#linuxcnc Logs

Apr 26 2022

#linuxcnc Calendar

01:23 AM randy: morning
01:39 AM -!- #linuxcnc mode set to +v by ChanServ
01:39 AM MrSunshine: roycroft, could it be some rounding error ? as 3d prints has lots of layers, say the initial layer heighht is 0.254 mm and octopi roundsit down to 0.25, layer height 0.3 for example in the gcode every layer will be 0.554 etc and a rounding error of 4 thousands on each layer, would end up at 0,66mm of a 50mm model .. tho its far from the huge error oyu got .. but noticed this problem myself with ocotopi and even mainsail that the models
01:39 AM MrSunshine: height is reported wrongly
01:39 AM MrSunshine: do not think it affects the actual print tho, atleast ive not noticed any problem
01:39 AM XXCoder: MrSunshine: it was not tiny error, it was way off
01:39 AM XXCoder: like 1/3 short
02:09 AM Deejay: moin
03:50 AM tcurdt: pcw--home it was suggested you could maybe help with the 7i92 ... I would appreciate a ping if you can
04:36 AM randy: tcurdt: did you tried to use mesaflash with the "--fix-boot-block" yet?
04:36 AM randy: i needed that once when i played around with custom bitfiles on my 7i92
04:39 AM tcurdt: randy no, not yet. thx for the pointer!
04:48 AM tcurdt: I am little hesitant to try more at this stage ... the potential to brick the card isn't particular enticing
04:48 AM tcurdt: I just found the 7i92_fallback.bit
04:48 AM tcurdt: I would guess I could try to flash that
04:48 AM randy: the video is in german, but you will see the commands how to reflash the card. a "normal" fw + the fallback fw
04:48 AM randy: https://youtu.be/WYoe1mRrcGg?t=362
04:48 AM tcurdt: my mother tongue is German so - that should work :-D
04:48 AM randy: ja dann sollte es kein problem sein <scnr> :)
04:50 AM tcurdt: do I understand correctly that --fallback selects which configuration (of the two) to write to
04:50 AM tcurdt: so as long as I don't ever write with --fallback I will always have a fallback config and cannot not brick my card
04:51 AM tcurdt: does that sound correct?
04:51 AM Tom_L: pretty sure that's right
04:52 AM Tom_L: i'm not sure which bit file you need to load so i'd wait for pcw--home
04:53 AM Tom_L: JT-Shop might know
04:56 AM randy: tcurdt: i did it the other way around - wrote fallback bit file, reboot to fallback, tested it (readhmid) and after that, flashed the normal one
04:57 AM Tom_L: the fallback should already be loaded
04:57 AM tcurdt: if I understand the flashing LED correctly the fallback is currently loaded and something is wrong with the other config
04:58 AM Tom_L: so wait until you know what bit file you need then load it
04:58 AM tcurdt: so my guess is I should be able to write the fallback to the user config
04:58 AM Tom_L: i'm not sure why you would do that
04:58 AM tcurdt: just to get back to the default state
04:58 AM tcurdt: my assumption is that the two configs are the same by default
04:59 AM tcurdt: but ... that's just an assumption at this stage :)
04:59 AM Tom_L: not always true
05:00 AM tcurdt: for now I just want to see it init led not blink and having a user config loaded
05:00 AM Tom_L: mine had functions loaded like stepgen encoders etc
05:00 AM Tom_L: have a little patience :)
05:00 AM tcurdt: it's hard - but I try my best :)
05:01 AM Tom_L: since i don't know which bit file you need and you don't...
05:02 AM tcurdt: I probably need to create the bit file myself anyway
05:03 AM tcurdt: but I'll wait :)
05:04 AM randy: tcurdt: if you need help, i have ISE running vor "backing" bitfiles already
05:05 AM tcurdt: oh oh - I don't even know what ISE is yet :)
05:05 AM Tom_L: tcurdt, what hardware do you plan to connect to the 7i92?
05:06 AM Tom_L: you shouldn't need ISE unless you're building a custom bit file
05:06 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/emc/xilinx/xilinx14_install_index.php
05:06 AM Tom_L: but that's how you do it in case you ever wanna know
05:07 AM Tom_L: that is also on the wiki
05:07 AM tcurdt: given the darn shortage I have to be a little creative ... I have a mach3 parallel bob on the way ... not great but well
05:07 AM Tom_L: and was probably written prior to ethernet cards
05:07 AM tcurdt: worst case I'll try to create my own board
05:07 AM Tom_L: so you want some sort of bob bit file
05:08 AM tcurdt: yes, for now that seems to be the way
05:09 AM tcurdt: is there no command line way to build the bit file? does it have to be this Xilinx Design Suite thingy?
05:10 AM Tom_L: yes because it's vhd
05:10 AM Tom_L: but the zip you downloaded with your card has all them
05:10 AM Tom_L: already built
05:10 AM Tom_L: mesaflash --device 7i92 --addr 10.10.10.10 --write bitfilename.bit
05:11 AM tcurdt: what's "vhd"?
05:11 AM Tom_L: would be how to write the file to the board
05:11 AM tcurdt: yeah, I figured that
05:11 AM Tom_L: a fpga language
05:11 AM Tom_L: there is .v and .vhd files one is verilog and the other is vhdl
05:11 AM Tom_L: he doesn't use .v files
05:11 AM tcurdt: and there is no command line compiler for that?
05:11 AM Tom_L: no
05:12 AM Tom_L: it goes thru several processes to get to be a bit file
05:12 AM tcurdt: urgh
05:12 AM Tom_L: pin assignments etc
05:13 AM Tom_L: did you change the default ip address?
05:13 AM Tom_L: 192.168.1.121 is default
05:13 AM tcurdt: and ISE is a paid product?
05:13 AM Tom_L: it's free
05:13 AM Tom_L: but eol now
05:13 AM Tom_L: 14.x is the last version
05:13 AM tcurdt: I am using 10.10.10.10
05:13 AM Tom_L: 14.7 iirc
05:14 AM Tom_L: ok good
05:14 AM Tom_L: most users don't need ISE webpack
05:15 AM Tom_L: it's like a 6g download
05:15 AM tcurdt: urgh
05:15 AM Tom_L: and you shouldn't need it
05:15 AM Tom_L: afk...
05:16 AM tcurdt: but that would mean I cannot assign the pins as I want (IIUC?)
05:17 AM * tcurdt wonders what mesa will do in the long run if ISE is EOL
05:19 AM tcurdt: I cannot seem to find it here https://www.xilinx.com/support/download/index.html/content/xilinx/en/downloadNav/design-tools.html
05:26 AM JT-Cave: morning
05:32 AM randy: tcurdt: for the mach3 "china bob" there should be a bitfile already. its called 7i92_ChinaBob IIRC
05:33 AM tcurdt: randy at least it's not in the zip I downloaded
05:34 AM JT-Cave: I've never seen a 7i92 bit file by that name
05:34 AM tcurdt: there is a _5ABOBx2D
05:34 AM tcurdt: which sounds like a possible match
05:35 AM randy: or this, yes
05:35 AM tcurdt: not sure what the x2D stands for
05:35 AM JT-Cave: if it's a 5 axis bob then use 7i92_5ABOBx2D.bit
05:36 AM tcurdt: what does the x2D mean?
05:36 AM JT-Cave: x2 means it's a double ie both headers have stepgens
05:36 AM JT-Cave: D is dpll
05:37 AM JT-Cave: you could use my mesa configuration tool to flash and build your configuration
05:38 AM tcurdt: so x2 means "have the pins mapped to both both connectors"?
05:38 AM tcurdt: what's "dpll"?
05:38 AM randy: tcurdt: in 7i92_5ABOB_Enc.pin or 7i92_5ABOBx2D.pin you will see the pin layput
05:38 AM randy: (http://www.mesanet.com/software/parallel/7i92.zip)
05:38 AM tcurdt: I would have guessed 7i92_5ABOBx2D.pin
05:38 AM tcurdt: :)
05:39 AM tcurdt: what's the _Enc one?
05:39 AM tcurdt: JT-Cave got a link to your tool?
05:39 AM randy: the second pinheader is "only" for encoders (6 encoders in total)
05:40 AM JT-Cave: https://gnipsel.com/mesa/mesact/index.html
05:42 AM tcurdt: JT-Cave wow - mesact looks fancy!
05:42 AM JT-Cave: I put a lot of work into it
05:51 AM tcurdt: JT-Cave Cool. I will check it out.
05:51 AM tcurdt: JT-Cave Any particular reason you have the debs in the repo? I'd recommend to have them as release assets instead. In a repo is usually not such a great idea.
05:53 AM tcurdt: JT-Cave what's https://github.com/jethornton/7i92 ?
06:01 AM JT-Cave: I just started making releases so it's new to me
06:02 AM randy: which is really cool, btw
06:05 AM JT-Cave: I don't know what a release asset is...
06:08 AM tcurdt: JT-Cave you can attach the release assets (debs in this case) here https://github.com/jethornton/mesact/releases
06:31 AM -!- #linuxcnc mode set to +v by ChanServ
06:55 AM tcurdt: JT-Cave can you explain the difference betwen mesact and the 7i92 repo? they seem to the very least overlap in terms of features, no?
06:56 AM JT-Cave: 7i92 was first along with all the others
06:56 AM JT-Cave: mesact is all the rest rolled into one and is my latest work
06:59 AM JT-Cave: 7i92 was the first one to have support for more than one board so paved the way for mesact
07:08 AM tcurdt: ah ok!
07:08 AM tcurdt: looking really good
07:08 AM tcurdt: still too many questions marks finish the config ... but it's pretty cool
07:09 AM tcurdt: about axis and joints ... my machine has two y motors ... does that mean 2 joints but 1 axis?
07:10 AM tcurdt: and should I better drive the motors separately? or share the stepgen?
07:17 AM tcurdt: JT-Cave why does mesaact require a more recent mesaflash?
07:18 AM JT-Cave: for the 7i96s
07:19 AM JT-Cave: dual motor gantrys use two joints with the same axis
07:22 AM tcurdt: JT-Cave so should the older version work with the 7i92? not eager to compile it myself.
07:24 AM JT-Cave: yes
07:24 AM JT-Cave: it's easy to build and install and only takes a few minutes
07:26 AM tcurdt: JT-Cave the 7i92_5ABOB_Enc.bit isn't listed in the dropdown ... does it come with just the fix set of configs?
07:27 AM Tom_L: getting things sorted out?
07:28 AM tcurdt: still waiting for pcw--home ... although almost feel confident enough to just try it :)
07:28 AM tcurdt: right now looking at the config ... and wha JT-Cave built
07:29 AM tcurdt: still a little overwhelmed with all the question/information I need to put in :)
07:29 AM -!- #linuxcnc mode set to +v by ChanServ
07:32 AM tcurdt: like ... Position Feedback? Position Offset? Spindle Encoder Scale and PID Settings?
07:32 AM Tom_L: some of them you won't know yet
07:33 AM Tom_L: position feedback is for encoders
07:33 AM tcurdt: ...and I assume the restriction of the encoders to 2 is because of x2D bit file? (as effectively only one of the parports is used)
07:33 AM Tom_L: probably
07:33 AM Tom_L: i'm out the door..
07:42 AM JT-Cave: tcurdt, I just pushed a new deb to the 7i92 repo
07:43 AM tcurdt: yup, that's the way to do the assets :)
07:43 AM tcurdt: JT-Cave but just to confirm - I should rather be using the mesact, right?
07:45 AM JT-Cave: yes, that's where most of my effort is going to now
07:46 AM JT-Cave: mesa cards all use pid and the default is what you want
07:46 AM JT-Cave: pressing F1 on any tab brings up limited help for that tab
07:46 AM tcurdt: JT-Cave where does mesact get the bit files from?
07:47 AM tcurdt: ah .. found it
07:47 AM tcurdt: the in "holding" still need work/support?
07:48 AM JT-Cave: yes
07:49 AM tcurdt: ah ... bummer :)
07:49 AM JT-Cave: 99% of the firmware needed is available to use, the rest are pretty special... thinking of adding a button to just show all dunno
07:49 AM tcurdt: I am pretty sure I want 7i92_5ABOB_Enc ... and it's in "holding"
07:51 AM JT-Cave: you need encoders on one header?
07:56 AM JT-Cave: just added 7i92_5ABOB_Enc.bit
08:01 AM tcurdt: JT-Cave while the plan is to have close loop stepper/servos (so they should already take care of the PID) I still think it would be nice to connect the encoders so LinuxCNC also has a real idea what's going on
08:01 AM tcurdt: not entirely sure this makes sense or will work
08:01 AM tcurdt: but it seems more useful than to just use one of the par ports
08:02 AM tcurdt: "just added 7i92_5ABOB_Enc.bit" ... oh - was it that simple?
08:15 AM * JT-Cave heads out to start his chicken chores
08:24 AM tcurdt: JT-Cave chickens? :) ... did you just update the deb?
08:25 AM tcurdt: (of mesact)
08:27 AM tcurdt: JT-Cave I don't think the board options are working for the _Enc config
09:08 AM pcw--home: tcurdt: if you ge a "bad boot block" warning, the first thing to do is to upgrade to the latest mesaflash
09:08 AM pcw--home: https://github.com/LinuxCNC/mesaflash
09:15 AM tcurdt: hey! pcw--home ... I was planning on doing this "mesaflash --device 7i92 --write 7i92_5ABOB_Enc.bit --fix-boot-block" but I wanted to check with you first
09:17 AM tcurdt: Did I understand this correctly that as long I don't use "--fallback" there always be a fallback to fall back to ... and won't brick my card (until restored via JTAG)?
09:17 AM tcurdt: pcw--home Oh, and can you confirm the difference between 7i92_5ABOBx2D.bit and 7i92_5ABOB_Enc. ?
09:19 AM pcw--home: You can read the associated .pin files to determine a configuration pinout
09:21 AM pcw--home: I would not use the "Fix boot block" option until you try with the latest mesaflash
09:21 AM tcurdt: ok!
09:23 AM pcw--home: It most likely means you have a newer 7I92 that happens to have a 32 MBit flash chip which the old mesaflash cannot deal properly with
09:25 AM tcurdt: I think I will just try with the new one then ... but that one I will have to build myself, correct? https://github.com/LinuxCNC/mesaflash/issues/73
09:26 AM pcw--home: the build instructions are at the bottom of the page
10:24 AM tcurdt: pcw--home "./mesaflash --device 7i92 --addr 10.10.10.10 --write 7i92_5ABOB_Enc.bit --reload" did the trick .. the LED is no longer blinking :) (yay!)
10:24 AM tcurdt: "./mesaflash --device 7i92 --addr 10.10.10.10 --readhmid" also looks fine
10:35 AM tcurdt: for a double y machine I would use y plus y tandem, right?
10:39 AM pcw--home: you mean a gantry machine?
10:44 AM tcurdt: this one https://wiki.printnc.info
10:48 AM roycroft: in linuxcnc parlance, you would have a y axis with two joints
10:49 AM tcurdt: roycroft and the 2nd one is called a "tandem"?
10:49 AM roycroft: each of them is called a joint
10:50 AM roycroft: neither is called a tandem
10:50 AM roycroft: essentially, axis = line of travel, joint = motor
10:51 AM roycroft: the joints may or may not be synchronized
10:51 AM tcurdt: tandem is what is offered in the pncconf wizard ... hence my wondering
10:52 AM tcurdt: axis vs joint is clea
10:52 AM tcurdt: r
10:52 AM roycroft: ok, perhaps that refers to synchronized joints as a tandem
10:52 AM tcurdt: how would that sync look like is the question
10:53 AM tcurdt: and should the above machine use such a sync or not
10:56 AM t4nk_freenode: I'm disappointed in you, tcurdt ... you caved, and instead of tapping in the bits with a used doorbell, like a man, ... you decided to buy a mesa card :|
10:57 AM tcurdt: t4nk_freenode lol ... what? :)
10:58 AM t4nk_freenode: :b
11:04 AM tcurdt: t4nk_freenode what used doorbell was that? :)
11:06 AM tcurdt: I was really tempted to go down the remora route (with all the mesas not being available) - but I found one ... at least the 7i92
11:06 AM tcurdt: I took that as a sign :-p
11:07 AM t4nk_freenode: what did it cost you btw?
11:07 AM tcurdt: it's not a good time to talk about money atm ;)
11:09 AM tcurdt: around 10 EUR more than shipping from the US
11:09 AM tcurdt: which is ok
11:16 AM t4nk_freenode: 80euros?
11:17 AM tcurdt: t4nk_freenode I wish .... that's not the price when you ship from the US
11:19 AM t4nk_freenode: ah well, as long as it works, right?
11:20 AM t4nk_freenode: really curious what it actually cost though, but... sometimes it's best to just forget about it, and move on
11:20 AM t4nk_freenode: use it, and be happy
11:24 AM tcurdt: t4nk_freenode 125 EUR ... IIRC shipping from the US would have been around 115 EUR
11:24 AM tcurdt: t4nk_freenode happy? as soon as I have linuxcnc configured ;)
11:25 AM t4nk_freenode: lol
11:25 AM tcurdt: so that might still take a while ;)
11:26 AM solarwind: Hey guys, what solders bond to steel or galvanized?
11:27 AM t4nk_freenode: good question, solarwind, would also like to know
11:28 AM solarwind: bronze brazing rod is great, but it'll always cause the zinc to boil off
11:28 AM solarwind: I'd like to avoid that
11:28 AM t4nk_freenode: probably steel from a welding machine
12:06 PM JT-Shop: tcurdt, tandem is just a word that Chris used it means nothing
12:07 PM tcurdt: JT-Shop hmmm ... but it's a different option in the config
12:07 PM tcurdt: in the wizard - that is
12:07 PM JT-Shop: which wizard?
12:08 PM tcurdt: JT-Shop pncconf wizard
12:08 PM JT-Shop: a multi joint axis can have 2 or more joints and all you do is assign the same axis to the joints
12:08 PM tcurdt: alright
12:08 PM JT-Shop: I don't use it much so I'm no help with Chris's tool
12:09 PM tcurdt: as long as it's ok to ignore I am OK too :)
12:09 PM tcurdt: end the end - it would just be for start anyway
12:09 PM tcurdt: I am pretty sure I will end up editing the files by hand anyway
12:09 PM JT-Shop: if you want them to home coordinated you simply make the homing sequence of one or more joints negative and all with the same sequence
12:10 PM * JT-Shop goes back to work
12:11 PM tcurdt: talking about homing ... I'll have boat load of end stops and I am wondering if there is a smart way to safe some inputs
12:13 PM tcurdt: my plan was to have a have a min and max end stop per joint (xyy) and a homing stop per axis (xyz) ... but that's 2*3 + 3 = 9 endstops ... and I was thinking of reporting back the endcoder info of all 4 motors ... that's a whole log of inputs
12:15 PM tcurdt: s/log/lot/
12:17 PM solarwind: t4nk_freenode I think silver bearing solders might
12:18 PM solarwind: I'm also trying to figure out how to make copper braid
12:18 PM solarwind: Out of finely stranded copper from scrap 00 gauge SOOW wire
12:18 PM solarwind: Don't need a lot, just a bit for good electrical contact for ground clamps and such
12:20 PM solarwind: It's fascinating how conductive copper is. I've made SAE battery post terminal clamps to get the maximum surface contact between the copper and lead and it makes quite a bit of difference when you're drawing like 3,000A
12:22 PM roycroft: it seems it would be hard to make a nice, neat copper braid out of that stripped wire
12:22 PM XXCoder: unless you make a auto-braider
12:23 PM roycroft: i would take a copper plate, solder the wires to it in groups - each group being the contents of one conductor of the sjoow, having flattened out the wires, clamp that in a vise, and then braid the bundles by hand
12:23 PM roycroft: keeping them flat and in order is the challenge
12:23 PM roycroft: neither of which is necessary for it to function properly
12:24 PM roycroft: but it something i would certainly want to do if i were making them
12:24 PM XXCoder: https://www.walmart.com/ip/Anself-Automatic-Electronic-Quick-Hair-Braiding-Tool-Black/203197679
12:24 PM XXCoder: wonder if it'd work lol
12:24 PM roycroft: but i would likely just go to the auto parts store and buy braided ground cables
12:24 PM roycroft: xxcoder: braiding is trivial
12:25 PM roycroft: keeping the wire bundles flat and non-overlapping while braiding not so much
12:25 PM solarwind: Orrrrr outsource it to my little sister
12:26 PM solarwind: but yeah, for the purposes of a ground clamp, it doesn't need to be perfect
12:26 PM roycroft: or go to the auto parts store and buy some ready-made, and use your salvaged copper for something else :)
12:26 PM solarwind: It's just that sourcing such an item is difficult
12:26 PM solarwind: What use does it have in automotive?
12:26 PM roycroft: they're used all the time
12:26 PM roycroft: to ground the engine to the chassis, for example
12:26 PM roycroft: the engine vibrates
12:27 PM roycroft: so a copper mesh ground strap works well to keep it grounded
12:27 PM solarwind: I'm talking about something that's about this size: https://m.media-amazon.com/images/I/51GtTQR6ZyL._AC_SL1200_.jpg
12:27 PM solarwind: most of the copper braid I find is for electronics desoldering use which is waaayyy too small
12:27 PM solarwind: roycroft ok thanks, I didn't know it was used for that purpose. I'll go and ask them
12:28 PM roycroft: oh, you're talking about the business end of the clamp itself
12:28 PM solarwind: no, the copper that's connecting the top and bottom part
12:28 PM solarwind: the flexible copper braid
12:28 PM solarwind: Better image: https://m.media-amazon.com/images/I/41c5yc5clFL.jpg
12:29 PM solarwind: I don't want to destroy a clamp just to salvage the copper braid strip
12:30 PM roycroft: https://www.amazon.com/dp/B08X4N6251?th=1
12:30 PM roycroft: that's the kind i'm talking about
12:30 PM solarwind: Yeah that would be great, but it looks very small
12:30 PM solarwind: I guess I could parallel them
12:30 PM roycroft: 8" long
12:31 PM solarwind: Oh, ok then that might do nicely
12:31 PM roycroft: and has lugs for 6ga wire
12:31 PM solarwind: thanks
12:31 PM roycroft: so at least 50z
12:31 PM roycroft: 50z
12:31 PM roycroft: dammit
12:31 PM roycroft: 50a
12:31 PM roycroft: my fingers are not working well today
12:31 PM roycroft: but three of those for $5.50
12:31 PM solarwind: Yeah that solves the problem, thanks!
12:31 PM roycroft: you'd spend probably 20 minutes each making a braid
12:34 PM solarwind: My ground clamp itself seems to get hot at the contact point because the actual surface contact area is low. It's probably causing the welder to increase the voltage to compensate and maintain current. For MIG, that's a problem.
12:35 PM solarwind: Problem goes away when I use a copper braid salvaged from a ground clamp from the above picture. Just need a few more of those
12:36 PM solarwind: It's a solid brass clamp too, but nothing comes close to the conductivity off annealed copper. It's such a magical element
12:37 PM tcurdt: in theory it should be possible to share at least the input pin for the min/max endstops .. can linuxcnc deal with that?
12:38 PM tcurdt: in combination with direction it should be clear whether that signal is a min or max endstop
12:38 PM CaptHindsight[m]: solarwind: may you be blessed with all annealed copper wiring
12:39 PM solarwind: CaptHindsight[m] I'm in Canada, I haven't seen aluminum wiring used anywhere
12:40 PM solarwind: Funny thing is, people build their induction heaters using copper pipe for plumbing from the hardware store, but that's not made out of C110 copper and has significantly lower conductivity than pure copper.
12:40 PM solarwind: Probably doesn't matter if you use thick wall type M pipe, or at least L
12:41 PM CaptHindsight[m]: solarwind: have skyrocketing rental prices hit industrial space there yet?
12:41 PM solarwind: no
12:41 PM solarwind: They increased the rental for this year by $100/month due to "inflation"
12:42 PM solarwind: but I'm already paying less than residental rental price for same sq footage
12:42 PM solarwind: Also landlord forgot that he increased the price and refunded me the "extra" that I paid lol
12:42 PM solarwind: But fuel prices are insane
12:43 PM solarwind: It also heavily depends on the area. The area that I'm in is not "ideal" in terms of proximity to highway and so on. It doesn't bother me, but a lot of people care about that
12:44 PM solarwind: And further out of the city, it hasn't been affected at all as far as I know. Our population density is _very_ low so most people have lots of farm land and if they need to do any kind of industrial type of work, they just use the land they have already.
12:45 PM solarwind: And if you have friends, they'll even let you use part of their warehouse space for free/small favours/very low cost. Like I have a backup industrial space from a friend if I ever need to get out of my unit just because I did all his electrical work and fixed his air compressors and all that.
12:46 PM CaptHindsight[m]: good deal
12:46 PM solarwind: I'm talking > 2,000 sq•ft of space for free
12:46 PM solarwind: But the industrial unit I have right now is great. The parking lot however needs maintenance because I only have cushion tire forklifts and aside from the discomfort, the potholes make it dangerous for loads
12:47 PM CaptHindsight[m]: small industrial space is difficult to find here. <1K sq meters
12:47 PM solarwind: The guys the landlord calls to fill the potholes in every year always screw it up. They just pour some asphalt in there which gets squished like putty under a forklift tire
12:47 PM solarwind: I should just fill them in myself using some reinforced concrete
12:47 PM CaptHindsight[m]: yeah so it just changes the shape of the holes
12:48 PM solarwind: It's so annoying
12:48 PM solarwind: I use my scissor lift to get on top of the roof so I can fly my RC aircraft and the potholes make it so annoying to drive over with a scissor lift
12:48 PM tcurdt: pcw--home Can I (ab)use a QCount as pin for an endstop? It feels like I am getting closer and closer to wanting my own bit file :-/
12:49 PM CaptHindsight[m]: tcurdt: https://forum.linuxcnc.org/38-general-linuxcnc-questions/38815-sharing-limit-switches-and-homing-on-a-single-input
12:49 PM tcurdt: CaptHindsight[m] thx!
12:50 PM CaptHindsight[m]: many posts on the forums about shared Home and Limit switches
12:50 PM JT-Shop: tcurdt, reduce the number of encoders frees up gpio
12:50 PM tcurdt: CaptHindsight[m] I wasn't aware it such a popular thing. I will read up! :)
12:51 PM pcw--home: You can use encoder pins as GP inputs but you must disable the encoder if you want to use them as outputs
12:52 PM tcurdt: JT-Shop I know - but it would be great to have real position info in linuxcnc ... not sure how useful it really for a Close Loop Stepper/Servo system ... just juggling with the possibilities here
12:53 PM tcurdt: pcw--home so the "counter" is not that import - only whether it's input or output
12:53 PM pcw--home: Closing the loop on step/dir systems usually only makes sense if you have linear encoders
12:55 PM JT-Shop: tcurdt, this is where you reduce the number of stepgens, encoders or PWM gens
12:55 PM JT-Shop: https://imagebin.ca/v/6f6Ulzu0cKsr
12:55 PM pcw--home: Yes, special inputs (quadrature encoders absolute encoders, UART RX pins etc) don interfere with use of the pin as an input
12:57 PM tcurdt: thx, pcw--home !
01:00 PM tcurdt: JT-Shop that's a part I don't get ... IIUC the bit file defines the number of stepgens and encoders (and their mapping) ... and one needs that ISE(?) software to compile that into a bit file ... how can the mesact change the number of encoders?
01:01 PM pcw--home: special purpose outputs must be disabled to use ans inputs or general purpose outputs
01:01 PM tcurdt: ah, so the bit file is fix - but there is also configuration component to it?
01:01 PM pcw--home: all I/O pins (of any firmware) support GPIO functions
01:02 PM tcurdt: so what defines "this pin belongs to an encoder"?
01:02 PM pcw--home: The special functions are enabled by the driver
01:03 PM * tcurdt feels a bit lost
01:03 PM tcurdt: the driver on the host configures them?
01:03 PM pcw--home: the driver command line specifies the number of each module enabled
01:03 PM pcw--home: (from 0 to N)
01:04 PM tcurdt: so the bit file just defines whether it's an input or an output and the driver sends the purpose of the pin?
01:04 PM pcw--home: All I/O pins are GPIO set as inputs until the driver set up the special functions
01:05 PM pcw--home: No, all pins are GPIO
01:05 PM tcurdt: I feel like I am asking stupid questions - is there something read up on this so I don't have to bother you?
01:05 PM pcw--home: the bit file determines the secondary pin function (stepgen, PWMgen, encoder etc)
01:06 PM pcw--home: you can read the hostmot2 manual page
01:07 PM pcw--home: if you disable all secondary functions, all pins can be GPIO and any direction
01:07 PM -!- #linuxcnc mode set to +v by ChanServ
01:08 PM pcw--home: secondary pin function is actually settable on a pin by pin basis
01:08 PM pcw--home: but to keep the driver command line simple you can only enable 0..N modules (encoders etc)
01:11 PM tcurdt: so ... without changing the bit file I could reduce the number stepgens and encoders and use them as GPIO - input or output ... as I like?
01:12 PM tcurdt: so basically the bit file implements the stepgen and encoders and so on but the driver assigns the pins?
01:13 PM tcurdt: where I have trouble is why it would need so many different bit files then 7i92_5ABOB_Enc vs 7i92_5ABOBx2D
01:14 PM tcurdt: but I need to go now ... and probably need to do some homework first to ask better questions
01:16 PM JT-Shop: tcurdt, different bit files put different special functions on different pins or is for a special daughter card like 7i77 or 7i76 etcc
01:16 PM tcurdt: JT-Shop but I thought the driver assigns the pins
01:17 PM tcurdt: the bit file only has n instances of the modules
01:17 PM tcurdt: which are compile time
01:18 PM pcw--home: The pinout is fixed by the bit file , The driver can only enable or disable the secondary (non GPIO) functions
01:19 PM tcurdt: aaaaah ... ok
01:19 PM tcurdt: so the special functions (if disabled) fall back to norm GPIO
01:20 PM pcw--home: Yes
01:20 PM tcurdt: so this took so long :)
01:20 PM tcurdt: thanks for the patience
01:28 PM -!- #linuxcnc mode set to +v by ChanServ
03:00 PM Tom_L: so tcurdt you got your mill up and running now right?
03:16 PM -!- #linuxcnc mode set to +v by ChanServ
03:38 PM -!- #linuxcnc mode set to +v by ChanServ
03:39 PM CaptHindsight[m]: https://www.youtube.com/watch?v=yGgNvV1_3hQ :)
03:40 PM CaptHindsight[m]: precision bend? :) https://youtu.be/yGgNvV1_3hQ?t=487
03:54 PM XXCoder: interesving video CaptHindsight[m]
04:21 PM JT-Shop: pcw--home, refresh my memory first column is the physical pin on the header and the second column is the gpio number https://pastebin.mozilla.org/eRss3Vmq
04:39 PM _unreal_: sweet just got my new rails/carriages for the new cnc
04:39 PM _unreal_: I should say NEW NEW rails
04:39 PM _unreal_: considering I'm replacing the Z rails from what I had before
04:39 PM XXCoder: what happened to old new rails
04:40 PM _unreal_: nothing
04:40 PM _unreal_: I'm rebuilding the X/Z axis setup
04:40 PM _unreal_: to a different system
04:42 PM _unreal_: the original z setup was built for a much longer z travel
04:42 PM _unreal_: I dont need all that travel now
04:44 PM _unreal_: and its build way more heavy duty
04:49 PM _unreal_: sweet almost done cutting the tabs with my cnc :)
04:49 PM _unreal_: I was not sure if I wanted to cut it on my bandsaw or just do a full cut so I elected t do tabs
04:49 PM _unreal_: at this point the next one I do with these new milling bit. I'll likely do no tabs regardless
04:50 PM _unreal_: still using the same bit I used yesterday going like a champ
04:50 PM _unreal_: I really cant wait to be able to just pick up the part
04:50 PM _unreal_: normally I fight tabs but this is cutting so clean
04:52 PM roycroft: i'm seriously considering getting a composite finish nailer for holding down parts on my cnc router
04:53 PM roycroft: and for other uses as well
04:53 PM roycroft: that would avoid the tab/double sided tape/whatever issues
04:55 PM CaptHindsight[m]: I wish they worked on metals parts
04:55 PM roycroft: that might be nice
04:55 PM roycroft: you can use ca glue for some metal parts in some circumstances
04:56 PM CaptHindsight[m]: the Russians are so poor they can't even get metal to make tools
04:56 PM CaptHindsight[m]: https://www.youtube.com/watch?v=Y8ctS8TMKL0
04:58 PM CaptHindsight[m]: another hack https://youtu.be/F0wjPqT2GLs?t=375
04:58 PM XXCoder: thats great for starting rough circles
04:58 PM XXCoder: or erm complete..
04:59 PM CaptHindsight[m]: rotary table for table saw
04:59 PM CaptHindsight[m]: lets CNC one!
04:59 PM CaptHindsight[m]: i should do some silly CNC's like that
05:00 PM _unreal_: hum
05:00 PM CaptHindsight[m]: tool post grinder for my bench grinder
05:01 PM CaptHindsight[m]: 2 axis table for saw and carve spheres, or for a drill press
05:01 PM _unreal_: UGH... Cant wait to pull this part
05:02 PM XXCoder: wood one wow its far more fancy than woodgears one lol
05:08 PM Tom_L: JT-Shop, is that the xml or the vhd file?
05:09 PM Tom_L: if i had to guess... xml
05:09 PM JT-Shop: pin file
05:10 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/emc/xilinx/xilinx14_install_index.php
05:10 PM Tom_L: dunno if this will help what you're doing but..
05:10 PM Tom_L: The first array in the .vhd file is the moduleIDType array with 32 entries and the module ID fields are:
05:10 PM Tom_L: 'TAG, VERSION, CLOCK, INSTANCES, BASE ADDRESS, NUMBER OF REGISTERS, STRIDES, MPBITMASK'
05:10 PM JT-Shop: no, I'm just parsing out the gpio from the pin file
05:11 PM Tom_L: probably the more technical version of what you're after...
05:11 PM Tom_L: i'm not sure i've ever looked at a pin file :)
05:11 PM JT-Shop: I'm just trying to use any free gpio in my mesact program
05:11 PM Tom_L: did you see bambi?
05:12 PM JT-Shop: the other day I did
05:12 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/rue/reindeer/reindeer17.jpg
05:12 PM XXCoder: cast mold?
05:12 PM JT-Shop: nice
05:12 PM XXCoder: awesome like last time saw it :D
05:12 PM Tom_L: the circle is supposed to be a loop for hanging ornament
05:13 PM Tom_L: needs a little cleanup but that's on him
05:13 PM XXCoder: I plan to do melting tests for pla sometime. I did very roough basics usinbg air heat device
05:13 PM Tom_L: he's making a pipe extruder
05:13 PM XXCoder: not sure how but want to cast pla for fun, using pla waste
05:13 PM Tom_L: using eddy currents to heat it
05:14 PM XXCoder: wheres injection point anyway
05:14 PM Tom_L: we were discussing that
05:14 PM Tom_L: probably in the neck area
05:14 PM Tom_L: http://ruemohr.org/~ircjunk/tempimage/p1230648.jpg
05:15 PM XXCoder: wonder if best to have it by middle of body side, and add thin air drains at all extremes
05:15 PM Tom_L: plans to use up old filament or leftover scraps
05:15 PM XXCoder: making thin "hair" thats easily removable
05:15 PM Tom_L: it's experimental
05:15 PM Tom_L: he likes to do things the hard way but usually succeeds
05:16 PM XXCoder: yeah. well unless plates isnt perfectly flush then it'll be fine besides from flash
05:16 PM XXCoder: out to walk the woofmaker. be back in 5-30 min
05:16 PM Tom_L: and if proof of concept works may make more molds for things
05:24 PM XXCoder: Tom_L: good luck :D
05:25 PM XXCoder: crafsman has video on making your own injection plastic stuff
05:25 PM XXCoder: I cant afford the minium "vitamins" like heat cylinder
05:27 PM roycroft: i have printed at least one of every size of my hardware boxes now, and verified that the dimensions are correct
05:27 PM roycroft: i'm working on the batch prints now, to ensure that they all print properly as well
05:27 PM XXCoder: roy thats awesome
05:27 PM roycroft: yeah, i'm glad i'm at the point where i can start cranking them out
05:28 PM roycroft: also, the first few boxes of drawer slides will allegedly be delivered today, so i might be able to get back to my cabinet models soon
05:28 PM XXCoder: roy yeserday I played around with melting scrap pla with heat air. it worked, but too awkward lol
05:28 PM XXCoder: I'll try oven later
05:29 PM roycroft: there are filament recycling machines
05:30 PM roycroft: https://www.filastruder.com/collections/filastruders-accessories
05:30 PM roycroft: looks like they're out of stock at the moment, but they have a cheap filament extruder
05:31 PM XXCoder: yeah im not making filament but casting it :)
05:31 PM roycroft: all you have to do is figure out how to grind up your scrap plastic and you're good to go with one of those
05:31 PM roycroft: if you're casting that's easier
05:32 PM XXCoder: yeah im thinking what to cast em though lol. maybe something fun shape
05:39 PM XXCoder: im not makijng any molds unlike tom so its likely kitchen food mold of some kind
05:42 PM XXCoder: hmm browsing a bit, theres silicne ones. cool, though since I start woith large pile of pla i doubt I can use those since it can be flattened or something
05:43 PM XXCoder: or can pla liquidify enough to be poured?
05:44 PM t4nk_freenode: well, _unreal_, that vid was absolutely useless ;)
05:44 PM t4nk_freenode: the only thing that was clear on it... were the bits
05:44 PM t4nk_freenode: lol
05:44 PM XXCoder: looks like no. must be pressured into form. aka inject mold
05:44 PM t4nk_freenode: you should ask the seller money for such advertisement
05:45 PM t4nk_freenode: I will say though... that seems like some very thick aluminium you got there
05:45 PM t4nk_freenode: how thick is it?
05:45 PM _unreal_: ok I'm all zeroed at this point.
05:46 PM _unreal_: now I just have to :S redo the CAM for the flipped part
05:56 PM _unreal_: music to my :S ears... now cnc drilling holes then milling
06:22 PM _unreal_: ok all showered
06:22 PM _unreal_: machine survived and is still tooling away drilling holes
06:22 PM _unreal_: I'm still SOOOO in love with my mini DIY dust collector
06:23 PM roycroft: was oregon just annexed by the brd?
06:24 PM roycroft: is olaf schulz my new leader?
06:25 PM XXCoder: interesting. I dont see any ads on youtube lol
06:33 PM _unreal_: FINALLY the drill cycle is done
06:40 PM _unreal_: I'm still using the same bit I was using the other day
06:40 PM _unreal_: LOVE these bits.
06:41 PM _unreal_: last night I purchased 3 more boxes of them thats 40 bits at $12 each. NOT bad
06:41 PM _unreal_: at all
06:42 PM _unreal_: considering how long the other bits I was and HAVE been using with great success up to this point.
06:42 PM _unreal_: estimated milling time for all of the holes 2 hors
06:42 PM _unreal_: hours
06:42 PM _unreal_: holes drilling was 40min
06:42 PM _unreal_: wonder what the long CUT out will be ?
06:49 PM _unreal_: ? no one around
06:50 PM Tom_L: nope
06:54 PM _unreal_: damn
06:55 PM _unreal_: I am having a hind sight moment and thinking I should have run the new milling bit over the last work peace before I cut the tabs and removed it. just to ensure it was cleaned up
06:55 PM _unreal_: :/
06:55 PM _unreal_: sigh
06:55 PM _unreal_: oh well
06:56 PM _unreal_: so far these bits are doing a reallllly nice clean job
06:56 PM _unreal_: never had such good results at this speed
06:56 PM Tom_L: you could have been done with it by now!
06:56 PM Tom_L: or cut the tabs and find out you screwed up
06:59 PM _unreal_: what also is amazing is these HDPE lead screw nuts are still holding up just fine
06:59 PM _unreal_: no play at all
06:59 PM _unreal_: and milling aluminum with them :)
06:59 PM _unreal_: I've got
06:59 PM _unreal_: ....
06:59 PM _unreal_: ..
06:59 PM _unreal_: ahh...
07:00 PM _unreal_: oooo 377 hours of run time on the new hdpe nuts I build
07:00 PM _unreal_: just using threaded rod
07:04 PM _unreal_: Tom_L, its interesting, drilling holes really heated up the aluminum plate. but milling is not. its cooled off a lot.
07:32 PM roycroft: my drawer slides have started arriving
07:33 PM roycroft: and since they came from amazon, each box of drawer slides is enclosed in a ginormous amazon box that is about 6x the volume of the drawer slide box
07:33 PM roycroft: so i have a tremendous amount of paper and cardboard to recycle
07:33 PM XXCoder: fun
07:33 PM roycroft: yes
07:33 PM roycroft: and as well, these slides do not conform to the universal 32mm standard AT ALL
07:33 PM _unreal_: I really have to admit these new bits are just fucking amazing
07:34 PM _unreal_: really leaving a clean surface
07:34 PM roycroft: i'm glad i did not complete my cabinet design until i got the first batch of them
07:34 PM roycroft: o
07:34 PM _unreal_: the finish it just muhhahaaa finger kissing good
07:34 PM roycroft: i'm going to have to do some careful measuring and dimensioning of them
07:34 PM roycroft: but they're about 1/2 the price of the ones i usually get, and other than the non-standard dimensions, seem rather sturdy
07:34 PM _unreal_: i'm still on the same bit I was using last night
07:34 PM _unreal_: these bits are CHAMPS
07:36 PM _unreal_: t4nk_freenode, you should get your self some of those bits I posted yesterday
07:36 PM _unreal_: they are just totally amazing
10:20 PM _unreal_ is now known as Guest8744
10:20 PM acer is now known as _unreal_