#linuxcnc Logs

Jan 08 2023

12:36 AM roycroft: my vinyl cutter arrived today while i was gone
12:47 AM roycroft: when i got home i brought it inside and notice the box was rattling
12:47 AM roycroft: i just opened it up, and it is used, parts are missing, and it was just dumped back in the box with no packing materials
12:47 AM XXCoder: cricut?
12:47 AM roycroft: no, vevor
12:47 AM XXCoder: ok thats good. I have a cricut i need to work to sell off but meh been lazy
12:47 AM roycroft: https://www.amazon.com/dp/B09FLHBR1B/?th=1
12:47 AM XXCoder: is it your first one?
12:47 AM roycroft: yes
12:47 AM XXCoder: ok
12:47 AM XXCoder: I guess questions is pointless then lol
12:47 AM roycroft: there are inkscape drivers for it
12:48 AM roycroft: that's the point
12:48 AM roycroft: it would pay for itself with that one job
12:49 AM roycroft: it's not a great cutter, but it should be ok for occasional use, if i get a new one with all that parts and that is not broken
12:49 AM XXCoder: yeah, I guess you'd have to return it and buy another I guess
12:49 AM XXCoder: if they dont ask for return that'd be awesome for spare parts
12:50 AM roycroft: i've already gotten a return authorization
12:50 AM roycroft: and i already ordered another one
12:50 AM roycroft: they want me to take the box of junk to a ups store and give it to them
12:50 AM XXCoder: thats good, delays is annoying though
12:50 AM roycroft: oh well
12:50 AM roycroft: i am patient
12:50 AM XXCoder: probably ship again and hope next one isnt saavy
12:50 AM roycroft: i'd rather wait than put up with the crap
12:52 AM XXCoder: makes sense yeah. buy is to different seller?
12:52 AM roycroft: it is from amazon
12:52 AM roycroft: as was the first one
12:52 AM XXCoder: unchecked return I guess
12:52 AM roycroft: yes
12:52 AM roycroft: and if the next one is the same i'll return it
12:52 AM XXCoder: is it even same model?
12:52 AM roycroft: eventually i'll get a new one
12:52 AM roycroft: i don't know
12:52 AM roycroft: it is supposed to be
12:53 AM roycroft: but i did not dig in enough to verify
12:53 AM XXCoder: not very important
12:53 AM roycroft: i didn't even unbox it - i just opened the box and was able to determine immediately that it's used and missing parts and damaged
12:54 AM roycroft: amazon have gotten really bad at not checking returns
01:03 AM XXCoder: not surpised
01:03 AM XXCoder: probably cheaper to try to find buyers who dont return em
01:06 AM sensille: i once bought a mini surface plate from amazon. it came in a wooden box. it was fine, but i decided to buy a larger one, so i returned it
01:07 AM sensille: 2 week later i got an unexpected package from amazon containing just the plate. they kept the box and returned the plate to me as "surplus"
01:08 AM XXCoder: uh was it still any good?
01:08 AM sensille: yes
01:08 AM XXCoder: wow because even small one it would need protectip for shipping. nice
01:08 AM sensille: so they were fully content with the box
01:09 AM XXCoder: strange
01:09 AM sensille: i bet during transport the plate fell out of the box, so they received a package with 2 items in it
01:10 AM sensille: one looked like the others on the shelf (a box) and the other was just a piece of stone i put in there to annoy them
01:12 AM sensille: just an example of the level of checking they do on returned items
01:31 AM XXCoder: might be the case yeah
02:03 AM XXCoder: with qtpyvcp you can even make your own ui
02:04 AM travis_farmer[m]: G'Morning, dang sister made pot butter yesterday, stunk up the whole house, and i didn't sleep worth a darn (headache). never did pot, and never will... not if it gives me a headache just from being in the same house...
02:07 AM XXCoder: you ,ay be actually allegric to it
02:08 AM travis_farmer[m]: could be...
02:12 AM sensille: maybe it is not too hard to add more keyboard shortcuts to qtdragon
02:43 AM Deejay: moin
02:44 AM travis_farmer[m]: i kinda wonder what my new ATC spindle has for pneumatic piston rings... i know the air nailers i work with need pneumatic tool oil, yet the air lines on my spindle for both the clearing air and the change air needs to be dry (no oil or water). makes me wonder what the longevity of the spindle will be.
03:10 AM sensille: nice: https://i.imgur.com/Wp7JQ0J.jpg (chinese touch probe vs. chinese micrometer)
03:11 AM XXCoder: fancy
03:11 AM XXCoder: I wouldnt try to certify it but plenty good for home shop
03:11 AM XXCoder: try moving it out and in few times then measure it again
03:12 AM sensille: quite consistent, +/- 0.001mm
03:12 AM XXCoder: thats decent enough
03:13 AM sensille: i'm sure the runout ruins everything
03:13 AM XXCoder: touch probe tend to have adjustments to center it
03:14 AM sensille: + spindle runout
03:16 AM XXCoder: yep, hopefully it can be adjusted when its held by spindle
03:20 AM sensille: and then always mount in with the same orientation?
03:21 AM XXCoder: can always mark it yes, and align to known feature
03:21 AM XXCoder: unless range is just tiny enough that you dont care
03:21 AM sensille: the collet can also contribute to runout. so many variables to control
03:21 AM XXCoder: like if runout is .002mm
03:21 AM sensille: probably
04:25 AM jpa-: travis_farmer[m]: probably polymer; even for long life, the total number of cycles will not be that huge
04:26 AM travis_farmer[m]: true... not as many cycles as a air nailer, i suppose... makes me feel better :-)
04:52 AM travis_farmer[m]: hmmm, just realized... my spindle outputs a tach signal (24V) that is 1 pulse per revolution, so i can use that to read the spindle speed. the problem is, it is 24V, and of course my 7i80HD-16 only takes in 5v, so i need an optocoupler to change the voltage. the ones i have, i will need to get the part number off them so i can lookup the datasheet, and see if they can work with a 24Khz signal...
05:00 AM jpa-: most optocouplers should be fast enough for that - low load resistance (high current) helps
05:00 AM travis_farmer[m]: ok, good :-)
05:04 AM travis_farmer[m]: these are the opto boards i use: https://www.ebay.com/itm/125585742375 and even zooming in, i can not make out the part number. maybe 837, and a lookup on digikey, for an example 837, says maybe 80Khz, so like you say, it should work. :-)
05:11 AM He[m]1: the product description on that makes absolutely 0 sense haha
05:14 AM travis_farmer[m]: how do you mean?
06:14 AM Tom_L: morning
06:16 AM travis_farmer[m]: hey Tom_L, did you catch the spindle pics i posted yesterday, as requested? :-)
06:21 AM Tom_L: yup
07:26 AM * pere have no idea
07:27 AM jpa-: travis_farmer[m]: i would say that is EL817
07:30 AM jpa-: with the 500 ohm pull-up on that board, should get below 10µs switching delay
07:53 AM ZincBoy[CAON][m]: Very likely that these are open collector outputs. If they are, you just need a pull up to 5V and they will work with your inputs. You can tell this by measuring the tach output with a meter. If it stays at 0V as you rotate the spindle with 0V and 24V on the supply lines, it is open collector and needs a pullup. If you see it go to 24V then it is push/pull or has an internal pull up and needs the isolator. Most of the time
07:53 AM ZincBoy[CAON][m]: these are open collector though.
07:54 AM travis_farmer[m]: that would be handy, if i don't need the isolation...
07:59 AM * travis_farmer[m] prepares to head out to the shop to finish the adaptor plate to mount the spindle...
07:59 AM He[m]1: travis_farmer the title is for a 24 to 5V board, yet the description seems to be for a 5 to 24V board
08:00 AM He[m]1: With a truth table that always outputs a low signal
09:24 AM Tom_L: JT[m]1, 7i73 pins, 7i48, 7i33 still 404
09:26 AM JT-Cave: ok
09:29 AM JT-Cave: the 7i73 pins I'm not sure where to put that as it's a user supplied pdf
09:29 AM JT-Cave: and crap the 7i48 and 7i33 are in a different directory...
09:29 AM JT-Cave: http://www.mesanet.com/pdf/motion/7i48man.pdf
09:30 AM Tom_L: maybe like someone suggested, store them in a git repo
09:32 AM JT-Cave: I could store 7i73 pins in a git repo but I think it's best to download from mesa so you always get the latest one in case pcw-home makes a correction to a manual
09:33 AM Tom_L: maybe for the user one
09:33 AM Tom_L: maybe ask pcw to move the other 2
09:34 AM JT-Cave: I can fix the paths
09:34 AM JT-Cave: manual = pdfs[self.manualsPB.text()]
09:34 AM JT-Cave: pdf_url = f'http://www.mesanet.com/pdf/parallel/{manual}'
09:35 AM JT-Cave: '7i87':'7i87man.pdf',
09:35 AM JT-Cave: I just need to move either motion or parallel to the dictionary value
09:36 AM Tom_L: i'll build after you push
09:40 AM Tom_L: you _could_ put the manual tab back and point it to the user's local dir. once they enter a dir, force that dir for the tab
09:40 AM Tom_L: otherwise ' no pdf found yet'
09:41 AM Tom_L: that could be a mess though
09:42 AM JT-Shop: yup
09:44 AM roguish[m]: good morning all
09:55 AM * travis_farmer[m4 uploaded an image: (495KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/ONNvaxrIQDgxvqYtMObIgWDG/20230108_104006.jpg >
09:56 AM travis_farmer[m]: mounted the spindle on a temporary adaptor plate, and ran the power cable to the VFD :-)
09:56 AM travis_farmer[m]: a lot of temporary stuff in that picture, to be honest...
09:57 AM ZincBoy[CAON][m]: Looking good.
09:59 AM travis_farmer[m]: what, no comments about the Z-axis home sensor attached with copper tape? ;-) actually it works for testing, but i will have to solid mount for use
10:00 AM roguish[m]: hell, I've used double sided sticky tape before for switches.
10:00 AM roguish[m]: do what ya gotta do.
10:00 AM travis_farmer[m]: i still would prefer solid mount ;-)
10:01 AM Tom_L: roguish[m], i found automotive trim adhesive strips hold longer
10:01 AM Tom_L: (happened to have a roll)
10:01 AM roguish[m]: it's only taperary, right?
10:02 AM Tom_L: right...
10:02 AM ZincBoy[CAON][m]: I saw the copper tape but I have done worse 🙂
10:02 AM travis_farmer[m]: i noticed too, the Z-axis box will be in the way for a rear tool rack/carousel. the spindle should really be adjusted up more. have to do a front change system, some how
10:02 AM ZincBoy[CAON][m]: I get a lot of use out of blue painters tape.
10:02 AM Tom_L: i'm just curious how he got that in the trailer
10:03 AM Tom_L: or is it like building a boat in the basement..
10:06 AM travis_farmer[m]: the trailer has a double-door on the right side that opens about 5 feet wide. it was added to get the machine inside
10:08 AM * travis_farmer[m] uploaded an image: (302KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/jdMbdHyeRYZJzkaDDxGqOgKq/IMG_0314.png >
10:08 AM travis_farmer[m]: old pic
10:08 AM travis_farmer[m]: can you also guess why the trailer has sat there for a while? ;-)
10:10 AM travis_farmer[m]: had one good tire when i got it, and somebody stole that wheel right off the trailer... now the remaining 3 tires are flat...
10:11 AM roguish[m]: jack it up and set it on blocks so the tires don't rot.
10:11 AM travis_farmer[m]: too late...
10:12 AM Tom_L: JT-Cave, help -> documents goes nowhere now too
10:13 AM Tom_L: i doubt he was planning to take the router to the lake anyway
10:17 AM travis_farmer[m]: lol, no... but when i finally get the new shed built, i will put this trailer on craigslist, free. maybe somebody will want a 29 foot dual-axel trailer frame
10:22 AM JT-Cave: ok
10:25 AM Tom_L: JT-Cave, http://tom-itx.no-ip.biz:81/~webpage/cnc/JT-SHOP/mesact/firmware_error.png
10:26 AM Tom_L: 7i92t
10:26 AM Tom_L: & others
10:27 AM JT-Cave: opps
10:35 AM JT-Cave: help > documents opens the mesact pdf for me
10:35 AM Tom_L: maybe the pdf is missing here?
10:36 AM Tom_L: where would it be?
10:37 AM Tom_L: i think i removed that usr/directory the pdfs used to be in
10:37 AM Tom_L: i forgot what it was now..
10:39 AM JT-Cave: let me look
10:40 AM JT-Cave: installed is broken, dev is ok so it's either not getting installed or there is a path issue
10:50 AM JT-Cave: yep the pdf never got installed after the big pdf removal process
10:50 AM JT-Cave: fixed now
10:51 AM pere: My draft MQTT component is ready for wider testing at <URL: https://github.com/LinuxCNC/linuxcnc/pull/2253 >.
10:53 AM Tom_L: help->docs works now
10:53 AM Tom_L: firmware dl works
10:54 AM Tom_L: 7i73 pins fails
10:56 AM Tom_L: 7i48 & 7i33 ok
10:56 AM JT-Shop: I though I removed pins lol
10:57 AM Tom_L: under smartserial
10:57 AM JT-Shop: oh I forgot to remove it from the menu part...
10:57 AM Tom_L: i think you're just making sure i recheck :)
10:58 AM JT-Shop: I removed it from the dictionary but not the list of dicts
10:58 AM Tom_L: yeah
11:12 AM isaacggw[m]: ZincBoy[CA, ON] I am thinking of buying this servo for my mill spindle (https://www.ebay.com/itm/153488071070?hash). Do you see any red flags?
11:22 AM _unreal_: wow My git page is really starting to get a lot of views
11:23 AM _unreal_: https://github.com/berin-aquaquad/orange-pi-5
11:23 AM jpa-: isaacggw[m]: the motor specs there seem quite thin (torque, speed, etc missing)
11:27 AM jpa-: but apparently found by model number elsewhere
12:10 PM travis_farmer[m]: so, my spindle manual recommends not running the spindle in reverse, as it can loosen the collet nuts during braking. is this really a thing? i never had that issue with my other spindle. though for woodworking, i can't really see a reason to run in reverse, as i "think" all my endmills are one direction anyway. just seems odd, not to have reverse... i mean, i can set an override in the VFD, but i am not entirely sure i need to...
12:12 PM CloudEvil: I would compute the effective loosening torque on the moment of inertia of the nuts and the braking decelleration.
12:13 PM jpa-: travis_farmer[m]: sounds weird, as the collet nuts would loosen on reverse acceleration, which is same as braking in forward direction
12:13 PM CloudEvil: (that is mostly a lie, I would start with it very very loose and see if I can make it spin off)
12:13 PM CloudEvil: Sounds apophrycal.
12:13 PM jpa-: if very loose, you can certainly make the nut spin off - but then again the tool wouldn't stay in place anyway
12:13 PM travis_farmer[m]: i thought it sounded odd
12:14 PM jpa-: often braking torques are higher than acceleration torques anyway, so that would be a reason to *always* run the spindle in reverse ;)
12:15 PM jpa-: (at least I have configured my VFD to accelerate smoothly, but brake very quickly, and many safety standards require a stop within 5 revolutions or similar)
12:15 PM travis_farmer[m]: and maybe i misread (or misinterpreted), always a possibility ;-)
12:16 PM travis_farmer[m]: yeah, i am reading the VFD manual now to figure out what i need to setup when i go back out
12:18 PM travis_farmer[m]: i wonder why my VFD has a RJ45 connector, and nothing about it in the manual... also very odd
12:19 PM jpa-: one good reason to run spindle in reverse (when up in air) is that it helps clear stringy plastic chips stuck on it :)
12:20 PM travis_farmer[m]: yeah, i can picture that
12:20 PM roycroft: probably rs-422, travis_farmer
12:21 PM travis_farmer[m]: could be, but if it has it, i wonder why there is nothing about it in manual
12:21 PM roycroft: 1. it may be a typical, poorly-written manual
12:22 PM roycroft: 2. it may be just a jack in your version, not electrically connected to the unit, but may be a feature in another model from the same manufacturer
12:22 PM roycroft: 3. china :)
12:22 PM travis_farmer[m]: true ;-)
12:33 PM * isaacggw[m] uploaded an image: (32KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/uUJoueKZsIXvjKXZyccCfFuA/image.png >
12:33 PM isaacggw[m]: I am now thinking that i will buy a servo from dmm-tech.
12:33 PM isaacggw[m]: I am thinking of using the ``86M AC`` servo with the ``DYN4-H01C2-00 driver`` . Will this driver work ok with linuxcnc as a spindle motor? It says step and dir control, so that is good. The other thing I am not sure about is if i can read the encoder data from the servo drive to my 7i96s?
12:35 PM pere: <URL: https://people.skolelinux.org/pere/blog/LinuxCNC_MQTT_publisher_component.html > :)
01:03 PM markgonz: hello, does anyone how to make axis gui full screen?
01:09 PM JT-Shop: https://forum.linuxcnc.org/21-axis/29735-axis-modifications
01:16 PM markgonz2: thanks JT
01:17 PM JT-Woodshop: YW
02:12 PM travis_farmer[m4: well, my VFD still works... connected the braking resistor, and when i turned it on, smoke started coming from the VFD, so i shut it down, and threw the braking resistor in a snow bank, never to be used again (clearly the wrong unit...). but the VFD still works... so i think i will call it a day before i break something... heading back inside...
02:13 PM ZincBoy[CAON][m]: Looks good to me
02:17 PM ZincBoy[CAON][m]: Did you connect the braking resistor to the correct terminals? You often need to disconnect a linkage before connecting the resistor. The other issue may be that your brake threshold voltage is set too low. This will turn the braking resistor on even when the bus voltage is nominal. You really want the braking resistor or your deceleration time will be excessive.
02:18 PM * Loetmichel had used 2kw halogen lamps as braking resistors a few times. Very nice when you can see every E-stop across the whole production line ;)
02:21 PM travis_farmer[m]: yes, correct terminals, according to the manual. there was no linkage, and the brake turn on is set by frequency. but the resistor is 180 ohm, and it needs a 50 ohm, if that makes a difference. and it was 500W, when the need was 300W. it is the one that was needed for the last spindle
02:22 PM ZincBoy[CAON][m]: A higher resistance is not a problem. You just can't go lower than 50 ohm if that is the spec. Power wise you want at least 300W. 500W is fine.
02:23 PM Tom_L: something smells fishy
02:24 PM ZincBoy[CAON][m]: Yes, if the drive was smoking with that resistor, something was hooked up wrong or the drive was set wrong.
02:24 PM travis_farmer[m]: well, i may try it again tomorrow, see if i can finish off the VFD. after i check it to make sure it is not a dead-short
02:24 PM Tom_L: does it have a smaller internal brake resistor?
02:24 PM travis_farmer[m]: i nearly threw up with worry... i hope the VFD survived
02:25 PM Tom_L: mine did and i had to change the settings to use the external one
02:25 PM ZincBoy[CAON][m]: Send a picture of the VFD terminals.
02:25 PM travis_farmer[m]: don't know, Tom_L
02:25 PM travis_farmer[m]: well, now i am inside...
02:25 PM Tom_L: the internal one did get hot iirc but the external one doesn't
02:25 PM Tom_L: it's been a while...
02:26 PM Tom_L: google 'your model brake resistor' i'm sure someone has done it
02:27 PM travis_farmer[m]: a copy of the VFD manual is here: https://www.tjfhome.net/community/index.php?threads/my-techno-isel-4-foot-square-cnc-router-table-build.16/post-138 with the connections on page 9
02:27 PM travis_farmer[m]: (last post on that page
02:29 PM travis_farmer[m]: i will take a picture of my connections on the VFD tomorrow
02:32 PM Tom_L: F0.09, F0.22, F0.23 for starters
02:32 PM * travis_farmer[m] uploaded an image: (17KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/dIHiKwPUEAyhdZfvVbFyWwbh/Screenshot%202023-01-08%20at%2015-31-23%20Fuling%20DZB200_DZB300%20Inverter%20User%20Manual.pdf.png >
02:32 PM Tom_L: als0 17, 18
02:32 PM travis_farmer[m]: i wonder if i had DC+ to DC-, instead of BR+ to BR-
02:33 PM Tom_L: seems you found the fire pit
02:33 PM Tom_L: put a meter on the resistor
02:33 PM ZincBoy[CAON][m]: Yes, that would do it. The manual is not clear as it covers a wide range of units.
02:33 PM Tom_L: it's probaly still good
02:33 PM travis_farmer[m]: the BR+ and DC+ are same terminals
02:34 PM travis_farmer[m]: i could have mixed the others (hangs head in shame)
02:34 PM ZincBoy[CAON][m]: It is the BR- that will be the important one.
02:34 PM travis_farmer[m]: yup
02:34 PM travis_farmer[m]: i will meter the resistor tomorrow
02:34 PM ZincBoy[CAON][m]: DC- is usually the negative bus voltage. BR- is switched to DC-
02:34 PM Tom_L: DC braking time 0.0s
02:35 PM Tom_L: so either disabled or immediate
02:35 PM Tom_L: better give it some time there
02:35 PM Tom_L: f0.23
02:35 PM travis_farmer[m]: i think it was set to max of 50s
02:36 PM Tom_L: should be set to a safe limit i would think
02:36 PM travis_farmer[m]: F0.22 was 100
02:36 PM ZincBoy[CAON][m]: DC braking is DC injection into the motor for deceleration and has nothing to do with the brake resistor on the bus.
02:37 PM Tom_L: current
02:37 PM Tom_L: may be a bit high for a start point
02:37 PM ZincBoy[CAON][m]: You want F4.16, Brake threshold voltage.
02:38 PM ZincBoy[CAON][m]: Ignore DC braking for now. It has nothing to do with your issue.
02:38 PM travis_farmer[m]: ok
02:38 PM ZincBoy[CAON][m]: Leave F0.22 set to 0.
02:38 PM travis_farmer[m]: ok
02:39 PM travis_farmer[m]: (taking notes)
02:40 PM ZincBoy[CAON][m]: F4.16 should be set to 120. Odd that they set this as a %. Most drives set this as a voltage to turn the braking resistor on at.
02:40 PM travis_farmer[m]: ok
02:41 PM ZincBoy[CAON][m]: It does not look like they allow you to set it too low so that should not be an issue.
02:44 PM travis_farmer[m]: well, it was good to hear the spindle turn while i was out there :-) at least it is progress, even if smokey...
02:51 PM travis_farmer[m]: if i ever get this brake resistor sorted out, then i can work on the analog control settings so LCNC can control it. would be a PITA to manually control it ;-)
02:57 PM Tom_L: is it 0-10v or +10 - -10v?
02:57 PM travis_farmer[m]: 0-10v
02:58 PM Tom_L: got a 10v source?
02:58 PM travis_farmer[m]: right on the VFD, there is a source
02:58 PM Tom_L: what card are you using, i forgot
02:58 PM Tom_L: 7i96s?
02:58 PM travis_farmer[m]: 7i80HD-16
02:59 PM Tom_L: what are you using for the analog signal?
02:59 PM Tom_L: that little card i mentioned?
02:59 PM travis_farmer[m]: PWM to one of these https://www.ebay.com/itm/165703924008
03:00 PM Tom_L: https://mesaus.com/product/spinx1/
03:01 PM travis_farmer[m]: out of stock
03:01 PM Tom_L: yours takes 12-30v in
03:01 PM JT-Woodshop: just sold the last one the other day
03:02 PM Tom_L: http://store.mesanet.com/index.php?route=product/product&product_id=205&search=spinx
03:03 PM Tom_L: out too
03:04 PM travis_farmer[m]: mine will work, with some tinkering. i have 12v availible in my control box, or i can take 24V from a terminal on the VFD
03:05 PM Tom_L: just check the max out before wiring it up
03:06 PM ZincBoy[CAON][m]: One of these should work fine. Though if I was starting from scratch these days I would just control the drive with modbus. Quite a bit better and lets you do much more with the internal parameters on the fly.
03:06 PM ZincBoy[CAON][m]: A USB->RS485 converter and you are good to go.
03:06 PM travis_farmer[m]: this VFD is limited with ModBus i was told
03:07 PM travis_farmer[m]: https://forum.linuxcnc.org/24-hal-components/47798-lcnc-2-9-debian-buster-10-mb2hal-fuling-dbz312-vfd
03:09 PM roguish[m]: travis_farmer: I started with a spinx1, then went to modbus.
03:09 PM roguish[m]: all ya need is on/off/ and speed
03:10 PM Tom_L: mo speed!
03:10 PM travis_farmer[m]: as well as knowing how to setup mb2hal
03:11 PM Tom_L: mine didn't have modbus enabled
03:12 PM ZincBoy[CAON][m]: mb2hal is great. Makes it easy to get the basics running quickly. I haven't encountered a situation where I need anything more than mb2hal. You aren't really doing hard real time things with the spindle anyway.
03:14 PM Centurion_Dan1 is now known as Centurion_Dan
03:15 PM travis_farmer[m]: i was quite confused on how to set up the config for it though. the more i read the docs, the more confused i got
03:17 PM roguish[m]: travis_farmer: KISS...........keep it simple stupid. don't get too carried away.
03:17 PM roguish[m]: ya got lots of help here
03:23 PM * JT-Woodshop wonders if roguish[m] ever tried out the fix to gcode
03:23 PM Tom_L: in a sub if i set S400 and do some things, can i save the previous S number prior then set it back after the S400
03:23 PM roguish[m]: JT-Woodshop: no, should I?
03:23 PM Tom_L: pretty sure that's a simple thing to do, just never have yet
03:23 PM roguish[m]: I can give it a try
03:24 PM ZincBoy[CAON][m]: Save the current rpm to a global #<_rpm> = #<_varname>
03:24 PM Tom_L: named or numbered var?
03:24 PM ZincBoy[CAON][m]: oops, #<_varname> = #<_rpm>
03:24 PM roguish[m]: JT-Woodshop: I did look at it.
03:24 PM Tom_L: named i presume
03:25 PM ZincBoy[CAON][m]: then just do S#<_varname> when you want to set it back.
03:25 PM Tom_L: ZincBoy[CAON][m], i wouldn't just save the s word to a var?
03:25 PM roguish[m]: JT-Woodshop: tell ya what, I'll get it to work, and you make it pretty.
03:25 PM JT-Woodshop: and?
03:25 PM ZincBoy[CAON][m]: #<_rpm> is the current S value.
03:26 PM JT-Woodshop: does Z behave for you now?
03:26 PM Tom_L: ok
03:26 PM roguish[m]: not yet. but I'll have a go at fixing it.
03:26 PM ZincBoy[CAON][m]: See: https://linuxcnc.org/docs/html/gcode/overview.html section 3.5
03:27 PM Tom_L: ahh there we go, thanks
03:27 PM ZincBoy[CAON][m]: I don't like using numbered global variables as there is too much chance of conflict. I much prefer to use named globals as the chances of conflict are pretty much nil.
03:28 PM Tom_L: so it can be any random name just so there is no conflict
03:29 PM ZincBoy[CAON][m]: It just needs the leading underscore to make it global: #<_name> is global, #<name> is not.
03:29 PM Tom_L: i was thinking of revisiting spindle orient with this new spindle motor to see if i can get it to work better than the old DC sherline motor
03:29 PM Tom_L: ok
03:29 PM Tom_L: so you don't need to declare it global just use '_'
03:30 PM Tom_L: just a test
03:31 PM ZincBoy[CAON][m]: Yes, you will get an error if you try to read a non-existent global. They have to be set to a value before they can be read. There is the EXISTS statement that can be used to work around this: o100 if [exists[#<_name>]] ...
03:31 PM Tom_L: shouldn't be necessary since S would be set prior
03:32 PM Tom_L: default S is 0
03:32 PM ZincBoy[CAON][m]: For #<_rpm>, it is one of the predefined so it should always exist.
03:36 PM travis_farmer[m]: ok, i copied the sample mb2hal to my config Github: https://github.com/travis-farmer/linuxcnc/blob/master/configs/my_LinuxCNC_router/mb2hal.ini , how do i change it to work with my vfd?
03:37 PM ZincBoy[CAON][m]: What RS485 interface are you using?
03:38 PM travis_farmer[m]: serial to start, but i have a TCP option if i choose
03:43 PM ZincBoy[CAON][m]: You have a ethernet to RS485 adapter or the drive has a ethernet port? I didn't see an ethernet option in the manual.
03:44 PM travis_farmer[m]: i have a ethernet to rs485 device
03:44 PM travis_farmer[m]: and a serial one
03:45 PM ZincBoy[CAON][m]: You need to setup either the usb or ethernet adapter in the first part of the INI file. I think the sample file is pretty clear on how to do this.
03:45 PM ZincBoy[CAON][m]: Next you need to figure out what registers you need to write/read in the drive.
03:45 PM travis_farmer[m]: i filled in that part already
03:50 PM ZincBoy[CAON][m]: The basics are the frequency, forward, reverse, start, stop. Register 1000H has the basic start/stop bits in it.
03:50 PM ZincBoy[CAON][m]: You would set MB_TX_CODE=fnct_06_write_single_register
03:51 PM ZincBoy[CAON][m]: FIRST_ELEMENT = 4096 (1000h)
03:52 PM ZincBoy[CAON][m]: NELEMENTS=1
03:52 PM ZincBoy[CAON][m]: HAL_TX_NAME=DriveBits
03:52 PM ZincBoy[CAON][m]: MAX_UPDATE_RATE=0.0
03:53 PM ZincBoy[CAON][m]: That will create a HAL integer value with the name DriveBits that follows the value in the 0x1000 register in the drive.
03:54 PM travis_farmer[m]: ok
03:55 PM ZincBoy[CAON][m]: You will also need to set the F0.01 register to 2 to configure the drive to accept the communication values from the modbus interface.
03:56 PM travis_farmer[m]: ok
03:58 PM ZincBoy[CAON][m]: The drive comm parameters also have to be set to match your serial config. Section F6 of the drive parameters.
03:58 PM travis_farmer[m]: ok
04:08 PM ZincBoy[CAON][m]: F0.03 needs to be set to 6 to allow the drive frequency to be controlled from the serial port but I can't find what register actually sets the frequency... You need to do some digging to figure that one out. It _might_ bit F0.07 but the manual is very unclear.
04:08 PM travis_farmer[m]: i think it is 2000H somehow
04:09 PM ZincBoy[CAON][m]: I think you are correct.
04:11 PM ZincBoy[CAON][m]: You need to have a second transaction in the MB2hal ini that is pretty much the same as the first but has 8192 (2000h) as the address and a diffrent hal_tx_name.
04:13 PM ZincBoy[CAON][m]: If you want the status of the drive, you can have a third transaction with MB_TX_CODE=fnct_03_read_holding_registers,
04:13 PM ZincBoy[CAON][m]: FIRST_ELEMENT=12288 (0x3000)
04:14 PM ZincBoy[CAON][m]: NELEMENTS=5 (max of 5 for this drive)
04:15 PM ZincBoy[CAON][m]: This will read address 0x3000 to 0x3004 and put each value into different hal pins.
04:16 PM ZincBoy[CAON][m]: Though you might want to start with 0x3002 as the range to 0x3006 are more useful for feedback to the UI.
04:16 PM travis_farmer[m]: ok
04:17 PM travis_farmer[m]: do i specify 5 hal names?
04:18 PM ZincBoy[CAON][m]: No. It will create something like name.0, name.1 or something like that. Can't remember what it was right now.
04:19 PM travis_farmer[m]: oh, ok
04:20 PM travis_farmer[m]: ok, i updated that github file (refresh), how's it look?
04:22 PM ZincBoy[CAON][m]: TOTAL_TRANSACTIONS has to match the number of transactions you have defined in the file, so 3 in your case
04:23 PM travis_farmer[m]: ok, fixed
04:23 PM ZincBoy[CAON][m]: The only other thing to be aware of is the serial port permissions. By default running as a user you won't have access to the serial port.
04:23 PM travis_farmer[m]: i have dialout access
04:24 PM ZincBoy[CAON][m]: You are good to go then.
04:24 PM travis_farmer[m]: ok, thanks :-)
04:24 PM ZincBoy[CAON][m]: Once you get the mb2hal to load, I would check the pins with HALSHOW to make sure it matches what you expect.
04:25 PM travis_farmer[m]: ok, sounds good. :-) now i gotta go to the store to get supper... :-(
04:39 PM justache is now known as justReddy
04:41 PM JT-Woodshop: sometimes it hard to get a straight answer from people... makes you wonder if they have been sniffing burlap
05:38 PM roycroft: or they've been talking to attorneys a lot
05:38 PM roycroft: or politicians :)
05:38 PM roycroft: but perhaps sniffing burlap is prerequisite for both of those professions
06:31 PM Tom_L: JT-Cave, #79
07:01 PM -!- #linuxcnc mode set to +v by ChanServ
08:28 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/configs/orient_error.png
08:29 PM Tom_L: any ideas?
08:43 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/configs/orient_error.txt
08:47 PM XXCoder: nonetype?
08:47 PM Tom_L: that's the error i got on the 2nd try
08:47 PM Tom_L: the pic is a first try
08:47 PM Tom_L: should have updated it
08:48 PM XXCoder: so its toolchange that returned 5
08:48 PM Tom_L: err no i got a different error the 2nd time but didn't update the pic
08:49 PM Tom_L: appears so
08:50 PM Tom_L: i may have a clue
08:51 PM XXCoder: found post anout it, reading
08:52 PM XXCoder: "The interpreter error causes emcTaskPlanExecute() to return 5 (INTERP_ERROR)"
08:52 PM XXCoder: may be a clue but then post is from 2015
08:53 PM Tom_L: back out to test
08:56 PM XXCoder: #define EMC_TASK_EXEC_WAITING_FOR_IO 5
08:56 PM XXCoder: other ponental, from emc.hh
08:57 PM XXCoder: trying to find same on github
08:59 PM XXCoder: https://github.com/LinuxCNC/linuxcnc/blob/master/src/emc/nml_intf/emc.hh#L269
09:06 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/configs/orient_error.png
09:06 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/configs/orient_error.txt
09:06 PM XXCoder: wrong tool number?
09:06 PM Tom_L: no
09:07 PM Tom_L: something in the orient glue isn't right
09:07 PM Tom_L: i removed alot of code since i did it on the sherline
09:07 PM Tom_L: the sherline was a 'special' case
09:08 PM Tom_L: and it never worked right due to the crappy motor
09:08 PM Tom_L: it tried
09:08 PM XXCoder: one of your patches was not undone?
09:09 PM Tom_L: waiting for IO may be the index pulse
09:09 PM XXCoder: what was spindle physically doing during that?
09:09 PM Tom_L: sitting idle
09:11 PM XXCoder: wouldnt it be rotating while orenting? or am I picturing it wrong
09:11 PM Tom_L: it would
09:13 PM XXCoder: so error happened before it could start doing that?
09:16 PM Tom_L: soon as i executed the M6
09:16 PM Tom_L: it remapped M6 and started thru my orient code
09:16 PM Tom_L: which obviously has a bug
09:16 PM XXCoder: would have to go though the code then. fun
09:17 PM Tom_L: the old one is in my configs commented out
09:17 PM Tom_L: i left it on purpose so i wouldn't have to start from scratch
09:18 PM Tom_L: i may be required to use a secondary PID loop for it
09:18 PM Tom_L: On spindle.N.orient disconnect the spindle control and connect to the orient-pid loop:
09:49 PM Tom_L: last time i had to use both PID and use mux2 to switch between
09:50 PM Tom_L: mux2.0.sel <= iocontrol.0.tool-change
09:51 PM Tom_L: XXCoder, if you're curious
09:51 PM XXCoder: interesting. so that works now?
09:51 PM Tom_L: no, i haven't tested the changes
09:51 PM Tom_L: i'll finish up tomorrow
09:52 PM Tom_L: i was trying to do it without the 2nd pid but i'm not sure you can
09:52 PM XXCoder: ok. have good night :)
09:52 PM Tom_L: just something to do
09:55 PM * Bleepshop returns with junque.
09:56 PM Tom_L: useful i hope
09:56 PM Bleepshop: That I'm honestly waiting for it to explode or the feds to show up swearing it was Hunters or something.
09:56 PM XXCoder: synonyms for junque · counterfeit · deception · fabrication · forgery · hoax · phony · scam · sham ...
09:56 PM XXCoder: hmm
09:57 PM Bleepshop: Asus P9X79, i7 CPU, 32gb RAM, 3tb HDD, Asus/Ait Radeon 4830 IIRC. $50 off Craigslist. Looks like it was somebodies high dollar gaming machine 10 years ago.
09:58 PM XXCoder: funny that that laptop have more disk space than my pc
09:58 PM XXCoder: and sa,e ra,
09:58 PM XXCoder: *same ram
10:02 PM Bleepshop: This is a desktop, Cooler Master case big enough to bury an old PC or a couple of small dogs in.
10:02 PM XXCoder: doh not laptop. lol ok
10:02 PM XXCoder: damn thats amazing deal
10:02 PM XXCoder: id do a bits zeroing on disk in case
10:05 PM Bleepshop: It's got a sloppy, non-booting ubuntu slapped on the drive with no other data.
10:05 PM Bleepshop: I' zero the drive, stick it in my video NAS and replace it with an SSD.
10:06 PM XXCoder: cool
10:06 PM XXCoder: my pc still uses small ssd and spinning rust disk
10:06 PM XXCoder: 256 gb and 2 tb respectively
10:10 PM Bleepshop: Holy Carp! This is an old Cooler Master HAF X case. LOL
10:16 PM Bleepshop: Looks like a HAF 922 from the pics I can track down.
10:16 PM Tom_L: lcnc box or other?
10:21 PM Bleepshop: I dunno. That Optiplex 790 is fast and stable and I'm thinking about making this my new gaming machine. It's be replacing an AMD A10-5700 box.
10:22 PM Bleepshop: And then I could try the A10 with lcnc. LOL
10:24 PM Bleepshop: Just got done flashing the bios from ver2002 to ver4801 which is 2 years of stability fixes according to Asus. LOL
11:43 PM travis_farmer[m]: Don't you hate when you wake from a perfectly nice slumber... and can't get back to sleep? so here i am, back awake at 42 minutes past midnight, with no sensation of sleep...
11:56 PM Bleepshop: travis_farmer[m4: That's why I keep a large mallet on the nightstand. ;D
11:57 PM travis_farmer[m]: i can't afford to lose anymore braincells ;-)