#linuxcnc Logs
Jan 15 2022
#linuxcnc Calendar
12:08 AM XXCoder: awesome
12:16 AM XXCoder: did you know credits scene with that paper had a umm interesting easter egg
12:16 AM roycroft: yeah and i got to make a couple hundred dollars overtime which i can really use
01:33 AM Deejay: moin
05:07 AM Tom_L: morning
05:07 AM Tom_L: 21°F
05:12 AM JT-Cave: morning
05:13 AM JT-Cave: N/A°F
05:15 AM Tom_L: tilt!
05:16 AM JT-Cave: hmm I see rain on the camera
05:18 AM Tom_L: got a speck of snow but quit by ~10 last night
05:19 AM Tom_L: ~30mph wind
05:20 AM Tom_L: https://earth.nullschool.net/#current/wind/surface/level/orthographic=-103.31,34.62,1287/loc=-99.933,35.130
05:21 AM JT-Cave: looks like our winds are from the east which is not normal
05:21 AM Tom_L: yeah
08:13 AM JT-Cave: damn can't find my instructions for changing my user name on the rpi
08:13 AM JT-Cave: guess I'll have to wing it
08:29 AM _unreal_: 61F here :) little colder then norm but I'll take it
09:05 AM -!- #linuxcnc mode set to +v by ChanServ
10:47 AM Tom_L: JT-Cave, it's under extras
10:47 AM Tom_L: https://gnipsel.com/linuxcnc/uspace/rpi4-ext.html
10:49 AM JT-Shop: I saw that one but for some reason it doesn't always work like the one where I created a new user pia and then changed pi to john then deleted pia
10:49 AM Tom_L: guess i haven't seen that one
10:50 AM Tom_L: oh yes.. that's the one i _do_ remember.. i didn't read the page just saw the 2nd line
10:50 AM JT-Shop: IIRC even though you logout as pi and login as root pi is still there and you can't change a user name unless it's not active
10:51 AM JT-Shop: I wonder if the history machine can find it lol
10:51 AM Tom_L: i may have a copy of it
10:51 AM * JT-Shop waddles out to the other shop to do some glue ups
10:52 AM Tom_L: lemme paste this
10:52 AM -!- #linuxcnc mode set to +v by ChanServ
10:54 AM Tom_dev: JT-Cave, JT-Shop: https://paste.debian.net/1227223/
10:55 AM Tom_L: look more like it?
11:04 AM _unreal_: I have not found the info I'm looking for online. I know any DC motor can be turned into a servo motor with an encoder. But are there anything to avoid when choosing a motor.
11:04 AM _unreal_: has anyone done much with DC brushed servo motors?
11:04 AM Tom_L: i did but didn't stay with it
11:04 AM Tom_L: my sherline spindle motor was brushed dc
11:05 AM _unreal_: well regardless what I'm looking at is this motor that came from a battery powered scooter. one of those over sized "Razor electric scooters" for kids
11:05 AM _unreal_: its a DC 24v 100watt
11:05 AM _unreal_: the thing I dont like about it is that its brushes are internal
11:05 AM _unreal_: the brushes are HUGE but internal
11:06 AM Tom_L: my first one was internal the replacement were not
11:07 AM _unreal_: mainly what I'm getting at is these encoders that I ordered are very nice. but they are more or less a 99% ONE time install. the base is designed in such a way that the "clips" crack just by installing the encoder. and I mean JUST pushing it down into place
11:07 AM _unreal_: how was the working life span?
11:08 AM Tom_L: i never wore it out but i don't use it constantly either
11:08 AM Tom_L: now i got 2 spare sherline spindle motors
11:08 AM JT-Shop: Tom_L, thanks
11:08 AM Tom_L: np
11:08 AM _unreal_: I dont think I'll wear it out if I end up using it. regardless its 3x more powerful the the other brushed servos
11:09 AM _unreal_: the other one's are rated for 30 oz/in
11:10 AM _unreal_: question. it has the standared DC brushed motor TWO big ass magnets glued to the walls of the the motor body. do you think it will harm the magnets IF I should take the motor apart if I should need to deal with the brushes?
11:10 AM _unreal_: its ODD. the brushes are mounted to the end cap that is also the drive shaft end of the motor
11:14 AM _unreal_: as far as the brushes go. what I dont like is that they seem to be a bit loose. when I turn the motor left/right. I can hear and feel the brushes slightly shift inside there holder. its only like a 2 degree before they have shifted but I dont...? like that. I dont know that it would cause any operation issues. in fact I tend to doubt it. BUT there is always the chance of something. I'm more or less speculating with the hopes of
11:14 AM _unreal_: someone pointing out an aspect I may have over looked.
11:35 AM _unreal_: and everyone leaves me
11:35 AM _unreal_: sigh
11:36 AM t4nk_freenode: poor guy ;)
11:40 AM _unreal_: just finished designing an encoder mount for this scooter motor
11:41 AM _unreal_: should be amazing when everything is said and done
11:41 AM t4nk_freenode: I want to 'design' another z-axis
11:41 AM t4nk_freenode: this time with square tubes and bearings
11:42 AM _unreal_: dude. just get liner rails
11:42 AM t4nk_freenode: that's cheating, make up your mind ;)
11:42 AM _unreal_: 12mm
11:42 AM Tom_L: dude. that would be too simple!
11:42 AM t4nk_freenode: hehe
11:42 AM _unreal_: 12mm g something something rails
11:42 AM _unreal_: use your cnc mill to "mill" the main plate
11:43 AM _unreal_: then its just mounting the hardware to it and preforming the swap out surgery
11:43 AM _unreal_: but depending on design, then your Z becomes the most rigid part of your machine
11:44 AM _unreal_: a set of 9-12mm 10" long or what ever you need is dirt cheap.
11:45 AM perry_j1987: i just upgraded the z axis on my 6040
11:45 AM _unreal_: ! isnt that an evolving version of cheating? if your using the MACHINE to do the work. to install better hardware. its still DIY
11:45 AM perry_j1987: was going to make the z axis used on printNC routers but i ended up just getting a prebuilt z axis for cheaper off amazon
11:45 AM _unreal_: t4nk_freenode,
11:45 AM _unreal_: pre-built. now thats cheating
11:46 AM perry_j1987: still had to make adapter plates :)
11:46 AM _unreal_: I can see an MFG'ed A: or some kind of rotory axis being purchased.
11:46 AM _unreal_: :p
11:46 AM perry_j1987: got one of them too heh
11:46 AM _unreal_: the new machine I'm designing is going to be amazing
11:46 AM _unreal_: when I build it
11:47 AM _unreal_: I just dont plan on ANY construction, till I settle on a motor drive system.
11:47 AM perry_j1987: clearpath servos?
11:47 AM perry_j1987: or do you have a budget lol
11:47 AM _unreal_: I'm ether using the BRUSHED servos, the BLDC motors. or a set of the stepper motors I have. I have drivers for everything so its not an issue.
11:48 AM _unreal_: perry_j1987, I built the ve7it brushed servo controllers
11:48 AM perry_j1987: close the loop on the steppers
11:48 AM _unreal_: and as of a few days ago finally figured out why they failed
11:48 AM _unreal_: some how I fucked up and shorted one of the brushed motor wires to the case
11:48 AM _unreal_: shorted everything out
11:48 AM _unreal_: with +30v
11:49 AM perry_j1987: figured you'd give yourself an interesting time eh?
11:49 AM _unreal_: amazingly I dont believe it harmed any of the electronics
11:49 AM _unreal_: I'll know for sure when I finally get one of these computers up and running with a serial console so I can "log" into the motor controllers
11:50 AM _unreal_: if they suddenly "function" then all is well. IF they still do odd ball stuff. then I prob blew the DSPIC's :(
11:50 AM _unreal_: I dont think I did though
11:51 AM _unreal_: it looks like I may have damaged my optical limit board. but thats no major biggie, its just an LN393 AND LM339. but I dont know if I messed up the optical interrupters or not? I have not tested them.
11:51 AM _unreal_: :/
11:52 AM _unreal_: problem is both of those chips are no longer MFG'ed and are hard to find. and I dont know what a drop in replacement chip may be?
11:53 AM _unreal_: so its a work in progress
11:53 AM _unreal_: but like I said right now I'm trying to figure out if the servo controllers I build are in good health or not.
11:54 AM _unreal_: I built it I want to use it.
11:56 AM _unreal_: the BLDC's would be my runner up BUT they are weaker. 20 oz/in but rated for 2150 RPM. if I gear them down with what I have. thats a 6:1. That gives me 180 OZ/IN
11:57 AM _unreal_: the bldc's would give me close to 30"/min rapids
11:57 AM _unreal_: sigh
11:58 AM perry_j1987: lm393 are mfg still
11:58 AM t4nk_freenode: just buy a decent motor with encoder?
11:58 AM perry_j1987: got a reel of them sitting here
11:58 AM _unreal_: digikey and mouser dont have them
11:58 AM perry_j1987: you can get em on aliexpress even
11:58 AM _unreal_: ali would not suprise me
11:59 AM _unreal_: t4nk_freenode, I have decent motors with encoders
11:59 AM t4nk_freenode: then fit those, and be done then?
11:59 AM _unreal_: I have 3 choices
11:59 AM _unreal_: brushed servos, bldc servos, steppers.
12:00 PM _unreal_: if I'm looking at strongest, it would be stepper, brushed, bldc as far as what I have on hand
12:00 PM perry_j1987: close the loop on the steppers and your good to go
12:00 PM _unreal_: dont need to do closed loop stepper
12:01 PM _unreal_: besides if I was going to do that I would get a stepper controller that takes closed loop and steppers WITH closed loop
12:01 PM _unreal_: but I dont need that
12:03 PM roycroft: weird
12:03 PM roycroft: we had a small tsunami on the oregon coast a couple hours ago
12:03 PM roycroft: but now a new tsunami watch has been issued
12:04 PM jdh: from tonga
12:04 PM roycroft: yes
12:04 PM roycroft: and it was from a volcano eruption, not an earthquake
12:04 PM roycroft: which is also weird
12:04 PM roycroft: very unusual
12:04 PM roycroft: fortunately the tsunami was very minor - more like a king wave
12:05 PM roycroft: no damage, just flooded parking lots and the like
12:06 PM perry_j1987: came across this just now https://www.toolholdershop.com/cnc_machines_accessories/testing_bars/er20_testing_bar_l_60_mm randomly
12:06 PM perry_j1987: how do you get that in and out of the er nut without it being a collet that can compress a bit to tilt to one side
12:08 PM roycroft: there is probably just enough clearance to be able to snap it in like you do with a collet
12:09 PM _unreal_: perry_j1987, I would assume it has internal relief to allow it to snap in. nifty though
12:10 PM _unreal_: would make for a great probe
12:10 PM _unreal_: if your dealing with tooling metal
12:16 PM _unreal_: I'm torn between milling this adapter or 3d printing it
12:16 PM _unreal_: :/
12:16 PM _unreal_: My 3d printer tends to make stuff on the small side
12:17 PM _unreal_: and I have yet to figure out exactly why but its like its always 1-3% small
12:24 PM _unreal_: screw it 3d printing it
12:25 PM _unreal_: just "scaled" the output up 2% it will be interesting to see how it turns out
12:25 PM _unreal_: I should do a "test" to solve what the scale offset is?
12:26 PM perry_j1987: what adapter
12:27 PM _unreal_: I just drew in cad a "mount" for this rotory encoder
12:27 PM _unreal_: its to mount to this electric scooter motor
12:29 PM _unreal_: perry_j1987, kids electric scooters are tossed away endlessly because the PB batteries go bad. mainly because no one keeps them plugged in to maintain the PB battery.
12:30 PM _unreal_: and rather then just buy a new dirt cheap battery they buy a whole new scooter
12:30 PM _unreal_: so the goodwills end up with them and sell them for often cheap.
12:30 PM perry_j1987: ive never seen electric scooters at goodwill around here
12:30 PM perry_j1987: those hoverboards or actual scooters?
12:31 PM _unreal_: ahh no
12:31 PM _unreal_: er yes scooters
12:31 PM _unreal_: you know the kids razor scooters only the electric versions
12:31 PM _unreal_: be it a knock off brand or the razor. regardless they are at good wills often
12:32 PM _unreal_: the motors are always really powerful DC brushed motors
12:32 PM _unreal_: rated at 24v 99% of the time
12:32 PM _unreal_: two 12v lead acid batteries drive them
12:33 PM _unreal_: perry_j1987, the 30 oz/in when I was testing brushed motor was more then powerful enough to drive the other CNC machine I had been building that is sitting at work.
12:34 PM _unreal_: really wishing I had never took it there but just made room in my room :/
12:34 PM _unreal_: oh well
12:35 PM _unreal_: speaking of that cnc machine build. I just got the last part TOOLED. a company was able to do it FINALLY
12:36 PM _unreal_: they got it like 90% tooled and could not reach the rest but I can work around that
12:37 PM perry_j1987: im gearing up to make my first fixture plate
12:37 PM _unreal_: I can AT LEAST do what I need now. and start the build process of the Z for that machine
12:37 PM _unreal_: dude I have two aluminum fixture plates
12:37 PM _unreal_: I use them on the laser etcher mainly
12:37 PM perry_j1987: im tired of doing one at a time on the vise
12:38 PM perry_j1987: got a tool length setter probe now and extended z axis so i can reach down to the bed and then back up to the tool length probe
12:38 PM perry_j1987: its going to get REAL nice here soon
12:39 PM _unreal_: My main use for them which has never happend is the other cnc machine that I've been building. that bild has been stalled for almost a year on one part.
12:39 PM _unreal_: but regardless ya I made two fixture plates that are 1/4" aluminum with 1.5" 1/4" holes in a GRID across both plates
12:40 PM _unreal_: currently other then laser etcher setups I've only use then for HAND building fixturing projects :/
12:41 PM _unreal_: this 3d print is going to be more of an experiment. its 100% infill. so its GOING to take the full weight of the motor on the 3 mount points
12:42 PM _unreal_: and it holds the encoder
12:42 PM _unreal_: Its annoying but the HARD mounts on the scooter motor are only on one end. the shaft end and in a triangle setup.
12:43 PM _unreal_: I could build a body holder but that could prove to be problmimatic. I'm going to be using 4"x4" steel tubing as part of the main base for the new cnc build
12:44 PM _unreal_: so I have a choice of cut a big ass hole in the steel tube to fit/hide the motor under the body of the machine. OR mount it external. I'd rather mount it hidden.
12:44 PM _unreal_: ve7it, hello
12:44 PM _unreal_: ve7it, i'm printing a mount for that DC brushed scooter motor right now for the encoder
12:45 PM perry_j1987: there's guy on youtube 3d printed brackets that bolted onto. his rollerblades with pusher electric motors off the back heh
12:45 PM _unreal_: ya I saw that. interesting but kind of sketchy
12:45 PM perry_j1987: ya heh
12:46 PM _unreal_: I dont view those as a reasonable build concept
12:47 PM perry_j1987: t-shirt weather this tuesday
12:47 PM perry_j1987: will be up to 36
12:47 PM _unreal_: ya got to love being in the tropics :)
12:47 PM _unreal_: its like 70 outright now
12:47 PM _unreal_: sigh... cant wait for this print to be done
12:48 PM perry_j1987: snowing now
12:48 PM _unreal_: 75% done
12:48 PM _unreal_: ya its raining sun shine here
12:48 PM _unreal_: not a cloud in the sky
12:48 PM _unreal_: after this print I may take my daughter to the park or something
12:51 PM _unreal_: 85%
12:53 PM _unreal_: 98%
01:02 PM _unreal_: sweet... the encoder body will fit
01:02 PM _unreal_: now I just need to find some more NANO screws so I can mount the encoder
01:05 PM roycroft: we're having sunny breaks this morning, and no more rain until at least mid-week
01:15 PM _unreal_: and thats it. the encoder is now mounted
02:22 PM perry_j1987: mmm coffee
03:24 PM JT-Shop: yea got the gouge roughing jig done and in place... now to fix all my bowl gouges that had been sharpened by hand
03:25 PM XXCoder: nice
03:30 PM roycroft: i need to make a gouge sharpening jig for my belt grinder
03:31 PM roycroft: it's on the list, but the list is long, and it's not near the top
03:31 PM XXCoder: if you are soon done with list, it means you forgot a lot of stuff needed to do lol
03:32 PM roycroft: i'll never be done with the list
03:32 PM roycroft: well, when i die, i should imagine my list would be voided at that point
03:32 PM roycroft: but until then, there will always be a list
03:33 PM roycroft: and maybe i'll be mean and assign some list items to my estate
03:41 PM _unreal_: hello
03:42 PM _unreal_: ve7it, thre is an RB2425K CAP on my brushed servos. not sure what its for but the scooter brushed motor does not have one. are those caps sized by any chance? or should I just add one that is close in size...
03:43 PM ve7it: _unreal_, hey
03:43 PM _unreal_: that caps is an IMB 0.47uF 50v metal film cap
03:43 PM _unreal_: ve7it, I got the new encoders mounted to the motors today. I just got back. took my daughter to the playground
03:44 PM ve7it: noise reduction to kill off some of the brush hash
03:44 PM ve7it: (pictures of mounts?
03:44 PM _unreal_: I have not taken any yet
03:44 PM _unreal_: but just 3d printed mounts to allow me to mount the encoders
03:44 PM _unreal_: I set them to 500 PPR
03:45 PM ve7it: what did you use on the scooter motor?
03:45 PM _unreal_: and I'm NOT removing them again. its too risky
03:45 PM _unreal_: ?
03:45 PM _unreal_: use?
03:45 PM ve7it: I am interested in how you mounted the encoder on the scooter motor
03:45 PM _unreal_: I just designed mounts I could 3d print that would mesh with the existing mounting holes
03:47 PM ve7it: front or back of the motor?
03:47 PM _unreal_: give me a sec and I'll post a photo
03:48 PM _unreal_: https://drive.google.com/file/d/1ULqcSCD7Ek14wrXVRLao5wOD7Ji01xr_/view?usp=sharing
03:48 PM _unreal_: ve7it,
03:49 PM _unreal_: I have 11mm of shaft left to mount too
03:49 PM _unreal_: more then enough
03:50 PM ve7it: cool... was that a 250W motor?
03:50 PM _unreal_: no 100
03:50 PM _unreal_: the OPA opamps are rated for 250 watts
03:50 PM _unreal_: 30v, 8amps
03:51 PM _unreal_: constant and what 12amps burst?
03:51 PM ve7it: nice motor
03:51 PM _unreal_: those motors are standard on those razor and razor like elec motors
03:51 PM _unreal_: scooters
03:51 PM _unreal_: the kids scooters
03:51 PM _unreal_: those scooters are endlessly being tossed or given to second hand and sold for cheap
03:52 PM _unreal_: I'm not a fan of the mount but what ever
03:53 PM _unreal_: if the body was not so big I would have just printed an "adapter" to convert it to nema23
03:53 PM ve7it: yea... I have some larger versions, but never figured out how to mount encoders... I was thinking of drilling and extending the rear shaft, but putting the encoder on the front is not bad
03:53 PM _unreal_: ya
03:53 PM _unreal_: I'm happy with it
03:53 PM roycroft: _unreal_'s suggestion about extending my spindle mounting plate is a good one, but not for the reason he mentioned
03:54 PM roycroft: or, at least, not as i understood his reason
03:54 PM ve7it: they have radial versions of the encoder (wires out the side)
03:54 PM roycroft: i need to extend it so that i have clearance when machining thick pieces
03:54 PM _unreal_: the wires are not an issue
03:55 PM _unreal_: ve7it, I'm going to loop the belt passed the mount that is at the other end from the wires
03:55 PM _unreal_: could not be a better setup. and amazingly the encder JUST fits
03:56 PM _unreal_: roycroft, thats what I was getting at CLEAREANCE
03:58 PM roycroft: i've been playing with the model and i see that now
03:58 PM roycroft: but it's really not as bad as you thought
04:02 PM roycroft: https://roycroft.us/Z-Axis-Assembly-Preliminary.jpeg
04:03 PM roycroft: with things adjusted like that, i have about 3.5" clearance from the spindle nose to the bottom bracket
04:03 PM roycroft: i need more than that, but not a lot more
04:03 PM roycroft: and i think the spindle is mounted very sturdily in that configuration
04:05 PM roycroft: i also think i could have gotten rails that are 50mm shorter
04:05 PM XXCoder: meh you would want those if you ever want to use reamer
04:05 PM XXCoder: reamer tend to be very long
04:05 PM _unreal_: ve7it, looks like the closest CAP I have in that range is a 1uF cap
04:05 PM _unreal_: ceramic
04:06 PM _unreal_: ve7it, I assume that will work fine
04:06 PM roycroft: yes, and although i don't anticipate using a reamer in my router, i may want one at some poit
04:06 PM roycroft: point
04:07 PM roycroft: one of the design parameters is to keep the x axis crossmember as low as possible, for rigidity
04:07 PM roycroft: but that does not mean that the top of the z axis cannot extend well above the x axis crossmember
04:08 PM _unreal_: roycroft, you really want to "slide' the main z carriage all the way to the TOP and then extext the mount down just long enough that your spindle shaft is just below or just above the bottom of the entire Z. That way you can maximize your tooling area
04:09 PM CaptHindsight[m]: https://www.linearmotiontips.com/what-you-need-to-know-about-the-2-1-ratio/
04:09 PM XXCoder: unreal yeah when I was designing mine, I had that in mind.
04:09 PM _unreal_: roycroft, generally all Z's have an extended cantilever setup.
04:10 PM CaptHindsight[m]: https://www.machinedesign.com/mechanical-motion-systems/article/21836017/linear-bearings-understanding-the-21-ratio-and-how-to-overcome-the-stickslip-phenomenon
04:10 PM roycroft: yes, that's what i'm measuring for now
04:10 PM CaptHindsight[m]: https://www.youtube.com/watch?v=tI0dF27hJiU
04:11 PM roycroft: i think that what i need to do next is work on the x axis, then the y axis
04:11 PM roycroft: leaving the z as is for the moment
04:11 PM _unreal_: roycroft, the ideal setup take into account the maximum load that the max of the largest BIT meaning shaft of the bit, and what kind of LOAD it can yield and what deflection you can expect depending on the materials used. AKA thickness size o the metal. if there are extra supports if any etc....
04:11 PM roycroft: because only when i have the whole machine designed will i be able to measure clearance between the machine bed and the bottom of the z axis
04:12 PM CaptHindsight[m]: jump to ~11 minutes in to get passed the commercial
04:12 PM _unreal_: roycroft, your bearings should be able to take something like a 50NM load EACH. then times that by 4
04:12 PM _unreal_: roycroft, nonsense
04:12 PM _unreal_: you need only know what your designed tooling area is
04:12 PM _unreal_: depth/plunge
04:13 PM _unreal_: then measure off of your desired work area
04:14 PM _unreal_: if you want 12" well then the first thing it to identify the shortest bit you will ever likely use
04:14 PM roycroft: i don't yet know how high off the bed the x axis crossmember will be
04:15 PM _unreal_: or what ever bit you would expect to be using AT the maximum height. then ensure the spindle shaft WITH "bit" installed can reach that location in its linear travel
04:15 PM _unreal_: then in the bottom of the travel what do you expect to get down to. What is the shortest and logest bits you expect to use...
04:15 PM _unreal_: what is the height of the waste board
04:15 PM _unreal_: etc....
04:16 PM _unreal_: I designed ALLL of those aspects into my cad drawing and much more
04:16 PM _unreal_: that gives a TRUE feel for the working system and what can be expected or planned for.
04:16 PM _unreal_: roycroft, you design for it
04:17 PM _unreal_: first design for your spindle expectations with SAID max/min bits meaning how long and short they will be. and where you will expect then to land from the TOP of your waste board
04:17 PM roycroft: the thing i need to do next is decide the maximum thickness of material i want to be able to machine on this router
04:17 PM _unreal_: once you know that average then you can design for your rails in the z and how much travel you will need
04:18 PM CaptHindsight[m]: https://matrix-client.matrix.org/_matrix/media/r0/download/matrix.org/kzfErJoaluLddvjOkSsLNDTk BOB for Mesa 7i76e with DB connectors
04:18 PM roycroft: i already ordered the z axis rails and ball screw
04:18 PM _unreal_: once you know what then you can design your X axis and how wide it must be. so you can acheave your desired X travel based on CENTER of the Z carriage
04:18 PM roycroft: so that decision is made
04:18 PM CaptHindsight[m]: PCB date is from April 2018
04:18 PM roycroft: and yes, the reason i decided to design the z axis first was so i could calculate the width of the x axis
04:19 PM _unreal_: so if your z is 6" wide then on center is 3" +/- so if you want to move 6" left right for max working area. then you need your X to have .... 12" of travel
04:19 PM CaptHindsight[m]: roycroft: how does it take to drive from Portland to Seattle?
04:20 PM CaptHindsight[m]: how long, sorry
04:20 PM _unreal_: roycroft, thickness of material your going to mill does not matter...
04:20 PM roycroft: about 2 1/2 hours, depending on traffic
04:20 PM roycroft: _unreal_: it actually does
04:20 PM roycroft: because i cannot have the x axis crossmember closer to the spoilboard than the thickness of the material
04:20 PM _unreal_: you dont mill at the max thickness in one passs but rather in step down tooling passes
04:20 PM _unreal_: so thickness does not matter on the working material
04:21 PM _unreal_: now frame thickness of the machine matters very much :)
04:21 PM roycroft: if i want to machine something that is 6" thick my x axis crossmember cannot be 5" above the spoilboard
04:22 PM _unreal_: roycroft, ugh... thats what all that typing was about. start with your desired Z travel based on the average of the tooling bits your going to use. shortest to longest. shortest normally are the 1/8 V bits. those are like 1" long on average
04:23 PM _unreal_: once you know what your going to need/disire for z maxinum travel. then you design your rail length. and z carriage.
04:24 PM _unreal_: if you want a long Z travel. then the z carriage is going to be as long as it needs to be to match the travel of the rails. as expected per design.
04:25 PM roycroft: i was already anticipating that the shortest cutter i use will have at least 1" stickout from the spindle nose
04:25 PM _unreal_: for example. if you have 10" rails, and you want 5" of travel. then you are going to want the bearing cars to be spaced 5" apart. but when the z carriage is at the top the spindle may not reach below the z frame. so you must extend the z carriage DOWN as a cantilever design
04:25 PM _unreal_: stickout? eheh its called reveal
04:26 PM _unreal_: but stickout works
04:26 PM roycroft: it's a term i'm familiar with, and that i've seen others use in this context
04:27 PM roycroft: whether it's the standard term, i do not know
04:27 PM roycroft: but you understood what i was saying, so it worked, this time at least
04:27 PM _unreal_: standard no.... do you have an "engineers" standard reference book?
04:27 PM _unreal_: if not if you have a harbor freight near buy. they sell them
04:28 PM _unreal_: they are a great reference for everything machanical
04:28 PM roycroft: i have multiple copies of machinery's handbook, and a number of other machining/engineering books
04:28 PM _unreal_: ok.... o,O well reveal is a term that will be refrenced in them.... not likely so much "stick out"
04:29 PM roycroft: i did go to engineering school a century or so ago, and still have most of my textbooks, as well
04:29 PM _unreal_: a century... hum things could be fuzzying
04:29 PM _unreal_: LOL
04:29 PM roycroft: where they all are, i cannot say right now - most are in one of several piles that i have not the shelf space to organize
04:30 PM _unreal_: get the cnc going and you can mill a duzen shelves
04:30 PM roycroft: i don't need the cnc router for that, and it would not be big enough anyway
04:30 PM roycroft: i need two things, one of which i don't have:
04:30 PM _unreal_: ve7it, so I assume your developing some evil plans for encoders and those kind of motors after seeing my design.
04:30 PM roycroft: 1. dry timber
04:30 PM roycroft: 2. floor space to install the shelves
04:31 PM roycroft: the first item i pretty much have now
04:31 PM roycroft: i purchased a bunch of green douglas fir 2x12s a bit over a year ago
04:31 PM roycroft: they had a 30-40% moisture content when i purchased them
04:32 PM roycroft: they're stickered and inside my house, and are finally down to <10% moisture
04:32 PM ve7it: _unreal_, Always scheming
04:32 PM roycroft: so finally ready to use
04:33 PM roycroft: the floor space will continue to be an issue, and even when i build the three new bookcases, i'll not have enough shelf space
04:33 PM roguish[m]: capt_hindsight[m]: hey. look at the YouTube by Bishop-Wisecarver that you just referenced. go to 10:00 minutes into it and see the product.
04:33 PM roycroft: i'll need to purge some of the books, which is always difficult for me, and try to find some other floor space that i can free up
04:33 PM roguish[m]: I designed the original for them. Have the prototype on my shelf.
04:33 PM CaptHindsight[m]: roguish[m]: yeah why i said to skip to 11 min in :)
04:34 PM CaptHindsight[m]: sorry I usually include the start point in youtube links
04:35 PM CaptHindsight[m]: even non corporate produced vids have long intros
04:38 PM roguish[m]: capt_hindsight[m]: seriously, go out to 10 min, and stop it. check out the product.
04:38 PM roguish[m]: Utilitrack
04:41 PM CaptHindsight[m]: yeah, the cam bearings and tracks
04:41 PM CaptHindsight[m]: you like them?
04:42 PM CaptHindsight[m]: https://www.bwc.com/utilitrak-linear-guide-track.html
04:43 PM CaptHindsight[m]: https://www.hepcomotion.com/videos/discover-utilitrak/
04:43 PM CaptHindsight[m]: v-groove
04:44 PM CaptHindsight[m]: precision drawer slides
04:45 PM roguish[m]: capt_hindsight[m]: they have purpose. depends on what for. The original Utilitrack was BWC's rollers on a plate, captured in an extruded channel.
04:45 PM roguish[m]: It's evolved over the years.
04:46 PM roguish[m]: Hepco is a UK company that BWC partners with. market and sell each other's products. Hepco is wicked pricey.
04:58 PM Tom_L: roycroft, machinable thickness can be figured from the largest bit that will fit in the spindle with the largest flute length you can get
05:06 PM perry_j1987: just threw the testbar back in the lathe and it was out 0.1778 over the length of the bar
05:07 PM perry_j1987: got it dialed in again to zero needle movement
05:08 PM CaptHindsight[m]: perry_j1987: inch or mm?
05:09 PM perry_j1987: mm
05:11 PM perry_j1987: now to throw the boaring head in the lathe and make some new tool holders
05:11 PM _unreal_: ok both servo motors have been soldered up right
05:11 PM _unreal_: at this point I just need to create the data cable
05:12 PM _unreal_: for the encoder
05:12 PM perry_j1987: making a gang tool block that holds 3 er16 collet holders and a 5/8 insert holder
05:12 PM _unreal_: ve7it, this time I have REALLY good shielded data cable
05:12 PM _unreal_: wish it was round not flat but what ever
05:12 PM _unreal_: flat good for the drag chains but not so good for the motors them selves...
05:25 PM _unreal_: sigh....
05:25 PM _unreal_: I am not going to get these motors tested tonight thast for sure
05:25 PM _unreal_: I need to build an RJ45 breakout board
05:26 PM _unreal_: last thing I want to do. becuase that means cnc milling pcb's :/
05:26 PM _unreal_: one for each end of the high quality flex able shielded cable :(
05:27 PM _unreal_: I didnt want to use this cable but its just too good. and it combinds all the signals I Need
05:28 PM _unreal_: +/- 5v, + 6 signals A: B: encoder, +/- LIMITS, probe, and ESTOP
05:30 PM _unreal_: UGH.... I do not want to build a new PCB board design
05:42 PM CaptHindsight[m]: https://www.reddit.com/r/CNC/comments/s4p6y0/giant_6_axis_cnc/
05:43 PM CaptHindsight[m]: dual-Z gantry
06:35 PM andypugh: Is this thing on? :-)
06:35 PM XXCoder: no
06:38 PM andypugh: Ah, just the chap…
06:38 PM andypugh: Bear with mw whilst I craft some links
06:41 PM andypugh: Working on the direct-mount probe:
06:41 PM andypugh: https://photos.app.goo.gl/3DyURWc8mBSX5TZ66
06:41 PM andypugh: https://photos.app.goo.gl/XjAam1skv5z73MgZA
06:45 PM XXCoder: interesting
06:47 PM andypugh: The scribble underneath is another aspect of this, I am working on probe centre calibration in software,
06:50 PM XXCoder: :)
06:51 PM andypugh: If this works I will keep just one probe and shift it between holders. My mill currently has three spindles.
07:27 PM skunkworks: andypugh: https://photos.app.goo.gl/QcQF57iMDjawoZp27
07:27 PM skunkworks: finally getting back to that. (accupin supply)
07:27 PM andypugh: Lovely museum piece :-)
07:27 PM skunkworks: distilled down to https://photos.app.goo.gl/7YgQFuYbosUmi8pu8
07:27 PM skunkworks: and an arduino.
07:28 PM skunkworks: hopfully
07:28 PM skunkworks: Right!?
07:28 PM andypugh: The original is probably more stable. And cost 1000x more.
07:28 PM XXCoder: andy thats cool :)
07:30 PM _unreal_: andypugh, I see the cube has been solved in the background
07:30 PM skunkworks: andypugh: yes - when we actually ran the old control (for years) it positioned perfectly.
07:30 PM andypugh: Hmm, logging in as “shunned one” feels like a way to not encourage responses.
07:30 PM andypugh: _unreal_: Yes, lockdown project, re-learn the cube. I can do it in about a minute again. (Like I could when I was 14)
07:31 PM skunkworks: andypugh: your handwriting is awesome.
07:32 PM _unreal_: GOD I hate converting kicad to gerber
07:33 PM skunkworks: andypugh: that circuit was just the driver for the coils.. pretty much 2 op amps.
07:34 PM andypugh: Before you could buy an op-amp as an IC, I guess.
07:35 PM skunkworks: yes - the schematics have 62' dates
07:35 PM andypugh: Does the old control still run? It might be the oldest working CNC control in the world....
07:35 PM skunkworks: no.. We scraped it when a substantial amount of smoke came out of it.
07:35 PM CloudEvil: I am now imagining clicksprings 'oildest CNC in the world' series of videos
07:36 PM skunkworks: at the point emc2 was viable.
07:36 PM skunkworks: the linear scales are in the machine still - always wanted to get those running. Would help with some thermal expansion problems we work around.
07:37 PM skunkworks: plus a cool little project
07:38 PM skunkworks: https://electronicsam.com/images/KandT/conversion/rearcontrller.JPG
07:38 PM skunkworks: that wasn't all of the cards.. It was about 8ft tall.
07:38 PM skunkworks: lower right are the 3 accupin cards.
07:39 PM skunkworks: eh - towards the middle I guess. x,y and z
07:39 PM andypugh: That’s a beautiful museum piece. I am glad I don;t have the job of deciding to throw it away :-)
07:39 PM skunkworks: yah - it was a hard decition. it was so big.
07:40 PM skunkworks: Still don't understand how they did circular and linear interp.
07:44 PM skunkworks: it may have been analog.
07:45 PM andypugh: Seems rather likely.
07:47 PM skunkworks: we figure the to section of the control was the interpolation.. (the section bolted above the control panel)
07:47 PM skunkworks: interpolation
07:48 PM skunkworks: https://electronicsam.com/images/KandT/oldkandt.JPG
07:48 PM skunkworks: it was an option.
07:48 PM skunkworks: *top
07:48 PM andypugh: I do hope you always wear a white lab-coat when operating it?
07:49 PM skunkworks: I do! lol
07:49 PM roycroft: with a proper pocket protector?
07:49 PM skunkworks: I like the floor too -
07:50 PM skunkworks: that machine goes down about 2 foot into a pit.
07:50 PM skunkworks: so that flooring around the machine is removable.
07:52 PM skunkworks: ours isn't as pretty - but what came with the machine.
07:53 PM skunkworks: https://electronicsam.com/images/KandT/conversion/decking2.jpg
07:59 PM andypugh: I have an ambition to see your machine at some point.
08:28 PM _unreal_: god I hate doing PCB's
08:28 PM _unreal_: ok gcode is generated
08:28 PM _unreal_: 9pm... do I really want to do a cnc milling job :/
08:29 PM _unreal_: I do have PCB already setup on the cnc.I have but to swap in a BIT to do the job.....
08:29 PM _unreal_: sucks I have to do 6 of these boards :/
08:29 PM _unreal_: not that I want to but I need the Break out headers for this network cable.
08:32 PM _unreal_: I MUST say its a nice design. for its intended use, it gives a 4pin header, 2x 3 pin headers, and 2x 2pin headers, 4pin is encoder +/-v A:B, 3 pin headers are end stops +/- signals and +/-v for pull high/low, and two 2 pin headers are for Probe sig/grd, and ESTOP sig/grd. no one pulls those two signals HIGH
08:32 PM _unreal_: if you do WTF's wrong with you....
08:33 PM _unreal_: regardless of how I use the break out I have more then enough signal lines for each location
08:34 PM andypugh: I actually like designing PCBs.
08:34 PM _unreal_: I dont mind designing PCB's its the flatcam phase I detest
08:35 PM andypugh: Though I don’t enjoy making them enough to bother any more. Seeed / JLCPCB and wait a week is what I do noe.
08:35 PM _unreal_: granted I own fusion360 and it has eagle. I've never really learned EAGLE pcb
08:36 PM _unreal_: odd thing is years ago I was pirating EAGLE back in the early 2k's and it was the greatest thing. now its so over bloated its just confusing.
08:37 PM _unreal_: andypugh, I solved my major fail with building the ve7it BRUSHED SERVO
08:37 PM _unreal_: some how I damaged a wire by catching it with a screw. I shorted one of the motor wires to the case
08:37 PM _unreal_: it blew out a hand full of things
08:37 PM CloudEvil: :)
08:38 PM andypugh: I use KiCAD. Or sometimes the online JLCPB thing. Depending whether it’s a one-off or a real project.
08:38 PM _unreal_: blew out the encoders on the brushed servos I have :(
08:38 PM CloudEvil: :(
08:38 PM _unreal_: that really bumbed me out
08:38 PM _unreal_: I just today finished installing replacement modern encoders on both motors
08:39 PM _unreal_: now I'm building the aggressive shielding setup for the encoder
08:40 PM _unreal_: though I dont know that it ever had an issue. considering I dumped 30v onto my "5v" line
08:40 PM _unreal_: my original wiring may have been fine????
08:40 PM _unreal_: in any case I may as well shield everything unlike what I had setup before
08:43 PM _unreal_: CloudEvil, andypugh have you guys messed with BLDC or BRUSHED servos at all?
08:43 PM andypugh: BLDC, yes.
08:44 PM andypugh: To the extent that I wrote the “bldc” HAL component, and am (slightly) involved in the STMBL project.
08:46 PM _unreal_: andypugh, I have a hand full of options
08:46 PM _unreal_: I'm designing a new cnc machine to "match" my laser etcher
08:47 PM _unreal_: https://photos.app.goo.gl/1ejmwQjHGskbBvob9 andypugh
08:47 PM CloudEvil: My knowledge is limited to having bought some brushless motors with that aim, buit not actually done it.
08:47 PM _unreal_: I need a bigger home cnc machine that has an equal/better work area to my laser etcher
08:48 PM _unreal_: I have endless parts, so I have plenty of stepper motors, I have these amazing brushless servos, and the ve7it brushed servo project
08:48 PM _unreal_: my brushed servos are 20 oz/in but rated for 2150 RPM
08:48 PM andypugh: I have a test rig rather like that, with STMBL and some 23-size motors, but it’s become buried in project strata.
08:49 PM _unreal_: they come from mega super yacht satelites
08:49 PM _unreal_: so they are nema23 there small but drive really big sats
08:49 PM _unreal_: 60" dishs
08:50 PM _unreal_: in pan,tilt,rotate setup
08:50 PM _unreal_: in that setup. they are really geared down
08:50 PM _unreal_: 6:1 ratio
08:51 PM _unreal_: regardless they are FAST pushing 180 oz/in in 6:1 gear ratio. my brushed servo system that I'm trying JUST get to work can run upto 250watts.
08:52 PM _unreal_: so I'm really kind of torn over what I should go with.
08:52 PM _unreal_: andypugh, the part that really kills me is some of the motors from the SAT's are really big with onboard encoders, BUT there max RPM is 800
08:53 PM _unreal_: the smaller motors 2150 rpm but both are rated for the same oz/in. the bigger motors I'm thinking might make a great 3d printer setup.
08:53 PM _unreal_: or laser
08:55 PM _unreal_: andypugh, I can not seem to find the photo of my new cnc machine concept :/
09:04 PM skunkworks: andypugh: the 5 year plan is to have a cncfest...
09:04 PM andypugh: Hold that thought, I am out of steam at 0248, time to sleep.