#linuxcnc Logs

Jun 30 2021

02:03 AM Deejay: moin
04:53 AM Tom_L: morning
04:53 AM XXCoder: yo
05:08 AM JT-Cave: morning
05:08 AM XXCoder: o/
07:27 AM drdoc: veegee_: don't y'all have Craters & Freighters?
08:54 AM perry_j1987: good morning
11:57 AM captain_morgan: today might be the day! first cut on my Avid. Did a bunch of reading last night and think I understand how to setup the dual motor/switch gantry homing, which I think was the last thing on my list....
12:06 PM Bleepshop: Anybody here use FreeCAD? I'm having an issue where the cursor coordinates, the unit and the grid are not working as expected...
12:28 PM Bleepshop: Aha! The the grid is NOT what I told it to be [0.5"], the units are in inches but the cursor coordinates are in milineters... WTH ???
01:13 PM rs[m]: I use freecad occasionally, but I never had any need for a grid or cursor coordinates
02:31 PM miss0r: re
02:42 PM jthornton: you get your bed frame figured out?
02:54 PM miss0r: jthornton: yeah. So much so that me and 7 friends are taking it and another car for a roadtrip & campingrun tomorrow at noon
02:59 PM jthornton: how did you solve the hinge support?
03:00 PM miss0r: man this will be hard to make understandable over IRC. but I will give it a go :D
03:01 PM miss0r: the issue was I have a lower part that only has three legs. and once that part is moved out from the wheel box it sits on in the remainding corner, it has no support there. The back then folds down and gets supported on the wheel box, but the issue was to support that one hinged corner.
03:02 PM Tom_L: miss0r, pic is worth 1k words
03:02 PM jthornton: aye
03:02 PM miss0r: Under that corner, mounted to the lower part I installed a hinged support arm, that is spring loaded, so the one end of it always runs along the top of the wheel box and in the fully extented state, it is located underneath both the lower part and the top part along the support frame in one side
03:03 PM miss0r: yeah, but getting a picture now is a nogo, as the car is already quite full of stuff for tomorrows trip
03:03 PM Tom_L: he's takin the 1k word route :)
03:03 PM miss0r: hehe
03:05 PM miss0r: I will follow up with a photo of the stuff when I get back then :)
03:05 PM jthornton: ok
03:05 PM miss0r: I did take a few photos of my rear sparewheel mount, though: https://imgur.com/a/FFSI57h
03:05 PM jthornton: have a good trip
03:06 PM miss0r: I think we will. the other car on the trip is a 1965 volvo amazon, in horrible condition. and I must say it warms my heart that there is a chance my car won't be the most unrelieable one on the trip :D
03:14 PM * Tom_L starts on another cad file
03:16 PM jthornton: then it will be a real adventure
03:17 PM Tom_L: you've had a couple of those 'adventures'
03:26 PM jthornton: aye, for sure
03:31 PM jthornton: I need to invent an adjustable blade holder for the stripper
03:32 PM Tom_L: you must really have alot of wire
03:33 PM jthornton: down to 4-5 boxes actually
03:41 PM jthornton: most of the big long stuff is stripped
04:10 PM Eric__: I'm going to have to filter "code of conduct" into the trash, aren't I?
04:11 PM Eric__ is now known as unterhausen
04:22 PM captain_morgan: if it'll stop you from coming to a public forum and complaining about a document that basically says 'dont be a jerk or make assumptions about people' please do
04:25 PM unterhausen: I would do it because of the people that want to reserve the right to be a jerk
04:27 PM unterhausen: I know I have been a little unfriendly to people that have spent years on the ML talking about how someone should make it more like what they want
04:28 PM unterhausen: although the main culprit has now gotten over that and embraced the existing paradigm
04:28 PM roycroft: so i don't knwo what's going on with this new policy, but proofpoint are flagging all the posts about the new policy as spam, while continuing to pass all other posts from emc-users to me as normal
04:28 PM unterhausen: that's funny
04:28 PM roycroft: i wasn't aware of this until i checked my spam quarentine a while ago looking for something else
04:28 PM roycroft: and i found dozens of posts to emc-users about the new behavior rules
04:29 PM unterhausen: the policy didn't seem like it required any discussion to me
04:31 PM captain_morgan: and here we were.. not talking about it until you seemed to need to discuss it
04:32 PM roycroft: the policy apparently does not apply here anyway
04:34 PM enleth: oh, that point about "not answering beyond the scope of the question" is gone, isn't it?
04:35 PM unterhausen: yeah
04:35 PM unterhausen: I thought some of the complaints about the ML were interesting, I hadn't noticed
04:36 PM unterhausen: I don't think it's particularly unfriendly. Newbies do occasionally ask questions that would be difficult to answer on there
04:38 PM enleth: right now the code looks very reasonable to me - initially, though, there was this veiled passive hostility vibe in it that I feel in the way many codes of conduct are worded
04:39 PM enleth: not sure if what I mean is clear, it might be a very subjective feeling
04:40 PM enleth: but there's this whiff of "behave just *so*, or else" in those
04:59 PM unterhausen: the silly thing is there basically isn't anything the community can do to someone
05:13 PM unterhausen: who is the author of the machinekit 9 axis trajectory planner
05:14 PM unterhausen: I'm curious if I have the mk email list discussion of it
05:19 PM andypugh_: unterhausen: That would be Rob Ellenberg
05:20 PM unterhausen: thanks
05:20 PM andypugh_: Here's a rough breakdown of the differences between the Tormach fork of LinuxCNC and upstream. The Tormach fork happened in the 2.6 era, but a lot of the changes were ported to 2.7 as well (and the state tag work for 2.8). Currently, the Tormach fork is similar to LinuxCNC 2.7, but with some additions:
05:20 PM andypugh_: Extended work offsets (currently 500 offsets)
05:20 PM andypugh_: Blending across all motion types / axes (using bi-arcs)
05:20 PM andypugh_: State-tag / restore on abort changes (so that the interp state matches what the user sees as the last line when stopping a program early). This may be in 2.8 w/ the recent state tag port.
05:20 PM andypugh_: Finer single-block stepping (related to state tag work, may already be in 2.8)
05:20 PM andypugh_: Rigid tapping overshoot compensation (using known spindle acceleration to offset the hole depth)
05:20 PM andypugh_: Fanuc-style lathe tool / wear offsets
05:20 PM andypugh_: Easier start-from-line (handles cutter comp, tool changes, offsets, lead-in moves) with toolpath preview
05:20 PM andypugh_: Customized the 4th axis toolpath display for Tormach hardware (specify a fixed rotation center, toolpath / live plot / XYZ limit box rotates to emulate what a user would see on a typical XYZA mill)
05:20 PM andypugh_: Context for G-code error messages (gcc-like stack trace through G-code subroutines and remaps with source file / line number / caller)
05:20 PM andypugh_: Code polish in various places:
05:20 PM andypugh_: Replaced hand-coded coordinate transforms in interp / canon with a simple transform library
05:20 PM andypugh_: Performance optimizations in emcmodule (ported geometry calculations to C++)
05:20 PM andypugh_: Separate toolsetter / probe inputs in motion (since both can be connected at once on recent machines)
05:21 PM andypugh_: unterhausen: I am not sure if MK has it. Pathpilot has it.
05:21 PM unterhausen: okay, thus my confusion
05:21 PM * Tom_L wonders why you need 500 work offsets
05:22 PM unterhausen: andypugh, so we should be able to request the code from tormach?
05:25 PM andypugh_: Yes. And, in fact, I got the impression that Rob was going to look at how hard a merge would be.
05:51 PM perry_j1987: whats new
05:51 PM Tom_L: nothing under the sun
05:51 PM XXCoder: you
05:52 PM perry_j1987: sitting infront of air conditioner at the shop trying to cool down
05:55 PM jthornton: getting ready to make pizza
05:56 PM * jthornton eyeballs the dwindling pile of wire to be scrapped
05:56 PM Tom_L: yum
05:56 PM XXCoder: its fairly cool here now'awesome
05:56 PM XXCoder: cloudy
05:56 PM perry_j1987: no longer 119 heh
05:56 PM perry_j1987: that was crazy highest i remember on the west side of mountains was 109
05:57 PM perry_j1987: back in late 90s or very early 2000s
05:57 PM XXCoder: was 112 here but yeah insane. 112f is just too much lol
05:57 PM XXCoder: broke records
05:57 PM XXCoder: and not by a little either
05:58 PM XXCoder: unfortunatley it means smokey hellscope is coming earlier this year
06:00 PM jthornton: the weekend looks cool here
06:01 PM XXCoder: here sat will have small spike of temp to 88f besides that nice 7s-low 80s
06:01 PM jthornton: there will be some BlueWing miles put on this weekend
06:01 PM jthornton: low 80's and sunny here
06:13 PM andypugh_: It’s been raining for weeks here...
06:48 PM Roguish: andypugh_, could you share an example of the lincurv comp? i'm having a heck of a time with my spindle vfd.
06:57 PM Roguish: andypugh_, ok, just read through the comp itself. think I get it. looks like it's a set of up to 16 pairs of points, with a linear interpolation between the closest 2 points to the given input point. correct?
07:18 PM andypugh_: Yes, that’s right
07:19 PM Roguish: andypugh_, thanks. thought so. boy the spindle with vfd setup and tuning information sure is spread out..
07:20 PM andypugh_: Night all
07:37 PM -!- #linuxcnc mode set to +v by ChanServ
08:29 PM -!- #linuxcnc mode set to +v by ChanServ
09:41 PM veegee_ is now known as veegee