#linuxcnc Logs
Mar 08 2023
#linuxcnc Calendar
01:23 AM Deejay: moin
03:55 AM travis_farmer[m]: G'Morning ☕️
04:05 AM sensille: my 6040 is slightly out of square. is there a good reason to fix that in hardware (very hard) and not compensate for it in software?
04:51 AM Tom_L: morning
04:52 AM jpa-: sensille: spindle vs. X/Y tramming cannot be fully compensated in software (it will show up as staircase shape when facing with larger endmills), but for other small out-of-squaredness matrixkins should work fine
04:55 AM rmu: jpa-: you mean spindle axis not parallel to z axis
04:56 AM sensille: i meant just x/y squareness
04:56 AM jpa-: rmu: no
04:56 AM sensille: and matrixkins is already the answer to my followup question, thanks :)
04:57 AM jpa-: spindle axis vs. Z axis paralleness can be compensated fairly well by moving X/Y by a small amount when Z changes
05:01 AM rmu: jpa-: then you could in theory flatten your bed to be perpendicular to the spindle and compensate the rest in software
05:01 AM jpa-: yeah, that is true
05:20 AM -!- #linuxcnc mode set to +v by ChanServ
05:24 AM JT-Cave: morning
05:24 AM JT-Cave: fedex now says the modem is in transit as of 2pm yesterday...
05:25 AM JT-Cave: now what did Tom_L tell me yesterday
06:41 AM Lcvette[m]: JT-Cave: when it says its in transit.... its a not wrong situation.. it is in fact in transit, but they have lost track of where in transit it is and when it is next scanned they will know and update. likey got tossed on a wrong truck
06:42 AM Lcvette[m]: seems to happen kinda frequently last couple yeara
06:42 AM Lcvette[m]: s/yeara/years/
06:51 AM * travis_farmer[m] gets some breakfast 🍴
06:59 AM JT-Cave: Unfortunately we are unable to retrieve your tracking results at this time. Please try again later.
07:00 AM JT-Cave: yup they lost it...
07:02 AM travis_farmer[m]: probably a disgruntled FedEx driver tossed it in the woods somewhere
07:06 AM Tom_L: JT-Cave, you find my notes?
07:08 AM JT-Cave: no
07:08 AM JT-Cave: yes
07:08 AM Tom_L: circle one?
07:36 AM JT-Cave: aye the issue tracker will work well as you test
07:37 AM JT-Cave: just did the jog slider label updates
07:40 AM JT-Cave: and fixed the 7i87 groupbox size
07:40 AM * JT-Cave heads to the shower to start his chicken day
07:41 AM Tom_L: i'll check on it later this AM
07:54 AM unterhaus_: My last fedex shipment, they didn't even admit they had the package until the day of delivery
07:55 AM unterhaus_: Something is going on with them. Although that was an amazon shipment. OTOH, the post office never does that with amazon shipments
07:57 AM Lcvette[m]: Tom_L: which mesact are you testing?
07:57 AM unterhaus_: mesact2
07:59 AM unterhaus_: no reason to try it at this point
08:57 AM -!- #linuxcnc mode set to +v by ChanServ
09:11 AM JT-Shop: damn modem has been in Memphis since the 5th...
09:12 AM unterhaus_: frustrating
09:41 AM JT-Shop: I filed a ticket at starlink and they said give it 24 hours and if it doesn't arrive they will ship me another one...
09:41 AM roycroft: my phones sat in mpls for over a week because of the "weather", which from mpls to spokane was dry and unseasonably warm
09:42 AM roycroft: i'm beginning to think that logistics companies just don't care at all about deadlines any more
09:42 AM JT-Shop: I cut the thing in two again and used the jointer to slowly remove the mahogany strip checking for square often
09:42 AM roycroft: does it line up now?
09:42 AM JT-Shop: last bit with my hand plane and my sanding block
09:43 AM JT-Shop: still have the other side to clean up then size the replacement strip
09:43 AM roycroft: well i'm glad you're fixing it instead of tossing it in the firewood bin
09:44 AM JT-Shop: thanks for the suggestions I was really bummed out about it
09:45 AM roycroft: yeah, that's a lot of work and lot of material to toss
09:49 AM unterhaus_: not that I was ever a perfectionist, but making guitars made it so I never give up because of a mistake
09:49 AM unterhaus_: I used to think I was doing it wrong and then gibson ran an ad showing someone in their custom shop repairing a new guitar
09:50 AM unterhaus_: They didn't mention that's what the guy was doing, but it was
09:50 AM unterhaus_: he also didn't look particularly happy
09:52 AM roycroft: no snow on the valley floor yet
09:52 AM roycroft: but schools are closed or on late start all around here
09:52 AM roycroft: and we may still get snow later today
10:07 AM unterhaus_: today and tomorrow are supposed to be nice and then 3" of snow on friday. More snow on Monday and more on next Friday
10:08 AM unterhaus_: Wish I hadn't looked
10:08 AM unterhaus_: I'm glad I can't see the 15 day forecast
10:22 AM JT-Shop: I "think" the first piece is ready to glue in... I'll take a break and look it over again close before opening the glue bottle
10:24 AM JT-Shop: good news is the return "kit" is on the radar now and tracking shows it's in transit back to starlink
10:26 AM roycroft: we're going to get up to seasonal temperatures by this weekend
10:26 AM roycroft: and we'll have the occasional sunny break as well
10:33 AM JT-Shop: friday is supposed to be somewhat nice
10:47 AM * JT[m] uploaded an image: (346KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/DqfeXVHZjizxuBXEjArllBRJ/20230308_104534.jpg >
10:47 AM JT[m]: Second attempt PASS
10:47 AM JT-Shop: there is actually two wrong strips in the upper one but only one sticks out...
10:50 AM roycroft: that looks good
10:50 AM roycroft: congratulations
10:51 AM JT-Shop: thanks a lot for pushing me
10:51 AM roycroft: i'm happy to do it :)
10:51 AM roycroft: i've been working on a new design method in solidworks
10:52 AM roycroft: i've been doing mostly bottom-up design
10:52 AM roycroft: but i've been watching some interesting top-down design videos, specific to cabinet making, where the designer creates a bunch of reference planes as a skeleton
10:52 AM roycroft: and then snaps all the parts to the planes
10:53 AM roycroft: it makes it really easy to resize the cabinets
10:53 AM roycroft: it takes a different way of thinking, but it's a good method, i believe
10:54 AM roycroft: at first i thought it would apply to the lamp designs i'm doing, as there will be multiple variations of the same basic design
10:54 AM roycroft: but i'm thinking now it might also make my cnc router a lot easier to design - i won't have to go editing a bunch of parts every time i make a minor change to one part
10:55 AM roycroft: that's been a big hassle, and if i'm honest, a reason i haven't focusted as much on it as i should
10:57 AM JT-Shop: yup if your designing parts that need to change size it's a different method than designing an assembly that's static
10:58 AM Tom_L: roycroft, or make them table driven
10:58 AM Tom_L: pretty easy to do
10:59 AM * travis_farmer[m] is rendering a new YouTube Video, making chips! :-)
11:02 AM Tom_L: corn or potatoe chips?
11:02 AM travis_farmer[m]: wheat ;-)
11:02 AM sensille: or pringles == wood chips
11:04 AM Loetmichel: roycroft: sounds like how Solidworks operates. if you set the right restrictions to your draft lines that you use to make the 3d model you can resize one line without the rest of the model falling apart. It will be moved accordingly.
11:12 AM roycroft: i've made table-driven drawings
11:12 AM roycroft: and assemblies
11:13 AM roycroft: all those hardware boxes that i 3d printed last year were table-driven
11:13 AM roycroft: and as simple as those are, it was easy to do
11:13 AM roycroft: i've found it a bit harder to keep track of things when doing semi-complex assemblies, though
11:14 AM roycroft: plus, my table-driven stuff has still been bottom-up designs
11:15 AM roycroft: coming into this with decades of 2d autocad experience, bottom-up is the most natural way to design, but top-down will be much more efficient once i get used to that approach, i think
11:26 AM travis_farmer[m]: YouTube is processing the video...
11:27 AM unterhaus_: roycroft, which laser did you buy?
11:27 AM roycroft: i got a laserpecker 1 pro
11:28 AM roycroft: it was cheap and very portable
11:28 AM roycroft: i just got the tripod mount - i did not want to pay extra for auto height adjusting stand or the filters or the fancy case
11:29 AM roycroft: hopefully it will work decently
11:29 AM roycroft: enough to get me going for a while
11:29 AM travis_farmer[m]: https://www.youtube.com/watch?v=qU_rFcrkBM0 my full voice-over didn't seem to record... so i added music and speed up the boring sections, so it wasn't a 45 minute video. :-)
11:29 AM roycroft: it was just a few dollars more than buying a electric branding iron
11:54 AM unterhaus_: ha, the clock downstairs is an hour ahead and I forgot. It sets itself
12:05 PM -!- #linuxcnc mode set to +v by ChanServ
12:17 PM roycroft: it's jumping the gun a bit, isn't it?
12:17 PM roycroft: summer time does not start until this weekend
12:19 PM roycroft: and in interesting time news, since we've segued into that, an international commission are looking into establishing an official time zone for the moon, so that the various space missions to the satellite will be better able to coordiate their doings
12:20 PM roycroft: since a lunar day is ~28 earth days, i wonder how many hours there will be in a day in that time zone
12:20 PM roycroft: or perhaps the length of an hour will be different
12:22 PM roycroft: also, clocks run faster on the moon than on earth
12:22 PM roycroft: because of the gravitational time warp
12:33 PM unterhaus_: no wonder all the people that live on the moon are always feeling rushed
12:34 PM roycroft: i'm also not sure about the concept of months on the moon, since there are only ~13 lunar days in one earth year
12:34 PM roycroft: and a lunar year is the same as an earth year
12:34 PM roycroft: so i think we can skip the month bit
12:59 PM unterhaus_: I got to go change the brake pads on my bicycle. I used them long enough I wore out the spring and it's making scary noises
01:00 PM roycroft: stoppage is an important feature on a bicycle
01:02 PM unterhaus_: they are on the rear, so maybe a little less important. Scary noises is the worst part
01:22 PM unterhaus_: okay, I'm glad I work on my bikes myself, there would be pictures of my (lack of) brake show on reddit
01:22 PM unterhaus_: the one side had no pad material left
01:49 PM XXCoder: JT-Shop: nice work
01:54 PM XXCoder: roycroft: more people noticed lack of patreon updates
01:56 PM XXCoder: looks like it might be simply because there was no patreon only video and updates lately
02:00 PM JT-Shop: thanks
02:01 PM JT-Shop: second repair is being glued up... one more easy repair then time to resaw
02:14 PM unterhaus_: harbor freight gave up on me and won't send me coupons
02:18 PM Rab: Hmm, I now realize they've stopped for me as well. I think I just ran out of stuff that would merit a drive to HF, coupon or no.
02:21 PM unterhaus_: maybe it's because they are having a parking lot sale. I'm going to be next to there, so I'll have to check it out
02:29 PM roycroft: i still get their coupons, but i have an hf credit card
02:30 PM roycroft: i got it when i bought my winch, which was on sale already, and getting the cc got me an additional 10% off
02:34 PM unterhaus_: we have a new store and it doesn't even smell like rubber tires
02:36 PM unterhaus_: I just bought something using paypal, and the website totally garbled my address
02:36 PM unterhaus_: the vendor's website
03:40 PM JT-Shop: a guy is coming up Saturday to buy my 48" box and pan brake!
03:52 PM * JT[m] uploaded an image: (341KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/QoHiHXiaLsDEkSsapBCPBvzb/20230308_155131.jpg >
04:01 PM roycroft: i had one of those 3-in-1 sheet metal machines, 36" wide
04:01 PM roycroft: it mostly took up space in my fabrication shop
04:01 PM roycroft: i finally sold it, and i'm very happy to have done so
04:05 PM JT-Shop_ is now known as JT-Shop
04:05 PM JT-Shop: I have one as well and it needs to go too
04:15 PM XXCoder: roy sheet metal machines? bender or?
04:27 PM roycroft: bend/shear/roll
04:27 PM roycroft: it's a nice idea
04:27 PM roycroft: to be able to do all three things with one machine
04:27 PM roycroft: the problem is that it doesn't do any of them well
04:27 PM roycroft: the finger brake is probably the best part of it
04:28 PM XXCoder: what seems mitutoyo mics now have plastic thumble now
04:29 PM XXCoder: roy ouch jack of all trades, master of none
04:29 PM XXCoder: https://youtu.be/--ZGbiejZ84 (video about mics)
04:29 PM JT-Shop: did that photo of the horse cart show up?
04:30 PM travis_farmer[m]: yes on matrix, i think
04:30 PM * JT-Shop takes 5 after gluing the last fix on the chunk
04:30 PM JT-Shop: don't look like it made it here
04:31 PM * JT[m] uploaded an image: (341KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/wvebLGdeYQFZNVQcuErctlrj/20230308_155131.jpg >
04:31 PM travis_farmer[m]: yeah, that one
04:31 PM JT[m]: Try again
04:31 PM JT-Shop: there it is... got the basic frame tacked together for the rumble seat
04:32 PM travis_farmer[m]: i think the bot had a bit too much to drink... it has been a little slow to post lately ;-)
04:33 PM roycroft: the person i sold mine to is a metal sculptor
04:33 PM roycroft: he does mostly free-form stuff
04:33 PM roycroft: the 3-in-1 machine will probably be fine for him
04:35 PM XXCoder: interesting
04:40 PM mits[m]: Does anyone ever use fixturing pins with absolute machine coordinates so you can recut them into materials easily?
04:41 PM roycroft: i suspect lots of people do
04:41 PM mits[m]: it's really the first use case I've thought of for absolute machine coordinates in a program
04:41 PM ZincBoy[CAON][m]: I use fixture pins with work offsets.
04:42 PM ZincBoy[CAON][m]: Absolute coordinates are for things like tool changes or tool setters.
04:42 PM mits[m]: hmm
04:42 PM ZincBoy[CAON][m]: Not usually a good idea to reference an absolute offset in a normal program.
04:42 PM mits[m]: how would that work, if you move the origin of your work then the fixture pins move, right?
04:43 PM ZincBoy[CAON][m]: I have multiple sets of fixtures, each one would be located at a work offset. The work is always relative to the pins. The pins can then be probed to set the work offset.
04:44 PM mits[m]: oh
04:44 PM mits[m]: is that a feature of something in gcode native, or do you have gcode "macros" built up to do that work?
04:45 PM ZincBoy[CAON][m]: For example, if you have a pin at 0,0,0 in your work. The pin can be anywhere in your work envelope. Once you set your G54/55/56... to 0,0,0 at that pin location your work will be machined correctly.
04:45 PM ZincBoy[CAON][m]: No macros. Just the work offsets.
04:45 PM mits[m]: ok
04:51 PM Tom_L: practically everybody uses work offsets in industry
04:54 PM ZincBoy[CAON][m]: Yes, and the limit of 9 offsets in linuxcnc is a bit limiting. Makes it harder to do arrays.
04:54 PM JT-Shop: I remember asking way back when why you couldn't use g53...
04:54 PM XXCoder: workoffsets everywhere
04:55 PM travis_farmer[m]: i used work offsets just today... :-)
04:55 PM XXCoder: I remember at job before my last one, they had large slab of alum with keystone sides.
04:55 PM XXCoder: many many sheet parts use that
04:55 PM XXCoder: no need to change coord, just place sheet and run
04:57 PM Tom_L: G53 might be more practical on a very small machine
04:58 PM JT-Shop: I use g53 on the plasma to return the torch to z0
04:59 PM XXCoder: some machines use specific gcode to move in machine coord
04:59 PM XXCoder: not g53
04:59 PM ZincBoy[CAON][m]: It still limits you with zeroing your work. On my old Taig I only ever used G54 but it was still critical to indicate in work.
05:01 PM Tom_L: would have to be on every line of gcode too
05:01 PM JT-Shop: yup
05:02 PM XXCoder: yeah I prefer work offsets
05:02 PM XXCoder: you dont worry about placement when doing cam
05:02 PM XXCoder: just remember its zero point
05:03 PM XXCoder: and setup you set g54 or whatever and test it, run part thats it
05:03 PM XXCoder: oh you need new part? just do setup again. dont need to touch cam
05:07 PM Tom_L: i have 2-3 parts set up on a plate and all i do is set G54 on the corner of the plate
05:08 PM XXCoder: multipart with single coord certainly is fine, though always be sure if set again, its at same places
05:08 PM XXCoder: (relivate to g54)
05:14 PM XXCoder: man latest clickspring is cool
05:14 PM XXCoder: (clips)
05:25 PM Tom_L: the offsets are in the cad as well for each part
06:57 PM redlegion_ is now known as redlegion
06:57 PM rene-dev56 is now known as rene-dev5
08:12 PM mits[m]: Does anyone understand what's going on here?
08:12 PM * mits[m] uploaded an image: (126KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/aZIhswljWPcRalxyuPgjOPTI/image.png >
08:12 PM mits[m]: Program is stopped, MDI, G0 X0 Y0, when I hit Go this happens
08:13 PM * mits[m] uploaded an image: (10KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/XuYMSwAvfZFBnTfFdlyFZKGu/image.png >
08:14 PM XXCoder: the position of head + tool length gets it too far up
08:14 PM XXCoder: typical reason for that problem
08:14 PM ZincBoy[CAON][m]: Your gcode causes moves that go beyond the travel limits.
08:14 PM mits[m]: I get that, but I'm not moving Z
08:15 PM mits[m]: Nor am I running the program, so why is it saying it's a problem with line 9403
08:15 PM mits[m]: which happens to be an M2
08:15 PM ZincBoy[CAON][m]: You have z moves in the program. Something isn't happy. Sometimes the parser doesn't show the error on the line that it occurs.
08:16 PM ZincBoy[CAON][m]: Any remaps of gcode?
08:16 PM mits[m]: but I'm not running the program
08:16 PM XXCoder: it happening on M2 sounds like end of program script issue, like parking after program is done
08:16 PM mits[m]: None that I'm aware of
08:16 PM XXCoder: not sure then
08:17 PM ZincBoy[CAON][m]: When you load the program the parser runs the gcode virtually. That is how the backplot works.
08:18 PM ZincBoy[CAON][m]: But I agree that it would most likely be a end of program issue.
08:18 PM mits[m]: This program also ran correctly and had no limit problems until I tried to do this G0 X0 Y0 after finishing
08:18 PM mits[m]: I'm not sure what I can do... is this a "restart linuxcnc to fix it" sort of situation? is there no other way to clean up?
08:19 PM ZincBoy[CAON][m]: You have G0 X0 Y0 in your end script?
08:19 PM mits[m]: no
08:19 PM mits[m]: I'm typing it into the MDI
08:20 PM mits[m]: you can see it in the screenshot
08:20 PM ZincBoy[CAON][m]: Try G53 G0 X0 Y0
08:20 PM mits[m]: do you mean G54? I'm not trying to G53 this
08:20 PM mits[m]: or are you specifially saying to G53
08:21 PM ZincBoy[CAON][m]: Trying the G53 helps isolate if it an offset issue with the G54 work offset.
08:21 PM mits[m]: ok
08:21 PM mits[m]: that worked
08:22 PM ZincBoy[CAON][m]: Okay. It seems your G54 offsets are such that you can't reach x0 y0 then.
08:22 PM ZincBoy[CAON][m]: Why you get that error with the MDI call I am not sure.
08:23 PM XXCoder: whats your current g54 offset numbers?
08:24 PM * mits[m] uploaded an image: (49KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/PTunKnrTtcvUudtKuGyRahsm/image.png >
08:25 PM mits[m]: I am at machine 0 0 and then very near the top... so that seems right
08:26 PM XXCoder: thats coords current for g54, I meant the zero coord for g54. sorry, my brain wont work well
08:26 PM XXCoder: ie where is g54 zero point is
08:26 PM XXCoder: (in terms of machine coord)
08:27 PM mits[m]: I don't know how to view that other than to crack open the params file
08:27 PM mits[m]: is there a decent way?
08:27 PM XXCoder: not sure havent opened linuxcnc for a while
08:27 PM XXCoder: zinc help? :D
08:29 PM XXCoder: ZincBoy[CAON][m]:
08:34 PM mits[m]: I have an idea of using (MSG, #<_stuff>).... trying
08:34 PM ZincBoy[CAON][m]: In axis turn on "show machine limits"
08:34 PM ZincBoy[CAON][m]: You will get a bounding box that shows the g53 limits of the machine.
08:35 PM ZincBoy[CAON][m]: From there you can visually confirm where you are and if the requested move would violate the limits.
08:36 PM * mits[m] uploaded an image: (30KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/iroqtwmAXEwFLEtLKJeLnvxh/image.png >
08:37 PM * mits[m] uploaded an image: (22KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/iYUagvMDjNGoPrPJStxoiVCT/image.png >
08:37 PM mits[m]: that seems fine to me
08:38 PM XXCoder: if red box is workspace then tool went outside it for short time
08:38 PM ZincBoy[CAON][m]: Your tool is at G53 X0 Y0? What is your Y axis travel?
08:39 PM mits[m]: I have MAX_LIMIT set to 0.4
08:40 PM mits[m]: I think that was because my home switch is also my limit switch, so I wanted my 0 to be away from it
08:40 PM XXCoder: you can set homing switch to be at negative space
08:40 PM mits[m]: if that makes sense at all
08:40 PM XXCoder: and soft limit set to 0
08:41 PM XXCoder: supposing you have workspace of 0-400, switch would be at -1, it triggers homing, then it moves back to 0 set soft limit there (made up numbers but you get idea)
08:41 PM mits[m]: I wonder why I wouldn't have done it that way... this was pncconf to start but I modified these so long ago
08:41 PM djdelorie: on my machine, the 0,0 points are all about a half inch away from the limit switches...
08:42 PM ZincBoy[CAON][m]: Shouldn't really matter as the g53 space should never really be used for anything. The only important value is the max - min to give the travel distance.
08:42 PM ZincBoy[CAON][m]: What is your min_limit?
08:42 PM djdelorie: unless the G54 space maps to the G53 space
08:42 PM ZincBoy[CAON][m]: for X and Y.
08:43 PM mits[m]: X MIN is -259.6
08:43 PM ZincBoy[CAON][m]: Only if G54 offsets are all zero. Which they are not in this case.
08:43 PM mits[m]: Y MIN is -294.6
08:45 PM ZincBoy[CAON][m]: That should be fine then. I think the issue that XXCoder pointed out is more likely the problem. The machine went outside of the limits. This should never happen.
08:45 PM mits[m]: I don't know how it could have... dang
08:45 PM mits[m]: ok
08:45 PM XXCoder: yeah. since it goes to 90, and max is 0.4
08:45 PM XXCoder: erm 0 not 90
08:46 PM mits[m]: Also I feel like I did this home at zero thing because I had some glitch happen when I tried to home and the limit switch was already tripped
08:46 PM mits[m]: like it kept erroring
08:46 PM mits[m]: but maybe that was false correlation
08:47 PM XXCoder: wait you use negative numbers workspace, so 0.4 would be beyond 0
08:47 PM XXCoder: this is why im "retired", my brain have issues and I cant think though things properly anymore.
08:48 PM mits[m]: are you saying this makes more sense now, or less sense now?
08:49 PM XXCoder: to be sure, on your x workspace, it is -259.6 to 0 correct?
08:49 PM XXCoder: with switch homing trigger at 0.4?
08:50 PM XXCoder: in ideal setup, homing switch is touched only once when its being powered up and set,
08:50 PM mits[m]: no, I think it's the other way around
08:50 PM mits[m]: -259.6 to 0.4 for the MIN/MAX
08:50 PM mits[m]: and then HOME at 0
08:51 PM mits[m]: I do not know why I did this.... it makes no sense when logically thought out
08:51 PM XXCoder: you probably switched 0 and 0.4
08:52 PM mits[m]: sure, but the 259.6 is .4 less thant 260mm
08:52 PM mits[m]: so I was trying to have a 260mm space
08:52 PM mits[m]: I remember doing that part
08:53 PM mits[m]: I had a nice number, just pulled a little bit off and put it on the other side of zero
08:53 PM XXCoder: then it should be -260 to 0, with .4 as homing
08:53 PM mits[m]: yeah
08:53 PM XXCoder: assuming it can travel the distance without hitting other end
08:53 PM mits[m]: sure yes
08:53 PM XXCoder: workspace is always slightly smaller than actual machine travel
08:55 PM XXCoder: I know its olvious but sometimes need to be sure lol
08:55 PM mits[m]: The only thing I can think of was that I thought I had an issue where it kept giving me errors if the machine was started and it was on the limit switches
08:55 PM mits[m]: and then I wouldn't be able to jog it off of them
08:55 PM mits[m]: yeah, I appreciate the foundational statements
08:55 PM mits[m]: to make sure I'm understanding this
08:56 PM XXCoder: hmm what direct did it go to back off after homing?
08:56 PM XXCoder: if it didnt back far enough, or wrong way, it would be still on switch
08:57 PM mits[m]: ehn.... this was a while ago (year or so?) so I can't remember
08:57 PM mits[m]: I'm going to try to switch this back... I guess I should just scrap the part and restart the machine
08:57 PM djdelorie: I've always had problems with the limit switch homing. I ended up using "different direction" for the second move, so it sensed when it came off the switch instead
08:57 PM mits[m]: it thinks it went out of bounds somehow
08:58 PM djdelorie: wait... you *did* successfully home the machine, yes?
08:58 PM ZincBoy[CAON][m]: I did the same thing. Seemed easier than figuring out the other methods.
08:59 PM ZincBoy[CAON][m]: The error was a soft error so the machine was not moving when it was thrown. Limit switches would be a different error.
09:00 PM mits[m]: I did yes
09:00 PM djdelorie: whew
09:01 PM djdelorie: homing defines the volume the machine can physically move in
09:01 PM djdelorie: soft limits defines a slightly smaller volume you can work in
09:01 PM djdelorie: between touching off (to set a working coordinate system) and what your program does, there's a volume it expects to move in
09:02 PM djdelorie: if that volume exits the soft limits volume, you get the error you see
09:02 PM djdelorie: if you hadn't homed, that would have put the soft limits volume in the wrong spot
09:03 PM djdelorie: after you run a program, the MDI is affected by whatever changes the program did, too
09:03 PM djdelorie: so the 0,0 you moved to in the MDI might not have been the 0,0 you thought you were moving to
09:04 PM djdelorie: at least with 0,0 you aren't affected by the inch/mm setting being wrong, like I usually am ;-)
09:14 PM roycroft: note to self: never travel to mars on a spacecraft djdelorie designed
09:15 PM djdelorie: you'll get there in 1/25.4th the time!
09:17 PM roycroft: and from takeoff until the edge of the martian atmosphere, that would be fine
09:17 PM roycroft: but it's that last little bit that concerns me
09:17 PM djdelorie: perfect is the enemy of good enough :-)
10:09 PM roycroft: while you're sitting in your ivory tower it may look like it's "good enough"
10:10 PM roycroft: when i'm crashing onto the surface of mars, forgive me if i disagree with your assessment
10:11 PM mits[m]: is there a compile time flag for linuxcnc to disable imperial support? 😄
10:11 PM mits[m]: if not, why not? 😄
10:11 PM djdelorie: did your machine get switched to inches? ;-)
10:11 PM mits[m]: nono
10:11 PM mits[m]: I just want to guarantee it
10:12 PM mits[m]: also I'm just sassing
10:13 PM djdelorie: I did ask a similar question some time ago, here... if you could add some gcode after a program exits to reset the mode, so MDI wasn't so influenced by the program
10:14 PM XXCoder: well theres g20 and g21 to make it for sure too
10:14 PM djdelorie: at least when you usually use inches and it switches to mm, a "G0 X60" doesn't break your machine
10:14 PM djdelorie: XXCoder: I know about those, the question was, how to run them automatically when a program ends (M2)
10:15 PM XXCoder: some cam put it right below gcode header
10:15 PM djdelorie: on my 3d printer, there's a whole directory full of gcode scripts that the printer runs automatically on various events
10:15 PM djdelorie: I don't trust the cam to always do this ;-)
10:16 PM XXCoder: that is why you walk in at starting all tools on first run
10:16 PM djdelorie: and the default axis gui doesn't have a big YOUR IN METRIC MODE warning light
10:16 PM XXCoder: i wonder if that is possible
10:16 PM XXCoder: metric/inches led lights
10:17 PM djdelorie: iirc the other guis do, just not the default. It has a "mode" box that just lists all the active G and M codes
10:17 PM djdelorie: G20 is in there but it's easy to miss a 10pt text in a paragraph of them...
10:18 PM djdelorie: I'm still running my machine off a temporary Fedora laptop, so I haven't (and/or can't) yet started customizing the GUI
10:22 PM mits[m]: maybe remap G20 to nothing
10:23 PM djdelorie: then any metric-based programs won't work
10:23 PM mits[m]: and then blame the user when their project comes out tiny
10:23 PM djdelorie: heh, the number of 3d prints that came out 1:25.4 scale...
10:28 PM roycroft: i'm not a fan of unit conversions
10:28 PM roycroft: i prefer just working in the same units from start to finish of a project
10:31 PM djdelorie: well, me too, but my preferred unit is "inch" and most software doesn't respect that
10:32 PM roycroft: the lamps i'm designing currently are a lot easier to design in mm, but the tooling i need to manufacture them is much easier to procure and much more affordable in inches
10:33 PM roycroft: so i'm using inches for the designs
10:34 PM djdelorie: in FreeCAD, if you set the default units to "inch", the CAD parts default to inch but the CAM parts still default to mm
10:34 PM djdelorie: so I have to type "in" after everything anyway, to be sure
10:36 PM djdelorie: and the linuxcnc exporter has its own defaults, separate from the main app
11:14 PM roycroft: do folks do guilloché work with a drag graver in a cnc mill/router?
11:29 PM unterhaus: in earlier times, emc had trouble with metric units
11:30 PM unterhaus: inches were baked in somehow
11:30 PM XXCoder: hardcoded stuff I guess
11:30 PM unterhaus: it wasn't the best code
11:31 PM unterhaus: well, there is still some in lcnc
11:31 PM unterhaus: as our favorite new wave artist pointed out