#linuxcnc Logs
Jul 06 2022
#linuxcnc Calendar
12:34 AM randy: morning
12:56 AM Deejay: moin
02:43 AM pere: hi
04:42 AM Tom_L: morning
04:45 AM t4nk_freenode: 'moaning
04:53 AM XXCoder: mooning
04:58 AM t4nk_freenode: hehe, that's the spirit
05:00 AM travis_farmer: G'Morning
05:30 AM JT-Cave: same plus green apple trots, which I thought was over with yesterday but no...
05:31 AM XXCoder: sucks
05:31 AM travis_farmer: speaking of, i have had those particular trots... wonder if it is something going around...
05:32 AM travis_farmer: mine may be food poisoning though...
05:35 AM JT-Cave: in my case I walked through a cloud of Bacillus Thuringensis while spraying the tomato plants
05:37 AM travis_farmer: hmm, so poisoning either way...
05:44 AM * JT-Cave wanders back to the couch now that the chickens are out
05:46 AM travis_farmer: i am surprised you haven't built up a device that lets the chickens out from a few taps on your phone ;)
07:07 AM * Tom_L tries to avoid eating green apples
07:16 AM travis_farmer: mine wasn't green apples, actualy... it was contaminated jellybeans. one tasted a very tart, and was sticky... didn't notice until i ate it... should have spit it out.
08:10 AM pere: anyone here willing to help out with the spanish translation of the documentation, <URL: https://hosted.weblate.org/projects/linuxcnc/linuxcnc-docs/es/ >. A lot could be done by confirming the migrated strings (marked fuzzy) while ignoring the automated proposals (also marked fuzzy).
08:53 AM robert1 is now known as rmu
09:53 AM GTKplusplus[m]: Sigh, found out why probe screen is not working
09:54 AM GTKplusplus[m]: It can't find handlers for... Anything
09:54 AM GTKplusplus[m]: But it's not an ini issue. Which kinda stumps me because I just downloaded the stock files from GitHub
10:00 AM ZincBoy[CAON][m]: You have the right python path in your INI?
10:01 AM ZincBoy[CAON][m]: Example: my ini:
10:01 AM ZincBoy[CAON][m]: EMBED_TAB_NAME=PSNG
10:01 AM ZincBoy[CAON][m]: EMBED_TAB_LOCATION = ntb_preview
10:01 AM ZincBoy[CAON][m]: EMBED_TAB_COMMAND=gladevcp -c probe -x {XID} -u psng/python -H psng/psng.hal psng/psng_nextgen3.glade
10:04 AM roguish[m]: _discord_554508768402997249[m]:so you're running the psng version of probe_screen? is it ok? i've been running the v2 version.
10:16 AM ZincBoy[CAON][m]: I more or less completely re-wrote it. There were a few things I needed to add like probe calibration and low battery/fault detect. Plus the way the python/LinuxCNC interface was implemented was non ideal. I plan to commit it back when I am happy with it.
11:05 AM GTKplusplus[m]: I had
11:05 AM GTKplusplus[m]: but I used the probe screen you linked and it worked first try
11:05 AM GTKplusplus[m]: now I need to test the new tool change, I had a different macro setup
11:05 AM GTKplusplus[m]: and see if I need to do any post changes for it
11:09 AM GTKplusplus[m]: I just tested it and it's fine, although there is quite a long delay between each gcode command being sent, for reasons I haven't looked into yet
11:41 AM roycroft: well, the day is finally here
11:41 AM roycroft: after waiting 3 years, today i haul our stuff out to the fair
11:41 AM roycroft: and tomorrow i head out there for the duration
12:16 PM ZincBoy[CAON][m]: IIRC that is why I re-wrote it. There were a bunch of fixed delays in the code and I changed it to polling.
12:17 PM GTKplusplus[m]: I'll happily beta test your code if you are willing to share
12:19 PM GTKplusplus[m]: (or I could go look for them myself, if they are in the python part of the code
12:19 PM GTKplusplus[m]: * the code)
12:19 PM GTKplusplus[m]: I can't see why they would exist in the gcode parts
12:23 PM ZincBoy[CAON][m]: They are in the Python code. The calls to run Gcode and Ocodes are the main culprits.
12:24 PM ZincBoy[CAON][m]: My version is included in the config I uploaded to my conversion thread on the linuxcnc forum.
12:24 PM ZincBoy[CAON][m]: https://forum.linuxcnc.org/media/kunena/attachments/2195/VK45_v2.zip
12:26 PM ZincBoy[CAON][m]: Note there are bugs with this! It is still a work in progress but it works for the basics.
12:36 PM GTKplusplus[m]: Thanks, I'll check it out
12:37 PM GTKplusplus[m]: I'm fine with just trying to fix the delays. Not that it really matters on my kind of machine, but it's still nice to avoid time wastes
12:37 PM ZincBoy[CAON][m]: I only tested it on my system so there are a bunch of assumptions that may not be correct for your setup.
12:38 PM -!- #linuxcnc mode set to +v by ChanServ
01:34 PM pere: Is <URL: https://github.com/LinuxCNC/linuxcnc/pull/1802 > correct, or did I misunderstand something?
01:39 PM ZincBoy[CAON][m]: I am not sure the at_pid component was ever pin compatible with the standard pid component? There are pins required here that are not in the standard pid.
01:41 PM pere: I guess the compatibility was only one way.
01:48 PM -!- #linuxcnc mode set to +v by ChanServ
01:51 PM pere: In my setup, the missing pins in at_pid was index-enable and command-deriv, both as far as I can tell, added to the pid component after 2011.
05:04 PM Tom_L: 101F
05:13 PM _unreal_: WHAT A PAIN IN THE ASS that was.
05:13 PM _unreal_: My filiment broke inside the extruder body.
05:14 PM _unreal_: I also just installed one of the hardened print tips.
05:14 PM _unreal_: brb
05:22 PM MikeM[m]: @Tom_L Read through you configs as well as did some research on PYVCP. I think I have something usable for my machine now. Thank you!
05:23 PM _unreal_: ok that was odd. my computer was acting very odd.
05:23 PM _unreal_: just rebooted
05:23 PM * MikeM[m] uploaded an image: (164KiB) < https://libera.ems.host/_matrix/media/r0/download/jauriarts.org/MpNcmNrZxrzxDHFdVbeWCQiP/LinuxCNC_Screenshot_from_2022-07-06_18-17-06.png >
05:32 PM _unreal_: printing the foot of the base again so I can use it as a "washer" plate to the backside of the jug I'm going to mount this new dust cyclone to.
05:35 PM t4nk_freenode: rdbx is doin great ey, _unreal_? ;)
05:35 PM t4nk_freenode: you're only at 50%
05:36 PM t4nk_freenode: loss :))
05:54 PM Tom_L: MikeM[m], i like showing the tool number because i preset them all up front instead of touchoff each one
05:56 PM MikeM[m]: The issue with that is I use VCarve, and each tool change is a separate GCode file, so I think the tool number would always be 1 for me.
06:00 PM Tom_L: i figured something like that or you wouldn't be using a touch plate
06:24 PM CloudEvil: https://i.imgur.com/0GIrMUW.jpeg 'I've completed the frame for my CNC router'
06:26 PM XXCoder: lol
06:27 PM CloudEvil: I guess it's about as stiff as 3m*2*50mm tube.
06:27 PM CloudEvil: I got bored before actually comptuing how stiff that is.
06:27 PM CloudEvil: And how much it would deflect under its own weight
06:28 PM CloudEvil: Appropriate thickness box section would make it lots less flexy
06:30 PM XXCoder: good finals question for engineering? lol
06:32 PM CloudEvil: :)
06:32 PM CloudEvil: 'design a control and actuation system to make the table not perceptibly move when sat on'
06:33 PM CloudEvil: ^invisible
06:35 PM Tom_L: JT-Cave, doin any better?
06:35 PM Tom_L: is there a way to set the Axis gui screen size on startup?
06:58 PM tcurdt: I just found out that linuxcnc can also drive steppers without configuring pids ... given that servos/closed loop stepper have their own pid that somewhat made sense
06:58 PM tcurdt: but I am wondering if there are drawbacks
07:16 PM pcw--home: PID is always used (but sometimes the built in PID in the driver is used, when the driver is set for position mode)
07:17 PM pcw--home: The built in driver is not as good dealing with jitter so typically Mesa setups use an external PID comp
07:18 PM pcw--home: (this could be eliminated if the built in PID was improved, but this is a fair project)
07:27 PM tcurdt: pcw--home "built in PID in the driver" ... what driver do you mean exactly?
08:02 PM _unreal_: finally broke down and got an MPG pendent wheel just ordered it
08:34 PM roycroft: i have a couple pendant wheels
08:34 PM roycroft: they seem like they can be rather useful
08:43 PM _unreal_: ok just finished setting up the new cyclone I designed and printed
08:43 PM _unreal_: without question. better air flow
08:43 PM _unreal_: My suction is much better
08:44 PM _unreal_: as to particals.... yaaa still passing shit.
08:44 PM roycroft: my cyclone got clogged when i was running some ash through my thicknesser recently
08:44 PM roycroft: so the main dust collector bag filled up
08:44 PM roycroft: which was exactly what i was trying to avoid
08:44 PM roycroft: i have to get that sorted - i think the filter bag may be getting clogged, reducing airflow
08:45 PM XXCoder: fun
08:45 PM roycroft: the next phase in my dust collection upgrade was to replace the rest of the flex tubing with rigid ductwork, but i think i may skip that phase and move on to phase iii next
08:45 PM ZincBoy[CAON][m]: I had one of these on my router and it worked great for years: https://www.clearvuecyclones.com/product/cv06-mini-system/
08:46 PM roycroft: which is replacing the filter bag with a filter cannister
08:46 PM XXCoder: zinc yeah
08:46 PM _unreal_: at this point. I dont want to but I'm prob going to grab some zip ties and design and print a simple cage frame and wrap the main intake internally with some bug screen. :( dont want to but I got to keep the big shit out of the blower
08:46 PM XXCoder: I wouldnt be able to afford that though. I would just 3d print my own
08:46 PM XXCoder: maybe 10 dollars total
08:46 PM roycroft: that should triple the surface area for air to flow through, at least
08:46 PM _unreal_: I have two dust collectors at work
08:47 PM XXCoder: yeah larger surface area makes it a lot more effective
08:47 PM XXCoder: needs slower air movement to work. which means it works much longer
08:47 PM _unreal_: a small portable one :( that I have to fix the fucking bearings. FUCKING ASSWHOLES barrowed it and got it wet.
08:47 PM _unreal_: then tried to hid it on me INSIDE fucking plastic bag. STILL WET.
08:47 PM XXCoder: barrowed into your machine or what?
08:48 PM ZincBoy[CAON][m]: Bill Pentz's site has the plans to make your own cyclone. You can use that to base the 3d print design on.
08:48 PM XXCoder: yeah or download
08:48 PM XXCoder: lemme link to interesting one
08:49 PM roycroft: i have this cyclone: https://www.ebay.com/p/26033166793
08:50 PM roycroft: and until recently it worked fine, including when using the thicknesser with wide boards and big cuts
08:50 PM roycroft: i'm thinking it's the filter bag on the dust collector that's getting clogged and slowing the airflow, which is causing the cyclone to clog
08:51 PM XXCoder: finally found it
08:51 PM XXCoder: https://www.thingiverse.com/thing:1447214
08:51 PM XXCoder: its hose-inline cylone thing
08:51 PM roycroft: the dust collector has a 1.5hp motor, which should be sufficient to run with a single machine at a time
08:51 PM XXCoder: so you can get rid of some of it on way to main one, if long one lol
08:52 PM ZincBoy[CAON][m]: I don't think that will do much. It is much too small for the airflow in a shop vac system.
08:52 PM XXCoder: https://www.youtube.com/watch?v=FEvztl8UPPk
08:52 PM roycroft: i love it
08:52 PM XXCoder: it works fine, it vaccums into coke bottle
08:52 PM ZincBoy[CAON][m]: Something like this is much more likely to work: https://www.thingiverse.com/thing:2375150
08:52 PM roycroft: 0% chance of rain in the forecast for the next week
08:53 PM roycroft: we had a wee bit of a drizzle early this morning, but now it's perfect weather for the fair
08:53 PM XXCoder: very light drizzle also. rest of week is just cloudy
08:54 PM XXCoder: ZincBoy[CAON][m]: whats interesting is theres nothing stopping you from having more than one
08:54 PM roycroft: mostly sunny to sunny here through the 15th
08:54 PM roycroft: as of now
08:54 PM XXCoder: small local one then main centered one
08:54 PM roycroft: and highs will be in the mid 20s over the weekend, and then up in the 30s starting monday
08:54 PM roycroft: the fair ends sunday, so that could not be more perfect
08:54 PM XXCoder: lol found my 4 year old comment there
08:56 PM XXCoder: comment is very obselete. youtube killed that video stabilizer, or as I call it, video autovomic
08:56 PM XXCoder: *autovomit
08:56 PM ZincBoy[CAON][m]: You really only want one. The problem is you get loss in each one and cascading them is going to cause excessive drop. Shop vac based systems are marginal at best and can't tolerate too much flow drop. You can run them in parallel to a point but you still need to size them to get the right airflow.
08:57 PM XXCoder: interesting
08:57 PM XXCoder: though it can be very handy for local vaccum machine though!
08:58 PM XXCoder: your link is excellent one
08:59 PM XXCoder: glad I know the zip trick, that one has so many fules
09:02 PM ZincBoy[CAON][m]: Anyone working on dust collection I highly recommend reading Bill Pentz's site. He has a wealth of knowledge on dust collection and it can save you a ton of time and wasted effort. https://billpentz.com/woodworking/cyclone/equipment.php
09:10 PM ZincBoy[CAON][m]: There is even a spreadsheet that lets you design your own to match your requirements: https://billpentz.com/woodworking/cyclone/cyclonespreadsheetac160629.xls
09:11 PM XXCoder: saved
09:15 PM _unreal_: there I now designed an INTAKE cage to 3d print :(
09:16 PM _unreal_: .... not 100% sure how I will get into place... but I'll make it happen. then two-3 zip ties and shit will NEVER get into my blower again
09:17 PM _unreal_: and I will have gained a clensing issue
09:18 PM _unreal_: long story short. the new cyclone setup DOES work better then its predecessor. my design is better. and it DOES clearly want more air flow.
09:18 PM _unreal_: meaning bigger blower or stronger vacuum
09:18 PM _unreal_: if I get to it I'll take a picture or two
09:18 PM _unreal_: later
09:23 PM roycroft: i'm not sure i've checked out that site before, but i did do a fair amount of research on dust collection ssytems
09:23 PM XXCoder: it seems very detailed roy
09:24 PM XXCoder: only did basic analsys of site
09:24 PM roycroft: and the recent improvements i've been doing on my system have worked exactly as anticipated, until this recent failure
09:24 PM roycroft: but i have bookmarked billpenz.com, and will spend some time there when i have the time to do so, after the fair
09:25 PM XXCoder: on fair my bro is too busy, even when he plan to go to vancouver
09:25 PM XXCoder: I dont care to go by myself so im out I guess
09:26 PM roycroft: i'll be heading out again in the morning, probably around 9am, for the duration
09:26 PM roycroft: i could have stayed there tonight, but i usually come back into town wednesday night so i can pick up last minute supplies and the like to take out on thursday
09:26 PM roycroft: the fair starts on friday
09:26 PM XXCoder: nice
09:26 PM XXCoder: just too far for me lol
09:27 PM XXCoder: dont want to drive that long when dizz
09:40 PM roycroft: yeah, it's 20 minutes for me
09:40 PM roycroft: pretty easy
09:40 PM XXCoder: 4 hours for me I think
09:40 PM roycroft: that sounds about right
09:41 PM roycroft: it's a bit over 4 hours from here to seattle
09:41 PM roycroft: tacoma is half hour south of seattle, and the fair is 20 minutes west of here
09:49 PM roycroft: you know, speaking of dust collection
09:49 PM roycroft: yesterday i needed to make a new platen for vacuum veneering, etc.
09:50 PM roycroft: it consisted of routing a bunch of grooves in some melamine with a round cutter, and then rounding over the edges
09:50 PM roycroft: instead of using one of my usual routers, i pulled out the festool trim router that i recently bought
09:50 PM roycroft: with that i was able to do the entire job with absolutely no visible dust residue at all
09:51 PM XXCoder: nice
09:51 PM roycroft: yeah
09:51 PM roycroft: the festool stuff is wicked expensive, but it really is worth it for the extraction alone
09:52 PM roycroft: i've never used a router without making a mess before
09:52 PM roycroft: my usualy strategy when doing a lot of routing is to wait for good weather, set up sawhorses outside, and be prepared to sweep up the mess on the driveway later
10:00 PM Centurion_Dan1 is now known as Centurion_Dan
10:01 PM roycroft: https://roycroft.us/Bench/Tail-Vise-Setup-1.jpeg
10:01 PM roycroft: https://roycroft.us/Bench/Tail-Vise-Setup-2.jpeg
10:01 PM roycroft: as examples :)
10:09 PM XXCoder: interesting]
10:14 PM roycroft: i was actually very pleased with how that tail vise worked out
10:14 PM roycroft: the joinery is pretty tricky
10:14 PM roycroft: and i nailed it, absolutely spot on, the first time
10:14 PM roycroft: i did spend about a day and a half laying it out and machining it
10:15 PM roycroft: but matt estlea, whose plans i based my bench on, screwed it up big time when he made his
10:16 PM XXCoder: cringeville https://www.reddit.com/r/Machinists/comments/vsmy6e/why_would_you_do_this/
10:16 PM pcw--home: tcurdt: LinuxCNCs software stepgen component and Mesas stepgen part of the hm2 driver
10:18 PM roycroft: i didn't need to see that tonight
10:18 PM roycroft: i'm in a great mood
10:18 PM XXCoder: lol sorry
10:18 PM roycroft: that's ok
10:19 PM roycroft: it hasn't changed my mood
10:19 PM roycroft: i closed the link fast enough :)
10:19 PM XXCoder: lol