#linuxcnc Logs

Mar 29 2017

#linuxcnc Calendar

01:36 AM chopper79: Does anybody know if there is a 3D model of a Mesa 7i76?
01:42 AM IchGucksLive: morning from Germany
01:43 AM IchGucksLive: chopper79: are you building an animation
01:43 AM IchGucksLive: why not simple do a rectangle 2d masking plate messurment
01:44 AM IchGucksLive: XXCoder: hi
01:50 AM IchGucksLive: chopper79: still on
01:51 AM IchGucksLive: till later ;-)
02:32 AM Deejay: moin
05:06 AM XXCoder: heys
05:13 AM XXCoder: heh https://img-9gag-fun.9cache.com/photo/a1ML4Zv_700b_v1.jpg
05:35 AM jthornton: morning
05:36 AM XXCoder: yo
05:43 AM Tom_L: hey
05:49 AM Deejay: whoohoo ;)
06:29 AM {HD}: why are gx16-2 more expensive than gx16-4s...ugh!
06:31 AM archivist: because 4 pin is made in bazilions
06:34 AM Longbow: hello, any glide "experts" here?
06:34 AM archivist: glade?
06:34 AM Longbow: I have a custom.glide panel GUI that is working fine. Now I have problems connecting HAL pins
06:35 AM Longbow: yes glade
06:35 AM Longbow: :)
06:35 AM archivist: see error message and any changes in the source
06:36 AM Longbow: How do I connect the buttons in the .glade file to HAL?
06:36 AM Longbow: where do I have to put the "net" command?
06:36 AM Longbow: I tried in the postgui_call_list but it will not connect
06:36 AM archivist: so look at error messages
06:36 AM archivist: wee why
06:37 AM archivist: see
06:37 AM Longbow: I tried with this in my.INI file: EMBED_TAB_COMMAND = gladevcp -x {XID} -H settingsTab.hal settingsTab.glade
06:37 AM Longbow: then if I put a net command in the .hal file the GUI wont load ?! no errors
06:38 AM archivist: there will be error messages somewhere
06:38 AM Longbow: it just wont load my custom panel
06:38 AM Longbow: gmoccapy it is
06:39 AM archivist: start it in a terminal
06:39 AM Longbow: Where am I supposed to place the net commands? in postgui_call_list.hal or in another .hal file?
06:40 AM Longbow: When I was using AXIS gui I just placed them in the postgui_call_list and it worked
06:40 AM archivist: in the right place, not just in one file
06:41 AM Longbow: where is this right place? :)
06:41 AM Longbow: which file?
06:41 AM archivist: I cannot know YOUR situation,
06:42 AM Longbow: so you are saying that both methods should work?
06:42 AM archivist: no I am saying, some research and thinking required for your system
06:43 AM archivist: there is no universal right answer
06:44 AM archivist: not enough info given for anyone in here to make a guess
06:47 AM Longbow: I see
06:47 AM Longbow: if I use the terminal I get an error now
06:47 AM Longbow: settingsTab.hal:2: Pin 'settingsTab.checkbutton4' does not exist
06:50 AM Longbow: there it is: <object class="GtkCheckButton" id="checkbutton4">
06:50 AM Longbow: Am I missing something? :)
06:56 AM Longbow: id == PIN ?
07:04 AM archivist: 'settingsTab.checkbutton4' !="checkbutton4" part of it might be
07:06 AM archivist: reading and diagnosing error message is a required skill these days, for any software not just linuxcnc
07:13 AM Longbow: I figured it out!
07:15 AM Longbow: there is a GtkCheckButton and and there is HAL_checkButton
07:15 AM Longbow: Doh!
07:15 AM Longbow: now works
07:32 AM skunkworks: Yay!
07:33 AM skunkworks: https://www.youtube.com/watch?v=Ss0bzMXl1to
07:46 AM XXCoder: wtf
07:46 AM XXCoder: I guess its holding spindle or something in real would, or would if it did exist
07:46 AM XXCoder: or 3d printer hotend
08:02 AM archivist: optos arrived...attempt n Renishaw arduino OMM
08:06 AM DavenportGuy: skunkworks: u do dis?
08:07 AM skunkworks: DavenportGuy, no.. Just was searching for videos
08:08 AM DavenportGuy: its wonderful whoever did it
08:08 AM skunkworks: The vismach is very well done
08:09 AM skunkworks: DavenportGuy, Iowa?
08:09 AM DavenportGuy: machine
08:09 AM skunkworks: ah
08:09 AM DavenportGuy: Are there any good irc for manual machinists?
08:10 AM archivist: none that I know of except here
08:10 AM skunkworks: Hmm - I don't know.
08:10 AM archivist: some of us do more manual than cnc :)
08:10 AM skunkworks: we only have a couple of lathes and a shaper that is manual... Now..
08:11 AM skunkworks: when we got the matsuura - we got rid of the brown and sharp mill
08:11 AM archivist: aw
08:12 AM skunkworks: (it was pretty bad...)
08:12 AM archivist: I have used an old one
08:12 AM skunkworks: http://electronicsam.com/images/brownsharp/scan0219.jpg
08:13 AM skunkworks: it did have all the bells and whistles.. Jut pretty wore out. we did a lot of work with it.
08:13 AM archivist: at the museum we had a 2a iirc universal
08:13 AM skunkworks: dad got it in the 80's from a local tech school for $100
08:14 AM skunkworks: it was heavy ;)
08:15 AM DavenportGuy: oh lawdy we had those in shop when i was in school
08:16 AM DavenportGuy: they were worn out then
08:16 AM DavenportGuy: worst mill table I ever used
08:16 AM DavenportGuy: in the loose spot the table would wiggle on the ends about an 8th inch if not locked down
08:18 AM skunkworks: yes - this one was tight on the end of the travels - loose in the middle
08:19 AM skunkworks: constantly messing with the gib depending on what you were milling
08:19 AM skunkworks: DavenportGuy, uh - what school?
08:20 AM archivist: you just need a scraper and bucket of elbow grease to get the ways back in shape
08:20 AM skunkworks: we were thinking of throwing it on the K&T and just milling it. (would be way better than it was ;) )
08:21 AM archivist: and a copy of machine tool reconditioning :)
08:21 AM DavenportGuy: highschool, in indiana
08:21 AM skunkworks: ah - this was in western wisconsin
08:21 AM DavenportGuy: probably where we got the machines
08:21 AM DavenportGuy: hahaha
08:22 AM archivist: kids of today will be "trained" of chinese glue guns
08:24 AM skunkworks: :) I remember a local pattern maker said 'kids these days need to make a program for everything!) (talking about cnc)
08:30 AM archivist: I was amused at andy moaning about pattern making and needed a gluegun !!
08:31 AM archivist: I think wood is better and faster for that
08:36 AM archivist: and he moans about time for subtractive milling, traditional is additive with bits of wood...3d :)
08:43 AM gonzo_: some of the younger eng's here were (a couple pof years ago) banging on about 3d, and how it would revolutionalise everything. Couldn't help myself, so I mentioned this even newer tech, where you take a lumop of solid mateiral and cut away, (using a computer controled machine) all the bits you didn't want. They thought that was great!
08:44 AM gonzo_: it does seem that most of the benifits thof 3d pri ting that the non mech people see, are the ideas of local manufacture and sendong of designs rather than finished parts.
08:45 AM archivist: er wot, bits of paper were posted in the olden days
08:47 AM gonzo_: yep, imagine that!
08:48 AM archivist: or sent down the street by messenger boy
08:50 AM gonzo_: I can see 3d being useful for protiotyping or very customised parts, but the materials available seem so limited. And I can't see it being cost effective for mass production
08:50 AM archivist: company I worked for had a messenger "boy" in his 50's/60's
08:50 AM gonzo_: (prob a conversation that has been had in here plenty of times....)
08:51 AM archivist: latest thing since sliced bread will take over the world, except you just cant beat the original
09:43 AM jdh: I got some small parts 3d printed out of 316. cost way too much and surface finish was not good (but not spec'ed either)
09:49 AM jdh: https://imgur.com/yy6OGU3
09:49 AM jdh: guess how much they cost!
09:49 AM archivist: lots, sintered look
09:53 AM jdh: I have seen newer parts they made. better surface finish, but not great.
09:55 AM jdh: I made some mods to them at home on my g0704 and the metal cuts like normal 316
09:55 AM cradek: they kinda look cast
09:55 AM cradek: what did they cost?
09:57 AM jdh: about $500 each
09:57 AM _methods: sweet jeebus
09:57 AM cradek: wow
09:58 AM archivist: I would have filed from solid for less
09:59 AM _methods: hehe
09:59 AM cradek: they do look tricky to machine
10:01 AM archivist: they look a bit like make on the end of a stick and last face is the part off
10:03 AM nallar is now known as Ross
10:06 AM jdh: I made the initial one out of brass shimstock, scissors, pliers
10:06 AM jdh: it is just a gas deflector
10:08 AM archivist: so it could be pressed from annealed ss sheet
10:19 AM jdh: sure
11:09 AM IchGucksLive: hi
11:12 AM IchGucksLive: jdh: what is 326
11:12 AM IchGucksLive: 316
11:12 AM IchGucksLive: PC PS
11:13 AM jdh: stainless steel
11:13 AM IchGucksLive: printed part out of staninless
11:13 AM IchGucksLive: cool man
11:14 AM jdh: this one can print Ti also
11:14 AM IchGucksLive: im off till later 1hr or so Ti is on the university here i will ask them if thex can do also stainless
12:31 PM insomnia is now known as Clownshoes
01:20 PM IchGucksLive: hi
01:29 PM IchGucksLive: the G51 workaround is done in almost 2 lines
01:39 PM IchGucksLive: Gn8
03:08 PM terry_: Hi all
03:09 PM terry_: Trying to add an encoder to the spindle of my mill. Followed example in docs. Mostly works.
03:10 PM terry_: Counter mode, no B channel (currently, did have), On the rising edge of the A channel, it will increment the count by 1. So far so good.
03:12 PM terry_: On the falling edge of the A channel, it will increment the count be some 10s or 100s... usually 30 to a few hundred.
03:13 PM terry_: The hal O-scope shows a clean signal on the A channel. (and B when it was connected.)
03:13 PM terry_: Using a Mesa 5i23 card.
03:14 PM Deejay: welcome to the black mesa transit system
03:14 PM terry_: I've been on quite a ride so far, does it get better?-)
03:22 PM _methods: make sure you grab a crowbar
03:23 PM terry_: Is this a known mesa issue?
03:28 PM DavenportGuy: terry_: is it possible the encoder is the issue? have you tried swapping A and B channels and checking if the problem moves?
03:29 PM DavenportGuy: I never had such an issue on any cards I have
03:34 PM terry_: I tried swapping channels, same.
03:34 PM terry_: In quadrature mode, it works fine.
03:36 PM terry_: Is there any reason that a mill spindle must be in counting mode? as per the example in the docs.
03:36 PM terry_: Could I not run it in quadrature mode, as that works?
03:37 PM jdh: quad would be the normal mode imo
03:37 PM terry_: Oh, good.
03:38 PM terry_: I had just been following the example in the docs, and it said use counting mode, Ach only.
03:38 PM jdh: url?
03:39 PM terry_: one moment...
03:42 PM terry_: http://linuxcnc.org/docs/2.7/html/examples/spindle.html#_spindle_feedback
03:42 PM terry_: I was reading the installed off-line docs, but that looks the be the same.
03:45 PM jdh: that seems to be a fairly specific example for a low count encoder on a parallel port
03:45 PM terry_: Yes. I did not see another example.
03:46 PM terry_: Is there a better example?
03:49 PM jdh: dunno offhand. the HostMot2 page has a lot of raw info
03:50 PM jdh: not example type stuff though.
03:56 PM terry_: Just found: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?SpindleOrientation
04:06 PM terry_: stepping out for 1/2 hour or so ... will leave this open for any comments.
04:25 PM * JT-Shop just noticed it's 5 o'clock at my sisters house
04:34 PM Deejay: gn8
04:41 PM terry_: Thanks everyone, bye.
04:42 PM Simonious: If I've got a part in Solidworks that needs the same cut out of opposite faces and a mirror operation won't work.. how?
04:43 PM kyle____: maybe you can make a second cut feature out of the same sketch?
04:43 PM * Simonious ponders
04:43 PM * Simonious pokes
04:43 PM kyle____: if the cuts are on parallel planes then you should be able to move the start point and end points of the cut
04:44 PM Simonious: they are parallel
04:44 PM kyle____: don't have SW in front of me right now but I think that's doable?
04:44 PM Simonious: I'm pretty novice in SW, but I'll poke atit
04:45 PM andypugh: Sometimes the modeller just can’t do a mirror or array. (Well, in Inventor at least).
04:45 PM Simonious: I think most of the weak link here is between the keyboard and the chair.
04:46 PM andypugh: Sometimes starting the feature again in a different way makes it work <shrug>
04:46 PM Simonious: screw it.. I'll just draw the sketch on both sides
04:46 PM Simonious: thought maybe I'd learn something, but it isn't cooperating
04:47 PM andypugh: So, say I am machining the casting that is the body of this part.
04:47 PM andypugh: https://goo.gl/photos/u7rME6J62QYkCkcb8
04:48 PM Simonious: This should be trivial, but the tool acts as on obstacle. In fairness I learned how to 3d model in sketchup, so I'm wired to think differently.. it'd be trivial there.
04:48 PM andypugh: I have a horizontal milling machine.
04:49 PM andypugh: If I machine it all square and make the large bore, how can I then set it up to machine the alignment slot in the base to be absolutely perpendicuar and central to the large bore?
04:49 PM roycroft: solidworks has a fairly steep learning curve
04:49 PM roycroft: but once you get it it makes total sense
04:49 PM roycroft: and is very easy to work with
04:49 PM Simonious: thanks roycroft.. and I don't use it every day
04:50 PM Simonious: I like it a lot, it can dance circles around things that sketchup gives up on.
04:50 PM andypugh: This is pretty much exactly the same puzzle as drilling a radial hole directly in the centre of an existing bore on a vertical mill.
04:50 PM Simonious: but.. I struggle with it.
04:52 PM kyle____: Simonious, you could do that actually
04:52 PM roycroft: i don't use it every day either
04:53 PM kyle____: make the cut, start a new sketch on the other side and just convert entities
04:53 PM roycroft: and i sometimes still struggle with it, although i've been using solidworks for several years
04:53 PM kyle____: or whatever it's called; there's a button that projects shapes onto your sketch plane
04:54 PM Simonious: kyle____: hehe, that sounds like a good way to go.. I'm done with it for now and moving on. just used two sketches.
04:54 PM Simonious: it's a shame
04:54 PM Simonious: makes for more work in the future if I change things
04:55 PM kyle____: meh, unless you're working on a big team, it doesn't matter too much how you get it done
04:55 PM kyle____: I think the converted entities might be constrained to the geometry they were created from...? probably wrong
04:55 PM kyle____: but maybe you won't have to do too much
04:55 PM kyle____: wait, yeah, they are
04:55 PM kyle____: if you change the original sketch or cut feature, it should update the other sketch at least
04:55 PM kyle____: just not the cut
05:25 PM Tom_L: you can't project the cut to the other face?
05:27 PM Tom_L: or at least the profi.e
05:27 PM Tom_L: l
05:36 PM Tom_L: Simonious, tie the 2 sketches together with parameters
05:39 PM Tom_L: you could change your whole part around using parameters if it's something you will make alot of
09:25 PM Cromaglious_: I so need to make some wipers for the cnc 3040... and plan on ball screw upgrades
09:44 PM Cromaglious_: doing a 5x g0x270; g0x0 rep.. then homing... X was out 0.088mm
09:46 PM CaptHindsight: Cromaglious_: 3040 like these? https://imagebin.ca/v/2TYvESEACWSs
09:47 PM pfred1: sometimes my machine doesn't home right but it runs OK
09:48 PM pfred1: I've got no idea how it manages that either
09:48 PM CaptHindsight: pfred1: why doesn't home? bad switch?
09:48 PM Cromaglious_: CaptHindsight, basically... no front overhang, striaght up gantry sides
09:50 PM CaptHindsight: out of the box these came with red oxide coated fasteners https://imagebin.ca/v/2aq2qZUtbOIM
09:51 PM {HD}: lol, nicw.
09:51 PM {HD}: nice*
09:51 PM CaptHindsight: and >1mm of backlash in the ballscrews https://imagebin.ca/v/2aq3eWhTxbGY
09:51 PM {HD}: ick
09:53 PM CaptHindsight: https://imagebin.ca/v/2apzmk7rpwvC but after some rework, cleaning and anodizing they are ok https://imagebin.ca/v/2apzmk7rpwvC
10:10 PM os1r1s: Does 2.8 have the new conversational screens?
10:25 PM Cromaglious_: I would love Tnuts made for the table
10:26 PM Cromaglious_: os1r1s, i haven't found them running 2.8.1 here
10:26 PM Cromaglious_: using gmoccapy as well
10:26 PM os1r1s: Cromaglious_ I thought that was one of the big features?
10:31 PM Cromaglious_: os1r1s, look for NGCGUI TCL scripts
10:33 PM pfred1: popcorn T nuts!
10:33 PM pfred1: I've made T nuts
10:33 PM pfred1: for those odd thread sizes
10:35 PM os1r1s: Pathpilot is very pretty
10:35 PM os1r1s: Quite different than Axis
11:21 PM Cromaglious_: 6x more 0.064 out
11:25 PM {HD}: Well, fix it!
11:38 PM Cromaglious_: {HD}, I am... waiting for them to show up
11:38 PM Cromaglious_: NGCGUI is neat...