#linuxcnc Logs

Jun 13 2023

#linuxcnc Calendar

12:36 AM Deejay: moin
01:46 AM Fritz[m] is now known as fritz62250[m]
03:26 AM travis_farmer[m]: Morning ☕️
03:26 AM XXCoder: .,..,... .,,..,., .,,.,,.. .,,.,,.. .,,.,,,,
03:27 AM travis_farmer[m]: ???
03:27 AM XXCoder: its binary numbers lol
03:28 AM XXCoder: meaning hello
03:28 AM travis_farmer[m]: i figured either that or Morse Code ;-)
03:33 AM * travis_farmer[m] eagerly awaits his 7i83, that is still stuck in Arizona...
04:14 AM miss0r: g'day mates
04:15 AM XXCoder: how goes things there
04:15 AM miss0r: its been so sunny here and we've broken half a dusin weather records for days without rain. Hence the "outback" greeting :)
04:16 AM miss0r: lol now the danish weather service has issued a "heat wave" warning to top it all off
04:17 AM XXCoder: here it finally entered proper spring weather... in june lol
04:17 AM XXCoder: may have been late very late stage spring into early summer
04:17 AM miss0r: we should split the difference, so we could've both had moderate weather :D
04:18 AM XXCoder: indeed lol
04:18 AM travis_farmer[m]: 59F with a high of 65F, with rain predicted
04:18 AM miss0r: I need to figure out to grind this busted cam axle I took home with me here
04:19 AM miss0r: well. cam bushing, rather
04:19 AM miss0r: doesn't matter. The point is, the cam itself is realy busted
04:20 AM miss0r: and I think I will have to make a grinder jig for my cnc mill
04:21 AM miss0r: do any of you know how to calculate cam profiles?
04:21 AM XXCoder: not me sadly
04:24 AM travis_farmer[m]: no idea. i checked google, and all i found was for an engine camshaft
04:24 AM miss0r: theres a tiny rim around the edge of the cam that is not worn down. I guess I will have to measure and "wing it"
04:25 AM XXCoder: you have shadow measuring thing?
04:25 AM XXCoder: that'd be most accurate way
04:26 AM XXCoder: https://www.sme.org/globalassets/sme.org/technologies/articles/2018/03---march/lead-image-1-starrett_he400-w-m1cusa1-768x432.jpg
04:26 AM XXCoder: this thing
04:26 AM miss0r: sadly I dont. When I've had some worn out splines or something like that, I would just turn down the end of the shaft till I get clean splines, and then sit it on top of a scanner
04:26 AM miss0r: Then use computer based protractors ect
04:27 AM miss0r: but this cam sits in the middle of a bushing, so that model won't work - as I plan to regrind this part
04:27 AM XXCoder: that thing probably cost a lot anyway
04:28 AM miss0r: its from a vintage production machine, that still runs 24/7 - so I couldn't get spare if I wanted to
04:28 AM miss0r: well... right now it runs 0/0...
04:45 AM JT-Cave1: morning
04:48 AM Tom_L: morning
04:48 AM JT-Cave1: 53°F
04:48 AM JT-Cave1: feels nice
04:48 AM Tom_L: 64
04:50 AM miss0r: 82 here - sweating bullets
04:50 AM JT-Cave1: that's hot if it's C
04:51 AM JT-Cave1: check finally cleared the bank... that put a hole in that account lol
04:51 AM travis_farmer[m]: yeah, like 180F LMAO
05:01 AM JT-Cave1 is now known as JT-Cave
05:01 AM miss0r: lol
05:01 AM -!- #linuxcnc mode set to +v by ChanServ
05:01 AM miss0r: i took the liberty to convert it to F for you metric denying fellas :)
05:01 AM JT-Cave: lol
05:38 AM JT-Cave: rooster just crowed
06:00 AM * travis_farmer[m] eats a Jimmy Dean breakfast bowl...
06:43 AM * JT-Cave breaks fast with cantaloupe cheese strawberries and pills
06:44 AM travis_farmer[m]: oh crap... forgot my morning pills... better take those...
06:47 AM travis_farmer[m]: there, breakfast is complete :-)
06:54 AM travis_farmer[m]: i have been trying to figure out a way to set my laser focus height, without crashing the laser, or the 3D probe while lasering. best i can think of is to mount the sensor so it senses above the laser tip, and use a setup block of a known height to touch off, pause while i remove the setup block, then move to laser focus height over the part. i could also use a second tool-height sensor, and just slide it under an adjustable bolt be
06:54 AM travis_farmer[m]: the same thing. is there an easier way?
06:56 AM JT-Cave: mesact2 test for axis letter(s) changed seems to be working
06:57 AM JT-Cave: if you know the offset of the laser then your g code can set the correct height after you probe the material
06:59 AM travis_farmer[m]: Hmmm, yeah, i could just mount the probe in a spindle tool-holder, probe down, pause to remove probe, and then move to offset... probably the best way...
07:06 AM * JT-Cave needs MesaCT2 testers
07:23 AM JT-Cave: I think I'm at the last test for joints and axes added in mesact2...
07:32 AM * JT-Cave starts his chicken day
08:04 AM -!- #linuxcnc mode set to +v by ChanServ
08:05 AM * JT-Shop tries out debian 12 lxde
08:12 AM * dudul5811[m] uploaded an image: (17KiB) < https://libera.ems.host/_matrix/media/v3/download/jauriarts.org/soFkQqLOQnvGZXNaEWiZUPWB/Screenshot_from_2023-06-13_16-40-21.png >
08:12 AM dudul5811[m]: hi friends. i want to test if my new Parallel port is working or not. i used parallel port tester ptest.hal and connected the pin2 and ground of port to a voltmeter. it shows 0 and when i click on the pin 02 , it does not change. does this indicate a problem ? or i have done something wrong ?
08:21 AM JT-Cave: wow it's been a long time since I wrote ptest.hal... first thing I'd check is the address of your parallel port matches what is used in ptest.hal
08:22 AM JT-Cave: next I'd google testing parallel port outputs to get the correct wiring
08:52 AM JohnGuan[m] is now known as johnguan0[m]
10:03 AM travis_farmer[m]: well heck... was out working in my shop, when a branch fell and took out the power-line supplying the garage (and my shop)... power company is supposed to be here in a bit to splice it back together. in the meantime, i turned off the main breaker to the garage in case of any surges when power is restored. nearly lunch time anyway.
10:05 AM * flyback rubs ve7it on travis_farmer853's power line
10:05 AM flyback: *CANUCKED*
10:05 AM flyback: *CANUCKED*
10:05 AM flyback: *CANUCKED*
10:05 AM flyback: *CANUCKED*
10:05 AM roguish[m]: travis_farmer: break time....
10:06 AM travis_farmer[m]: yeah, it is now, LOL
10:07 AM travis_farmer[m]: flyback: what is "ve7it"?
10:07 AM flyback: a canadian in the channel
10:08 AM travis_farmer[m]: ok, now that statement is even more confusing... LOL
10:08 AM flyback: here you go
10:08 AM flyback: https://www.urbandictionary.com/define.php?term=Canucked
10:09 AM flyback: so whenever stuff goes wrong I rub a canadian on it
10:09 AM flyback: CAUSE IT'S CANUCKED
10:09 AM flyback: I usually use veegee but he escaped
10:09 AM travis_farmer[m]: umm, ok... LOL
10:10 AM travis_farmer[m]: lucky the house has a standby generator :-)
10:10 AM flyback: nice
10:10 AM * flyback has a 1400w pull start one
10:10 AM flyback: still at my folks house I need to bring it down to mine
10:11 AM travis_farmer[m]: 15,000w LP gas genny :-)
10:11 AM flyback: nice
10:11 AM flyback: mine was $150 :P
10:11 AM travis_farmer[m]: i bought it off FB marketplace for $250, and fixed it
10:11 AM flyback: NICE
10:11 AM flyback: what was wrong with iht
10:12 AM travis_farmer[m]: just had a cable unpluged way inside the unit
10:12 AM * JT-Shop has a 2 cyl diesel 10kw 3 phase gen
10:12 AM flyback: nice, nice
10:13 AM travis_farmer[m]: i wish the garage power came off the house so it was on the genny
10:15 AM travis_farmer[m]: i suppose i may as well make lunch...
10:17 AM spray[m] is now known as zmrdko0[m]
10:17 AM roguish[m]: I have a transmission tower in my back yard, I just need to throw a couple of lines up to the 7kva wires.
10:17 AM roguish[m]: that's 57 kva
10:19 AM flyback: hopefully you have some really fatass grounding spikes around the antenna
10:27 AM JT-Shop: lxde is ok but the file manager is missing some good features so when we get back from the dairy shack I'll install debian 12 mint
10:28 AM JT-Shop: lxde is not too bad compared to some of the other ones lol
10:29 AM roycroft: with 57kva you should be able to suck some power via induction, and not have to go through the trouble of stringing wires
10:34 AM roguish[m]: roycroft: hey, what kind of finish would you use for kitchen cabinets? not paint.
10:35 AM roguish[m]: or melanine.
10:36 AM roycroft: what material will be used for the construction?
10:36 AM roguish[m]: good ply
10:36 AM roycroft: i either use veneered plywood or mdf and a polyurethane finish, or paint
10:37 AM roycroft: i don't use polyurethane much, but for kitchen cabinets it's a good finish to use
10:37 AM roycroft: it's not repairable, but it's quite durable
10:37 AM roguish[m]: https://store.ashbylumber.com/departments/sheetgoods-|01.html
10:37 AM roguish[m]: this is a very good place.
10:39 AM roguish[m]: also have this place close by. https://www.macbeath.com/
10:39 AM roguish[m]: https://www.macbeath.com/products/plywood
10:39 AM roycroft: macbeath.com have pre-finished plywood
10:40 AM roycroft: i'd look into that for the cabinets
10:40 AM roguish[m]: you once mentioned that the finish may be un-repairable
10:40 AM roycroft: it will be expensive, but it will save you much time and grief
10:40 AM roycroft: yes
10:41 AM roycroft: it's a problem with kitchen and bathroom cabinets - a finish that is durable enough for those environments is going to be unrepairable
10:41 AM roguish[m]: check pricing a few days ago. at least $150 a sheet.
10:41 AM JT-Shop: discount tire direct is taking the tires back with just a few dollars shipping
10:41 AM JT-Shop: nap time now
10:41 AM roycroft: $150/sheet isn't bad
10:46 AM roguish[m]: roycroft: is edge banding difficult to do?
10:49 AM roycroft: it is not difficult to do, but i prefer solid edging when possible
10:50 AM roycroft: solid edging is more durable
10:51 AM roycroft: what i would recommend, if you don't have your design completed yet, is considering pre-finished plywood carcases with solid wood face frames
10:52 AM roguish[m]: I really don't want to use face frames.... seems like doubling the work.
10:54 AM roguish[m]: and I just want to make simple kitchen cabinets. not heirloom stuff..... not going to gold plate the pig.
10:54 AM roguish[m]: mixed metaphore.... gold plate the turd, lip stick on the pig.....
11:05 AM roycroft: if you want to do euro-style frameless cabinets, then edge banding would be fine, but know that it's not going to be as durable as a solid wood face
11:06 AM roycroft: i don't want to spend all your money, but a proper edging system with hot glue like the festool conturo will be much more durable than iron-on edging
11:06 AM roycroft: the conturo is $3600, though
11:07 AM roycroft: if you bought it, used it for the one job, and then sold it, you would probably be able to recover about $3250 or more, but it may take some time to sell - it's not in high demand
11:08 AM roycroft: you may also be able to find a cabinet shop who would apply the edging for you
11:14 AM roycroft: any medium to large cabinet shop will have a hot glue edge banding machine that can apply and trim it very quickly
11:21 AM roguish[m]: roycroft: thanks, i'll check around\
11:34 AM Hufo[m] is now known as hufo0[m]
12:13 PM Tom_L: JT-Shop, had plans to test but the main pc is borked and i'm not in the mood to deal with it atm
12:13 PM Tom_L: it's been acting up a while
01:24 PM roycroft: my shaper products all shipped today
01:24 PM roycroft: they told me it would be a week before they shipped
01:25 PM roycroft: so they should all arrive well before the fair, and i won't have to worry about them getting stolen
01:36 PM travis_farmer[m]: ran into an odd situation... i have analog sensors for air PSI (0-5v) that i run into the analog inputs of my 7i70... since rewiring, the wire transit from sensor to analog input is a touch longer, but i don't think it is long enough for line-loss to be a consideration. but a sensor that should read about 120psi is reading about 92psi. i think i need some different sensors... ModBus or 0-10v would be better...
01:46 PM travis_farmer[m]: or i can use my Arduino ModBus project to read the sensors close-up, and send the ModBus...
02:16 PM roycroft: https://www.shapertools.com/en-us/trace
02:16 PM roycroft: early kickstarter price for that will be <$100
02:16 PM roycroft: it might be worth it
02:17 PM roycroft: maybe i'll finally get a kickstarter account
02:24 PM roycroft: https://www.toolkaiser.com/
02:24 PM roycroft: but there is that, which is cost-free
02:27 PM skunkworks[m]: that would require me to be a good sketcher.. I am not
02:31 PM NoGodDamnIdea: why yall got [m] after your name
02:31 PM travis_farmer[m]: we are special :-)
02:32 PM Anatoly3yb[m] is now known as anatoly3yb0[m]
02:32 PM roycroft: not really
02:32 PM roycroft: with the shaper software, you trace around a part with a felt tip pen, place the frame over the tracing, and take a picture
02:32 PM travis_farmer[m]: i'm not special? :-(
02:33 PM roycroft: with the toolkaiser software you just set the part on a piece of white paper and take a picture
02:33 PM roycroft: no sketching skills neede
02:33 PM roycroft: d
02:33 PM roycroft: i don't know if you're special or not, travis_farmer
02:33 PM travis_farmer[m]: oh, different conversation...
02:33 PM roycroft: you're you, and you are an individual
02:33 PM roycroft: arguably everyone is special in that respect
02:35 PM * roycroft is really hoping there is no rioting in miami today
02:35 PM travis_farmer[m]: same here
02:37 PM roycroft: there's a good chance that _unreal_ is participating in the protests there
02:40 PM miss0r: Grinding the cam in the CNC mill now. The client will get a big fat bill for the thorough cleaning I will have to do after...
02:54 PM Unterhaus__: apparently there are more media reps than protestors, so probably wouldn't be much of a riot
02:55 PM roycroft: hopefully
02:55 PM roycroft: kary lake is there, though, trying to stir things up
02:55 PM roycroft: kari, rather
03:29 PM Unterhaus__: probably more rabble rousers there than rabble
03:44 PM miss0r: My god this is a mess.. the bill will be through the roof :D
03:51 PM Tom_L: is github down?
03:52 PM Tom_L: no, it's my local connection
03:52 PM travis_farmer[m]: not for me
04:00 PM Tom_L: how to list eth mac address?
04:06 PM roycroft: in debian?
04:06 PM Tom_L: i got it
04:06 PM Tom_L: but yes
04:06 PM roycroft: ip maddress will list all the mac addresses
04:06 PM roycroft: or ip link list
04:06 PM Tom_L: i had to swap mb and the eth is different naming
04:07 PM Tom_L: what is docker0?
04:07 PM Tom_L: i had that on the previous one too
04:08 PM roycroft: a pseudo interface for docker instances
04:09 PM Tom_L: ok hopefully this will work
04:09 PM roycroft: nothing is working for me today
04:09 PM roycroft: i hope you have a better day :)
04:10 PM Tom_L: i think the sata buss was crapping out on the other mb or some other conflict
04:11 PM -!- #linuxcnc mode set to +v by ChanServ
04:11 PM Tom_dev: ok so far so good
04:14 PM Tom_L: JT-Shop, is mesact2 working from a deb yet?
04:15 PM JT-Shop: I've not tried to build a deb yet, getting close to solving the update ini section
04:15 PM Tom_L: ok got an error right off
04:15 PM JT-Shop: so it's ready to try that
04:15 PM JT-Shop: ok, I'll try in a bit and see what it is
04:15 PM Tom_L: put a different MB in the main pc and things seem better
04:16 PM JT-Shop: finally back to working on it instead of hanging out on corvetteforum.com lol
04:16 PM Tom_L: trying to get back up to speed on this... it's been a while
04:16 PM travis_farmer[m]: not saying you were corvette obsessed... but... ;-)
04:17 PM JT-Shop: passed 500 mile break in mark today on the way to the Dairy Shack...
04:18 PM JT-Shop: well I'm still a corvette fan :)
04:18 PM Tom_L: how many more before you can 'drive' it?
04:18 PM Tom_L: 1500?
04:19 PM JT-Shop: you "drive' it after 500 miles you can 'track' it after 1500 miles
04:20 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/JT-SHOP/mesact2/Mesact2_error.png
04:20 PM Tom_L: wow that looks like a doozie
04:21 PM Tom_L: opening a saved ini
04:22 PM Tom_L: i'm betting the format is different from 1.x to 2.x
04:22 PM JT-Shop: ok
04:23 PM Tom_L: i haven't saved any 2.x ini yet
04:23 PM roycroft: i bought my pickup last october and just passed 500 miles on it
04:23 PM Tom_L: what's the Open1..3G about?
04:24 PM Tom_L: on the file menu?
04:26 PM JT-Shop: https://github.com/jethornton/mesact2
04:27 PM JT-Shop: copy the sample configs and you can quickly open them from the tool bar
04:27 PM JT-Shop: see the readme
04:27 PM Tom_L: i will in a bit.. just shut down to go to a few things
04:27 PM JT-Shop: ok
04:27 PM Tom_L: at least the pc is back in order. that was the goal
04:31 PM JT-Shop: I know how that goes for sure
04:41 PM roycroft: i'm doing my first debian 12 upgrade now, on a machine that isn't public-facing, isn't used that much, and is running on a hypervisor, so i have a snapshot of debian 11 that i can easily restore if necessary
04:41 PM roycroft: the way today has been going i wonder it was a good idea to even attempt this
04:43 PM Tom_L: JT-Shop, the pc is off but if i cloned that i should have the examples right?
04:43 PM Tom_L: just copy the sample over to the lcnc/config directory
04:53 PM JT-Shop: yes the examples are in the config for now
04:54 PM JT-Shop: and yep copy them to /configs
04:57 PM travis_farmer[m]: KFC for supper :-)
04:59 PM JT-Shop: I can't remember if I need to install SSH server or not...
04:59 PM travis_farmer[m]: i always do, just in case
05:01 PM travis_farmer[m]: signing off for supper :-)
05:01 PM JT-Shop: k
05:04 PM JT-Shop: Tom_L, did you try and flash a card or read hmid or something?
05:05 PM JT-Shop: oh yea flash.py readhmid is the source of the error
05:07 PM roycroft: it boggles my mind that any modern os installer would not install an ssh server by default, but many do not
05:22 PM JT-Shop: I just noticed the on/off switch on the air pump is a variable speed... roycroft what speed do you run it at when engraving or cutting?
05:25 PM roycroft: full speed
05:25 PM roycroft: the more air the better, for what i've done so far
05:26 PM roycroft: i think the one that xtool sell has a simple on-off switch
05:26 PM Tom_L: JT-Shop, no i'd just finished swapping MB and got ethernet working again
05:27 PM roycroft: but that's the only difference between that and the one on amazon for less money than the xtool version
05:30 PM Tom_L: JT-Shop, readhmid works
05:33 PM JT-Shop: roycroft, thanks
05:33 PM JT-Shop: Tom_L, your error was from readhmid...
05:33 PM Tom_L: i got it loading the 1.x ini i'd saved before
05:34 PM JT-Shop: 1.x loads here too
05:34 PM Tom_L: 7i96s_d.bin is the default right?
05:35 PM srk_ is now known as srk
05:35 PM Tom_L: load & verify seem to work
05:37 PM Doc[m] is now known as jubilationt0[m]
05:38 PM JT-Shop: I think so...
05:38 PM Tom_L: i just did it
05:39 PM JT-Shop: did you look on the info tab of the firmware tab?
05:39 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/JT-SHOP/mesact2/mesact2_error2.png
05:39 PM Tom_L: loading a 1.1.2 config
05:40 PM JT-Shop: post up that config for me
05:40 PM Tom_L: in that same dir
05:40 PM Tom_L: ^^
05:41 PM skunkworks[m]: wow - I really assumed that most cnc controllers have some sort of path following tolerance... I think I might be wrong.. Linuxcnc is pretty awesome.
05:42 PM skunkworks[m]: someone commented on one of my videos that fanucs don't really have it - you run it slower. (because you cannot change the 'parameter' that adjusts the path following tolerence.
05:43 PM skunkworks[m]: well - I guess I could see that from a company stand point - all machines must follow that path within X.XXX amount.
06:01 PM Tom_L: skunkworks[m], most button pushers wouldn't know to check that anyway
06:04 PM Tom_L: https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=05b204a24a425137eaa9af26986bf88ae4bed3e2
06:08 PM skunkworks[m]: sure...
06:09 PM XXCoder: skunkworks yep didnt see such in any larher machines I ran
06:09 PM XXCoder: I dont know if a61 has it or not
06:09 PM XXCoder: considering how expensive a61 is. probably does.
06:09 PM Tom_L: i _think_ the controls my kid runs are yasnac
06:10 PM Tom_L: and fairly dated ones at that
06:12 PM JT-Shop: wow that's an old file :)
06:12 PM Tom_L: the pdf or the ini i posted for you?
06:13 PM JT-Shop: the ini
06:13 PM Tom_L: yeah it probably is old
06:14 PM JT-Shop: I get the error as well, look in the morning
06:14 PM Tom_L: i thought i'd saved it using a newer mesact
06:14 PM Tom_L: but maybe not
06:14 PM JT-Shop: says 1.0.0
06:16 PM Tom_L: being a bit rusty, i just grabbed one and opened it
06:20 PM JT-Shop: crap trying to find out the fedex location to drop the tires off and I bet it's not walgreens
06:24 PM Tom_L: sending the all seasons back?
06:24 PM JT-Shop: yup, don't need two sets
06:25 PM Tom_L: did they keep the rims you swapped or you have a spare set now?
06:26 PM JT-Shop: I have the stock rims in the boxes
06:29 PM JT-Shop: yuck the fedex ship center is in Sikeston...
06:29 PM JT-Shop: oh well
06:30 PM JT-Shop: I think I'll call them and see if they will pick them up
06:39 PM JT-Shop: night
08:47 PM mits5782[m]: I'm trying to wrap my brain over how linuxcnc handles encoders when 'ignoring' them
08:47 PM mits5782[m]: specifically I'm thinking about a handwheel, which is inactive while the machine is running a program
08:48 PM mits5782[m]: if you move it while the program is running, why doesn't linuxcnc make a move when the program stops?
09:01 PM Tom_L: mits5782[m], it is disabled in a
09:01 PM Tom_L: 'auto' mode
09:02 PM Tom_L: http://linuxcnc.org/docs/stable/html/man/man9/motion.9.html#AXIS%20PINS
09:02 PM Tom_L: axis.L.jog-enable IN BIT
09:02 PM Tom_L: When TRUE (and in manual mode),....
09:02 PM mits5782[m]: sure, but when it exits auto mode
09:02 PM mits5782[m]: why doesn't it "catch up" ?
09:03 PM mits5782[m]: I'm asking this because I want to let my pendant adjust the eoffset of the Z axis
09:03 PM Tom_L: nothing happens when it is disabled
09:03 PM mits5782[m]: but the jogwheel is affecting eoffset all of the time
09:03 PM roycroft: it doesn't catch up because all it does is send pulses
09:03 PM roycroft: if you ignore the pulses then nothing happens
09:03 PM srk- is now known as srk
09:04 PM mits5782[m]: even if I detach the pin and then reattach it when I want to... there's a jump in the motion
09:04 PM Tom_L: axis.L.eoffset-enable IN BIT
09:04 PM Tom_L: Enable for external offset (also requires ini file setting for [AXIS_L]OFFSET_AV_RATIO)
09:04 PM Tom_L: keep that disabled until you need it
09:04 PM Tom_L: i personally don't use that
09:04 PM mits5782[m]: I am
09:05 PM Tom_L: so i don't know all the 'in's and out's' of it
09:05 PM Tom_L: there's also an eoffset-clear bit in
09:06 PM mits5782[m]: yeah, the problem is that I don't want to clear it... I just want to ignore updates for a period of time
09:06 PM mits5782[m]: because the jog wheel is being used for a different axis
09:07 PM Tom_L: use some logic to hold the eoffset-enable off until you need it
09:07 PM Tom_L: for which ever axis you need
09:07 PM mits5782[m]: I already have it on, if I turn it off the axis will move
09:08 PM Tom_L: i'd compare it to my pendant axis select switch
09:08 PM Tom_L: the mpg is pulsing but it's only going to the selected axis
09:08 PM mits5782[m]: okay... so the problem as I see it is
09:08 PM mits5782[m]: it's not pulses at this layer
09:08 PM mits5782[m]: it's a count pin
09:08 PM mits5782[m]: a number
09:08 PM Tom_L: but if it's disabled is it counting?
09:09 PM Tom_L: hal show will tell that
09:09 PM mits5782[m]: yes, because the count pin is an output from something
09:09 PM mits5782[m]: in my case, it's a mesa card counting the encoder pulses
09:09 PM mits5782[m]: but it could also be encoder hal component
09:10 PM Tom_L: what are you using external offsets for?
09:10 PM mits5782[m]: depth of engraving into a PCB... the copper is so thin that I set it while cutting
09:11 PM mits5782[m]: I've tried doing a touch-off but I'm getting really inconsistent depths based on feel of a pin going under the cutter tip
09:12 PM ZincBoy[CAON][m]: The PCB engraver I used to use dealt with this by having the spindle be spring loaded. A shoe would ride on the PCB and set the hight.
09:26 PM roycroft: that's pretty common on engravers
09:26 PM * roycroft recently got some spring-loaded diamond-trip engravers for his mill, as a matter of fact
09:28 PM JB[m]1 is now known as jbxo0[m]
10:04 PM Scorpio[m] is now known as scorpio71600[m]
10:10 PM roycroft: it seems so wrong that the vegas golden knights won the stanley cup by defeating the florida panthers
10:10 PM roycroft: not only is neither team canadian, neither of them are even from a part of the world that gets icy in the winter
10:12 PM CaptHindsight[m]: roycroft: https://www.youtube.com/watch?v=8KX-ZFfCn6s&t=1s
10:20 PM roycroft: when i saw eric idle he conducted an audience sing-along of the philosophers' song
10:24 PM * roycroft is watching a perry mason episode, and wonders if perry will win his case
10:34 PM Thorhian7490[m]: Anyone know where cost effective (relatively speaking) ethercat AC servos and drives can be had?
10:56 PM CaptHindsight[m]: thorhian: yes
10:56 PM CaptHindsight[m]: thorhian: whatshisname up in Toronto found some nice ones, let me check
11:06 PM CaptHindsight[m]: solarwind_: ^^^
11:23 PM solarwind_: Thorhian7490[m] leadshine ones are the most cost effective by far
11:23 PM solarwind_ is now known as solarwind
11:24 PM solarwind: Thorhian7490[m] if you don't need Ethercat, you can cut your cost to 30%
11:24 PM solarwind: Thorhian7490[m] https://www.automationtechnologiesinc.com/products-page/ac-servo-motors/ac-servo-motor-kits-750w-220vac
11:25 PM solarwind: They price gouge you on the cable. Just get the 3m one and extend it yourself if necessary. But _do_ get the cable because the backup battery is part of that assembly and the connectors are weird. That keeps track of its position even if fully powered off and someone bumps the table or whatever.
11:27 PM solarwind: Leadshine just introduced an upgraded version of that servo kit. Same features/specs, but the software is quite a bit better. Call the automationtechnologies guy (he's pretty nice and responsive) and ask him to get you the EL7 series: https://www.leadshine.com/products/servo/servo-drive/el7-series.html
11:27 PM solarwind: It's the exact same price, but newer model with some other upgrades, so why the hell not
11:28 PM solarwind: There's also the EL8 series (he has ties with leadshine, so he can get you whatever product you want for a reasonable price). The EL8 series has dual encoder inputs, so you can connect your glass scales directly to the drive if you like.
11:29 PM solarwind: If doing a machine conversion, you normally want the higher torque/lower speed motors as opposed to the general purpose/high RPM ones that come in the kit. So ask him to get you those instead so you don't have to gear them down with a belt
11:30 PM solarwind: That reduced inertia ratio helps a lot.
11:31 PM solarwind: Thorhian7490[m] if you don't need ethercat, these prices will blow your mind: https://www.omc-stepperonline.com/e6-series-ac-servo-motor
11:31 PM solarwind: I have them and they work exactly as advertised
11:33 PM solarwind: Thorhian7490[m] depending on what you're building, the real advantage of ethercat is the CiA 402 protocol
11:34 PM solarwind: So unless you really need that, analog or step/dir is just as capable for machine control, so I would rethink the ethercat requirement
11:34 PM mrec: which one do you have?
11:35 PM solarwind: mrec I have all of them
11:35 PM mrec: did you run them on speed or just slowly?
11:35 PM solarwind: I shoudl write a blog post in detail about these
11:35 PM solarwind: but I ran and tested all of them and they all perform very well
11:35 PM mrec: did you run them on spead with a little bit load?
11:35 PM solarwind: The leadshine have "auto tuning"
11:35 PM solarwind: mrec yes
11:35 PM mrec: speed*
11:35 PM mrec: decelerating is fine?
11:35 PM solarwind: yes
11:36 PM solarwind: super sharp
11:36 PM solarwind: you can see the graph in the tuning software
11:36 PM mrec: I'm using some JMC IHSV motors they induce power to the line and take out the power supply here
11:36 PM solarwind: you can adjust all parameters to get exactly the response you need
11:36 PM mrec: I'd probably need some brake in the line
11:36 PM solarwind: But I keep my inertia ratios low
11:37 PM solarwind: it's not a challenging thing for any servo
11:37 PM solarwind: with the stepperonline ones, they don't have 1 click autotuning software, so you have to know how to do that yourself. Works great once you do it
11:37 PM mrec: well I have the integrated ones here it's challenging for them, the external ones are more flexible (my delta ASD-A2 are fine)
11:37 PM solarwind: I have the integrated ones too, zero issues
11:38 PM solarwind: I have a few of these: https://www.omc-stepperonline.com/integrated-servo-motor
11:38 PM solarwind: they all work within their spec
11:38 PM mrec: I have issues at very low speed with JMC IHSV motors
11:38 PM solarwind: that's a tuning problem
11:38 PM mrec: we were in contact with JMC they advised to get a stronger motor
11:38 PM mrec: iSV57T-180S
11:38 PM mrec: kind of this one
11:39 PM solarwind: yeah that solves it by reducing inertia ratio
11:39 PM mrec: it was only for the Z axis with some gear and belt.
11:39 PM solarwind: that's the easy brute force way
11:39 PM solarwind: go even bigger
11:39 PM mrec: I tried to tune but it didn't help and it's not so easy to reproduce either it just passes out sometimes
11:39 PM solarwind: For a bridgeport class machine, I put 1kW on each axis, no gear down belt
11:40 PM solarwind: lol iSV57T-180S for Z axis? On what machine?
11:40 PM solarwind: that thing makes 0.6Nm
11:40 PM mrec: it's a small machine
11:40 PM solarwind: Either gear it down (they sell the precision planetary gears), or just use a stepper
11:41 PM solarwind: closed loop steppers are much better at high torque, and cheaper
11:41 PM mrec: it's fine as long as I run it a little bit quicker :-) just very slow movements cause it (eg. when using the handwheel with a high resolution)
11:41 PM mrec: I still have some linear scales here which I want to install but I'm working on another machine now
11:41 PM solarwind: Yeah sticktion and inertia ratio
11:41 PM solarwind: that's what's causing that
11:42 PM solarwind: switching between the coefficient of static and dynamic friction and a very high inertia ratio
11:42 PM mrec: it would be nice if those integrated motors would route out the encoder pins that would be handy for homing
11:42 PM solarwind: you can
11:42 PM solarwind: disassemble the back
11:42 PM solarwind: they're an SPI encoder
11:42 PM solarwind: I decoded it, but have to find my notes
11:43 PM mrec: there are different versions of it
11:43 PM mrec: I have motors with ABI here
11:43 PM solarwind: they're all braindead easy
11:43 PM mrec: I also disassembled them
11:43 PM mrec: it would be nice to have some direct interface for this, yes I can jumpwire it's not a big deal
11:44 PM solarwind: Not the target audience for this motor
11:44 PM mrec: so do some homing with a hall effect sensor and fine tune it with the encoder
11:44 PM solarwind: But very doable
11:44 PM solarwind: Oh the RS232 interface can read out encoder counts
11:44 PM mrec: on the other side I have the linear scales with index pin :) the machine is idle now since I'm doing other things now
11:44 PM mrec: yes the RS232 can read it
11:45 PM mrec: those connectors are flimsy for a vibrating CNC machine
11:45 PM solarwind: The protocol is not published, but I can probably reverse engineer it over a weekend
11:45 PM mrec: it should be easy yes
11:45 PM mrec: some commands and get the values back
11:46 PM mrec: I'm currently reworking my smd feeder protocol ... RS485 master and bunch of slave connections with a smart protocol
11:46 PM solarwind: But yeah no issues with this motor, I have the exact same one
11:46 PM mrec: I'm using those Padauk controllers on each slot to identify where smd feeders are connected