#linuxcnc Logs
May 05 2017
#linuxcnc Calendar
12:40 AM rebecca: http://imgur.com/a/23047 my first try at depth map engraving and 2 sided milling.
12:55 AM evilren: neat
12:56 AM rebecca: thanks
12:57 AM rebecca: the sides were not perfectly aligned but quite close. i might try using pins next time
01:00 AM evilren: i try to use a reference edge
01:03 AM evilren: like, face back of stock and one side at least, push up against an edge parallel to X axis that is square with the machine
01:03 AM rebecca: ahhh, ok
01:03 AM evilren: then set origin based on the faced edges
01:04 AM evilren: when you flip, keep the faced back up against the square edge
01:04 AM * rebecca nods
01:04 AM evilren: and when you set the origin, used the same side, which is now on the opposite side because flipped
01:04 AM evilren: you dont really have to face the whole side, just a point so the top and bottom are the same
01:05 AM rebecca: that's a good point. it's good to avoid making more work than necessary
01:06 AM evilren: if you try and use the same side after you flip, youre depending on accurate measurement of the stock
01:06 AM evilren: like, if you flip and use right side both times, so opposite sides of the stock
01:07 AM * rebecca nods
01:08 AM evilren is now known as renesis
01:49 AM IchGucksLive: morning from Germany
01:51 AM IchGucksLive: rebecca: still on
01:53 AM IchGucksLive: till later Bye
02:21 AM rebecca: wut?
02:22 AM Deejay: moin
03:58 AM IchGucksLive: Hi
03:58 AM IchGucksLive: rebecca: still on =
03:59 AM IchGucksLive: itss wet windy and cold in my part of Germany
04:01 AM IchGucksLive: rebecca: you shoudt change your strategie on depth maps milling
04:02 AM IchGucksLive: rebecca: first make a Reakl 3D part out of it like https://youtu.be/27IA3I3zEUM Blender
04:02 AM IchGucksLive: then use a simple CAM to get the ZigZag path
04:03 AM rebecca: okay
04:03 AM IchGucksLive: better to go full depth one path at Ballbit with given precision
04:03 AM IchGucksLive: rebecca: hi
04:03 AM IchGucksLive: is it ALLoy or steeel
04:04 AM rebecca: yeah, i just constructed the depth map in a 2d vector graphics program then exported a bitmap from it.
04:04 AM IchGucksLive: ok
04:04 AM rebecca: it's Al
04:05 AM rebecca: my CNC is a cheap little 3020
04:05 AM IchGucksLive: my channel on more howtos https://www.youtube.com/channel/UCmwhCL92ZQRSVqFe3jGwJSQ
04:05 AM IchGucksLive: rebecca: then look on to estlcam that will give you a real powertool for milling Hard Stuff on ypour 3020
04:06 AM rebecca: ooooh, 4th axis. nice
04:07 AM IchGucksLive: https://youtu.be/UdvROAlCMMQ this is on a mashine that has more then 0,1mm backlash
04:07 AM rebecca: i try to keep my toolchain opensource where possible
04:07 AM IchGucksLive: https://youtu.be/TSvDz0gIbOY
04:08 AM IchGucksLive: on linux yopou can use plenty of stuff
04:08 AM rebecca: yeah
04:08 AM rebecca: although i'm presently using bCNC + grbl
04:09 AM IchGucksLive: Heekscad for instance is a good idee on your part
04:09 AM rebecca: but i'm looking to upgrade that setup
04:09 AM IchGucksLive: you will find the best as you are here at first hand
04:10 AM IchGucksLive: rebecca: full force 3D modell milling 4 axis https://youtu.be/sqx9w80diKo
04:10 AM Crom: ok... made a arduino uno speedometer for the car until the Vehicle Speed Sensor gets here...
04:11 AM IchGucksLive: Crom: does it has police permission
04:11 AM IchGucksLive: to non overspeeding
04:11 AM Crom: doesn't matter here
04:11 AM IchGucksLive: you shoudt combine it with a F-18 hesdsup display
04:12 AM rebecca: heekscad looks interesting, thanks
04:12 AM IchGucksLive: rebecca: the ppa https://launchpad.net/~neomilium/+archive/ubuntu/heekscnc-devel
04:13 AM IchGucksLive: the best posts are hidden and need to be added manuel
04:14 AM IchGucksLive: ok im off till later
04:19 AM gonzo_: I was looking at arudino for a car speedo
04:20 AM gonzo_: originally I was going to take a prop sense pulse, use a F-V chip, scale it in analogue and drive a PWM cct to run an RC servo for the dial pointer
04:21 AM gonzo_: but the arduino seems a nice soln
04:21 AM gonzo_: (Or do it with a pic)
04:21 AM archivist_herron: toolchain is a bit easier with aaarrrguino
04:43 AM Loetmichel_: meh. some neighbour seems to mow his lawn^W hayfield with a waaay to small petrol lawnmower... for the last 2++ hours i hear *FUUUUL THROOTLE* *SCREAMING ENGINE* *choking it down to near stall* *lift up* *SCREAMING ENGINE* *repeat ad infinitum*
05:20 AM gonzo_: some people have no mechanical sympathy at all
05:21 AM gonzo_: I've just had to listen to the chassis building people doing some mods to a prototype in the office. And thinking they need to learn how to use a drill or buy some sharper bits
05:22 AM gonzo_: I've not played with arduino yet, but my impression is that they have taken a lot of the pain out of getting a hello world test prog, up and running
05:22 AM gonzo_: though I have plenty of pic code that I can strip to make a skeleton
05:23 AM gonzo_: though, .cn arduino copies are cheaper than even the base PIC I would use
05:24 AM archivist_herron: I made a few k PIC based clocks, raw assembler, recently played wit arrguino to make scales and a probe interface that measures probe contact force
05:27 AM archivist_herron: the variation in probe contact force with direction is interesting http://www.archivist.info/cnc/Arduino_Probe_Interface/
06:20 AM XXCoder: oh yeah
06:20 AM XXCoder: touchoff pad arrived lol
06:21 AM XXCoder: its very solid
06:22 AM jthornton: morning
06:22 AM XXCoder: hey jt
06:22 AM XXCoder: I'm wondering if I can just wire led and battery to touchoff
06:23 AM XXCoder: so it lights up when touch
06:23 AM jthornton: on the printer?
06:23 AM XXCoder: cnc rputer lol
06:23 AM XXCoder: still, would be interesting
06:24 AM XXCoder: can touchoff with it yeah
06:24 AM jthornton: just use the dowel method
06:24 AM XXCoder: dowel method?
06:25 AM jthornton: lower tool to below the height of a dowel then slowly raise till the dowel just passes under then set the tool height to the diameter of the dowel
06:25 AM XXCoder: ahh
06:26 AM XXCoder: new one to me
06:26 AM XXCoder: for my cnc router I usually just do paper move test
06:26 AM jthornton: https://www.youtube.com/watch?v=VmRUJk0ptN4
06:26 AM XXCoder: since its not tool holder type anyway
06:27 AM XXCoder: simple.
06:28 AM jthornton: https://www.youtube.com/watch?v=_8HQqvZSqXw
06:28 AM jthornton: Big Al
06:28 AM gcardinal: thanks for all the tips - got my mill running yeasterday with test linuxcnc code.
06:28 AM XXCoder: yummy
06:29 AM gcardinal: here is my electric box https://www.dropbox.com/s/ier4zuvkejk3jsk/Photo%2005-05-2017%2C%2008%2056%2058.jpg?dl=0 (outside box currently)
06:29 AM jthornton: https://www.youtube.com/watch?v=-AbK6VeWZnc
06:29 AM jthornton: that's Pearl
06:29 AM XXCoder: is it dead?
06:29 AM jthornton: no
06:30 AM XXCoder: gcardinal: nicely laid out
06:30 AM XXCoder: I need to figure how to do mine
06:31 AM jthornton: it's upside down :O
06:31 AM gcardinal: XXCoder thank you! I tried to copy what I have seen at tormach etc.
06:31 AM gcardinal: jthornton yes :P
06:32 AM XXCoder: I STILL haven't gotten m3 board bolt ons
06:32 AM jthornton: I finally got my tx switches to work with the Q100 so I can arm it
06:33 AM XXCoder: boom
06:40 AM jthornton: http://gnipsel.com/images/chicks/coop03.jpg
06:40 AM jthornton: almost ready for them to move in
06:55 AM gcardinal: next huge step will be figuring out how to connect Mesa SPINx1
06:56 AM XXCoder: jt nice
06:56 AM gcardinal: my motor supports reverse so the big dream is to be able to control it using linuxcnc. maybe even adding sensor so I can do rigid tapping... now sure how realistic that is
06:57 AM XXCoder: mine wont be able to do rigid tap
06:57 AM XXCoder: just one sensor and 2 magnets
06:57 AM gcardinal: and retrofitting isnt worth it?
06:59 AM XXCoder: heh my machine is cheap cnc router
06:59 AM XXCoder: (not super cheap ones out lately)
07:00 AM gcardinal: ah so regular 2.2kw spindle?
07:01 AM XXCoder: thats huge compared with mine lol
07:01 AM XXCoder: 400w
07:01 AM XXCoder: thats why rpm is enough for it
07:07 AM jthornton: you could thread mill with it
07:09 AM XXCoder: indeed
07:09 AM XXCoder: not that I will do it. well not too likely anyway
07:12 AM jthornton: I want to try it one day
07:14 AM XXCoder: yeah
07:14 AM XXCoder: would be interesting to makle wood bolt and nut
07:16 AM jthornton: yea that would be neat
07:19 AM XXCoder: yeah
07:19 AM gcardinal: XXCoder ah okay, I had shapeoko2 a few years ago with 400w - made some stuff with it
07:19 AM gcardinal: but wood dust was killer for meg
07:19 AM XXCoder: mines bit more solid than shapeoko but yeah that things pretty cool
07:20 AM XXCoder: https://www.youtube.com/watch?v=gOWVCvER3i4
07:20 AM XXCoder: fairly boring videp
07:21 AM XXCoder: 2:09 is where it shows more of machine
07:23 AM gcardinal: XXCoder cool - but it looks like you can run a much more powerfull spindle on it
07:24 AM gcardinal: here is my from a few years back https://www.instagram.com/p/_9CMa9I0l1/
07:24 AM XXCoder: im limited by holder size
07:24 AM XXCoder: 52mm or 43mm
07:24 AM XXCoder: I would have to make new holder if bigger.
07:25 AM gcardinal: ah yeah, I had good run with Kress and Suhner - last one was 1.6kW and just an awesome spindle with low RPW down to 2500
07:25 AM gcardinal: both are 43mm
07:25 AM XXCoder: I wanted tachmeter so it can have feedback to increase power if say 3000rpm its stalling
07:26 AM archivist_herron: 2500 is fast for some tap sizes
07:26 AM XXCoder: I think minium speed for my spindle is 2k rpm
07:26 AM gcardinal: yeah hard to get slow on not a mill - most routers are quite fast
07:27 AM jthornton: I tap at 300-500 rpm
07:27 AM gcardinal: the one I have now runs at 100-7500 rpm. 2 kw
07:27 AM XXCoder: well 2k is WAY better than my first one - whoch is grinder. 27000 rpm! lol
07:27 AM XXCoder: if you look there you can see the ugly failed project
07:27 AM XXCoder: I want to revisit that project later
08:04 AM Jymmm: OH man, OSHA would go ape shit over that "splitter"! Even the whole machine is moving at every stroke! lol https://www.youtube.com/watch?v=CEPNcl531sc
08:09 AM Tom_itx: wonder how clean that floor is
08:11 AM sync: clean enough
08:12 AM Loetmichel_: Jymmm: so what? let them go apeshit... ever seen an angle grinder with the shield still attached that is older than a day?
08:12 AM gcardinal: auch
08:12 AM gcardinal: lol :P
08:15 AM Jymmm: Loetmichel_: But an angle grinder isn't a whole splitting machine that it itself is moving! lol
08:15 AM TurBoss: Hello
08:15 AM Loetmichel_: Jymmm; that was obviously a demo of the machines. all new painted and stuff
08:16 AM Loetmichel_: and if the girl would have wound down the brakes instead let it sit on the wheels it wouldnt have moved at all
08:16 AM Loetmichel_: ... doesent alleviate the fact that there was no protection for her hands at all
08:16 AM Loetmichel_: ;)
08:17 AM Jymmm: Loetmichel_: Sure, but the manual feeding of it wasn't. One little misplacement of the raw material and "You'll poke your eye out kid" with bamboo spinters in your face, eyes, etc.
08:17 AM Loetmichel_: i know
08:18 AM Loetmichel_: lets just say "workers are cheap in china"
08:19 AM Jymmm: True enough.
08:20 AM Tom_itx: https://www.youtube.com/watch?v=oMOBqRVDOYQ
08:20 AM Loetmichel_: ... that said: i usually have an accident about once a year that needs some stiches... not like i would be the epitome of work safety ;)
08:20 AM Loetmichel_: Tom_itx: classic
08:21 AM Jymmm: Loetmichel_: What, the stray death-ray signals coming to bite you on the ass for keeping them all locked up?
08:21 AM Loetmichel_: usually more CNC mill or hand tools related
08:22 AM gregcnc: progressive die stamping with manual feed built in
08:22 AM Jymmm: Loetmichel_: Oh, so just clumbsy. gotcha ;)
08:33 AM Loetmichel_: wouldnt dismanlte a working still... if it were allowed in germany and i had one. (hinthint)
08:33 AM Tom_itx: ok, time to go
09:52 AM MacGalempsy: {HD} did you get your parport figured out?
09:53 AM {HD}: No. I wacked out my ankle and cannot make it to the shop.
09:53 AM MacGalempsy: anyone got an opinion on going with another dual w5590 MD or doing two single processor computers?
09:53 AM MacGalempsy: MD = MB
09:56 AM robotustra: Does anybody tried to run linuxcnc from Up-board?
10:04 AM JT-Shop: the H81M-S2H GSM is a good MB
10:05 AM JT-Shop: and the E45 M1-M PRO is good
10:09 AM MacGalempsy: JT-Shop: I have this dell workstation and the integrated MB fan is pretty noisy. it was supposed to do SLI on the dual quadro fx4800, but it only did it with the original software patch. apparently the MB chipset is not supposed to run SLI naturally
10:10 AM MacGalempsy: so I thought maybe a new build using a better board, but then thought, maybe make 2 computers out of one
10:18 AM MacGalempsy: its for a windows box, not for a linuxbox
10:19 AM FinboySlick: MacGalempsy: What's the dell workstation? 5500 ?
10:19 AM MacGalempsy: 7500
10:20 AM FinboySlick: I have a couple 5500 here and wondered about SLI. I guess it's a similar situation.
10:21 AM MacGalempsy: the stupid thing is that it worked with the Dell fx drivers, but not the newest ones
10:21 AM MacGalempsy: supermicro boards are supposed to support SLI
10:22 AM FinboySlick: I generally like Supermicro stuff.
10:22 AM MacGalempsy: but not sure how/if they will mount up right in this dell tower
10:22 AM MacGalempsy: so the search continues for the right MB
10:22 AM FinboySlick: If it's anything like the 5500, not likely. They use weird BTX format.
10:23 AM FinboySlick: Chassis is a tank, mind you.
10:23 AM FinboySlick: No flimsy gage there, reminds me of the old SUN 250.
10:26 AM MacGalempsy: yeah. this thing weighs more than a minifridge
10:26 AM FinboySlick: Cheap computing power is nice and all, but I miss the old days where manufacturers took a lot of pride in offering quality-built chassis.
10:27 AM MacGalempsy: there are a ton of fans in there. once I ran a test and all of them fired up when it was at maxium test speed. it sounded like a jet engine
10:29 AM MacGalempsy: tomshardware says a standard ATX should work fine
10:29 AM FinboySlick: https://i2.wp.com/unixadminschool.com/blog/wp-content/uploads/ou/2011/08/SUN-E250.jpg <-- Nostalgia.
10:31 AM MacGalempsy: forklift not included
10:31 AM MacGalempsy: thats pretty much how this one is
10:31 AM FinboySlick: Heh. It did take two people to put these in a rack.
10:32 AM MacGalempsy: I keep the box because its so big.
10:32 AM MacGalempsy: hopefully I can find one with SAS SCSI
10:32 AM MacGalempsy: or just use the card
10:32 AM FinboySlick: I have a couple of those actually.
10:33 AM MacGalempsy: yeah for these 2 ssd I needed it for 6.0
10:33 AM MacGalempsy: the board only came ready for 3.0
11:48 AM Cromaglious_: I oopsed on my mph and kph from knot conversion
11:49 AM Cromaglious_: Divided instead of multiplied
11:55 AM TurBoss: Hello
11:55 AM TurBoss: How does M98 M99 Translate to Linuxcnc?
11:56 AM TurBoss: I have this subs
11:56 AM TurBoss: http://dpaste.com/1MN7XMY
11:58 AM cradek: TurBoss: http://www.linuxcnc.org/docs/html/gcode.html
11:58 AM cradek: TurBoss: click on the line for Subroutines
11:58 AM TurBoss: I'm in
11:58 AM cradek: TurBoss: and bookmark that page! single best doc page for our gcode dialect.
11:59 AM TurBoss: cradek: what I should use? I'm newbey in G-Code
11:59 AM cradek: sorry I don't understand your question
11:59 AM skunkworks: All of them - of course..
12:00 PM TurBoss: what I should use instead of m98/m99?
12:00 PM cradek: if you want subroutines and you're using linuxcnc, you have to use the linuxcnc subroutine declaration and calling codes
12:00 PM TurBoss: ok
12:00 PM archivist: whatever codes you need for the job in hand
12:00 PM TurBoss: decalaration is fine
12:00 PM cradek: TurBoss: on that page, click on "o sub" and read the docs about how to do subroutines
12:01 PM TurBoss: Ok "o bub
12:01 PM TurBoss: *"o sub"
12:01 PM TurBoss: Ok I fount it
12:02 PM TurBoss: thank you!
12:02 PM cradek: welcome
12:02 PM TurBoss: I think I did understand it
12:03 PM archivist: this has an example http://www.archivist.info/cnc/standard_helical_n_teeth.ngc
12:04 PM TurBoss: wow
12:05 PM archivist: insiderearofskull CAM
12:06 PM archivist: a simpler one for clock wheel teeth http://www.archivist.info/cnc/standard_clock_n_teeth.ngc
12:07 PM archivist: they both show how you can do maths as well
12:16 PM archivist: TurBoss, the first bit of code is doing this https://www.youtube.com/watch?v=HAtziCsUj5Q
12:17 PM archivist: weeee 666k views :)
12:17 PM TurBoss: will check , busy
12:17 PM TurBoss: worked like a charm
12:20 PM IchGucksLive: hi all from a cloudy Germany
12:27 PM roycroft: cloudy it is here as well
12:27 PM roycroft: it was 28 yesterday, and we'll be lucky if it gets up to 14 today
12:35 PM diginet: C or F?
12:35 PM roycroft: c
12:35 PM archivist: kelvin
12:35 PM IchGucksLive: diginet: it will get 44C tomorrow in New Deli
12:35 PM diginet: archivist: I only use Rankine
12:36 PM LeelooMinai: I was thinking, for tapping small holes with cnc (up to 5-6mm) - you think if I put a helical coupler (like for motor shafts, etc.) between the spindle and the tap, and try rigid taping, that would work/help?
12:36 PM archivist: diginet, real fun emulating a steam cycle in pascal :)
12:36 PM diginet: as in the language pascal?
12:36 PM diginet: I loe pascal
12:36 PM diginet: love, even
12:37 PM archivist: yes did an odd steam engine in turbo pascal many years ago
12:37 PM diginet: fuuun
12:37 PM diginet: ObjPas is a joy to use IMO
12:38 PM LeelooMinai: Pascal - not case-sensitive, verbose - bad:)
12:38 PM archivist: the governor turned out to change the engine cycle incorrectly
12:40 PM diginet: it's relaly not verbose, that's a canard
12:41 PM LeelooMinai: Well, it's pretty objevtively more verbose than c-style syntax languages.
12:41 PM LeelooMinai: It's more like visual basic in terms of how it writes.
12:42 PM IchGucksLive: There is a large BBq in town i think my connection will wobble
12:42 PM Loetmichel_: i only "speak" pascal, modula , half a dozen assembler, half a dozen basic dialects and a bit of fortran and java. no C* tho
12:42 PM IchGucksLive: as town traffic is limited
12:43 PM Loetmichel_: and i dont want to learn it. already enough work to do right now without anyone coming around "can you look at my code?"
12:44 PM LeelooMinai: So, no comments about the helical coupler? My thinking was that it can compress/extend vertically a bit, but should be rigid in terms of torque forces, no?
12:45 PM LeelooMinai: I know that there are "tapping heads", but they seem to have scary prices.
12:45 PM archivist: LeelooMinai, there are better more reliable couplers
12:45 PM LeelooMinai: Right, I guessed so, but $$$
12:46 PM archivist: oldham couplers
12:47 PM LeelooMinai: Hmm, I saw those on ali - not sure what's the idea there - wouldn't the plasticy insert cause backlashing problems?
12:47 PM Loetmichel_: from experience: they dont
12:47 PM LeelooMinai: You meant this kind, right? om/wp-content/uploads/2013/02/jaw-coupling.jpg
12:47 PM archivist: it is a tight fit if well made
12:47 PM LeelooMinai: Err: http://www.designworldonline.com/wp-content/uploads/2013/02/jaw-coupling.jpg
12:48 PM Loetmichel_: http://www.cyrom.org/palbum/main.php?g2_itemId=15011&g2_imageViewsIndex=1 <- working WAAAAY better then the original ones:
12:48 PM Loetmichel_: http://www.cyrom.org/palbum/main.php?g2_itemId=14193
12:48 PM Deejay: Loetmichel_ will make you some if you ask nicely
12:48 PM Loetmichel_: http://www.cyrom.org/palbum/main.php?g2_itemId=15014&g2_imageViewsIndex=1
12:48 PM archivist: the ones with a large X like http://www.designworldonline.com/wp-content/uploads/2013/02/jaw-coupling.jpg are resiliant for vibration
12:49 PM gregcnc: there are standard tension compression tapping units for CNC
12:49 PM LeelooMinai: My first reaction would be that they seem to be more play-prone than helical ones, but it would be only a guess.
12:49 PM Loetmichel_: they are not
12:49 PM Loetmichel_: the plastic insert is pretty stiff
12:49 PM LeelooMinai: As they seem to be universally more expensive, I guess that does not happen - not sure why.
12:50 PM LeelooMinai: Don't plastic "flow" under pressure though?
12:50 PM LeelooMinai: plastics*
12:50 PM SpeedEvil: depends on the plastic
12:50 PM Loetmichel_: Deejay: please refrain to put more work on my already full day.
12:50 PM SpeedEvil: And the pressure
12:50 PM gregcnc: even metal deflects, all depends how much you're willing to put up with
12:50 PM Loetmichel_: exactly
12:50 PM archivist: keep within spec
12:51 PM Loetmichel_: there are many plastics that have LESS deflection in this configuration than a spiral aluminium couplign
12:51 PM LeelooMinai: Why is the middle part plastic and not metal anyways?
12:51 PM Deejay: Loetmichel_, but you still have the weekend ;)
12:51 PM Loetmichel_: LeelooMinai: do ease out any misalignment in the shafts
12:51 PM Loetmichel_: -d+t
12:51 PM archivist: LeelooMinai, so it can be a tight fit
12:52 PM Loetmichel_: it acts as a "spring" of sorts , just one with a real high stiffness
12:53 PM gregcnc: but how does this coupling relate to tapping?
12:53 PM LeelooMinai: Right, ok, I see how it works for motor shafts + linear stage screws, but I was thinking abusing them for tapping.
12:53 PM archivist: er wot
12:53 PM Loetmichel_: should work well also
12:53 PM archivist: I dont think so
12:53 PM gregcnc: what kind of spindle?
12:53 PM LeelooMinai: archivist: My original question was if I can abuse helical one for tapping small holes with cnc.
12:54 PM Loetmichel_: at least the bigger ones with an 6 point star in the middle instead of a 4 point star
12:54 PM Loetmichel_: you cant, it will break
12:54 PM Loetmichel_: the star type things will not
12:54 PM Loetmichel_: they will just deflect a bit
12:54 PM archivist: use linuxcnc properly and solid tap
12:54 PM LeelooMinai: As I understand for tapping one want tiny play in axial direction, but rigidity otherwise.
12:55 PM Loetmichel_: IIRC you can even use them up to a few ° of misalignment if you dont care about replacing the plastic inserts regulary
12:55 PM archivist: most important is use machine taps not hand taps
12:55 PM LeelooMinai: archivist: Properly how? It cannot know what the tap vs z-axis misalignment is.
12:56 PM LeelooMinai: Right, I ordered machine-type ones.
12:56 PM gregcnc: the machine should be square?
12:56 PM archivist: they are ground concentric
12:56 PM LeelooMinai: I mean thread misaligment as the tap progresses in Z
12:56 PM archivist: fix machine errors rather than add flex
12:57 PM gregcnc: the holes should be square?
12:57 PM LeelooMinai: Er, I don't know any more what you are saying:)
12:57 PM archivist: they are ground, there should be no misalignment
12:57 PM IchGucksLive: ack
12:57 PM LeelooMinai: So, why do they make those heads with little play?
12:58 PM LeelooMinai: I mean even if one has perfect CNC, a tap can have misperfections, no?
12:58 PM gregcnc: in the real world there are tolerances
12:58 PM archivist: for use where they dont have full control
12:59 PM LeelooMinai: I guess I can make tests after I calibrate all the axis and see if I break those taps.
12:59 PM Loetmichel_: LeelooMinai: they are usually in there when the spindle cant stop in an instant, the z axis can.
01:00 PM Loetmichel_: or when there is no encoder on the spindle to be sure the rpm and the z movement matches perfectly
01:00 PM LeelooMinai: A, ok, that reminds me, that my VFD has some terminals for breaking resistor - maybe I want to add it for tapping.
01:01 PM LeelooMinai: Or braking even
01:01 PM Loetmichel_: do you have a spindle encoder?
01:01 PM LeelooMinai: No, I have one encoder now, and two coming, but they are for steppers - I think they are rated for max 6000 rpm, so I cannot use them for spindle.
01:02 PM Loetmichel_: then you HAVE to have some kind of "z relief" or the taps WILL brake
01:02 PM Loetmichel_: break
01:02 PM LeelooMinai: I was thinking of adding some kind of spindle rpm beter using led + phototranbsistor, but that's not the same.
01:03 PM Loetmichel_: i did it once by losening up the z axis nut so it had about 5mm backlash against the z sled
01:03 PM LeelooMinai: I just remembered that archivist is one of those people that assume that everyone has NASA like equipment and standards:)
01:03 PM Loetmichel_: works well if you dont have any of these tapping collets
01:04 PM LeelooMinai: Loetmichel_: That's what my helical coupler idea was about (z relief), but right, they can possibly break.
01:05 PM LeelooMinai: Unless someone makes them from steel:)
01:05 PM Loetmichel_: btw. what kind of spindle do you have?
01:05 PM Loetmichel_: its not that easy to have THAT kind of torque AND low rpm to tap a hole right
01:05 PM LeelooMinai: Not a big one - 800W ER11 - mostly because I am limited by 110V in my room.
01:06 PM IchGucksLive: LeelooMinai: with VFD or only speed control
01:06 PM LeelooMinai: But it's AC, so I checked some forces for tapping and from those I thought that sub 6mm I may be able to tap, something.
01:06 PM gregcnc: threadmill
01:06 PM LeelooMinai: I have GS2 VFD from Automation Direct.
01:07 PM LeelooMinai: It's supported in linuxcnc and is pretty nice imo.
01:07 PM IchGucksLive: did you check the Tourch parameter onnd the Amp one max
01:07 PM Loetmichel_: standard chinses800w with 24krpm?
01:07 PM gregcnc: high speed /no torque spindle, no encoder = no tapping
01:07 PM Loetmichel_: they have in NO way enough torque or slow speed to turn even a 3mm tap
01:08 PM LeelooMinai: I looked at the torque required and then at AC motor curves.
01:08 PM LeelooMinai: Loetmichel_: Are you sure? 3mm requires tiny torque...
01:08 PM LeelooMinai: It's an AC motor - it's not so bad at low speeds.
01:09 PM * Loetmichel_ used the motor/gearbox/chuck of a 12V battery drill in first gear to do tapping on my CNC mill
01:09 PM Loetmichel_: LeelooMinai: i have 2 of these 800W spindles
01:09 PM Loetmichel_: they cant really turn WITHOUT any load at lower than 3kRPM
01:09 PM archivist: I have done a lot of battery drill tapping, and broken a few taps
01:09 PM Loetmichel_: and you'll need more like 100RPM for tapping
01:10 PM LeelooMinai: 100rpm is lowest I can run them at right now - but did not measure the torque at that speed - not sure how to be honest.
01:10 PM Loetmichel_: archivist: i used it in the 43mm holder for the Kress on a gantry router
01:11 PM LeelooMinai: I have hook type scale - mabe I could use it + some kind of lever and measure - not sure...
01:12 PM Loetmichel_: worked pretty well, especially because i used 2 microswitches that would activate the battery drill motor in forward and reverse by utilitising the "z play" that battery drill chuck had inherently because of its "hammer drill" capabilities
01:12 PM IchGucksLive: ok im off the connection is to week here
01:12 PM Loetmichel_: LeelooMinai: trust me, you dan stop it by hand when turning 100 rpm. even when turning 1000rpm
01:12 PM Loetmichel_: been there, done that
01:13 PM LeelooMinai: In the worst case, I guess, one could add a secondary slim slow-speed spindle for taping.
01:13 PM Loetmichel_: thats what i did
01:14 PM LeelooMinai: Probably just shaft + motor and belt/gears on top?
01:14 PM Loetmichel_: removed the high speed spindle, replacesd it with the "head" part of a battery drill
01:14 PM Loetmichel_: including chuck
01:14 PM LeelooMinai: Right, I guess that's a project for some future - but I would keept the spindle.
01:15 PM DaViruz: how a bout a tap holder that sits in the high speed spindle that has a toothed belt sprocket that gets driven by another motor
01:15 PM DaViruz: for tapping
01:15 PM DaViruz: not that i like to over complicate things..
01:15 PM LeelooMinai: Heh, that's and interesting idea - at least it would be in same axis as the main speendle.
01:16 PM LeelooMinai: Would also generate electricity for VFD? :p
01:16 PM DaViruz: if it's a permanent magnet motor yes, a very tiny amount
01:16 PM DaViruz: if it's an induction motor it won't
01:17 PM LeelooMinai: A, ok then - never saw anyone doing it though.
01:17 PM Loetmichel_: DaViruz: they are induction motors
01:18 PM Loetmichel_: no permanent magnets in them
01:18 PM LeelooMinai: Right, it's just 2 pole, 3 phase AC spindle.
01:18 PM gregcnc: induction motors will also generate
01:18 PM DaViruz: well you can make induction motors generate
01:18 PM DaViruz: but they won't by their own accord
01:18 PM gregcnc: most will
01:19 PM LeelooMinai: At 100rpm it won't damage anything probably.
01:19 PM LeelooMinai: Loetmichel_: Did you measure runout on those spindles btw?
01:19 PM Loetmichel_: the chinses 800W ones?
01:20 PM Loetmichel_: my two have less than 0,05 mm
01:20 PM LeelooMinai: I measured 35 microns, but that was before I read that those ER11 collets requires big torque for fastening.
01:20 PM Loetmichel_: thats the smalles i can see on my dial
01:20 PM gregcnc: what's the runout of the taper?
01:20 PM LeelooMinai: A ok, so I guess that's normal.
01:20 PM LeelooMinai: I did not measure the tapper - I guess I could try.
01:21 PM LeelooMinai: taper
01:21 PM LeelooMinai: But they advertise them for less than 2 mil or so runout, so 35 microns is 1.2 so seems reasonable.
01:22 PM LeelooMinai: ER11 is kind of limiting though - makes me a bit sad, but with 110V cannot use anything bigger (or cannot find)
01:23 PM DaViruz: how about a transformer?
01:24 PM LeelooMinai: The 110V in rooms is limited to max 16A I think, that's the main problem.
01:25 PM DaViruz: diesel generator then!
01:25 PM LeelooMinai: That GS2 drive I have is the largest one they have, so I guess that's the limit more or less - cannot use 2.2kW spindle.
01:26 PM LeelooMinai: Right, well, could run one in the basement:) BTW, I actualy ran the air hose from the baement (inside walls) to my room, so I have remotely controlled compressor in the basement for air for my CNC.
01:27 PM LeelooMinai: Used one of those RF on/off mains switches.
01:27 PM LeelooMinai: Work pretty well.
01:30 PM Loetmichel_: LeelooMinai: you CAN use a 2.2kw spindle
01:30 PM Loetmichel_: you just have to soft start it and cant use its full power for more than a few seconds
01:30 PM LeelooMinai: Just received nice crowbar wrench set from ali - now I can use my tension wrench to tighten that er11 collet right.
01:30 PM LeelooMinai: Loetmichel_: Sound wonderful:)
01:31 PM Loetmichel_: uh, WHAT?
01:31 PM LeelooMinai: Loetmichel_: What?
01:31 PM LeelooMinai: Er, crow foot, not crowbar - lol
01:32 PM Loetmichel_: an ER11 needs less than 20 Nm to be tighened correctly... if you use a crowbar to tighten it you REALLY overdoing it
01:32 PM Loetmichel_: IIRC
01:32 PM Loetmichel_: ah, crow foot...
01:32 PM Loetmichel_: that changes things
01:32 PM LeelooMinai: Loetmichel_: I meant this: https://www.amazon.ca/TEKTON-2580-8-Inch-Crowfoot-Wrench/dp/B000NPR2BC/ref=sr_1_sc_2?ie=UTF8&qid=1494007328&sr=8-2-spell&keywords=tecto+crow+foot
01:32 PM LeelooMinai: SO I can attach them to the tension wrench.
01:32 PM Loetmichel_: i saw you standing there with a 3 foot long bar tightening um the ER11 ;)
01:33 PM LeelooMinai: Still 20Nm is pretty big - limit of my wrench in fact.
01:33 PM LeelooMinai: Right, I guess crowbar is not a good tool for this.
01:33 PM Loetmichel_: i tightenn mine with 2 normal 13mm and 17mm spanners
01:34 PM Loetmichel_: short single sided ones on top of that
01:34 PM Loetmichel_: by hand
01:34 PM LeelooMinai: That's what I did, but, I will feel better actually knowing that I use the right force.
01:34 PM Loetmichel_: and i never had any mill bit slipping
01:34 PM Loetmichel_: ... i do have a bit bigger hands and arms tho ;)
01:35 PM LeelooMinai: That's not necessarily good - ruining th thread on the spindle is pretty bad.
01:35 PM Loetmichel_: hmm?
01:35 PM LeelooMinai: Thought probably collet thread would go first when overtightening.
01:36 PM Loetmichel_: i dont OVERtigthen them
01:36 PM LeelooMinai: Loetmichel_: I meant that if you have a lot of force in hands, you risk applying too much - inless you have super-human tough power.
01:36 PM Loetmichel_: you possibly cant with 12 to 14cm long spanners without any leverage
01:36 PM LeelooMinai: I read an article ones where they took few mechanics claiming that they have perfect sense of torque and there were differences of 100% in reality:)
01:37 PM Loetmichel_: 100% is pretty close tho
01:37 PM LeelooMinai: Extra 100% that was.
01:37 PM Loetmichel_: but as we say here in germany: "Augenmaß und Handgewicht sind des deutschen Meisters Pflicht!"
01:38 PM Loetmichel_: ( no idea how to translate THAT ;)
01:38 PM LeelooMinai: I already did with google translate so no need:)
01:39 PM LeelooMinai: Sounds like what overconfident mechanics in Germany would say.
01:39 PM Loetmichel_: what did it spew out
01:39 PM LeelooMinai: "Eye and hand weight are the duty of the German master!"
01:40 PM Loetmichel_: its actually what our master in my first apprenticeship drilled into us
01:41 PM gregcnc: recommneded collet nut torque varies with collet and nut type
01:41 PM Loetmichel_: it means "true measuring by eye and felling forces/weights by had right has to be a duty of every german mechanic that is called a "master""
01:41 PM LeelooMinai: Well, ok, if that's you professional work, I can see that after 20 years you have perfect sense of what 20Nm torque is. In my case I first have to imagine a lever, then shorten it to something reasonable, like 30cm and then use force in Kg, and probably still would not be able to apply the exact force.
01:42 PM gregcnc: people who are using torque wrenches often fail to use them properly
01:42 PM archivist: get a torque meter :)
01:42 PM SpeedEvil: Loetmichel_: And also that true master should know that it doesn't matter, and they should use the appropriate measuring tool anyway, even if they can guess
01:42 PM Loetmichel_: its wider meaning is tho that a german mechanic is supposed to KNOW what xx Nm feels like
01:42 PM LeelooMinai: archivist: I have a torque wrench.
01:43 PM SpeedEvil: I want a torque wench to do up my nuts properly.
01:43 PM Loetmichel_: without applying any measurement device
01:43 PM Loetmichel_: so that they cant be fooled by malfunctioning measurement devices
01:43 PM archivist: but for tapping you can see a tap spring when approaching its limit
01:43 PM LeelooMinai: Loetmichel_: What if that mechanic just does not have a "feel" for those things - cannot he/she be a good mechanic any more? :)
01:44 PM Loetmichel_: i suppose so
01:44 PM Loetmichel_: you can learn that "feel"
01:44 PM archivist: those without feel, brake stuff often
01:44 PM Loetmichel_: one takes longer the other one shorter
01:44 PM archivist: break
01:44 PM LeelooMinai: Yes, but even then - you will probably not be able to feel the same very day.
01:44 PM roycroft: a mechanic cannot work effectively without having a "feel" for such things
01:44 PM LeelooMinai: You can be tired, or overnergetic and boom.
01:45 PM Loetmichel_: its like that mechanic on your car that says "please give me that 13mm spanner there"... and there is a 15mm. and the nut he wants to loosen up is 10mm ;)
01:45 PM SpeedEvil: Or it can be cold
01:46 PM SpeedEvil: A good mechanic can know what things are most of the time by feel, and still measure when it's important.
01:46 PM LeelooMinai: In any casde, I probably kept undertightening that nut - I guess that's better than overtightening.
01:46 PM Loetmichel_: SpeedEvil: exactly
01:46 PM gregcnc: unit for torque aren't "that feels about right"
01:46 PM Loetmichel_: LeelooMinai: better for the collet, worse for your mill bits/workpiece
01:47 PM LeelooMinai: Realisticly, it seems to me most people don't use any torque wrenches... even people who do some professional work.
01:47 PM Loetmichel_: gregcnc: it should be. at least any mechanic should have a "feel" for when he is approaching breaking point
01:47 PM gregcnc: bull
01:48 PM CaptHindsight: most mechanics I know have a leftover bolt and nut tray...
01:48 PM CaptHindsight: since the car co's tend to use too many extra fasteners
01:48 PM LeelooMinai: I got that wrench because I am a bit OCD, but when I once mentioned that I want to tighten my bolts on CNC efvenly I was told it's not necessary and overkill:)
01:48 PM archivist: spares for later rebuilds
01:48 PM SpeedEvil: Loetmichel_: when it's yielding and you can feel that, that is often too late
01:48 PM gregcnc: find a mechanic like that assembling new cars
01:48 PM Loetmichel_: CaptHindsight: yeah, pretty embarrassing if you dismantle your car and have some bolts leftover
01:48 PM DaViruz: gregcnc: so what about undoing screws, no feel for that either?
01:48 PM DaViruz: just snap them if they don't yield?
01:49 PM Loetmichel_: SpeedEvil: it is. thats why i said "approaching"
01:49 PM SpeedEvil: Loetmichel_: there is literally no feel before yield.
01:49 PM LeelooMinai: Right, or dismantle some electronic device, put it back, and you discover few extra parts left on the table.
01:49 PM SpeedEvil: Other than guesses about the strength of the fastner
01:49 PM archivist: some engines actually have to be taken to yield point on the bolts
01:49 PM SpeedEvil: yeah
01:49 PM SpeedEvil: one-use bolts
01:50 PM LeelooMinai: And it helps how?
01:50 PM Loetmichel_: SpeedEvil: old engines have "stretch-bolts" in the head
01:50 PM gregcnc: how do you know if it's going to snap or about to yield? either it comes out, snaps, or the head yields
01:50 PM DaViruz: new ones as well
01:50 PM DaViruz: there can be stretch bolts for pretty much anything
01:50 PM Loetmichel_: THERE the old saying of "tighten it until it yields and then a quarter back" really works
01:50 PM Loetmichel_: bot ONLY on these
01:50 PM LeelooMinai: So you tighten them until they break, and then back a bit? :)
01:51 PM sync: if you loosen it you lost all clamping force
01:51 PM gregcnc: been on both sides of loosening fasteners to know, I don't know
01:51 PM sync: LeelooMinai: it makes the clamping force independent from the tightening angle
01:51 PM Loetmichel_: LeelooMinai: these bolts have a small section thats not hardened at all
01:51 PM LeelooMinai: Heh, and there's already saying for my joke - sad.
01:51 PM Cromaglious_: Ugh stretch bolts.... U switched to cb performance panzer cases and non stretch studs pn my bus engines
01:51 PM Loetmichel_: they deform plastically there without breaking when approaching rated torque
01:51 PM LeelooMinai: Loetmichel_: So it's like built-in spring?
01:52 PM gregcnc: springs, spring back. torque to yield fasteners do not
01:52 PM Loetmichel_: not spring, more like a plastic deformation
01:52 PM Cromaglious_: Leeloominai yep
01:52 PM LeelooMinai: Well, some springs to not come back after you stretch them I guess.
01:52 PM gregcnc: because you yielded them
01:53 PM DaViruz: to be honest i don't get the point of stretch bolts
01:53 PM Cromaglious_: I prefer cupped washers for springs and non deforming studs
01:53 PM Loetmichel_: DaViruz: they stem from a time when there were not that many torque wrences
01:54 PM Loetmichel_: +h
01:54 PM archivist: DaViruz, they guarantee even and correct force
01:54 PM LeelooMinai: So bolt+torque limiter
01:54 PM LeelooMinai: In one
01:54 PM Loetmichel_: yes
01:54 PM DaViruz: more so then a torque wrench?
01:54 PM sync: Loetmichel_: it is the other way around TTY bolts are a recent development
01:54 PM Loetmichel_: one time use tho
01:54 PM Cromaglious_: You tighten to a set length
01:54 PM DaViruz: or rather a properly torqued screw would
01:54 PM sync: DaViruz: you can get the torque to tighter tolerances
01:54 PM sync: or rather clamping force
01:55 PM DaViruz: i see
01:55 PM sync: as it is not dependant on the µ of the thread
01:55 PM Loetmichel_: sync: those bolts were there since the late 40ties in car engines
01:55 PM Loetmichel_: until the late 60ies i would thing
01:55 PM Loetmichel_: think
01:55 PM gregcnc: many things influence tightening torque
01:55 PM sync: tty? nah boi
01:55 PM DaViruz: more like until 2017
01:56 PM Loetmichel_: then they started to be replaced by standard bolts and torque wrenches
01:56 PM LeelooMinai: They probably use one-time metal glue now:)
01:56 PM archivist: noo
01:56 PM archivist: Loetmichel_, yo wrong
01:56 PM Loetmichel_: which part
01:56 PM DaViruz: a modern head is full of stretch bolts
01:56 PM DaViruz: heck, a modern engine
01:56 PM gregcnc: https://en.wikipedia.org/wiki/Torque-to-yield_fastener
01:56 PM archivist: stretch is recent, 1980's
01:56 PM DaViruz: way more then just the head itself
01:57 PM Loetmichel_: archivist. cant be
01:57 PM Loetmichel_: had them on my old magirus
01:57 PM Loetmichel_: thats built in '61
01:57 PM Loetmichel_: and on some even older cars
01:57 PM Loetmichel_: never seen them today tho
01:57 PM Loetmichel_: as cylinder/head bolts
01:57 PM Cromaglious_: My 69 vw bus had them as well
01:57 PM LeelooMinai: Hmm, seems they are are good when vibrations are involved
01:58 PM archivist: in honda and BL engines from the 80's
01:59 PM LeelooMinai: So high forces+vibrations = being at yield is beneficial (?)
01:59 PM Loetmichel_: anyways, nowadays you have torque wrences and recommended torque in the cars repair handbooks
02:00 PM SpeedEvil: https://www.andrewsfasteners.uk/products/direct-tension-indicators-washers-14399/ are neat
02:00 PM LeelooMinai: They need to make something that changes color according to torque applied.
02:00 PM Cromaglious_: Torque to yeild bolts are stretched a bit, so they're plastic deformation range, however they have a range of spring back.
02:01 PM SpeedEvil: LeelooMinai: see above
02:01 PM LeelooMinai: Right, so I guess in this plastic range vibrations are less harmful.
02:02 PM SpeedEvil: LeelooMinai: http://www.appliedbolting.com/squirters.php
02:02 PM LeelooMinai: SpeedEvil: Er, I don't know how they indicate...
02:02 PM SpeedEvil: The washer squirts out its vital fluid when you've tugged on it enough.
02:02 PM LeelooMinai: A, lol - pretty weird,. but ok.
02:02 PM Cromaglious_: Non yield bolts without cup washers can be loose cold, tight hot, or anywhere between. I prefer cupped washers or deformation bolts
02:03 PM Cromaglious_: S/or/over/
02:04 PM Loetmichel_: Cromaglious_: had that once in extreme
02:04 PM Loetmichel_: had to mount a few dozen M8 metal bolts in concrete
02:04 PM LeelooMinai: So, what... they have liquidy paint in some tiny hole and when the hole is squeezed to some tension it squeezes the liquid part?
02:04 PM Loetmichel_: screwed the first one in with an 1/4" drive ratchet driver and 13mm nut...
02:04 PM Loetmichel_: one handed
02:05 PM Loetmichel_: *KNACK*
02:05 PM Loetmichel_: head sheared off
02:05 PM Loetmichel_: "HU?"
02:05 PM SpeedEvil: LeelooMinai: yes
02:05 PM LeelooMinai: Loetmichel_: I had deck screws that would break the head off when you used one hand and a philips screwdriver...
02:05 PM Loetmichel_: screwed the rest bolt out, took a second out of the box... screwed in: KNACK
02:05 PM Cromaglious_: Tormach uses cupped spring washers for their tool changer. You van add stretch range by stacking washers alternating cup. You can increase force by stacking cup same way
02:06 PM Loetmichel_: yes. tuned out there were hardened to glass hard
02:06 PM Loetmichel_: and the concrete wasnt exactly flat
02:06 PM roycroft: some deck screws are designed to break off like that
02:06 PM Loetmichel_: so there were some "tilting forces" on the heads
02:06 PM LeelooMinai: Tormach always reminds me that NYC CNC guy - recently he put a video of some "oops" he had, was pretty funny.
02:06 PM Loetmichel_: <- thought for a moment he got super powers ;)
02:07 PM Cromaglious_: That was a good video, I'll be seeing him next month at the BarZ summer bash
02:08 PM Loetmichel_: i mean i can rip off any M8 8.8 head with a half inch ratched and some steel tube on it
02:08 PM LeelooMinai: I loled at the pumpkin one - way to break an expensive probe tip on a fruit.
02:08 PM Loetmichel_: but with a 15cm short 1/4" ratchet one handed?
02:08 PM roycroft: i still have mixed feelings about that nyc cnc guy
02:08 PM roycroft: he comes off as really snarky
02:08 PM roycroft: but he also comes off as really sincere
02:08 PM roycroft: and i can't decide which he really is
02:08 PM Loetmichel_: whats snarky?
02:09 PM roycroft: know-it-all rude, condescending, etc.
02:09 PM Loetmichel_: ah
02:09 PM Cromaglious_: John is really down to earth and a REALLY nice guy
02:09 PM Loetmichel_: hmmm
02:09 PM LeelooMinai: Well, he is what he is - as long as I can get something useful from his videos, that's ok. His smile though at the end always gives me a pause:)
02:10 PM LeelooMinai: He seems pretty admitting that he has a lot to learn all the time.
02:10 PM Cromaglious_: He is a smiler
02:10 PM roycroft: that's the thing
02:10 PM roycroft: he can be both smug and self-deprecating at the same time
02:10 PM Loetmichel_: i should be quiet then... my wife gifted me a tshirt: http://www.cyrom.org/palbum/main.php?g2_itemId=16158&g2_imageViewsIndex=1
02:10 PM FinboySlick: roycroft: Maybe that's the NYC bit ;)
02:11 PM Cromaglious_: He alwYs has a smile on when i see him
02:11 PM roycroft: it could be
02:11 PM roycroft: i've considered that
02:11 PM LeelooMinai: But it's a "polotician" kind of smile - pretty robotic imo:)
02:11 PM Loetmichel_: " i am no 'know it all'... i REALLY know it all!"
02:11 PM Loetmichel_: :-)
02:11 PM FinboySlick: roycroft: It'll hopefully fade as he left NYC ;)
02:11 PM FinboySlick: Loetmichel_: Yeah, but you're also german, it cancels out.
02:11 PM FinboySlick: ;)
02:11 PM Cromaglious_: Do remember those welding goggles he had in a video last summer?
02:12 PM LeelooMinai: But, since I had to learn Fusion360, initially his videos were helpful.
02:12 PM LeelooMinai: He is not really an expert on it though, but ok for beginners.
02:12 PM Cromaglious_: Wirh the spinny leds? I made those
02:12 PM DaViruz: well it figures the german don't know what snarky means ;)
02:12 PM Cromaglious_: Hah
02:13 PM Loetmichel_: DaViruz: did you expect different? ;)
02:14 PM Loetmichel_: FinboySlick: i think she wanted to indicate that i am a bit "know it all" even for german standards...
02:26 PM sync: Cromaglious_: idk, I don't really like him
02:26 PM sync: he never seems to know what he does
02:29 PM guru_florida: So I was getting "main task took 120ms" errors which would stop the machine. I may have just found the issue.
02:30 PM guru_florida: I made a small circuit that translates modbus USB commands to analog 10v signals (and reads back RPM too). I disabled this and went back to panel but the errors *still* happened....I kept the USB device connected though.
02:31 PM guru_florida: I disconnected the modbus device (arduino like board) but left the usb-serial dongle on the usb cable and left that where it was near the VFD. Still got the errors.
02:32 PM guru_florida: Unplugged the usb cable from computer and removed it...Voila! I havent gotten an error since.
02:32 PM guru_florida: Seems like interference into the USB cable? The cable is shielded. usb-serial board was not though.
02:32 PM guru_florida: Could that interference really cause the computer to lock up for 120ms?
02:32 PM guru_florida: thoughts?
02:33 PM archivist: usb itself has latency in the design
02:33 PM SpeedEvil: But driver trying to find a device plugging and unplugging may take nonzero time
02:33 PM SpeedEvil: if falsely triggered by EMI
02:33 PM guru_florida: That's what I was wondering.
02:34 PM guru_florida: But USB is a differential signal pair
02:34 PM guru_florida: shouldnt that prevent it?
02:34 PM Loetmichel_: no
02:34 PM Loetmichel_: not if the interference is strong enough
02:34 PM guru_florida: gotcha
02:35 PM archivist: USB finding a device is a long time
02:35 PM guru_florida: Linuxcnc was not configured to read the serial port so any delay would have to be kernel time, not user space time or that would get properly preempted.
02:37 PM pcw_mesa: "main loop" delays should not stop the machine
02:38 PM pcw_mesa: they hold up the user interface but should not affect the real time portions of LinucCNC
02:39 PM guru_florida: Well it does, I keep the bash shell open and I see the error happen and the motion stop and spindle stops and the machine "Power" icon goes to off position.
02:40 PM pcw_mesa: But no real time errors?
02:41 PM pcw_mesa: I can see a modbus spindle stop on a comms timeout
02:42 PM Crom: vok got my linear bear holes for the new sopindle mount turned on the lathe
02:42 PM Crom: now waiting on the ball screw
02:43 PM pcw_mesa: in other words on a simple config I can launch linuxcnc from a shell, start a gcode program and hit control Z
02:43 PM pcw_mesa: and the program interpretation continues
02:44 PM LeelooMinai: I wish in linuxcnc in all the places (hal meter, config, scope, etc. where one chooses pins/signals/parameters from the list, there was a "fast filter" field - scrolling through hundreds of signals to find one is pita.
02:45 PM pcw_mesa: if i restart linuxCNC (say fg 1) I will get a complaint that the mainloop took X seconds but nothing stops/is-broken
02:46 PM pcw_mesa: LeelooMinai: often better to use halcmd for that sort of thing
02:46 PM LeelooMinai: pcw_mesa: To add a channel to scope?
02:47 PM pcw_mesa: then you have smart completions, command history, script files etc etc
02:47 PM LeelooMinai: I mean in dialogs like "Select Channel Source" - I am through VNC to make matters worse.
02:48 PM guru_florida: No realtime errors
02:48 PM pcw_mesa: so what stopped linuxcnc?
02:49 PM LeelooMinai: In GUIs I like those fields that are real-time and filter as you start typing letters - they are super-useful.
02:49 PM guru_florida: I have really good Latency test results after doing a bunch of stuff on the Wiki page <10micros, usually less than 1us
02:50 PM LeelooMinai: Hmm, maybe I should test my latency too.
02:50 PM guru_florida: It's not a modbus spindle, it's an analog one, but the little circuit I made using an arduino I coded modbus to analog output.
02:51 PM guru_florida: So I know that linuxcnc is commanding the spindle to stop. Remember linuxcnc switches off (I.e. the little orange tri-state button on the top goes unset)
02:51 PM pcw_mesa: latency is not terribly important with hardware stepgens/encoders
02:52 PM LeelooMinai: Less than 10 micros? Is that max interval on badse thread?
02:53 PM pcw_mesa: <100 usec is fine for normal configs <500 usec is file if the DPLL is used
02:53 PM LeelooMinai: Mine shows 29 - I guess I should read that wiki too.
02:53 PM pcw_mesa: no base thread normally with hardware I/O
02:53 PM LeelooMinai: Servo one is much higher though - seems like 1 ms.
02:54 PM pcw_mesa: hardware stepgens use servo thread only
02:54 PM LeelooMinai: Right, I don't really get how the servo thread is related to hardware, etc.
02:55 PM LeelooMinai: Er, I guess I read this latency-test dialog wrong - I think it's telling me deviations from pre-set 1ms and 25 micro seconds intervals.
02:55 PM pcw_mesa: LinuxCNCs position waypoints are updated at the servo thread rate
02:55 PM guru_florida: Yes, I have a Mesa card. Regardless, I do recommend reading the Wiki and doing some of the latency stuff...worked wonders for me. I switched to software GL...made the biggest difference. Go figure. lol. HardwareGL is bad on realtime. Higher CPU now but I have a quadcore so no problems...I cant tell the difference.
02:56 PM LeelooMinai: So the driver just executes simple moves, and the thread on the PC is doing all the geometry, etc. and samples at that 1ms interval?
02:56 PM pcw_mesa: also the servo and stepgen control loops are updated at the servo thread rae
02:57 PM guru_florida: I was getting a few "path following errors" as well. Usually on homing. I am using steppers so I dont get how open loop control could have path following errors...I am guessing it is if the waypoints are not updated fast enough and Mesa hardware keeps going.
02:57 PM guru_florida: Those path following errors are gone too since usb cable removed.
02:57 PM LeelooMinai: I have 2.3 μs jitter
02:57 PM guru_florida: Thats very good!
02:57 PM guru_florida: that worse c ase?
02:58 PM LeelooMinai: Right, Max Jitter on the servo thread
02:58 PM {HD}: I have 2.3 x 10k
02:58 PM pcw_mesa: if you get following errors with the USB stuff enabled it sounds like it is interfering with real time
02:58 PM guru_florida: HD try the Wiki. Specifically disable C stepping in cpu, isolcpu if on multiple cores, and software GL.
02:59 PM LeelooMinai: What is C stepping?
02:59 PM guru_florida: Could be I suppose, because with Latency test I cant run the VFD through metal to generate the required EMI to cause issues
02:59 PM {HD}: Ive not done any efficiency modifications
03:00 PM LeelooMinai: {HD}: Er, you have 20ms jitter?
03:01 PM LeelooMinai: I think I have less jitter when moving spindle by hand:)
03:02 PM LeelooMinai: Which wiki page is that btw? Cannot find it.
03:06 PM LeelooMinai: BASE_PERIOD does not apply with hardware mesa drivers, right?
03:17 PM TurBoss: archivist: cool video
03:17 PM TurBoss: the subs worked like a charm
03:17 PM TurBoss: !
03:22 PM guru_florida: Minai: I think so. I have only a 1ms period, the microsecond one is gone. If I recall, that is the SERVO_PERIOD one.
03:23 PM guru_florida: C stepping is a feature of the CPU that allows it to throttle speed down when demand is low. Unfortunately that means higher latency and even more when the C stepping is being changed. So basically disabling C stepping is forcing the processor to stay at the highest CPU speed.
03:23 PM andypugh: This is quite a nice feature of the Autodesk Inventor Prep-for-3D-Printing thing: https://youtu.be/l2doDTG7vRQ?t=2m15s
03:23 PM LeelooMinai: A, ok
03:23 PM andypugh: I don’t know if other packages have the same thing, I would imagine so.
03:25 PM LeelooMinai: I am a poor hobbyist so use free Fusion 360, but it's pretty neat.
03:25 PM LeelooMinai: Not sure about 3D printing, but it has CAM.
03:26 PM andypugh: Fusion + Print Studio is quite nice, but only if you output G-code or have one of the supported printers. If your printer is unsupported and needs x3g then it’s a bit more awkward.
03:26 PM guru_florida: andy: what an awesome feature. Some pre-STL export features. I bought Simply3D and I am really happy with it. Much better prints. Costs money, but I say its worth it for less failed prints and better experience.
03:27 PM andypugh: Though currently the version of Print Studio on my Windows VM instantly crashes. But the one on the Mac works fine.
03:27 PM guru_florida: FYI not implying Simplify3D replaces whats in your vid clip...because it's two different things.
03:41 PM FinboySlick: Well, I might complain about how unproductive and inefficient our workplace is... But I just got handed a free beer.
03:42 PM LeelooMinai: Probably won't help with productivity.
03:43 PM FinboySlick: It's very typical of our management's approach to problems. If we can blame beer for non-productivity, there's no need to fix the real causes.
04:06 PM miss0r: EDM controller/powersupply is nearing completion: http://picpaste.com/pics/20170505_223208-I5NZPMnG.1494016567.jpg
04:08 PM FinboySlick: miss0r: You starting with a sinker, or are you going full blown wire?
04:09 PM miss0r: FinboySlick: sinker. The ambition is actualy just removing taps :) (I know it is overkill). the cabinet holds a PC and touchscreen running advanceHMI and two micrologix 1100 PLCs :)
04:11 PM FinboySlick: miss0r: I guess that depends on how many taps you break a day ;)
04:11 PM miss0r: FinboySlick: This is the content yesterday - just a crude mockup of locations and stuff before I started putting things together: http://picpaste.com/pics/20170503_220730-3fzcplDU.1494016882.jpg
04:12 PM FinboySlick: Plain water as the electrolyte?
04:12 PM miss0r: heh. deionized(?? spell) water
04:13 PM miss0r: i've had some luck extending the lifespan of the water with a fuelfilter and a small circulation pump, that also creates a jet across the area of work
04:14 PM FinboySlick: Hey, if you're going overkill, why not go full blown through-tool water with a spinning head? ;)
04:15 PM miss0r: hehe. I do have a spinning head. And I am actualy thinking about getting brass pipes in small dimentions to serve as electrodes, so I can make through water :D
04:16 PM miss0r: but - I have yet to figure out how to do it in practise. getting water into the electrode while it is spinning... without making a mess of things, that is
04:17 PM miss0r: so far i've only had to order one part for this build online. that is a 0-400v DC to 4-20mA linear converter
04:17 PM miss0r: damn expensive stuff :-/
04:19 PM FinboySlick: miss0r: Magnetic coupling for rotation. You don't need torque and it'll leave you with only one 'seal' towards the part.
04:20 PM FinboySlick: (which can leak as that only helps wash out your cut)
04:20 PM miss0r: yeah. That will be rev. B. :) For the time being I just want to see this though. my shop is a _mess_. When this is done i'll do some cleaning and get back on upgrading it :D
04:21 PM miss0r: thats it for tonight. i'll head to bed. see you around
04:27 PM Deejay: gn8
04:27 PM TurBoss: n8
04:50 PM phipli: JT-Shop2, https://ibb.co/b49UT5
04:52 PM XXCoder: phipli: good, now try EMU ;)
04:53 PM JT-Shop: in here at the moment :)
04:54 PM JT-Shop: nice, I pick mine up every morning and put them in another place while I clean out the brooder
04:55 PM JT-Shop: phipli: did she fly up there?
04:56 PM XXCoder: do you guys dump grass cuttings into chicken cage area?
05:01 PM JT-Shop: pine shavings
05:06 PM phipli: I was sat on the bench - she walked up there and then I stood up
05:07 PM phipli: trouble maker of the flock
05:07 PM phipli: XXCoder, not usually
05:07 PM phipli: they can eat too much grass and give themself problems
05:08 PM phipli: we give them some, but compost most of the grass clippings
05:08 PM XXCoder: malnutution and such?
05:09 PM XXCoder: because I remember that most common cause of rabbit and gerbils is eating just carrots, causing malnution issues
05:09 PM phipli: compacted crops
05:09 PM phipli: they bung themselves up
05:09 PM XXCoder: chop is some internal body part?
05:09 PM phipli: basically the same as what rabbits do to themselves, but with slightly different anatomy
05:10 PM phipli: crop = a storage sack in their digestive system
05:10 PM phipli: chomp = short for the chicken's name
05:10 PM XXCoder: ahh ok
05:13 PM LeelooMinai: Now I know why one of the robot in battlebots was called "Chomp" - it looked like a chicken a bit...
05:21 PM phipli: Nothing would stand a chance with that chicken
05:21 PM phipli: she's a right little madam
05:23 PM Crom: *car has speedometer again... what a pain in the rear end~
05:55 PM LeelooMinai: Any tips on calibration method? I have 1μm test indicator and regular indicator, two 1-2-3 blocks, and, hmm, from precise things that's probably it (plus micrometer, but probably not useful here).
05:55 PM LeelooMinai: I mean axis calibration in linuxcnc
05:57 PM LeelooMinai: If I place the indicator and move 10mm both ways and note the distance difference, is that enough? Or do I have to do some backslash detection, etc.?
05:58 PM LeelooMinai: Also, what can I expect with chinese ballscrews - what is "normal" in terms of errors?
05:59 PM LeelooMinai: I presume that if a ballscrew has bad spacing, it will be proportional along the length, right? Well, at least I hope so.
06:01 PM LeelooMinai: I also have callipers, but, I don't know - I guess I could move the jaw over longer distance, but they are 10μm so, probably would not get any accuracy.
06:01 PM LeelooMinai: accuracy gain that is.
06:03 PM andypugh: Talking Emu/Ostrich I love this advert. : https://www.youtube.com/watch?v=hjKd24UCPYY
06:06 PM andypugh: LeelooMinai: I have never bothered checking the accuracy of my screws. And so far all my parts work.
06:07 PM andypugh: In practice it is very unusual for features 1000mm apart to need to be spaces more accurately than 0.5mm.
06:08 PM LeelooMinai: You are talking about some industrial marine port gantry or what? :)
06:09 PM andypugh: And, it your screw is inacurate to a level that you can measure (unlikely) then you can correct it in the software
06:10 PM LeelooMinai: Er... I find that weird. I mean if I move 10mm on a home-made CNC, I cannot expect it to move 10mm with micron precision...
06:11 PM JT-Shop2: new chicken lesson learned, a flat spot is a roost
06:13 PM andypugh: LeelooMinai: You can, in one direction
06:16 PM andypugh: Hmm, I am almost certain that this claim is bogus: http://americanmachinist.com/beyond-cutting-zone/michigan-company-installs-worlds-largest-milling-machine
06:16 PM andypugh: 60 feet is long, but 12 feet is not very wide
06:22 PM LeelooMinai: andypugh: Just read that Mach3 has features like "screw mapping", so I guess people do even more than simple scale adjustment.
06:23 PM andypugh: LeelooMinai: Look at COMP_FILE http://linuxcnc.org/docs/2.7/html/config/ini-config.html#_axis__lt_num_gt_section
06:25 PM LeelooMinai: Right, ok, so it's there - do not really have means to make a map like this for whole axis though. Well, actually, I have 10 micron linear scales on them... so maybe.
06:26 PM LeelooMinai: Not sure if it's sane though for home-made stuff.
06:27 PM LeelooMinai: Maybe I should first do it, say, 3 times for some axis and see if I get some repeatable curve even.
06:30 PM LeelooMinai: I wish I had something, like a piece of metal, which has length precise to, say, 1 micrometer - like a length standard, not sure what the name is, so I could measure it with the tools I have to verify them.
06:33 PM sync: get some gauge blocks
06:34 PM LeelooMinai: I know about those, but no - those sets have $$$ prices, I just need one 10mm length standard or something.
06:34 PM LeelooMinai: I guess for imperial I could just measure my foot, but I use metric now for CAD.
06:35 PM JT-Shop2: 123 blocks
06:35 PM sync: you know that you can get them in single quantities
06:36 PM andypugh: 123 blocks are good.
06:36 PM andypugh: Or, micrometer length standards
06:36 PM LeelooMinai: I have two 1-2-3 blocks, forgot what they are, but probably 1 mil for cheap ones? That's 25 microns.
06:38 PM LeelooMinai: Hmm, They are KBC and I read now that they are 0.1mil
06:38 PM andypugh: http://www.ebay.co.uk/itm/100mm-Micrometer-Standard-Setting-Length-Calibration-Gauge-Moore-and-Wright-/252708254482?hash=item3ad695f312:g:XiMAAOSwo4pYa53H
06:38 PM LeelooMinai: So 2.5 microns - that's better, but I guess sub 1 micron would be preferable for indicator that is 1 micron by itself
06:40 PM LeelooMinai: Hmm: https://www.amazon.ca/Fowler-52-227-001-1-Individual-Micrometer-Measuring/dp/B00B5HEEZY/ref=sr_1_10?ie=UTF8&qid=1494025804&sr=8-10&keywords=micrometer+standard
06:40 PM LeelooMinai: I guess I can get something like this next time.
06:41 PM phipli: JT-Shop2, https://ibb.co/ccf6rQ
06:41 PM phipli: they look comfortable anywhere
06:43 PM phipli: Hum... I haven't actually talked CNC on here yet - someone mentioned chickens while I was still just lurking
06:43 PM phipli: I do have a CNC machine, I promise :)
06:43 PM LeelooMinai: Why would you talk CNC on #chickens
06:43 PM phipli: :)
06:45 PM andypugh: I have decided that is is now 1920 for 3D printing, if you compare it to cars.
06:47 PM andypugh: You can expect it to work much of the time, but you are still expected to understand the machine. It isn’t 1905 where every print is an adventure and likely to end in failure, and it isn’t 1980 where it is a total surprise if the appliance does not do the job every time despite you having no clue how it works.
06:49 PM Phiplii: as owner of a 1991 car...
06:49 PM Phiplii: every journey can provide a new surprise
06:51 PM malcom2073: I just put a 1991 vehicle on craigslist
06:51 PM malcom2073: Hoping to get a 1992 or newer to trade, like I need more projects
06:53 PM jdh: the analogy might be more apt for cheap 3d printers. real commercial ones seem more like 1990's cars
07:11 PM andypugh: Well, I was rather referring to 1980 cars _in_ 1980.
07:24 PM Cromaglious_: I had good luck with 1980 cars... Ford courier, i pit 438kmiles, 1980 dasher diesel 387kmiles, and a honda 1980 civic 290kmiles
07:28 PM Phiplii: :)
07:29 PM Phiplii: I've lost too many hours up to my elbows in bits of engine
07:30 PM Phiplii: some numpty put dissimilar metals in my engine's pipework
07:58 PM JT-Shop: say goodnight Gracie
08:02 PM phipli: Anyone ever had a disk fail the "self test"
08:03 PM phipli: is it the end of everything?
08:05 PM XXCoder: phipli: backup everything. now.
08:06 PM XXCoder: it might be fine for years even then but good to have backups anyway
08:06 PM phipli: tis an empty disk that I had on the side
08:06 PM phipli: was in a nas box
08:06 PM phipli: old disk - not important disk
08:07 PM XXCoder: ah ok
08:07 PM XXCoder: can use it for nonimportant task maybe
08:07 PM phipli: I'm wondering what happened to it
08:07 PM phipli: I've not had many disks fail over the years
08:08 PM phipli: Might just throw it out as it is only 500gb
08:09 PM XXCoder: yah thres fake chinese flash dirves bigger than that
08:09 PM phipli: yes... but they are actually physically 8gb usually...
08:09 PM XXCoder: theres real 512 gb ones I think
08:09 PM phipli: heh
08:10 PM phipli: how do you ever have the confidence to buy one?
08:10 PM XXCoder: well. just dont buy online
08:10 PM phipli: given the number of fake ones there were a couple of years back
08:10 PM XXCoder: go to frys or something
08:10 PM phipli: That would involve an expensive plane ticket
08:10 PM phipli: wrong continent
08:10 PM XXCoder: buy 1,000 of em
08:11 PM XXCoder: sell it your country. lol
08:12 PM phipli: got one of these the other day :
08:12 PM phipli: http://www.tomshardware.co.uk/samsung-sm961-ssd,review-33585.html
08:12 PM phipli: still can't believe they can make them for the price
08:13 PM phipli: 3100mb/s read speed!!!
08:13 PM XXCoder: yeah
08:13 PM phipli: unnatural I tell you
08:13 PM XXCoder: I read recently that ibm did 1 atom a bit storage
08:13 PM XXCoder: imange the density of information
08:13 PM phipli: sorry.
08:13 PM phipli: I can't
08:13 PM XXCoder: current hard drive and ssd need 100,000 atoms for one bit
08:14 PM XXCoder: one atom to hold one bit but actually need 20 more atoms for changing data and such but still far less
08:14 PM phipli: heh
08:14 PM phipli: it is all quite incredible
08:16 PM XXCoder: indeed
08:16 PM XXCoder: it blows moore's law variant for storage easily
08:17 PM XXCoder: because it is hard to cram storage better
08:17 PM phipli: spinning type disks kind of fell over with progress a while back
08:17 PM XXCoder: I remember one fiction book they used solid stuctured light to store data. it grows as it needs more room
08:17 PM XXCoder: so basically infinite room
08:17 PM phipli: but the speed gain of the various solid state stuff makes them worth while
08:18 PM XXCoder: yeah my pc has 250 gb solid state for programs and os
08:18 PM phipli: Just got a new machine - hence plugging in old unloved drives to see what is on them
08:18 PM phipli: boot drive is that m2 thingy I linked
08:19 PM phipli: will start transferring disks from the last machine soon
08:19 PM XXCoder: cool
08:19 PM phipli: but for now have just been taking the chance (spare bays) to check out unused old disks
08:19 PM XXCoder: I usually use computer for years
08:19 PM phipli: same
08:20 PM XXCoder: my old pc was 9 years old when I replaced it
08:20 PM phipli: 9 years since I last bought a desktop
08:20 PM XXCoder: its now linuxcnc pc
08:20 PM phipli: :)
08:20 PM phipli: hopefully this thing will do me 9 years too
08:20 PM phipli: the last one still isn't too bad
08:21 PM XXCoder: yeah
08:22 PM phipli: New internal card reader arrived in the post today
08:23 PM phipli: USB 3, so (haven't speed tested yet) should be quick for writing SD cards for single board linux machines
08:32 PM LeelooMinai: Ok, so I attached linear scale + 1 micron indicator to X axis and made some movements. Seems that the ballscrew is off about 20 microns per 10mm - that seems a lot to me. Not sure if that's reasonable yet.
08:33 PM LeelooMinai: Or say 1 mil per inch
08:33 PM LeelooMinai: No, 2 mil
08:33 PM phipli: whats that in chains?
08:35 PM LeelooMinai: " ≈ 0.41 × length of a dust mite (≈ 125 µm )"
08:35 PM LeelooMinai: So, if I try to set zero offset to a dust mite, it may be killed.
08:41 PM LeelooMinai: Hmm, seems they are listed on Ali as "C7" class.
08:43 PM LeelooMinai: But THK has C7 as 50 microns per 300mm
08:44 PM LeelooMinai: I guess first I would have to eliminate backlash factor anyways.
08:49 PM XXCoder: new clickspring!
09:00 PM andypugh: XXCoder: New today, or is that yesterdays?
09:00 PM XXCoder: may 3rd one
09:00 PM andypugh: LeelooMinai: What is your error at 50mm and 100mm?
09:01 PM LeelooMinai: andypugh: I think it was kind of useless measurement by now. I should start from measuring backlash really - just watching videos on how to do that.
09:02 PM LeelooMinai: First need to get sense what normalish backlash for Chinese ballscrew is though.
09:04 PM XXCoder: I love the clamp he uses. ancient design
09:04 PM Tom_itx: phipli, are those drives in the TB range yet?
09:04 PM andypugh: Backlash depends on the preload level (P0, P1, P2 etc). There are specs. Cumulative position error depends on class (C7, etc)
09:05 PM andypugh: Did you specify preloaded screws?
09:05 PM LeelooMinai: I did not specify anything, just got a package from Ali:)
09:06 PM andypugh: Have you read the (I guess) GTEN manual? It actually goes in to quite a lot of detail about different types of preload and error classes.
09:07 PM LeelooMinai: I only found out they are supposed to be C7 class
09:08 PM andypugh: I get the impression you are in the UK? Zapp Automation are not a lot more expensive than Ali-express and you get to talk to a human about what you need. http://www.zappautomation.co.uk/r1605-c5-ballscrew-868.html
09:08 PM LeelooMinai: I am in Canada
09:08 PM andypugh: Ah, my mistake
09:09 PM LeelooMinai: Can linuxcnc do anything sensible with backlash compensation for not-simple moves?
09:09 PM andypugh: That does mean that you can walk in to DMM to buy servo drives
09:09 PM andypugh: LeelooMinai: Can _anything_ do that?
09:10 PM LeelooMinai: No idea - that's why I asked:)
09:10 PM andypugh: The motion controller has no way to predict the cutting-force direction. So not really.
09:11 PM andypugh: You can try to always mill conventional, not climb, to minimise the effect
09:11 PM LeelooMinai: Right, so, it will be what it is, or one can somehow improve it a bit?
09:12 PM andypugh: You can use preloaded screws, or double-them up in opposition to get zero backlash.
09:12 PM andypugh: Or, you can pretend you did that, and get on with making stuff, and worry about it if, and only if, you find that there is a problem with what you are making.
09:13 PM LeelooMinai: That's what I probably should do, but OCD is talking.
09:13 PM andypugh: Anyway, in this time-zone, it’s late
09:23 PM LeelooMinai: I have a mist thing ob my Z axis, but for now used it only to blow air. If you had a machine in your room, and wanted to spray only "on demand" by hand, hwat kind of fluid would you use? It's in my bedroom, so cannot really use anything toxic.
09:23 PM XXCoder: LeelooMinai: few possibilities
09:24 PM XXCoder: wd40 distalled water
09:24 PM LeelooMinai: I thought about wd40, but how do I get in a liquid form, not compressed bottle...
09:24 PM XXCoder: shops tend to have those
09:24 PM XXCoder: looks like metal container thing, or larger plastic jugs
09:25 PM LeelooMinai: As to the water, somehow that does not strike me as a good idea - I mean it does not do anything for lubrication for example.
09:25 PM XXCoder: yeah theres such thing as de-ionized distalled water
09:25 PM LeelooMinai: When millin alu
09:25 PM gregcnc: http://www.homedepot.com/p/WD-40-128-oz-1-Gal-Penetrating-Lubricant-10010/100043123
09:26 PM XXCoder: it can still rust metal but slower
09:26 PM LeelooMinai: O, good, WD40 it will be then probably.
09:26 PM gregcnc: I cut a lot of alumium with just vacuum
09:26 PM XXCoder: yeah its best
09:27 PM XXCoder: mister or spray yourself
09:27 PM XXCoder: no cfo
09:27 PM gregcnc: if you dont' want it to stink get lamp oil
09:27 PM XXCoder: isnt lamp oil... flammable?
09:27 PM LeelooMinai: WD40 smells nice at least...
09:27 PM XXCoder: well more so than wd40 anyway
09:27 PM gregcnc: and WD isn't?
09:27 PM LeelooMinai: Never tried putting wd40 on fire somehow...
09:27 PM XXCoder: like I said..
09:28 PM gregcnc: WD40, lamp oil, kerosene are cousins
09:28 PM XXCoder: ahh ok
09:29 PM XXCoder: I was wrong on deionized water also, its bit corrosive
09:29 PM gregcnc: one of the lamp oils i used to buy went green to some kind of bio prduct but should still work
09:29 PM gregcnc: was going to say DIwater is worse that tap
09:30 PM LeelooMinai: Of course Canadian homedepot does not have WD40 in liquid form - typical.
09:31 PM gregcnc: http://www.homedepot.com/p/Klean-Strip-Heat-Odorless-Fuel-GKKH99991/100112454
09:31 PM gregcnc: but i've never used it to say it doesn't stink
09:31 PM gregcnc: I use this http://www.acehardware.com/product/index.jsp?productId=1275451
09:32 PM LeelooMinai: Hmm
09:32 PM LeelooMinai: What about, I don't know, risk of putting the machine on fire?
09:32 PM gregcnc: milk, tallow, bacon drippings it all works
09:33 PM gregcnc: than you have to water based
09:33 PM gregcnc: or high flash point
09:33 PM gregcnc: corn oil, peanut oil, whatever
09:33 PM LeelooMinai: Are any of those things non-toxic when in mist form?
09:34 PM gregcnc: read the MSDS
09:35 PM gregcnc: i've used the lamp oil drilling steel because it was handy and never ignited, but it was on my mind
09:37 PM gregcnc: I gave up on the china mister I had. i found a cheap accu-lube
09:38 PM Tom_itx: you didn't like those $5 misters?
09:38 PM XXCoder: any such mister that dont use air pressure?
09:38 PM XXCoder: or maybe just timed drip thing?
09:38 PM Tom_itx: doubtful
09:41 PM gregcnc: I needed a lot of air to get chips out so maybe that cause the whole room to will with mist
09:41 PM XXCoder: unless you has vaccum and filter setup
09:42 PM XXCoder: nice to have when has cnc router
09:42 PM XXCoder: mill not so muchj
10:08 PM {HD}: I like the look of the fog Buster but I have not tried anything yet.