#linuxcnc Logs

Jan 07 2025

01:27 AM Deejay: moin
02:03 AM lcnc-relay: <meisterdippel@> moin
02:39 AM lcnc-relay: <vibram@> hello
04:02 AM lcnc-relay: <oddname_skane@> contemplating if im just plain stupid or blind, reading the documentation on how to config the ini, but I cant find anything regarding homing direction..
04:02 AM lcnc-relay: <oddname_skane@> * wether I'm
04:09 AM Tom_L: mrec, andy and archivist both have but neither are here much now
04:09 AM Tom_L: morning
04:23 AM Tom_L: https://www.youtube.com/watch?v=xdE46yvckbM
04:23 AM Tom_L: https://www.youtube.com/watch?v=ZhICrb0Tbn4
04:45 AM lcnc-relay: <oddname_skane@> another "issue" steppers are allways energized, if I want to use enable, it means when i "turn on" the machine in the interface steppers är off
04:45 AM lcnc-relay: <oddname_skane@> how to invert it so that pressing turns the port off and steppers kick in
05:14 AM jpa-: after homing, you probably want to keep the steppers enabled because otherwise they could lose position
05:14 AM jpa-: (you could lower the drive current though, if the drivers support that)
05:27 AM lcnc-relay: <meisterdippel@> oddname_skane@: depending on the interface, there are inverted pins with the ending _not or -not
05:39 AM jpa-: (halshow is nice for browsing them)
05:52 AM lcnc-relay: <oddname_skane@> Using mesa if that helps 🙂
05:52 AM lcnc-relay: I just want them off when prepping code etc
05:52 AM lcnc-relay: <oddname_skane@> Or tinkering
05:52 AM JT-Cave: so just turn them off
05:52 AM JT-Cave: https://youtu.be/2IaTZV4OAww
06:14 AM JT-Cave: Tom_L, bugfix for mesact
06:18 AM mrec: Tom_L: I need a more efficient setup here, I used to transfer the discs from a lathe to a mill now I'm working on an extension for the mill so I can do everything at once
06:22 AM mrec: I think I will end up putting the Lathe into a VMC
06:24 AM lcnc-relay: <oddname_skane@> are they heave?
06:24 AM lcnc-relay: <oddname_skane@> * heavy?
06:31 AM mrec: is that question for me? the mini lathe has 40-50kg only
06:32 AM mrec: it fits in easily into the >5tons VMC
06:33 AM lcnc-relay: <oddname_skane@> yeah, the disc you need to transfer 🙂
06:34 AM lcnc-relay: <oddname_skane@> considered a cobot if its light enough?
06:34 AM lcnc-relay: <oddname_skane@> or old industry robot, they can be found with really good pricing 🙂
06:36 AM mrec: I would like to avoid transferring the discs
06:37 AM mrec: I did that manually back then it's okay but I just want to avoid it and do whatever I can on the machine directly
06:37 AM mrec: a good CNC lathe could do that but I don't have enough business in that area to justify such an investment at this time
06:38 AM mrec: the machine would just idle around most of the time.
06:39 AM mrec: the best option would be a small desktop gear hobbing machine but I haven't seen one that would suit my needs
06:42 AM lcnc-relay: <oddname_skane@> ahh
07:05 AM lcnc-relay: <oddname_skane@> hmm homing works fine, bu when home it should think its at 300,300 and then if i move xy 10mm it should read 290,290 but nope, its 0,0 and then when i move it still = +10mm
08:02 AM JT-Cave: https://www.youtube.com/watch?v=pw4cxcSVtV4
09:54 AM lcnc-relay: <t4kuy4@> so.. im running servo motors without feeding back the encoder signal to my mesa card. does ferror and min_ferror matter to me?
09:54 AM lcnc-relay: the default pncconf values seem rather high (min at 1mm, ferror at 10mm)
10:09 AM NetPipe: https://github.com/netpipe/cuckooClock/releases/tag/macos
10:33 AM roycroft: well, my axle seal tool is in portland now
10:33 AM roycroft: so maybe i'll get it on thursday
10:55 AM lcnc-relay: <oddname_skane@> @JT getting the hang of editing ini now, like this way alot more than pncconf 😄
11:01 AM JT-Shop_: nice
11:02 AM -!- #linuxcnc mode set to +v by ChanServ
11:07 AM lcnc-relay: <oddname_skane@> but one step forward, 2 steps backwards 😄
11:07 AM lcnc-relay: <oddname_skane@> tandem y, 1 axis suddenly got a binding issue of some kind
11:08 AM lcnc-relay: <oddname_skane@> not even the side I expected due to pulleys being loose
11:09 AM lcnc-relay: <oddname_skane@> but it could have made the gantry twist, making it bind
11:10 AM lcnc-relay: <oddname_skane@> the joys of troubleshooting 😄
11:16 AM lcnc-relay: <oddname_skane@> it has to do with the loose pullet atleast so guessing it just was bumped askew since worked wonders before 🙂
11:18 AM roycroft: a loose pullet? that would be jt-shop's domain. he might recommend a small lasso
11:18 AM roycroft: or perhaps a net
11:22 AM JT-Cave: treats work the best
11:22 AM roycroft: aah, the positive reinforcement tactic
11:22 AM roycroft: very good
11:22 AM JT-Cave: chasing a chicken only get you and the chicken some exercise
11:24 AM lcnc-relay: <oddname_skane@> * pulleY
11:24 AM lcnc-relay: <oddname_skane@> 😄
11:24 AM lcnc-relay: <oddname_skane@> _thinks rude words_
12:56 PM rdtsc-w: was just pushing my cart in the grocery store, and some guy throws a bottle of mayo at me, nearly hits me in the head... I yell to him: What the Hellmans?
01:04 PM roycroft: that's a strictly back east joke
01:04 PM roycroft: west of the rockies hellman's is marketed as "best foods mayonnaise"
01:05 PM mrec: the only thing that someone threw something after me was because she was too disappointed that she couldn't steal my wallet at the czech trainstation in prague
01:05 PM roycroft: folks who haven't been back east have never heard of hellmann's
01:05 PM mrec: a friend was not so lucky a few years later and didn't learn from my story his concert tickets and his money was gone.
01:06 PM mrec: that was like 15 years ago
01:23 PM bjorkintosh: rdtsc-w: why did he do that?
01:26 PM Tom_L: bjorkintosh, i think it was supposed to be a joke
01:26 PM bjorkintosh: Oh.
01:26 PM Tom_L: jokes don't always fly on irc
01:26 PM Tom_L: hellman is a brand
01:26 PM bjorkintosh: yes I have it in my fridge.
01:28 PM Tom_L: if it wasn't a joke, rdtsc-w probably had it coming to him :)_
01:31 PM roycroft: i got the joke
01:31 PM roycroft: i was just pointing out that it's a regional joke
01:32 PM roycroft: best foods bought hellmann's in 1932 but decided to retain the hellmann's branding
01:32 PM xxcoder: yeah I guessed it was some regional thing, didnt know hellman brand existed
01:32 PM roycroft: thus the two regions
01:32 PM roycroft: it's the same stuff though
01:32 PM roycroft: the internet claims the two products taste different, but professional tasters are split on that, which means it's the same or essentially the same
01:35 PM bjorkintosh: humans. they're just going to shit it out anyway, but what the tongue decides seems to be a BIG. DEAL!.
01:39 PM xxcoder: lol reminds me of this joke. found it at reddit https://www.reddit.com/r/Jokes/comments/61jl4p/all_the_organs_were_deciding_who_should_be_the/
02:39 PM roycroft: peter yarrow has left the building
02:39 PM roycroft: i had the privilege of meeting him a few years ago - he was a nice fellow
02:40 PM roycroft: and a great singer/songwriter
02:41 PM bjorkintosh: oh I didn't even realize he was still alive.
02:41 PM lcnc-relay: <skunkworks8841@> lol - it is going to take about 6 days to transfer 36TB of data between to external esata boxes.. (which is about what I expected)
02:41 PM lcnc-relay: <skunkworks8841@> *two
02:41 PM roycroft: paul is the only one left now
02:42 PM roycroft: mary travers passed about 15 years ago
02:56 PM memleak: skunkworks, transfer 36TB? Sheesh, RDMA that sucker lol
02:57 PM lcnc-relay: <skunkworks8841@> maybe a little less. Seems to be going faster..
02:57 PM lcnc-relay: <skunkworks8841@> lol - this is old crap that usually just works. (for offsite backups)
02:58 PM lcnc-relay: <skunkworks8841@> (and zfs is awesome)
02:58 PM memleak: bcachefs is trying to be the next zfs and it's in mainline now
02:58 PM memleak: never used it
02:59 PM memleak: (never used either actually)
02:59 PM lcnc-relay: <skunkworks8841@> I thought btrfs was supposed to be the next best thing?
02:59 PM lcnc-relay: <skunkworks8841@> (I am only comparing it to mdadmin - which had some really crappy failure modes)
03:00 PM roycroft: i've been using zfs for ages
03:00 PM roycroft: it's pretty much error-proof
03:01 PM roycroft: even if you don't use error-correcting memory
03:01 PM lcnc-relay: <skunkworks8841@> right - I have had some pretty yecky failures with crappy hardware and zfs just keeps plugging along.
03:01 PM roycroft: doing it right requires very heavy feeding, though - you only get 50% of drive capacity, and there is no such thing as too much ram for a zfs machine
03:02 PM lcnc-relay: <skunkworks8841@> right
03:02 PM lcnc-relay: <skunkworks8841@> RAM usage 79.30% (399.47 GiB of 503.74 GiB)
03:02 PM roycroft: and if it's a nas you really should have an identical replication machine, so your effective drive capacity is actually only 25% of your total storage
03:02 PM lcnc-relay: <skunkworks8841@> (proxmox server)
03:03 PM roycroft: but there will be no bit rot
03:03 PM roycroft: as long as you pay attention to the machine
03:03 PM lcnc-relay: <skunkworks8841@> right
03:03 PM roycroft: even if you don't, there will be no bit rot
03:04 PM roycroft: there will be a perfectly good, functional machine, and then there will be a disaster
03:04 PM roycroft: but that's the case with all computer hardware
03:04 PM memleak: https://www.phoronix.com/review/linux-611-filesystems/3 geometric mean shows bcachefs to be faster than btrfs and it's supposedly faster than zfs too
03:05 PM lcnc-relay: <skunkworks8841@> zfs never seems to be fast 😉
03:05 PM roycroft: zfs can pretty much be as fast as you want it to be
03:05 PM roycroft: it gets expensive though
03:06 PM lcnc-relay: <skunkworks8841@> we found out that we want spinny drives just for data storage - not for vm's
03:06 PM memleak: buT tHE hOMepAGe foR BcaCHeFS SayS itS FaSTer thAn zFs
03:06 PM memleak: :P
03:07 PM roycroft: a turtle can be faster than a rabbit
03:07 PM memleak: nothing is faster though than pied piper
03:07 PM rdtsc-w: have btrfs on a laptop... it likes to seem like everything is fine, then upon reboot, tell you there are unlinked files, wipe half of the kernel, and leave you scrambling for a rescue .iso
03:07 PM roycroft: the grim reaper can be
03:08 PM memleak: rdtsc-w, noted, holy crap
03:08 PM roycroft: the zfs copy-before-write policy makes that kind of corruption almost impossible
03:09 PM rdtsc-w: rescued it several times - unsure what the problem is exactly - gone so far as to issue a sync before and after lid close, disable suspend, etc.
03:11 PM rdtsc-w: waiting for it to happen again, at which point I'll probably reload it as ext4
03:18 PM roycroft: i use xfs on my servers that are not natively running zfs
03:19 PM roycroft: i find it to be extremely robust, and i have servers parked in places with flaky power and no battery backup, so robust is very important
03:22 PM memleak: for robust i just use ntfs :P
03:24 PM roycroft: most of my servers with the flaky power are actually vms configured with xfs running on proxmox hypervisors that use zfs natively
03:26 PM memleak: brb
03:30 PM lcnc-relay: <schoch_@> murderfs (reiser) ftw!
03:33 PM roycroft: i was never impressed with reiserfs, massacres notwithstanding
03:36 PM xxcoder: roy why not have some double conversion ups with lifepo4 batteries? it would clean crappy power up and have in least short time to shut safely down
03:37 PM bjorkintosh: what about plain old ext4fs?
03:37 PM memleak: https://www.phoronix.com/news/ReiserFS-Deleted-Linux-6.13 about freakin time
03:38 PM memleak: 32.8K lines of code, only for the amdgpu driver to grow by 32.8 million lines heh
03:39 PM roycroft: not my business and not my money
03:39 PM xxcoder: interesting. I'd do a read-only support rather than complete removal
03:49 PM CloudEvil: yeah. Removing is just annoying. Having to patch back in support for stuff to read old disks is really annoying
03:50 PM xxcoder: I suppose can always have file browser for that
06:15 PM lcnc-relay: <skunkworks8841@> ext4 is just fine.. Just not for high availability/fault tolerance. (well - not easy anyway)
06:16 PM lcnc-relay: <skunkworks8841@> i am sure my laptop is ext4
06:37 PM memleak: i use ext4 for rotating media, f2fs for nvme and USB
06:50 PM memleak: you have to use inode_checksum and sb_checksum for any kind of error checking though i think for f2fs, pretty reckless without lol
06:50 PM memleak: my installer has those options set by default so there's at least some protection
08:16 PM xxcoder: oh no one of primative tech buildings got destroyed
10:34 PM memleak: was it destroyed due to terrorist attack assisted by chatgpt?
10:35 PM xxcoder: nah time killed it. dead tree roots was finally rotten enough that rainy season it fell over the building
10:36 PM xxcoder: he reclaimed almost everything from it
10:36 PM memleak: well at least he was able to reclaim all the apples that were lost from the tree!
10:36 PM xxcoder: it was 2 years and 8 months since he built it
10:37 PM xxcoder: apples from long dead tree? 🤔
10:37 PM memleak: nothing wrong with a little fermentation :P
10:38 PM memleak: that's how horses get their drink on