#linuxcnc | Logs for 2018-06-27

Back
[00:08:01] -!- ve7it has quit [Remote host closed the connection]
[00:08:37] -!- ve7it has joined #linuxcnc
[00:39:45] -!- ziper has quit [Read error: Connection reset by peer]
[01:00:02] <Lcvette> .weather mars
[01:00:02] <Lcvette> ?
[01:00:03] <theCockerel> Yahoo! Weather - Mars, NW, DE: Partly Cloudy, 9°C (48°F), Humidity: 96%, Light breeze 3.1m/s (↙)
[01:01:14] <Lcvette> .weather wilmington, nc
[01:01:15] <theCockerel> Yahoo! Weather - Wilmington, NC, US: Cloudy, 22°C (71°F), Humidity: 97%, Light breeze 3.1m/s (←)
[01:01:37] <Lcvette> pretty neat
[01:03:12] -!- c-log has quit [Ping timeout: 265 seconds]
[01:04:34] -!- c-log has joined #linuxcnc
[01:09:05] -!- ve7it has quit [Remote host closed the connection]
[01:13:07] -!- c-log has quit [Ping timeout: 245 seconds]
[01:14:24] -!- c-log has joined #linuxcnc
[01:15:01] -!- holzjunkie has joined #linuxcnc
[01:45:57] -!- holzjunkie has quit [Ping timeout: 264 seconds]
[01:49:41] -!- IchGucksLive has joined #linuxcnc
[01:50:05] <IchGucksLive> hi all
[01:50:27] <IchGucksLive> geting a hot week in germany as we expect 34Deg
[01:50:45] <IchGucksLive> Lcvette, lots of wilmingtons in the USA
[01:53:40] <hazzy-lab> .weather
[01:53:40] <theCockerel> I don't know where you live. Give me a location, like .weather London, or tell me where you live by saying .setlocation London, for example.
[01:53:51] <hazzy-lab> you used to know :P
[01:54:00] <hazzy-lab> .weather Atlanta
[01:54:00] <theCockerel> Yahoo! Weather - Atlanta, GA, US: Scattered Showers, 23°C (73°F), Humidity: 85%, Light breeze 3.1m/s (↗)
[01:59:23] <roycroft> it's cooling off here
[01:59:28] <roycroft> we'll be in the mid-20s all week
[02:02:41] * hazzy-lab is jealous
[02:05:09] <hazzy-lab> gn8
[02:12:32] -!- asdfasd has joined #linuxcnc
[02:13:00] <IchGucksLive> im off 2 garden
[02:13:06] -!- IchGucksLive has quit [Quit: Leaving]
[02:27:55] -!- Lcvette has quit [Remote host closed the connection]
[02:41:31] -!- P1ersson has joined #linuxcnc
[02:43:12] -!- hendrik_cnc has joined #linuxcnc
[02:44:07] -!- Deejay has joined #linuxcnc
[02:44:24] -!- P1ersson has quit [Client Quit]
[02:44:47] <Deejay> moin
[02:49:33] -!- P1ersson has joined #linuxcnc
[03:31:55] -!- jthornton has quit [Read error: Connection reset by peer]
[03:31:56] -!- JT-Shop has quit [Read error: Connection reset by peer]
[04:12:17] -!- miss0r has joined #linuxcnc
[04:34:09] <miss0r> g'day
[04:35:39] <miss0r> Do anyone in here have experience with the procedure of changing out the language in the controller on an Okuma lh55-n lathe?
[04:42:53] -!- RyanS has joined #linuxcnc
[05:28:43] <RyanS> made a ball turner https://drive.google.com
[05:34:40] -!- asdfasd1 has joined #linuxcnc
[05:35:05] <miss0r> Nice. Isn't the orientation on the cutter a bit off?
[05:36:16] <miss0r> also, it seems you have found something to use the chinesium boring head for :D I have the same shitty bang bang model you do. It took me a while to turn it into something useable
[05:36:22] -!- asdfasd has quit [Ping timeout: 264 seconds]
[05:37:30] <RyanS> yeah the toolbit cuts like shit
[05:38:02] <RyanS> https://drive.google.com
[05:39:38] <XXCoder> I keep gettinmg flipping house rentals ads
[05:39:47] <miss0r> The toolbits that come with it are a joke. I mean - you 'can' get them to work, but you have to grind the sh*t out of them first
[05:39:49] <XXCoder> its getting REAL annoying
[05:40:36] <RyanS> last time I used it the bore was oversized, not sure if the boring head slipped
[05:41:15] <miss0r> RyanS: You should always use a dial indicator when adjusting a boring head. Especially when it comes to the chinesium ones
[05:42:38] <miss0r> XXCoder: Are you on the market for renting a house? :P
[05:43:10] <miss0r> In that case i might have something that is of intrest to you ! (using add voice)
[05:44:11] <XXCoder> no!
[05:44:19] <XXCoder> I own part of house lol
[05:44:28] <XXCoder> which means...
[05:44:36] <XXCoder> somehow they predict doom for my house.
[05:44:52] <miss0r> Trust the algorithm!
[05:45:33] <XXCoder> lol
[05:45:35] -!- RyanS has quit [Ping timeout: 276 seconds]
[05:46:18] -!- RyanS has joined #linuxcnc
[05:46:57] <miss0r> I need to get going on installing some cable here. I have to set up a three phase outlet for the EDM. (I HATE EXTENTION CORDS LIKE NOTHING ELSE)
[05:48:47] <RyanS> did I miss something exciting?
[05:48:58] <XXCoder> yeah]
[05:49:03] <XXCoder> so much drama
[05:49:18] <XXCoder> 400 lines in least. surpising stopped right befor eyou entered
[05:49:29] -!- ty2u has quit [Quit: It is our time to be alive]
[05:49:51] <miss0r> hehe
[05:50:01] <miss0r> Last thing I wrote you, RyanS:
[05:50:04] <miss0r> RyanS: You should always use a dial indicator when adjusting a boring head. Especially when it comes to the chinesium ones
[05:50:21] -!- ty2u has joined #linuxcnc
[05:50:33] <RyanS> damn, can't keep my eyes off the screen for a second
[05:50:34] <miss0r> also: fsck! I'm out of the appropriate installation cable
[05:51:26] <RyanS> there's this type of bit https://www.haythornthwaite.com
[05:52:23] <miss0r> now THAT looks like something that would work well
[05:53:47] <miss0r> The stresses are all the right places, if you use it that way
[05:53:54] -!- selroc has joined #linuxcnc
[05:54:01] <RyanS> or use the unused edges of my CCMTs http://thebloughs.net
[05:54:32] <miss0r> sure
[05:55:40] <miss0r> What have you used for bearings in there?
[05:55:55] <RyanS> I wonder if there's any other tools that use the other edge of ccmt
[05:56:25] <RyanS> igus flange bushings
[05:57:57] <miss0r> Alright :]
[05:58:09] <miss0r> Excellent choice, might I add
[05:58:30] <miss0r> so I suppose your boring head came with a straight shank, and not a taper?
[05:59:13] -!- gloops has joined #linuxcnc
[06:00:04] <RyanS> made the shank, I literally can't believe the thread fitted well
[06:00:31] <miss0r> :D
[06:01:35] <RyanS> igus flange bushings meant to be good for low rpm, high, oscillating loads
[06:02:15] <RyanS> not sure if it's vibration damping
[06:04:54] <miss0r> its not
[06:05:04] <miss0r> But they are very tight in tolerences
[06:08:57] <gloops> https://www.ebay.co.uk lot of machine there for that money
[06:09:15] <RyanS> urgh, stuff the ccmt tool holder.. ooo would a carbide bit work as an EDM die?
[06:10:43] <miss0r> gloops: yeah, but it looks like it needs a lick of paint
[06:11:14] <gloops> nice little..or big project for somebody, those ward lathes were really good
[06:13:49] <XXCoder> lot of rust also
[06:14:41] <miss0r> but that rust looks somewhat superficial
[06:14:57] <miss0r> My ventilation equipment just got here. 100mm system to be installed here in the shop
[06:15:05] <jesseg> lol https://videoflier.com
[06:15:25] <gloops> probably would have been lathered in oil and grease when it got put outside, bit of the luck the important surfaces not corroded
[06:15:42] <gloops> looks like stainless ways anyway
[06:16:32] <miss0r> Those are just waycovers
[06:16:42] <miss0r> Its regular'ol cast iron underneath
[06:16:46] <gloops> yeah they are on closer inspection
[06:17:53] * Loetmichel just got a delivery from pearl.de (german "gadget"/surplus seller)... aaand of course the microusb cable to proggram the fan was already defective. luckily one has a bunch of these around for the smartphones... i am just courious how long it will take for the meter stick and the 4 screwdrivers (one already in my pocket) to disappear from my desk... :-) -> http://www.cyrom.org
[06:17:53] * Loetmichel /main.php?g2_itemId=17137&g2_imageViewsIndex=1
[06:19:35] <miss0r> I give it 2 days :]
[06:19:44] <miss0r> By the end of the week they're gone ;)
[06:20:05] <RyanS> I need to think of some other leadscrew cover, the telescopic tube preventing the carriage getting close to the chuck
[06:22:02] <miss0r> god damnit! I've done it again. I get distracted while ordering materials. I've managed to order two air duct regulator valves and one shutter valve, but it should've been the other way around! >.<
[06:23:53] -!- selroc has quit [Quit: Leaving]
[06:26:10] <XXCoder> checklist twice, order once. ;)
[06:45:01] -!- diverdude has quit [Quit: Page closed]
[06:56:55] -!- diverdude has joined #linuxcnc
[06:58:18] <diverdude> hi...i am getting to a point where I have configured most of my CNC. However I still need to configure "Driver microstepping", "Pulley teeth" and "Leadscrew pitch" for all axes. There are no mention of these in the manual. How do I figure out the right numbers for these settings?
[07:04:11] -!- jthornton has joined #linuxcnc
[07:07:18] <XXCoder> diverdude: its because it depends on your machine
[07:07:42] <XXCoder> drivers itself has microstepping setting. say if its set to 4, linuxcnc microstepping must be set at 4
[07:08:23] <XXCoder> leadscrews one you simply put in pitch as found on your machine info
[07:08:40] <jthornton> not really, microsteps is only in the conf wizard linuxcnc only uses the end number
[07:09:02] <XXCoder> oops yeah lol
[07:17:29] <rmu> diverdude: stepper motor usually takes 200 "full steps" per revolution, each of these full steps can be divided into n microsteps, n can be arbitrary number, usually 2^something or 10. microsteps are configured at the stepper driver (if at all). leadscrew pitch: just look at it or turn it a few revolutions (machine off) and measure how far the axis moves. that is usually a round number like 3mm/rev.
[07:18:01] <XXCoder> chineseium its almost always 8mm/rev
[07:19:07] <XXCoder> forgot if its for leadscrew or ballscrews
[07:19:12] <rmu> diverdude: what you really want to configure is how many steps are needed per machine unit. 200 steps/rev mit 10 microsteps and 5mm pitch screw would be 200 * 10 steps per revolution, that moves axis 5mm, ergo 400 steps per mm.
[07:21:40] <gloops> simple way to check screw pitch - turn the screw 1 rotation, measure the travel
[07:22:08] <gloops> if it has pulleys - get counting
[07:23:44] <XXCoder> my gearing system, i copied from this https://thumbs.dreamstime.com
[07:26:08] -!- Lcvette has joined #linuxcnc
[07:26:50] <diverdude> rmu: right...i know this machine has 400 motor steps
[07:27:01] <diverdude> rmu: but the microsteps is not mentioned anywhere
[07:27:08] <gloops> you dont actually need to go through stepconfig every time to adjust this
[07:27:35] <gloops> in fact, running stepconfig after setup is a habit to get out of
[07:27:42] <jthornton> yup
[07:27:42] <diverdude> gloops: i know....but so far i have kind of ignored that part to just get something running....but now i need to make sure that it actually moves correctly
[07:28:00] <gloops> ok, have you found the INI file yet?
[07:28:11] <diverdude> gloops: yes i know where the ini file is
[07:28:20] <gloops> open that
[07:28:29] <jthornton> diverdude: what stepper drives do you have?
[07:29:01] <diverdude> jthornton: this one https://www.china-cncrouter.com
[07:29:48] <diverdude> gloops: ok, and then?
[07:29:50] <gloops> what is SCALE set at in the INI file
[07:30:22] <rmu> diverdude: if unsure, measure. put in SCALE of 100 or so, make a move of 100mm and measure how far it really moves, adjust SCALE accordingly.
[07:30:34] <diverdude> gloops: 160
[07:30:39] <gloops> SCALE = steps per unit (if your config is mm thats the number of steps per mm)
[07:31:11] <gloops> 160 ok, how is the machine behaving - is it moving twice as far as it should, half as far etc
[07:32:53] <gloops> your scale needs to be 80
[07:33:07] <jthornton> diverdude: page 17 gives you the scale number of 400
[07:33:32] <gloops> lol, or 400
[07:34:53] <gloops> 160 scale @400 steps would mean screw pitch of 2.5mm, which is unlikely
[07:34:58] <rmu> page 17 is the page with page number -16- ;-)
[07:35:10] <gloops> far more likely to be 5mm, - 80 SCALE
[07:35:47] <rmu> according to that page, SCALE should be 400
[07:36:15] <gloops> 1mm screws?
[07:37:16] <rmu> 200 steps per rev, 16 microsteps, 8mm/rev would give 400 per mm
[07:37:50] <gloops> well im going by diverdude saying the machine uses 400 steps
[07:38:29] <rmu> yes, 400 steps per mm ;)
[07:39:13] <gloops> 400 steps per rotation, 400 scale = 1mm per rotation
[07:39:40] <rmu> whatever, it is easily measured
[07:39:47] <gloops> yes its the scale thats 400, not the steps diverdude
[07:40:59] <gloops> remember if you change anything in the ini or hal you need to save it and restart linuxcnc
[07:42:44] <rmu> or use emccalib
[07:45:50] -!- Beachbumpete1 has joined #linuxcnc
[07:46:44] <Beachbumpete1> morning
[07:47:15] <Beachbumpete1> I managed to lunch a part last night ;)
[07:47:49] <gloops> what does that mean Beachbumpete1 ?
[07:47:51] <Beachbumpete1> all because I had edited the program and forgot to delete the first one that had an issue
[07:48:13] <gloops> ahh
[07:48:18] <Beachbumpete1> so It had been a long time since I had machined that particular part
[07:48:44] <Beachbumpete1> and when I went to open the program I could not remember which one was the new one
[07:48:48] <Beachbumpete1> I chose wrong LOL
[07:49:07] <Beachbumpete1> worst part was the material was my last piece of that size
[07:49:32] <Beachbumpete1> so I had to cut down a larger piece of stock to the correct size so I could machine the part using the CORRECT program
[07:50:06] <Beachbumpete1> now I am down two pieces of stock so gotta order more today.
[07:50:49] <Beachbumpete1> really annoyed with myself because I really try to manage my programs and put notes in the body so I can read them when I have not run the program in awhile to help me remeber the details
[07:50:54] <gloops> i end up with files like 'part-latest' 'part-latest improved' 'part-very latest' lol
[07:51:02] <diverdude> gloops: ack..the phone rang
[07:51:05] <Beachbumpete1> LOL I do the same thing
[07:51:20] <Beachbumpete1> mine is usually Final, latest final, etc. LOL
[07:51:26] <diverdude> gloops: should i set SCALE = 400?
[07:51:51] <gloops> diverdude have you taken any measurememts of travel at all yet?
[07:51:51] <rmu> there is a thing called "last modified" on a file
[07:52:04] <Beachbumpete1> I do need to go back thru the files that are on the machine and delete the bullshit ones
[07:52:05] <diverdude> gloops: no
[07:52:06] <jthornton> I date my PLC files when I modify them like clip-inserter-01-12-18.
[07:52:14] <gloops> rmu yes but in the blur of activity nobody checks stuff like that
[07:52:18] <diverdude> gloops: well i have actually
[07:52:42] <Beachbumpete1> I need to come up with a better means of tracking this stuff.
[07:52:43] <diverdude> gloops: i checked a small travel on Z axis with a ruler to see if it was correct and it was off
[07:52:43] <gloops> a very simple way - set the jog increment to 5mm, jog - measure
[07:52:52] <gloops> should be 5mm thereabouts
[07:53:02] <Beachbumpete1> I typically do not like to delete the old ones in case I need a reference of something
[07:53:03] <diverdude> gloops: true.,...i will try that
[07:53:32] <Beachbumpete1> my CAD files are the worst because especially here they are ALWAYS making changes to drawings
[07:53:41] <Beachbumpete1> sometimes the changes get reverted
[07:53:59] <Beachbumpete1> so I cannot really delete the previous ones without causing myself a LOT more design work.
[07:54:03] <rmu> put everything into version control. storage is cheap nowadays.
[07:54:26] <Beachbumpete1> I mostly use dates on the CAD files tho
[07:54:37] <miss0r> There we go. Now I have some ventilation running, so I don't get suffocated in kerosine fumes.
[07:54:46] <diverdude> gloops: yeah when i move 5mm it only moves something like 3
[07:54:59] <Beachbumpete1> right now we have a customer that has a custom closet and a breakfast/media bar and there must be a hundred files just for those two builds
[07:55:39] <Beachbumpete1> add to that the .pdf files that contain the drawings in different elevations/views
[07:56:02] <Beachbumpete1> and then add to that the 3D photorealistic renderings of each one and it gets a bit crazy rather quickly.
[07:56:09] <diverdude> gloops: 2-5 to 3mm is what its moving when it should move 5
[07:56:10] <gloops> ok diverdude try 400 scale
[07:56:40] <diverdude> gloops: ok so i configure ini file and set SCALE = 400.0, save it and restart linuxcnc?
[07:56:53] <gloops> yep
[07:58:21] <diverdude> gloops: aha that looks more correct
[07:58:52] <Beachbumpete1> really need an accurate means of measuring travel along the axis to calibrate no?
[07:59:07] -!- BitEvil has joined #linuxcnc
[07:59:07] SpeedEvil is now known as Guest60705
[07:59:18] <gloops> <rmu> 200 steps per rev, 16 microsteps, 8mm/rev would give 400 per mm
[07:59:46] -!- Guest60705 has quit [Ping timeout: 264 seconds]
[08:00:50] <gloops> if you need to fine tune that just change the 400 SCALE by small amounts, see which way its going, measure etc
[08:01:43] <Lcvette> index
[08:01:43] <c-log> Lcvette: The #linuxcnc index http://tom-itx.no-ip.biz:81
[08:04:57] -!- BitEvil_ has joined #linuxcnc
[08:05:37] -!- BitEvil has quit [Ping timeout: 256 seconds]
[08:06:45] BitEvil_ is now known as SpeedEvil
[08:36:05] <rmu> if you have a travel of say 500mm and you can measure ("read the tape measure") to +-0,5mm and ignore inconsistencies in the screw/ways/etc... you can calibrate SCALE accurate enough
[08:37:18] <rmu> usually that should be a "round" number ;-)
[08:40:22] <miss0r> remember the supreme skills where they drilled a pensil led? well, this one is also quite interresting: https://www.youtube.com
[09:03:11] -!- JT-Shop has joined #linuxcnc
[09:05:50] -!- syyl has joined #linuxcnc
[09:09:57] -!- P1ersson has quit [Ping timeout: 240 seconds]
[09:15:28] <Loetmichel> harhar. ONE day and boss notices about the new shelve: "there is no room for our 24" displays! Change that!"... lets see how it takes him to notice that he now can no longer put high things under it :-) -> http://www.cyrom.org
[09:25:04] <rmu> miss0r: this video needs a void de-exciter
[09:28:05] -!- P1ersson has joined #linuxcnc
[09:29:06] <JT-Shop> stupid stamps.com won't print Russian letters...
[09:30:17] <SpeedEvil> Make stamps.com great again.
[09:32:28] -!- ferdna has joined #linuxcnc
[09:37:29] -!- ziper has joined #linuxcnc
[09:37:58] -!- nick001 has quit []
[10:01:21] -!- IchGucksLive has joined #linuxcnc
[10:01:25] <IchGucksLive> hi all
[10:01:31] <IchGucksLive> gloops, firer
[10:01:50] <IchGucksLive> diverdude, WHY the Hell are you using microsteps
[10:01:59] <IchGucksLive> 400steps per mm
[10:02:13] <IchGucksLive> diverdude, this is fully inefficent
[10:02:34] <IchGucksLive> diverdude, reduce to SCALE 100
[10:02:47] <IchGucksLive> diverdude, so it is a 0,01mm per step
[10:04:37] <IchGucksLive> diverdude, if you got a 8mm pitch and direkt drive use 800steps
[10:04:55] <IchGucksLive> diverdude, so 4microsteps = SCALE 100
[10:05:57] -!- Roguish has joined #linuxcnc
[10:06:07] <IchGucksLive> diverdude, 200Steps/stepper Times 4 microsteps = 800Steps/rev DIV by 8mm ==== SCALE 100 -> 100steps/mm = 0,01 Precicion
[10:06:17] <IchGucksLive> gloops, on ?
[10:06:43] <gloops> hi
[10:06:52] <IchGucksLive> :(
[10:07:06] <gloops> whats up Ichs?
[10:07:11] <IchGucksLive> WHY are you suporting microsteps as you know how to do it right
[10:07:30] <IchGucksLive> a SCALE of 400 is realy bad
[10:08:34] <IchGucksLive> gloops, if he got microsteping so th hardware can pull better then use it
[10:09:07] <IchGucksLive> so 8mm pitch on standard stepper direct move is a solution 800 meaning 4microsteps
[10:09:16] <gloops> dont know what drivers are with that machine - can he change the step settings?
[10:09:39] <IchGucksLive> the only thing is G540 that cand be changed
[10:09:49] <IchGucksLive> all otheres can
[10:10:58] <IchGucksLive> gloops, 'part-latest' 'part-latest improved' 'part-very latest' THAT is not your belive i hope
[10:11:12] <IchGucksLive> part_10_V1
[10:11:25] <IchGucksLive> part_10_motorClamp_V1
[10:11:31] <gloops> yes a good system for file organisation
[10:12:04] <IchGucksLive> part_11_motorDistance_V3
[10:12:30] <IchGucksLive> part_11_motorAxisDistance_67mm
[10:12:43] <IchGucksLive> this are partnames as the drawing graps this
[10:12:58] <IchGucksLive> with the projekt name or number it fils together
[10:13:05] <IchGucksLive> to a part list
[10:14:31] <IchGucksLive> gloops, tomorrow trhe summer is calle off in the UK Hazel saying
[10:15:49] <IchGucksLive> im off 2 till later
[10:15:55] -!- IchGucksLive has quit [Quit: Leaving]
[10:30:29] <jthornton> interesting
[10:40:22] <gloops> Wetherspoon running out of John Smith's beer and Strongbow cider as carbon dioxide shortage hits
[10:42:33] <gloops> this is the longest unbroken hot spell here i can ever remember
[10:43:02] <gloops> we had the beast from the east freeze, then straight into record heat - and still in it
[10:43:23] -!- veek has joined #linuxcnc
[10:43:32] <gregcnc> the end is nigh?
[10:43:57] <gloops> probably just coincidence
[10:52:20] -!- veek has quit [Quit: Leaving]
[10:57:54] -!- pingufan has joined #linuxcnc
[10:59:46] -!- gloops has quit [Ping timeout: 264 seconds]
[11:12:26] -!- h_maximilian has joined #linuxcnc
[11:17:02] -!- h_maximilian has quit [Ping timeout: 265 seconds]
[11:28:44] <JT-Shop> amazing he declared that all drives are configurable for micro steps except one... which is just not true!
[11:31:42] <rmu> since when are microsteps bad.
[11:35:45] <rmu> JT-Shop: i think he meant that except for the gecko G540, you can deactivate microsteps on all drives. which is also not true.
[11:36:05] -!- FAalbers has joined #linuxcnc
[11:36:44] <JT-Shop> I thought that was what I said
[11:37:15] <Rab> rmu, since they reduce your torque, unless I'm missing something.
[11:38:05] <rmu> Rab: that's a misconception
[11:38:32] <rmu> torque from one full step position to the next full step position is not reduced
[11:40:25] -!- ferdna has quit [Quit: Leaving]
[11:43:17] <gregcnc> of course it
[11:43:19] <gregcnc> is
[11:43:43] <gregcnc> doh, not full step
[11:43:52] -!- h_maximilian has joined #linuxcnc
[11:44:17] <gregcnc> but what good is a programmed position if the torque just isn't there to hold it?
[11:45:08] <Rab> I guess it's a form of dithering with the potential to non-deterministically increase resolution.
[11:45:42] <Rab> But I am not sure that benefits all jobs.
[11:45:56] <rmu> motor runs smoother
[11:46:04] <rmu> electrical noise is reduced
[11:46:16] <rmu> vibration problems reduced
[11:46:17] <rmu> etc
[11:46:26] <Rab> Those are benefits.
[11:46:56] <gregcnc> the illusion of accuracy is what most people get from it
[11:47:59] <rmu> a stepper should be able to hold at least 1000 different mechanical positions with microstepping
[11:48:06] <JT-Shop> my G203v drives morph from 10 micro steps to full steps and the morphing zone is adjustable to get super smooth stepper at all rpm ranges
[11:48:45] <gregcnc> it really depends on how you're using it in the machine
[11:49:58] <rmu> of course, but anything that assumes the stepper can resolve more than about 0.1° is an illusion
[11:50:12] -!- gloops has joined #linuxcnc
[11:50:26] <gloops> Germany are OUT
[11:50:37] <rmu> gloops: 5min to go
[11:50:46] <gloops> Korea just scored
[11:50:54] <gloops> ahh no, off side
[11:51:28] <gloops> ref checking video
[11:51:36] <rmu> hahahaha...
[11:51:47] <rmu> gloops: https://twitter.com
[11:52:17] <gloops> KOREA 1 GERMANY 0
[11:52:49] <rmu> f*ing lag
[11:53:01] <gloops> cant read german rmu
[11:53:39] <gloops> first time since 1938 they exit at the group stage - slightly ominous
[11:53:46] <rmu> gloops: the table has all scenarios from 0:0 to 3:3 in both games
[11:55:05] <gloops> its over now Korea 2 Germany 0
[11:55:12] <rmu> wtf
[11:55:48] <gloops> caught on the break, keeper was attacking
[11:55:50] <rmu> i'm a minute behind it seems... so SWE and KOR
[11:58:28] <gloops> thats it
[11:59:20] -!- P1ersson has quit [Ping timeout: 256 seconds]
[11:59:42] <rmu> hihihi seems germany is on the market for a new head coach
[12:00:52] -!- P1ersson has joined #linuxcnc
[12:01:56] <gloops> finished bottom, bit of a shocker
[12:05:58] -!- P1ersson has quit [Ping timeout: 260 seconds]
[12:06:12] <jesseg> Loetmichel, that usb readout fan is bloody charming
[12:08:43] <Loetmichel> jesseg: its funny. and it cools the brain
[12:09:05] <Loetmichel> company has no A/C, only for the measurement chambers ;)
[12:09:17] <jesseg> Does it work pretty good, stable display of text, pretty bright, etc?
[12:10:18] <Loetmichel> all of that, slight jitter when crossing what i imagine is the "zero degree point" with marquee text.
[12:10:35] <Loetmichel> but its not very well balanced
[12:10:42] -!- unterhausen has joined #linuxcnc
[12:10:45] <jesseg> lol what was it - pretty cheap?
[12:10:47] <Loetmichel> it shudders on its gooseneck quite a bit
[12:10:54] <Loetmichel> 9.90 eur ;)
[12:11:14] <jesseg> ahh OK
[12:12:17] <Loetmichel> https://www.pearl.de
[12:12:21] <Loetmichel> REALLY cheap ;)
[12:16:57] -!- dan2wik has quit [Ping timeout: 240 seconds]
[12:18:51] -!- dan2wik has joined #linuxcnc
[12:18:51] -!- dan2wik has quit [Changing host]
[12:18:51] -!- dan2wik has joined #linuxcnc
[12:33:47] -!- P1ersson has joined #linuxcnc
[12:51:36] -!- IchGucksLive has joined #linuxcnc
[12:51:40] <IchGucksLive> hi all
[12:52:04] <IchGucksLive> seams germany in deep shock
[12:52:27] <IchGucksLive> noone here at lakeside completly empty while shoud be crouwded
[12:52:52] <gloops> bit of a shock that Ichs
[12:54:25] -!- P1ersson has quit [Ping timeout: 245 seconds]
[12:54:49] <IchGucksLive> gloops, it is as it is
[12:54:57] <IchGucksLive> diverdude, online ?
[13:03:28] <gloops> as someone here just said - germany always mess up in russia lol
[13:11:01] <skunkworks> Jeeze - the lengths you go through when your choice of cnc control sucks..
[13:12:03] <skunkworks> and more https://www.youtube.com
[13:12:57] -!- h_maximilian has quit [Quit: Leaving.]
[13:13:30] -!- h_maximilian has joined #linuxcnc
[13:13:57] <IchGucksLive> gloops, today the cables ariffed and the dragchains for the new cnc serie
[13:14:08] <IchGucksLive> lots of work todo
[13:14:08] <skunkworks> (mach4 + Ethernet smooth stepper (ESS) doesn't do threading - yet)
[13:14:32] <IchGucksLive> it will at mach5+
[13:15:01] <IchGucksLive> you can do this with 5A china bob and a simple 4hole encoder
[13:15:07] <IchGucksLive> on lcnc
[13:15:09] <gloops> what are you building Ichs>
[13:15:20] <IchGucksLive> same as always
[13:15:29] -!- Lcvette has quit [Ping timeout: 276 seconds]
[13:17:33] -!- hendrik_cnc has quit [Quit: Connection closed for inactivity]
[13:17:55] <IchGucksLive> gloops, army now in to fire
[13:18:22] <gloops> yeah, difficult to stop on the moors, its all dry heather
[13:18:45] <gloops> also an earthquake recorded in Surrey today
[13:18:48] <IchGucksLive> and that will smell for some years
[13:19:12] <gloops> it will clear out all the dead stuff
[13:19:17] <IchGucksLive> gloops, firetruck to bumphole detected
[13:20:01] <IchGucksLive> fire is natural all over the world but we lost a eye on that
[13:20:44] <gloops> children playing with matches probably
[13:21:27] <IchGucksLive> more older expirienced i think
[13:22:08] <IchGucksLive> gloops, i did take a look at the forcast here and they saying heateave untill wendsday
[13:22:30] <gloops> supposed to be cooler here at weekend, desperate for rain
[13:22:54] <IchGucksLive> i got a lake and running water bye the house
[13:23:45] <gloops> have got water supply but no hosepipes, have to carry it
[13:23:48] <gloops> for garden
[13:24:04] <IchGucksLive> im off till tommorrow Gn8
[13:24:11] -!- IchGucksLive has quit [Quit: Leaving]
[13:25:32] <roycroft> that's a rather impressive wildfire for the uk
[13:25:38] <roycroft> it rivals some of our wildfires
[13:26:00] <roycroft> i hope you get the rain that is needed to put it out soon
[13:26:29] <gloops> its doing that ground a bit of good roycroft
[13:27:25] <gloops> i was on a flat roof earlier, the tar was molten, 33C
[13:32:18] <Deejay> sticky under the shoes ;)
[13:32:35] <mozmck> 33C is not bad.
[13:32:39] <gloops> yes, cant put hand on anything, metal is too hot
[13:33:10] -!- h_maximilian has quit [Quit: Leaving.]
[13:41:33] -!- Motioncontrol has joined #linuxcnc
[13:42:01] <roycroft> we have a fire that's almost 100,000 acres already
[13:42:08] <roycroft> fire season started really early this year
[13:42:17] <roycroft> and it's expected to be an especially bad season
[13:42:38] <roycroft> which is pretty much how every fire season for the past 15 years or so has been
[14:01:47] <miss0r> forrest can hardly regenerate at the rate its burning at the moment
[14:02:06] <roycroft> not in the short term
[14:02:14] <roycroft> but it will recover over time
[14:02:28] <roycroft> the uk have a much longer history of forest management than we do here in the usa
[14:02:37] <roycroft> and i don't know how forests are managed there
[14:02:53] <roycroft> but we spent much of the 20th century here focusing on fire suppression
[14:03:24] <roycroft> resulting in a huge buildup of fuel in our forests, which caused some very spectacular fires
[14:03:42] <roycroft> we've shifted in the last couple or three decades to containment, not suppression
[14:04:00] <roycroft> and that's both more cost effective and better for overall forest health
[14:04:42] <roycroft> we still do suppression around developed areas, of course
[14:05:56] <mozmck> Selective logging also helps.
[14:06:51] <jesseg> I just won this by bidding at t=-1s https://www.ebay.com
[14:07:03] <roycroft> yes
[14:07:08] <roycroft> clearcutting, however, does not
[14:07:14] <jesseg> oh sorry wrong link
[14:07:38] <mozmck> Well, It's bound to at first - nothing to burn!
[14:07:46] <jesseg> this is the one I got: https://www.ebay.com
[14:07:52] <roycroft> and then all the erosion happens
[14:07:57] <gloops> the fire here is only heather, it will grow back, they burn it anyway in strip rotation, to renew it
[14:07:59] <roycroft> and we get massive mudslides and floods
[14:08:11] <gloops> i dont know why theyre fussing over it being on fire
[14:08:13] <roycroft> and then topsoil is all gone and nothing can grow for decaces
[14:08:24] <roycroft> because it's huge, gloops
[14:08:28] <mozmck> Yeah, I'm not a fan of clearcutting everything.
[14:08:29] <roycroft> rule of thumb:
[14:08:35] <roycroft> if you can see the fire from space, it's a big fire
[14:09:03] <roycroft> and that fire in the moors can be seen from space
[14:09:17] <gloops> a lot of small mammals will be dying though
[14:12:44] <gloops> the looks a grand old tool jesseg
[14:12:52] <gloops> rivet gun?
[14:13:31] <Wolf__> hole punch
[14:15:51] <Wolf__> https://www.ebay.com what they look like made of shiny chinesium
[14:16:41] <roycroft> i could use a metric sheet metal punch
[14:16:50] <gloops> probably some slack in that
[14:16:53] <roycroft> i have an american customary punch that is similar to that
[14:16:55] <gloops> wont last 5 minutes
[14:16:58] <roycroft> i don't use it a lot
[14:17:02] <roycroft> but it seems to work just fine
[14:17:24] <roycroft> the coating on the handles is the exact same color as that metric one
[14:17:33] <roycroft> and the case is the same size, shape, and color
[14:17:42] <roycroft> so they're probably identical, except for the dies
[14:18:25] <roycroft> oh, actually that one claims to be american customary
[14:18:43] <roycroft> i saw the metric markings on the case - those are probably metric "equivalents" (read: approximations)
[14:20:27] <jesseg> gloops, yeah my dad has an old whitney Jr no 5 sort of like that one but older.. I'm always borrowing it but it's a pain because I have to go to his place to get it so it's not handy, so I figured I'd better just man up and get my own :P
[14:24:40] -!- Lcvette has joined #linuxcnc
[14:29:01] -!- pcw_mesa has quit [Read error: Connection reset by peer]
[14:32:22] -!- pcw_mesa has joined #linuxcnc
[14:34:52] -!- ve7it has joined #linuxcnc
[14:54:33] -!- theCockerel has quit [Remote host closed the connection]
[15:01:04] -!- Motioncontrol has quit [Quit: Sto andando via]
[15:05:33] -!- ferdna has joined #linuxcnc
[15:20:55] -!- CaptHindsight has joined #linuxcnc
[15:22:27] <CaptHindsight> http://money.cnn.com
[15:24:04] <gloops> and that is some beer swilling
[15:24:06] <CaptHindsight> how truthy is this story ^^
[15:24:27] <gloops> seems true its all over the media
[15:25:19] <CaptHindsight> might be a bump in nitrogen prices next
[15:25:54] <gloops> good world cup results + heatwave
[15:27:00] <CaptHindsight> maybe you can get some from the Germans
[15:27:47] <CaptHindsight> after they finish drowning their sorrows
[15:27:50] <Tom_L> they'll just sell more beer cause everybody will panic and rush the stores
[15:27:50] -!- jerryq has joined #linuxcnc
[15:28:27] <CaptHindsight> when you look into the backstory the blame seems to be an artificial shortage caused by CO2 makers
[15:28:49] <Tom_L> like the 70's gas shortage?
[15:28:55] <CaptHindsight> yup
[15:29:28] <CaptHindsight> still plenty of oil, they just lower production and call it a shortage
[15:29:38] <CaptHindsight> nice trick
[15:29:53] <gloops> its the girls that do it - https://www.youtube.com
[15:30:05] <CaptHindsight> "Several major ammonia plants in Europe have closed for maintenance, leading to a shortage of carbon dioxide. "
[15:30:32] <gloops> yes a likely story CaptHindsight
[15:30:46] <gloops> i can see its started already with the EU - cutting the beer off
[15:31:55] <Tom_L> damn, i wouldnt' try to keep up with her
[15:32:26] <Tom_L> wonder what else she swallows
[15:32:39] <gloops> could be an expensive night
[15:34:47] -!- h_maximilian has joined #linuxcnc
[15:34:49] -!- h_maximilian has quit [Client Quit]
[15:37:03] <gloops> 5 in 50 seconds https://www.youtube.com
[15:37:11] <gloops> had to pause though...
[15:38:31] <Lcvette> any of you using touch problem and screens?
[15:39:48] <CaptHindsight> touch problem? problems with touch screens?
[15:40:10] <CaptHindsight> 1-point or with multitouch?
[15:45:20] <Lcvette> probe
[15:45:29] <Lcvette> sorry, auto correct
[15:47:32] <Lcvette> trying to make Linuxcnc similar in we as what I became accustomed to, or similar
[15:47:55] <Lcvette> using a problem for work offset setting
[15:48:02] <Lcvette> probe
[15:49:31] <Lcvette> my old system used a master tool method, where once the probe zeroed the work piece origin, it was ready to cut
[15:49:53] -!- phiscribe_ has quit [Ping timeout: 248 seconds]
[15:50:59] <cradek> I do tool lengths that way on my vmc
[15:51:11] <cradek> the tool table is the difference in length between the probe and that tool
[15:51:31] -!- phiscribe has joined #linuxcnc
[15:51:47] <Lcvette> yeah?
[15:52:00] -!- syyl_ has joined #linuxcnc
[15:52:25] <Lcvette> so do you have an actual physical tool length and a master too offer length?
[15:52:31] <Lcvette> offset
[15:52:54] <Lcvette> or you use the probe as your gauge length
[15:53:28] <Simonious_> .07
[15:53:30] <Simonious_> 4.07
[15:53:34] <cradek> I measure tool length to a fixture (actually a 123 block) on the table
[15:53:39] <Simonious_> also.. kicad doesn't like to let go of the mouse
[15:54:05] <cradek> use the probe to set the origin of a coordinate system (g59.3) to Z=0 when it touches the 123 block
[15:54:10] <Simonious_> and when it does.. it only half does much of the time, resulting in my project scrolling off the side of the screen while I am in a different window.
[15:54:21] <cradek> then use touchy's setting "tool touch off to fixture"
[15:54:25] -!- syyl has quit [Ping timeout: 245 seconds]
[15:54:25] <Simonious_> so many gotchas :/
[15:54:34] <Simonious_> at least it beats Eagle, right? :P
[15:55:19] <cradek> who cares what the numbers are, as long as the tools and probe match
[15:55:41] <Lcvette> right
[15:56:48] <Lcvette> but if the tlos are set to the probe and you ever change a tip.. have to remeasure you're tool crib
[15:57:30] <cradek> the "tool touch off to fixture" just runs G10 L11 http://linuxcnc.org
[15:57:50] <cradek> sure, I suppose that's true, but I don't ever break my probe :-)
[15:58:40] <Lcvette> I rarely do either
[15:58:58] <Lcvette> but it happens every once in a while
[15:59:29] <Lcvette> or it needs a recalibration
[15:59:47] <Lcvette> if it does, that's a big delay in work
[16:00:02] <cradek> with a good fast tool changer, touching off all the tools just takes a couple minutes
[16:00:35] -!- Beachbumpete1 has quit [Quit: Relax, its only ONES and ZEROS!]
[16:00:39] <cradek> I go down near the 123 block and then turn the wheel upward until a dowel pin rolls under the tool
[16:00:43] <cradek> then poke the button on the screen
[16:00:48] <cradek> takes 5 seconds
[16:02:35] <Lcvette> the way I used to do it was very good and simple and very easy to add tools or perform maintenance
[16:02:46] <Lcvette> I wonder if it could be implemented
[16:03:16] <cradek> once you read and understand the docs for G10L10 and G10L11 you'll be able to do it whatever way you want
[16:04:06] -!- fragalot has joined #linuxcnc
[16:04:51] <fragalot> miss0rhi:
[16:04:56] <fragalot> ._. hi*
[16:05:09] emc_ is now known as JT-Shop-2
[16:05:29] <JT-Shop-2> I use the same method cradek taught me years ago with the dowel
[16:05:58] <cradek> I learned it from stuart
[16:06:01] <cradek> wonder where he got it
[16:06:18] <fragalot> from a finnish cookbook that came out in the 1960's
[16:07:29] <JT-Shop> 21 second video showing tool touch off https://www.youtube.com
[16:08:05] <fragalot> interesting
[16:08:13] <fragalot> don't you damage the cutting edge doing that on endmills?
[16:08:27] <cradek> that's why you go up
[16:08:48] <fragalot> might still microscopically roll the edge
[16:08:49] <Lcvette> I prefer offline measuring
[16:09:09] <fragalot> Lcvette: assuming that is possible and your toolholder system is repeatable
[16:10:11] <cradek> because the dowel is round, as you get closer to it passing under, it becomes much more sensitive
[16:10:44] <cradek> JT-Shop: the next video is 3m55s with some joker screwing around with a piece of paper and editing numbers on a screen
[16:11:14] <cradek> I'm sure he's a pleasant and good-looking guy but he should watch your video :-)
[16:11:44] <Lcvette> you mean for like an er spindle?
[16:13:42] <fragalot> ER or R8 etc
[16:13:46] <Lcvette> yeah I had not thought about it until now but I guess the message boards would take a simple touch plate type touch of
[16:14:08] <Lcvette> R8 just use TTS
[16:14:30] <fragalot> if your mill is rigid enough to deal with the extra stickout
[16:16:47] <Lcvette> you have R8?
[16:16:53] <fragalot> iso30
[16:17:20] <Lcvette> that should be repeatable
[16:17:59] <fragalot> i'm not saying mine isn't, i'm just saying that not everyone has a true repeatble system, like ER spindles that you mentioned
[16:18:21] <fragalot> also - iso30 is ridiculously expensive compared to iso40 :P
[16:19:26] <Lcvette> for those they sold have an auto touch off after every tool change
[16:19:35] <Lcvette> solves that problem
[16:19:51] <fragalot> yup
[16:19:58] <Lcvette> I am looking for a repeatable tool crib solution
[16:20:22] <fragalot> the dowel method is still nice to know though for use on manual machines
[16:20:23] <fragalot> imho
[16:20:37] <Lcvette> sure
[16:20:45] <Lcvette> good knowledge
[16:21:04] <fragalot> what's wrong with offline measure / electronic touch-off for tool cribs?
[16:22:09] <Lcvette> just trying to figure out how to use the probe setup
[16:22:26] <Lcvette> I'm only used to what I know from machstdmill
[16:22:46] <Lcvette> which is the master tool method
[16:24:06] <Lcvette> but I definitely would want to have to remeasure or have to remember to remeasure 140 tools if I broke a probe tip
[16:24:21] <Lcvette> even if it only happened once a year
[16:24:32] <Lcvette> would not be a good day
[16:24:41] <fragalot> does your probe not have it's own machine offset?
[16:24:50] <fragalot> which you could re-measure off of a known reference surface
[16:25:03] <Lcvette> t that's what I'm unsure of in linuxcnc
[16:30:10] <Lcvette> it looks like maybe g43.2 could be used to create a master too mode in the past processor maybe
[16:32:02] -!- andypugh has joined #linuxcnc
[16:32:32] <cradek> of course you could use a tool offset with your probe
[16:32:42] <Lcvette> http://linuxcnc.org
[16:32:44] <cradek> I just prefer not to
[16:32:57] <cradek> something should be the zero offset
[16:33:23] <Lcvette> generally that is the spindle gauge line
[16:33:37] <Lcvette> a non moving target
[16:33:39] <cradek> that's fine, but you can't touch it to your work
[16:34:01] -!- jthornton has quit [Read error: Connection reset by peer]
[16:34:13] <fragalot> that's why your probe with a known offset from the spindle face touches your work
[16:34:22] <Lcvette> correct
[16:34:25] -!- jthornton has joined #linuxcnc
[16:34:36] * cradek shrugs
[16:34:44] <cradek> you can do all these things, you just have to pick a way
[16:35:57] <Lcvette> that's the way I've been doing it, it works well because I can replace a tool or a probe tip and recalculate all length offsets without remeasuring everything
[16:36:24] -!- unterhausen has quit [Ping timeout: 260 seconds]
[16:36:36] <Lcvette> single point of failure won't bring down the house so to speak
[16:37:20] <Lcvette> because both the tools and probe are calculated from the non moving spindle gauge line
[16:38:22] -!- tachoknight has joined #linuxcnc
[16:38:23] <cradek> cool
[16:39:02] <Lcvette> I didn't invent it
[16:39:05] <Lcvette> lol
[16:40:21] <gregcnc> wouldn't you just find the new length offset for the probe and carry on?
[16:41:54] <andypugh> I tried screen-printing for the first time today.
[16:42:00] <Lcvette> if that how it works in linuxcnc probing
[16:42:15] <cradek> andypugh: that car is ridiculous and I want a CRT dash like that now
[16:42:40] <andypugh> You can have one for £30,000 (I did look)
[16:42:45] <cradek> I have a whole mess of 2" green crts
[16:42:58] <cradek> I could use one for each replacement gauge
[16:43:23] <andypugh> I thought that re-doing the scan generators in digital would be a good project, and would make things much better.
[16:43:28] <Rab> cradek, round or square?
[16:43:34] <cradek> they're square
[16:43:47] <Rab> neat
[16:44:06] <cradek> somewhere I have several of the tiny camcorder viewfinders too
[16:44:23] <cradek> they're super sharp
[16:45:10] <Rab> A friend of mine has a thing that's like an 8-channel rackmount signal monitor of some type, with eight little square CRTs. Just waiting for some awesome application.
[16:45:26] <cradek> sweet
[16:46:22] <andypugh> cradek: Sounds like a clock….
[16:46:51] <andypugh> This is what I was screen-printing: https://photos.app.goo.gl
[16:47:22] <hazzy-lab> andypugh: That screen printing looks awesome!
[16:47:39] <cradek> neato
[16:47:41] <Rab> andypugh, very nice!
[16:48:20] <andypugh> (That’s only a subset of the supported feedback types, by the way. I ran out of room)
[16:48:23] <Rab> andypugh, is this commercially available (or planned to be)?
[16:49:07] <andypugh> I am not sure. Not from me. It’s open-source hardware, though, so batches get ordered every now and again.
[16:50:35] <hazzy-lab> andypugh: how much for the screen printed case option??? :D
[16:51:15] <andypugh> Free option. Plain or printed.
[16:51:42] <andypugh> (But I might not send the best prints :-)
[16:53:06] -!- fragalot has quit [Quit: Lost terminal]
[16:53:45] <hazzy-lab> That is certainly a value added service!
[16:54:30] <hazzy-lab> BTW, are the cases waterjeted? They look great
[16:55:09] <andypugh> Lasered. deanforbes_ had them made. £3 each
[16:55:21] <roycroft> do you print while it's still flat, or after it's folded up?
[16:55:32] <andypugh> A nice feature of the hand-fold design is that they can be painted and printed flat, and shipped flat too.
[16:55:41] <roycroft> oh, that's hand-folded
[16:56:26] <andypugh> Yes, the little webs make it pretty easy. Neat idea.
[16:56:45] <roycroft> yeah
[17:00:43] <Deejay> gn8
[17:00:57] -!- Deejay has quit [Quit: bye]
[17:01:29] -!- FAalbers has quit [Quit: Leaving]
[17:10:57] <JT-Shop> andypugh: cool
[17:19:06] -!- syyl_ has quit [Quit: Leaving]
[17:44:23] <Lcvette> Andy that's awesome!
[17:48:01] <Lcvette> is that a servo drive?
[17:50:11] <andypugh> Yes, follow the URL on the top to get all the details :-)
[17:51:48] <Tom_L> nice print andy
[17:51:58] <Tom_L> what paint did you use?
[17:52:30] <Tom_L> i used UV curable when i was making my programmers
[17:52:37] <andypugh> Just the basic water-based ink that came with the kit from Amazon
[17:52:44] <Tom_L> nice
[17:53:15] <andypugh> I tried to wash a bad print off, it wouldn’t come off. Even when not fully cured.
[17:53:31] <Tom_L> http://tom-itx.no-ip.biz:81
[17:53:36] <Tom_L> there's a batch under fire
[17:53:57] <Tom_L> i had no kit though
[17:54:24] <andypugh> Just this: https://www.amazon.co.uk
[17:54:47] <andypugh> And I have _loads_ of photo-sensitive goop left.
[17:54:48] <Tom_L> http://tom-itx.no-ip.biz:81
[17:55:20] <Tom_L> http://tom-itx.no-ip.biz:81
[17:55:25] <Tom_L> i cnc'd a cutout for the lids
[17:55:48] <Tom_L> yeah you don't use much goop for sure
[17:56:11] -!- phiscribe has quit [Ping timeout: 276 seconds]
[17:56:43] <Tom_L> i got the screen from a local blueprint shop
[17:56:48] <Tom_L> made the frame
[17:57:15] <Tom_L> those drives look interesting
[17:57:33] -!- phiscribe has joined #linuxcnc
[17:59:04] <Tom_L> did you make the case?
[18:05:09] <andypugh> deanforbes_: Made the cases from a design by Crinq (from the STMBL project)
[18:10:21] -!- Tom_itx has joined #linuxcnc
[18:10:21] -!- Tom_itx has quit [Changing host]
[18:10:21] -!- Tom_itx has joined #linuxcnc
[18:19:07] -!- jerryq has quit [Ping timeout: 256 seconds]
[18:20:47] -!- phiscribe_ has joined #linuxcnc
[18:20:51] <Lcvette> are you selling these?
[18:21:33] <andypugh> I had a batch made, and pre-sold them before ordering.
[18:21:46] <Tom_L> cost?
[18:22:58] <andypugh> £88 each
[18:23:09] -!- phiscribe has quit [Ping timeout: 264 seconds]
[18:24:17] <Tom_L> that's not bad at all
[18:24:17] <andypugh> (In kit form)
[18:24:56] <Tom_L> doing any assembled?
[18:25:40] <Lcvette> kit form
[18:25:40] <andypugh> No. Though the kit already has all the surface mount components and connectors: http://bodgesoc.blogspot.com
[18:27:02] <Tom_L> those came from laen's board service... forget what he calls it
[18:27:09] <Tom_L> the purple is a dead giveaway
[18:27:39] <andypugh> No, actually from China. I just chose purple.
[18:27:45] <Tom_L> really..
[18:27:58] <Tom_L> i think he went with purple so he could track where his boards end up
[18:28:41] <andypugh> https://www.pcbastore.com
[18:29:37] <Tom_L> http://tom-itx.no-ip.biz:81
[18:29:48] <Tom_L> that was his first rendition of 'purple'
[18:29:58] <Tom_L> more brown than anything
[18:30:13] <Tom_L> he ended up with more what you've got
[18:31:47] <andypugh> I didn’t want to use China really. But £47 per board, assembled, including all the surface-mount parts and connectors…
[18:32:02] <Tom_L> hard to beat for prices
[18:32:30] -!- nick001 has joined #linuxcnc
[18:33:02] <Tom_itx> https://www.dorkbotpdx.org
[18:33:11] <Tom_L> i think that's his site
[18:33:20] -!- nick001 has quit [Client Quit]
[18:33:45] -!- justanotheruser has quit [Quit: WeeChat 2.0]
[18:34:07] -!- justanotheruser has joined #linuxcnc
[18:35:34] <Tom_itx> https://oshpark.com
[18:35:41] <Tom_L> no, that is.
[18:37:03] <Tom_L> pretty sure he uses a US Mfg
[18:38:01] -!- Tom_itx has quit [Quit: Quit]
[18:38:11] <andypugh> Yeah. OSHPark is no good to me. Shipping from the US is a killer compared to China.
[18:38:27] <Tom_L> yeah i'm sure it would be
[18:38:47] <Tom_L> another one doning the same thing is #hackvana (Mitch)
[18:38:52] <andypugh> Though it might have saved some bother with the regulator chip used on STMBL, it was (incorrectly) tagged as export-restricted.
[18:38:54] <Tom_L> he's an aussie
[18:40:10] <Tom_L> he was commuting back n forth to china back then, not sure if he still is
[18:41:03] -!- gloops has quit [Quit: Leaving]
[19:05:41] -!- jerryq has joined #linuxcnc
[19:14:03] -!- HSD has quit [Quit: Leaving]
[19:14:10] <Lcvette> how do those drives work Andy?
[19:14:33] <Lcvette> perform taker
[19:14:40] <Lcvette> rather*
[19:19:26] <JT-Shop-2> it's past midnight for andypugh...
[19:20:43] <andypugh> I am still here, though.
[19:21:05] <andypugh> Have a look at the showcase on the LinuxCNC front page
[19:21:14] <JT-Shop-2> you're a trooper for sure
[19:22:18] <JT-Shop-2> that's a beast
[19:22:59] <JT-Shop-2> time to turn my hat round to the chef side
[19:35:50] -!- nick001 has joined #linuxcnc
[20:02:42] -!- asdfasd1 has quit [Ping timeout: 256 seconds]
[20:03:08] <Lcvette> Andy that's pretty cool!
[20:07:56] <andypugh> Yes, I am quite impressed. I take no credit for any part of STMBL
[20:08:35] <andypugh> I just decided I wanted some and found that the only way to make that happen at the time was to have a batch made
[20:09:53] <andypugh> (I have done a bit of work on the LinuxCNC driver side. They integrate really neatly when hooked in to the Mesa Smart-Serial bus)
[20:15:43] <andypugh> This is what the HAL looks like to connect up an STMBL through smart-serial: https://pastebin.ubuntu.com
[20:18:07] -!- flyback has quit [Ping timeout: 256 seconds]
[20:18:58] -!- pfred1 has joined #linuxcnc
[20:33:35] -!- flyback has joined #linuxcnc
[20:39:29] <Lcvette> sweet! guessing they tune up nice?
[20:42:14] <andypugh> You can tune them using Servoterm (like their own version of Halscope, which is actually a Chrome app) but it isn’t normally necessary once you have programmed the drive with the motor parameters (including torque constants and inertias. There are ways to use the drive to measure those.)
[20:43:56] -!- flyback has quit [Remote host closed the connection]
[20:46:14] -!- flyback has joined #linuxcnc
[20:46:32] <Lcvette> gotcha
[20:46:57] <Lcvette> my next set of drives will be self tuning lol
[20:47:55] <Lcvette> I would love to be able to replace just my drives but every where I spoke to said their drives are motor specific
[20:50:39] <Lcvette> DMM said they could replace the encoders on my motors
[20:52:09] <pcw_mesa> There may still be a way to fix the tuning/encoder sim issues on your drives
[20:53:08] <pfred1> do you know the way?
[20:54:16] <Lcvette> pcw_mesa: I don't think there are any more settings in the drive to tweak
[20:54:45] <Lcvette> what did you have in mind
[20:54:48] <pcw_mesa> perhaps doing what is done with linear scales would work ( dual PID loops ) mainly stepgen control but with a low bandwidth encoder feedback (I only or small P large I)
[20:55:18] <Lcvette> yeah?
[20:55:40] <pcw_mesa> since the FF1 error fix is slow (limited by acceleration)
[20:56:49] <pcw_mesa> wont improve the jerk bumps (that just drive bandwidth limited) but should fix any feedback noise
[20:57:14] <pcw_mesa> bbl Dinner!
[20:57:24] <Lcvette> very interesting!!
[20:57:29] <Lcvette> ok enjoy
[20:57:46] <Lcvette> very interested and excited at this new idea!
[20:57:53] <Lcvette> new hope
[21:06:35] <Lcvette> pcw_mesa: I went back and was looking at the PID loop tune for the stepgen and it actually had the same jerk points as the encoder feedback loop
[21:08:15] <Lcvette> index
[21:08:15] <c-log> Lcvette: The #linuxcnc index http://tom-itx.no-ip.biz:81
[21:25:23] -!- veek has joined #linuxcnc
[21:30:57] -!- veek has quit [Ping timeout: 240 seconds]
[21:46:22] -!- ferdna has quit [Quit: Leaving]
[22:04:20] -!- HighInBC has quit [Ping timeout: 255 seconds]
[22:07:54] -!- pfred1 has quit [Quit: ..]
[22:08:15] -!- andypugh has quit [Quit: andypugh]
[22:09:42] -!- MrHindsight has joined #linuxcnc
[22:09:42] -!- MrHindsight has quit [Changing host]
[22:09:42] -!- MrHindsight has joined #linuxcnc
[22:13:53] -!- CaptHindsight has quit [Ping timeout: 265 seconds]
[22:17:35] -!- HighInBC has joined #linuxcnc
[22:25:20] <pcw_home> Lcvette: yes those are mostly determined by the drive bandwidth which will not change (unless you change the filtering)
[22:27:10] <Lcvette> I don't think I fully understand how those all for together yet
[22:27:28] <Lcvette> I know more bandwidth is good
[22:27:39] <roycroft> bandwidth is overrated
[22:27:55] <Lcvette> is that the 500kpps?
[22:28:09] <roycroft> patience is a virtue :)
[22:31:05] -!- Roguish has quit [Quit: ChatZilla 0.9.92-rdmsoft [XULRunner 35.0.1/20150122214805]]
[22:34:19] -!- Laminae has joined #linuxcnc
[22:39:08] <Laminae> Hey amigos i'm back from the dead and here to shamelessly ask for suggestions... i'm wanting to switch to extrusion on my cbeam machine XL and have been looking at what other have done and found that Misumi produces a really cool option for 15mmx250mmx1000mm which is nearly ideal for my purposes
[22:39:32] <Laminae> Here is an imagine of my quick and dirty concept: https://imgur.com
[22:40:00] <Laminae> And i'm looking at using these two parts: https://us.misumi-ec.com
[22:40:21] <Laminae> https://us.misumi-ec.com
[22:40:46] -!- Akex_ has quit [Ping timeout: 268 seconds]
[22:40:59] <Laminae> So my ultimate question... do you think the tolerances are too close to slide those u-extrusions over that?
[22:41:46] <Laminae> I decided to go ahead and shoot misumi the same question and here is the response i received: You: hello, Would the tolerance be too tight to fit a CASA2.5-2020-500 (U-extrusion) over the butt ends of HFSQN4-15250-900 (t-slot) i am intending to make a cnc router bed and was intending on tapping the ends and securing through the u-slot but was anxious about the stated tolerance variance. Bob: Hello, My name is Bob. How can I help yo
[22:42:20] <Laminae> You: Awesome, thank you Bob: I'm seeing that the A tolerance for the U channel is +/- 0.38 mm where the size for the flat extrusion is +/- 0.3 mm. There's a chance it would work, but you'd really be rolling the dice. There wouldn't be any way we could guarantee it
[22:42:31] -!- Akex_ has joined #linuxcnc
[22:46:02] -!- veek has joined #linuxcnc
[22:49:03] <pcw_home> Lcvette: no, it is the frequency response of the drive/motor/mechanics, usually in the hundred Hz range on heavy CNC machines
[22:49:10] -!- unterhausen has joined #linuxcnc
[22:53:15] <Lcvette> ah
[22:59:42] <pcw_home> it would also be less of an issue if LinuxCNC has a Jerk limited trajectory planner though for a lot of precision machining operations its not really an issue since you try to maintain a constant feed rate where there's no tangential acceleration
[23:00:28] <pcw_home> s/has/had/
[23:01:43] -!- ziper has quit [Read error: Connection reset by peer]
[23:32:32] <Lcvette> the drive has some features in it for acceleration
[23:32:43] <Lcvette> trapezoidal and s curve settings
[23:33:51] <Lcvette> would those be helpful?
[23:37:06] <pcw_home> No, those are really only useful if the drive does the trajectory planning
[23:37:06] <Kevin`> Laminae: I suspct a hammer could make it fit - one way or the other. none of that assembly is intended to be precise, right?
[23:37:30] <Lcvette> gotcha
[23:37:51] <Kevin`> Laminae: also, it's helpful to have access to the ends of the slots
[23:38:37] <pcw_home> I suspect you could reduce ringing from excessive jerk by using the drives jerk limiting but it would increase the peak error
[23:39:50] <Laminae> Heh, had to argue with that!
[23:40:07] <pcw_home> But as I mentioned, that error is probably not very significant for most real tasks
[23:40:52] <Laminae> My alternative thought was to take one 1/4 in piece of 6061 (flat) and cut it so i could still access the extrusions and bolt to the holes though that
[23:41:19] <Laminae> The downside was that it would have to be much bigger have less torsional rigidity and be a whole lot more work
[23:45:11] <renesis> but 6061 is sexy