines running xp off the net. I just hope nobody ever puts an infected flash drive in there
09:25 PM skunkworks: we still have some xp machines at work running stuff...
09:25 PM Javabean: that is what silicon caulk is for
09:25 PM unterhaus_: I think turning off autorun is probably enough
09:26 PM Javabean: meh, i prefer 100% silicon caulk... somethings you shouldn't leave to software
09:26 PM unterhaus_: there is a sticker over the ethernet port that says "if you plug this into the network, IT will take away your birthday"
09:27 PM unterhaus_: the problem is that these computers are used to generate data, which has to bet taken off. It's a connundrum
09:27 PM Javabean: well, if you open the case you can plug into the header
09:27 PM unterhaus_: still have the infected drive issue
09:27 PM Javabean: and yes, i was that dude
09:28 PM skunkworks: ours are hooked to the network... just not on the intenet
09:28 PM unterhaus_: we can't have a network that isn't hooked up to the internet, as far as I know
09:29 PM unterhaus_: this is what you get when IT gets hacked and so the university brings in high-priced consultants
09:29 PM Javabean: oh, a tip for your IT peeps... don't superglue the network cables
09:30 PM unterhaus_: apparently network drivers dont' tell you if your cables are unreliable
09:30 PM Javabean: exactly
09:31 PM unterhaus_: it seems likely that they could tell though
09:31 PM Javabean: i don't know of such an option
09:31 PM unterhaus_: I mean in a perfect world
09:32 PM Javabean: in a perfect world?? network cables wouldn't "go bad"
09:32 PM unterhaus_: I have started just throwing old cables out
09:33 PM unterhaus_: we're still living with that first DEC guy screwing up the board layout and swapping wires wrong
09:36 PM unterhaus_: I guess it's not that easy to steal a timer channel when you are running linux
09:39 PM Bushman: so apparently world hates me...
09:39 PM unterhaus_: join the crowd
09:39 PM Bushman: in the end the image to gcode from github does not like my image
09:39 PM Bushman: the demo yoda works fine with my settings
09:40 PM Bushman: but my image comes out as noise
09:40 PM Javabean: dude, i thought i was ugly... you win
09:40 PM Bushman: it loads fine in the preview, only after converting it looks like noise
09:40 PM unterhaus_: that part is weird
09:41 PM skunkworks: Bushman: where is the file?
09:41 PM Bushman: in the same folder
09:41 PM skunkworks: I mean... Can I test it?
09:41 PM Bushman: sure, one sec need to upload it somewhere
09:43 PM unterhaus_: researchgate tells me someone recommended my paper about rock strength estimation
09:44 PM Bushman: http://bushman.pl/downloads/michan.png
09:44 PM Bushman: http://bushman.pl/downloads/michan_s.png
09:45 PM Bushman: i'm testing with the small one
09:45 PM unterhaus_: inkscape would eat that up
09:45 PM Bushman: it's exported from inkscape
09:46 PM unterhaus_: isn't there an svg-> gcode approach you can use?
09:46 PM Bushman: none that i know of
09:47 PM Bushman: i could convert to DXF and use F-Engrave but it doesn't support V bit tip diameter
09:48 PM Bushman: i do that for 0mm tip V bits
09:48 PM Bushman: but in this case i need to use flat tipped V bits
09:52 PM Bushman: the idea being to use progresivelly finer v bits from 3mm down to 0.1mm
09:53 PM unterhaus_: I wonder if fusion360 killed all that because they support svg pretty well. All the inkscape plug ins are old
09:53 PM Bushman: i need this to be roughly 30mm wide and 2mm deep, also inverted (the black protruding)
09:54 PM unterhaus_: so you want to cut everything but the black?
09:54 PM Bushman: i can't use fusion360... altho i'm doing this as a hobby, this machine is actually at my workplace
09:54 PM Bushman: unterhaus_: yea
09:55 PM Bushman: also slanted by the v bit
09:55 PM unterhaus_: you could design at home
09:55 PM Bushman: i don't have a working windows installation. sorry.
09:56 PM Bushman: one of my laptops came with windows 8 but i don't even remember the password
09:56 PM Javabean: i have heard you can wine fusion360... but i understand it works today, but might not in an hour
09:56 PM skunkworks: Bushman: http://www.electronicsam.com/images/KandT/testing/michan.png
09:56 PM unterhaus_: too bad the linux trick of changing passwords doesn't work anymore
09:57 PM skunkworks: have you tried F-Engrave?
09:57 PM Bushman: skunkworks: the built-in filter or the one from github?
09:57 PM skunkworks: built in filter
09:57 PM Bushman: skunkworks: built-in one doesn't support tip diameter
09:57 PM Bushman: i'm trying this one, as suggested by CaptHindsight
09:57 PM Bushman: https://github.com/robomechs/image-to-gcode
09:57 PM unterhaus_: skunkworks, can you invert it?
09:58 PM skunkworks: unterhaus_: maybe...
09:58 PM Bushman: skunkworks: also F-Engrave has the same issue, doesn't support tip diameter
09:59 PM Bushman: the built-in filter works for me but if i select V bit (whatever angle) it probably only cares about toll diameter and never asks for tip diameter and assumes 0mm
10:05 PM unterhaus_: Bushman, you can upgrade your win 8 computer to win 10 for free and pick a password you can remember
10:06 PM * Javabean shudders... win10
10:06 PM Bushman: um.. don't get me wrong...
10:06 PM Bushman: but there's a reason i forgot windows password
10:06 PM Bushman: i simply never use it
10:06 PM Tom_L: it's 'password'
10:06 PM Tom_L: shhhh
10:07 PM Bushman: all my everyday activity is 100% on linux (except work related stuff on company computers)
10:07 PM * Javabean isn't suprised... is actually more suprised that there are people that use windows these days
10:08 PM Tom_L: i use both
10:08 PM skunkworks: like? http://www.electronicsam.com/images/KandT/testing/fusion.png
10:08 PM Bushman: unfortunately freecad from debian repos is still not in version that supports 2.5D milling
10:09 PM Bushman: skunkworks: what am i looking at? to much on the screen at once
10:09 PM Javabean: its a debian xfce desktop
10:10 PM skunkworks: fusion in the background..
10:10 PM Bushman: yea, i've noticed... this machine has exactly the same (came with LCNC)
10:10 PM skunkworks: running in a windows 10vm
10:10 PM skunkworks: Bushman: yes - I installed the linuxcnc livecd on this laptop.. :)
10:25 PM unterhaus_: I do work that requires windows. Believe me, I would have left it behind
10:25 PM unterhaus_: it works fine though
10:27 PM Bushman: yea, but it's out of question i would prepare gcode files at home on a windows machine in fusion360 which i have never used only to find out i need to make a minor tweak to the paths and whole day of play on the CNC goe out the window.
10:28 PM Bushman: i can't install any software here that might be deamed illegal in light of being on a work computer either
10:31 PM unterhaus_: have you used freecad?
10:32 PM Tom_L: it's stuck in metric mode for the most part
10:32 PM Tom_L: i looked at it briefly
10:32 PM unterhaus_: the website is pretty slick
10:32 PM Bushman: yes, but this tool is available since 0.18 and debian (the one that comes with LCNC) has only 0.16 in the repos
10:33 PM Bushman: also PPA doesn't work
10:33 PM Bushman: so i can't use nightly builds
10:33 PM unterhaus_: I was going to ask about the ppa
10:33 PM Bushman: i have it at home but i use mate
10:34 PM Tom_L: what is PPA?
10:34 PM Bushman: anyway...
10:34 PM Bushman: thanks for the chat, unfortunately my time is up
10:34 PM Javabean: can you use ppa on debian?? i thought it was ubuntu-base needed
10:34 PM Bushman: Tom_L: it's a Ubuntu thing
10:34 PM Tom_L: mmm
10:34 PM Tom_L: i stopped using ubuntu
10:34 PM Bushman: like software source
10:34 PM unterhaus_: ppa is an alternate source of software
10:35 PM Bushman: you can add PPAs and new repositories are available for you to install packages
10:35 PM Bushman: but it creates FrankenDebian
10:35 PM * Javabean either builds from source or just grabs pre-built tar balls
10:35 PM Bushman: also, this particular PPA doesn't work
10:36 PM unterhaus_: that's weird, you would think they would fix it
10:36 PM Bushman: anyway. c'ya all folks some other time (probably tomorrow night)
10:36 PM * Bushman is gone
10:36 PM Javabean: l8r
10:39 PM unterhaus_: the ability to import solidworks into fusion 360 now costs $350 a year
10:40 PM Tom_L: they lure you in then stick it to ya
10:43 PM hellosquireel: Greetings i have some q's on stepper motors. Can anybody give me some quick advise
10:43 PM unterhaus_: granted, that's a commercial use license
10:43 PM hellosquireel: I'm working with a NEMA 17 with a stepper angle of 1.8 degrees
10:44 PM hellosquireel: Need to get the steps down to 1 micron per step. Wondering how to do the calculations and how to achieve this
10:44 PM Tom_L: you won't unless you gear it
10:44 PM hellosquireel: It has a gear ratio of 5:1
10:45 PM Tom_L: http://linuxcnc.org/docs/2.8/html/config/stepper-quickstart.html
10:45 PM unterhaus_: micron is linear?
10:45 PM Tom_L: a small metric unit of measure
10:45 PM hellosquireel: It's x and y but horizontal movement
10:46 PM unterhaus_: you have a lead screw attached to the stepper?
10:46 PM hellosquireel: Yea, it's for a movable platform
10:46 PM hellosquireel: Yes 5mm
10:47 PM Tom_L: read that guide to see what microsteps you need
10:47 PM hellosquireel: Will do. Thank you!
10:47 PM unterhaus_: 5mm is lead screw pitch
10:47 PM unterhaus_: ?
10:47 PM Tom_L: that's what he said
10:47 PM unterhaus_: that's what you though he said
10:48 PM Tom_L: that's what i thought he said
10:48 PM Elmo40: even i thought it is what he said
10:48 PM Tom_L: but he also said 5:1 ratio
10:48 PM Elmo40: and i just opened the tab!
10:48 PM hellosquireel: YEs
10:48 PM Elmo40: well then... i stand corrected
10:48 PM unterhaus_: he wants to move a micron, so 5mm could be his full travel
10:48 PM Tom_L: so confusing
10:48 PM unterhaus_: I'm working on a stepper stage where my full travel is less than 5mm
10:48 PM Elmo40: is it a laser lense?
10:49 PM unterhaus_: just wanted to ask instead of assuming
10:49 PM Elmo40: 5mm is tight.
10:49 PM Elmo40: and 1micron is possible
10:49 PM Tom_L: who said 5mm?
10:50 PM Tom_L: not hellosquireel
10:50 PM Elmo40: i said it
10:50 PM Elmo40: pulled the number out of my ass... and read up a few lines.
10:50 PM unterhaus_: I'm not having to read back too far before I see him say linear stage 5mm
10:50 PM Elmo40: but 5mm pitch? or 5mm full travel?
10:50 PM Tom_L: nema17 with 1.8 deg steps needed to move in 1micron increments
10:50 PM Elmo40: what is the lead screw pitch?
10:51 PM Tom_L: with 5:1 ratio also he said
10:51 PM Elmo40: still need the pitch
10:51 PM unterhaus_: I feel like he has left us behind and is solving his own problem with Tom's help
10:51 PM Tom_L: he didn't mention 5mm
10:51 PM Tom_L: god help him if he is
10:51 PM unterhaus_: yes, he did mention 5mm
10:52 PM Elmo40: unterhaus_, you mentioned it
10:52 PM unterhaus_: you just opened the window, remember?
10:52 PM Elmo40: you said pitch
10:52 PM Elmo40: but he didnt confirm
10:52 PM Elmo40: just said 5mm
10:52 PM Tom_L: yeah i stand corrected
10:52 PM Elmo40: and we are left ASSuming...
10:52 PM Tom_L: but now i sit
10:52 PM unterhaus_: I know, he's off reading tom's link
10:52 PM unterhaus_: and ignoring us
10:53 PM Tom_L: or just ignoring us
10:53 PM unterhaus_: probably wise at this point
10:53 PM Elmo40: so, fuck'it. until it is clarified, i too sit.
10:53 PM unterhaus_: what's the fun in that?
10:54 PM CaptHindsight: what pitch screw?
10:54 PM CaptHindsight: any other gearing?
10:55 PM CaptHindsight: yah wannah know the microstepping required for 1um/step?
10:55 PM Tom_L: no but he does
10:56 PM CaptHindsight: klosenuf
10:57 PM unterhaus_: fusion360 doesn't like to be ignored
10:58 PM Tom_L: well i don't either
10:58 PM unterhaus_: is 5mm a common lead screw pitch?
10:58 PM Tom_L: pretty sure yes
10:58 PM CaptHindsight: hellosquireel: if your leadscrew move the nut 5mm for 1 turn, and 1um is 1/5000th of 1 turn
10:58 PM Tom_L: iirc mine may have been 5mm
11:00 PM Tom_L: https://www.ebay.com/itm/US-HGH20-Linear-Guide-Rail-CNC-Kit-Ballscrew-RM1605-400-700-1000mm-Nut-BK-BF12/114023794768?hash=item1a8c590850:g:vm8AAOSwCXxcNbtr
11:00 PM Tom_L: RM1605 so yes
11:00 PM Tom_L: 16mm rod 5mm pitch
11:00 PM unterhaus_: it's about 4 tpi
11:00 PM CaptHindsight: but yer moter is 200 steps per rev then I'd reckin 5000 / 200 = 25
11:00 PM CaptHindsight: 25 micro steps per step
11:01 PM CaptHindsight: 5mm pitch is common for a z-axis
11:01 PM CaptHindsight: so is 2mm for Z
11:02 PM unterhaus_: so Cap, if I read you right, 25 microstepping gets 1 micron/step?
11:02 PM CaptHindsight: with a 5mm per rev screw, yes
11:03 PM CaptHindsight: 12.5 microsteps with a 0.9 deg stepper and same screw
11:04 PM CaptHindsight: 5mm / 1um = 5000
11:10 PM hellosquireel: Actually I was incorrect
11:10 PM unterhaus_: thomson has linear actuators with network connectivity
11:10 PM hellosquireel: the pitch is 0.9638 mm
11:11 PM hellosquireel: One of my coworkers determined that the lead screw moves 0.37 degrees each step
11:11 PM CaptHindsight: https://zalophusdokdo.github.io/StepperMotorsCalculator/en/index.html
11:11 PM hellosquireel: and each step the lead screw moves 4.819 microns
11:12 PM CaptHindsight: that is an odd pitch
11:12 PM unterhaus_: must be inches somehow
11:12 PM CaptHindsight: 1mm or 25 or 25 TPI
11:12 PM CaptHindsight: ends up 24.47 TPI
11:13 PM hellosquireel: Well the lead of the screw is 5 mm
11:13 PM unterhaus_: I'm tapping out
11:13 PM CaptHindsight: lead = pitch
11:14 PM CaptHindsight: 5mm per rev
11:14 PM hellosquireel: Okay
11:14 PM CaptHindsight: or 1mm per rev
11:15 PM hellosquireel: So, my specs are
11:15 PM hellosquireel: NEMA 17
11:15 PM hellosquireel: 18:1 GR
11:15 PM hellosquireel: 5 mm lead
11:15 PM CaptHindsight: GR is?
11:15 PM hellosquireel: gear ratio
11:15 PM CaptHindsight: ah hah!
11:16 PM hellosquireel: So, basically I want to make it so that I can get the steps down to 1 micron/step
11:18 PM CaptHindsight: 1/8 microstepping is 1/5760mm
11:18 PM CaptHindsight: https://zalophusdokdo.github.io/StepperMotorsCalculator/en/index.html
11:18 PM CaptHindsight: a calculator so you can see for yourself
11:18 PM CaptHindsight: Steps per millimeter - leadscrew driven systems
11:19 PM CaptHindsight: if you want exactly 1um per step you'll have to change pulley ratios
11:20 PM selroc: hi guys
11:20 PM hellosquireel: Thank you for the info!
11:20 PM hellosquireel: I'm not using a pulley system though
11:21 PM CaptHindsight: pulleys or gearing
11:21 PM hellosquireel: Gearing system
11:21 PM CaptHindsight: and backlash
11:23 PM hellosquireel: Haven't figured out backlash yet
11:24 PM Elmo40: but you want 1 micron
11:24 PM hellosquireel: YEs
11:24 PM Elmo40: you NEED to know the backlash of the system
11:25 PM CaptHindsight: Elmo40: what if you just ignore it? Perception is reality
11:25 PM Elmo40: 18:1 GR? that is super high!
11:26 PM hellosquireel: backlash at no load <= 1 degree
11:26 PM CaptHindsight: plus unless the machine is temp controlled it will expand and contract by more than 1um over a few deg C
11:26 PM Elmo40: so... with 18:1 and 5mm leadscrew, isnt it 11520 steps / mm?
11:28 PM Elmo40: or is it 35.56 steps / mm?
11:28 PM Elmo40: i am confused where to put the 18...
11:28 PM Elmo40: either way you do NOT get 1 micron / step
11:29 PM hellosquireel: Is there anyway to get 1 micron/step
11:29 PM CaptHindsight: he likely will reduce the motors output by 18
11:30 PM hellosquireel: It's a low speed system
11:30 PM Elmo40: ok, so it is 11,520 steps / mm if you are reducing by 18
11:30 PM unterhaus_: why do you need 1 micron/step?
11:31 PM CaptHindsight: 5mm per rev screw, then 1:18 gear reduction, 5 /18 = 0.2777 mm per motor rev
11:31 PM hellosquireel: The accuracy had to be within a micron I belive
11:31 PM unterhaus_: wonder what the linear error is of that leadscrew
11:32 PM CaptHindsight: 0.2777777mm / 200 steps per rev = 0.001388mm
11:32 PM unterhaus_: I have some lead screw driven stages, error map looked like a banana
11:32 PM CaptHindsight: unterhaus_: :)more than that
11:32 PM CaptHindsight: yes
11:33 PM hellosquireel: 0.001388 mm might be close enough
11:33 PM unterhaus_: well tuned linear motor with feedback does okay at that scale
11:34 PM CaptHindsight: what I use them for (linear servos)
11:34 PM hellosquireel: Thanks a lot for the info! Really appreciate it
11:34 PM CaptHindsight: hellosquireel: what are you positioning at 1um res?
11:35 PM hellosquireel: It's a cell culture
11:35 PM hellosquireel: There's 6-9 cell cultures on a tray and it moves x and y position
11:35 PM hellosquireel: on a platform
11:35 PM CaptHindsight: probing, picking placing?
11:35 PM unterhaus_: hopefully the 1 micron figure is overkill
11:35 PM CaptHindsight: what is the XY travel?
11:36 PM hellosquireel: Well the partner for the project said the movements had to be accurate within 1 micron because of the microscope
11:37 PM CaptHindsight: https://i.imgur.com/LOriL1y.png this stage is accurate enough to poke a red blood cell in the nucleus
11:38 PM hellosquireel: So, essentially the x and y coordinates will be programmed to move to different cell cultures based on the revs and positioning
11:38 PM CaptHindsight: hellosquireel: at your scale you're going to learn about screw tolerances and thermal expansion
11:38 PM CaptHindsight: hellosquireel: what is the XY travel?
11:39 PM hellosquireel: Will research that
11:39 PM CaptHindsight: https://postimg.cc/4H509rzz <-- maybe 10mm travel XY
11:39 PM CaptHindsight: I make machines for things like you describe
11:40 PM hellosquireel: Not 100% sure on the x and y travel because we have just re-designed our platform for the 4th time
11:40 PM CaptHindsight: just roughly
11:40 PM CaptHindsight: microscope slide size travel?
11:40 PM hellosquireel: That I can get
11:40 PM hellosquireel: I also have the dimensions
11:41 PM CaptHindsight: if it moves slowly then maybe a very fine pitch screw vs the gearbox
11:42 PM unterhaus_: gearmotor seems like a bad idea.
11:42 PM CaptHindsight: will be ok with scales and close the loop
11:44 PM hellosquireel: https://ibb.co/qF3DBVd
11:45 PM hellosquireel: https://ibb.co/cgRR3k6
11:46 PM CaptHindsight: http://www.baselabtools.com/Fine-Adjustment-Screws_c_86.html
11:48 PM CaptHindsight: https://www.kssballscrew.com/us/pdf/catalog/PLS_Detail.pdf
11:49 PM CaptHindsight: down to 100um pitch
11:49 PM CaptHindsight: 250um is a standard pitch for that supplier
11:50 PM hellosquireel: So, in that case we should be able to get down to 1 micron
11:52 PM CaptHindsight: how slow will you be moving?
11:54 PM hellosquireel: Speed isn't really an issue
11:54 PM hellosquireel: slow