#linuxcnc | Logs for 2016-11-24

Back
[02:26:15] <Deejay> moin
[03:43:37] <holymoly> Good morning peps
[03:46:06] <holymoly> We got a 80000 Euro CNC donated from Datron
[03:46:32] <holymoly> the machine is ok beside the following
[03:47:44] <holymoly> we ve got a 1,5kw spindle which does not match to the z axis mount -> have to make mount
[03:48:48] <holymoly> and we miss the orginal frequenzy converter for the spindel
[03:49:23] <holymoly> instead of the orginal FC we have a chinese FC
[03:49:47] <holymoly> I want to marry spindel and FC
[03:50:05] <holymoly> to regulate thr rpm
[03:50:24] <holymoly> If u dont believe i have pics
[03:54:12] <holymoly> so I need to find information what is happening at all :D
[05:51:49] <jthornton> morning
[05:51:54] <XXCoder> hey
[05:51:56] <Tom_L> morning jt
[05:51:58] <Tom_L> xx
[05:52:00] <Deejay> hey
[05:52:06] <Tom_L> de
[05:52:06] <_methods> happy turkey day
[05:52:06] <XXCoder> solved that issue jt?
[05:52:10] <Tom_L> me
[05:52:18] <XXCoder> happy penis bird day
[05:52:34] <jthornton> still arraigning the stuff in the shop
[05:53:08] <_methods> gotta go clean the ash out of the egg and load up with fresh charcoal for the day
[05:53:19] <jthornton> got one corner arraigned lol
[05:53:27] <XXCoder> fancy
[05:53:45] <Tom_L> good it's not an octagon :)
[05:55:05] <XXCoder> or round
[05:55:49] <XXCoder> circle can be discribved as having infinite sides or basically corners heh
[06:04:42] <jthornton> lol
[06:04:53] * jthornton goes to check on the coffee...
[06:08:58] <_methods> charcoal loaded
[06:09:19] <XXCoder> charcoal loaded coffee
[06:21:14] <_methods> i loaded coffee hours ago
[06:21:20] <_methods> priorities
[06:22:18] <archivist> can someone load recovery, just walked back from town
[06:22:40] * SpeedEvil wishes he'd saved in 1980.
[06:22:53] <XXCoder> 1984
[06:28:28] <archivist> I shall see what an iced bun and a cup of coffee does
[06:49:21] <_methods> yeah i wouldn't mind having about 20 years back
[06:49:29] <_methods> got up at 4am and went to the gym
[06:49:47] <_methods> recovery doesn't happen like it used to for sure
[06:52:24] <malcom2073_> When someone posts on facebook, "hey, I think I left my gun at X, has anyone been there recently and seen it?" *facepalm* this is why we can't have nice things.
[06:53:33] <XXCoder> gun arent toys
[06:53:43] <XXCoder> secure it well and always know where it is
[06:53:47] <malcom2073_> Tell that to some of these people
[06:53:52] <XXCoder> indeed
[06:54:16] <XXCoder> if I ever own a gun, I wouldnt want someone to kill with my gun when I lose it
[06:55:09] <SpeedEvil> small portable CNC mill with a steel block to manufacture a gun on demand.
[06:55:26] <XXCoder> theres cnc machine designed to mill lowers only
[06:55:30] <XXCoder> what a stupid thing
[06:55:40] <SpeedEvil> yeah, I know.
[06:55:42] <malcom2073_> Haha yeah, it's a moneygrab by that guy
[06:55:56] <malcom2073_> The 3d printed gun guy
[06:56:30] <XXCoder> with modern materal science unfortunately we can basically make non-ferric gun
[06:56:32] <sync> I like how you can jump through the hoops that way
[06:56:36] <XXCoder> cermtics and such
[06:58:46] <SpeedEvil> Well - non ferric is easy
[06:58:58] <SpeedEvil> many of the first cannons were bronze
[06:59:09] <XXCoder> those is slightly big and heavy
[06:59:22] <archivist> and casting was not that good, some exploded
[06:59:55] <archivist> I have seen a cannon that failed a proof test
[07:00:13] <SpeedEvil> Sure - but nickel based superalloys
[07:00:20] <XXCoder> I have seen cannon so small it could fit inside medicine container
[07:01:33] <XXCoder> https://youtu.be/zZ6P04zbico
[07:02:15] <XXCoder> 40 bucks apparently
[07:02:57] <XXCoder> and lol ending
[07:06:13] <XXCoder> I wonder what would have happened if steel was discovered 1,000 years earlier.
[07:13:43] <Tom_L> we'd probably be extinct
[07:14:09] <_methods> we would have had selfie sticks a lot sooner
[07:14:32] <XXCoder> selfies existed for long time
[07:15:18] <XXCoder> though those old wood box camera would be too heavy for stick lol
[07:15:34] <XXCoder> we have one, but no idea how to use it lol
[07:26:07] <archivist> XXCoder, it was discovered but then forgotten, there is an iron pillar in India
[07:26:32] <XXCoder> thats not quite steel but interesting nevertheless
[07:26:46] <XXCoder> theres so many stuff lost and rediscovered
[07:27:05] <XXCoder> like computer made over 1,000 years ago, it calculates orbits and such for astromoney
[07:27:25] <XXCoder> someone made a lego reproduction of it, pretty amazing.
[07:27:32] <archivist> https://en.wikipedia.org/wiki/Iron_pillar_of_Delhi
[07:29:20] <XXCoder> its in least 1,000 years old
[07:29:40] <SpeedEvil> It's also regularly cleaned
[07:29:54] <XXCoder> ah it says 1,600 years
[07:30:45] <XXCoder> interesting
[07:31:18] <XXCoder> https://www.youtube.com/watch?v=RLPVCJjTNgk
[07:31:19] <XXCoder> enjoy
[07:41:01] <Jymmm> Happy Thanksgiving
[07:41:31] <XXCoder> happy penis bird day
[07:41:56] <Jymmm> ?¿
[07:42:20] <XXCoder> https://media1.britannica.com/eb-media/05/96205-004-CFC438E3.jpg
[07:42:49] <BeachBumPete> Jymmm Happy Thanksgiving to you as well
[07:44:39] <XXCoder> is thankgiving day thurs or what
[07:44:43] <XXCoder> ah yes it is
[07:44:47] <XXCoder> black friday lol
[07:48:17] <_methods> turkey on the egg
[07:49:47] <BeachBumPete> Man I still can't seem to figure out how to stream online video from my laptop to my home theater receiver :(
[07:52:49] <_methods> chromecast?
[07:54:17] <BeachBumPete> I dunno man I already stream music and pandora from my phone to the stereo easily but I want to watch youtube videos on it while I am surfing the net and be able to control what I watch etc.
[07:55:03] <BeachBumPete> the stereo has an android ap that allows you to setup everything but for some reason I don't get anything in there coming from my computer
[07:55:26] <BeachBumPete> It apparently uses DLNA link
[07:56:44] <_methods> hmm
[07:57:08] <_methods> i use chromecast for streaming video
[07:57:23] <_methods> my reciever only does music streaming
[07:57:26] <BeachBumPete> the funny thing is the app recognizes my home network and I can access all my music
[07:57:43] <_methods> yeah ive never been able to get mine to play dlna video
[07:57:45] <_methods> only music
[07:58:21] <BeachBumPete> is chromecast free or built into Chrome?
[07:58:44] <_methods> well you have to pay for the chromecast
[07:58:52] <_methods> but the app is built into chrome
[07:58:57] <BeachBumPete> yeah I have been streaming audio forever on it now
[07:58:57] <XXCoder> chromecast the device yeah
[07:59:22] <_methods> they're only like $30 or something like that
[07:59:28] <BeachBumPete> but I wanted to basically have the video on the bigscreen along with the audio
[07:59:38] <_methods> the only downside i've found to the chromecast is that if your internet is out it won't work
[08:00:13] <BeachBumPete> the TV has built in youtube and netflix etc but I wanted to control it without having to reach for the remote all the time ;)
[08:00:25] <XXCoder> roku 3 rocks
[08:00:34] <_methods> yeah i have mine setup so i can control all my stuff from my phone or tablet
[08:00:44] <XXCoder> they keep stuff current, unlike any tv firmware crap
[08:00:44] <Jymmm> _methods: Even google translate, though you have downloaded the language pack, will not always work offline.
[08:01:03] <_methods> yeah
[08:01:07] <BeachBumPete> yeah my phone controls the receiver and all that too
[08:01:16] <BeachBumPete> I actually love that
[08:01:20] <_methods> yeah me too
[08:01:21] <Jymmm> _methods: Google speach will not work offline either
[08:01:28] <BeachBumPete> I keep TONS of music on my cell
[08:01:31] <_methods> i always have a tablet or phone within arms reach
[08:01:44] <XXCoder> heh my most favorite device is kindle paperwhite
[08:01:45] <Jymmm> IF you want to stream, use VLC
[08:01:47] <XXCoder> it rocks
[08:01:48] <BeachBumPete> its nice to be able to stream to the onkyo
[08:01:50] <_methods> oh i don't keep anything on my devices all my stuff is on server
[08:02:32] <BeachBumPete> is there a way to have say youtube videos on your laptop and stream it live to the TV/receiver
[08:03:01] <Jymmm> IF you can stream to your tv, why not just pull directly?
[08:03:12] <BeachBumPete> how?
[08:03:28] <BeachBumPete> the TV has DLNA feature too but I am not familiar with how to use it
[08:03:40] <Jymmm> RTFM?
[08:03:49] <BeachBumPete> heh bite me
[08:03:57] <Jymmm> google or youtube your tv model
[08:04:01] <Jymmm> I was serious
[08:04:26] <BeachBumPete> I have tried to figure it out but so far no dive...
[08:04:28] <BeachBumPete> dice
[08:04:31] <BeachBumPete> hehe
[08:04:34] <Jymmm> There are thousands of tv's out there, how are we to know what is needed for YOUR tv, but it's in the manual
[08:04:49] <BeachBumPete> this is true
[08:04:56] <XXCoder> review the flower movie
[08:05:05] <Jymmm> so... Read the fscking manual =)
[08:05:15] <BeachBumPete> I honestly think it is more of a network issue than the TV etc.
[08:05:27] <_methods> it might be a codec thing
[08:05:34] <Jymmm> If you can stream from your phone to your tv, I doubt it
[08:05:37] <BeachBumPete> Oh hell breakfast is served :)
[08:05:40] <_methods> if your reciever doesn't have the right codec it cant play the video
[08:06:02] <Jymmm> Yeah, you might need to update your tv
[08:06:30] <XXCoder> if there is even an update
[08:06:31] <Jymmm> it's just a computer in it
[08:06:48] <Jymmm> If he RTFM, he might find out =)
[08:07:22] <XXCoder> review the fanny manual?
[08:11:38] <XXCoder> roll the ford manually
[08:19:41] <Jymmm> concrete batteries, interesting
[08:24:22] <BeachBumPete> well I guess I better RTFB ;)
[08:28:17] <Tom_L> tried all other ways first ehh?
[08:28:39] <BeachBumPete> probably not or I might have found one that worked ;)
[08:28:41] * Tom_L is crying like a baby but the onions are sliced
[08:29:50] <tjtr33> When I use the PID component for a velocity, I think the Dgain term affects acceleration. If so, a PID component for position affects what? 'precision' or 'tracking tightness'?
[08:32:09] <Tom_L> kinda the same thing
[08:32:29] <archivist> the loop, how it tracks
[08:32:56] <malcom2073_> Onions are about the only thing I use a food processor for
[08:33:23] <Jymmm> Tom_L: Next time... slice them near a window, it REALLY helps
[08:33:38] <Tom_L> not if it's blowing back in your face
[08:33:58] <Jymmm> Tom_L: That's the best actually, it displaces the onion
[08:34:31] <Jymmm> Also, if you have REALLY strong onions, toss them in a collander and rinse under water.
[08:34:37] <Tom_L> 35f
[08:34:39] <tjtr33> put a matchstick between your teeth https://www.reddit.com/r/LifeProTips/comments/1buzxm/lpt_avoid_onion_tears_by_putting_a_match_stick/
[08:34:57] <tjtr33> try it
[08:34:58] <Jymmm> Tom_L: Freezeing your ass off is better than crying imo
[08:35:45] <BeachBumPete> I just don't eat onions.....unless its a bloomin' onion :)
[08:35:57] <Jymmm> tjtr33: I suspect that just promote you breathing thrue your mouth than your nose
[08:36:02] <tjtr33> yep
[08:36:08] <Tom_L> BeachBumPete i don't eat them raw either
[08:36:20] <_methods> mmmm bloomin onion
[08:36:20] <Jymmm> tjtr33: But that doens't do shit for watering eyes =)
[08:36:35] <BeachBumPete> We had dinner last night at Outback LOL
[08:36:42] <_methods> i love that place
[08:36:45] <Tom_L> they're not what they used to be
[08:36:49] <_methods> cheese fries
[08:36:52] <Jymmm> nope
[08:37:04] <tjtr33> Jymm you didnt try it, it does help watering eyes
[08:37:07] <Jymmm> I hate the sauce they serve with blooming onion too
[08:37:26] <tjtr33> and a sharp knife helps, as crushing ,akes it worse
[08:37:44] <Tom_L> they're only good when they're so hot you can barely eat em
[08:37:55] <Tom_L> after that it's just soggy grease
[08:38:01] <sync> https://upload.wikimedia.org/wikipedia/commons/thumb/4/43/PID_en.svg/400px-PID_en.svg.png tjtr33 that is what is happening
[08:38:07] <Jymmm> tjtr33: I'll try it. I keep a diamond steel in the drawer =)
[08:38:51] <Jymmm> I can't stand a dull knife
[08:39:48] <BeachBumPete> I honestly am not a huge fan of the place but it had been a few years since we ate there so we tried it again. I had the Tilapia with lump crab on top and it was.....ok
[08:39:59] <BeachBumPete> but the Bloomin onion was yummy hehe
[08:40:11] <Jymmm> BeachBumPete: and $38 later...
[08:40:23] <Jymmm> ...for "ok" that is.
[08:40:24] <BeachBumPete> hell I wish it was only $38
[08:41:03] <Jymmm> I'll goto paces and just have the appetizers, at least I know I'll enjoy it.
[08:41:22] <BeachBumPete> ya know I have done that several times and it is great
[08:41:36] <Jymmm> BeachBumPete: I meant $38 per person
[08:42:00] <BeachBumPete> we go to Conchy Joes often and just get their Conch Fritters, Calamari, and some cups of their amazing Clam chowder....SOOOO Good
[08:42:31] <BeachBumPete> altho I don't think the waitresses enjoy that too much they think they are gonna get a low tip haha
[08:42:44] <Jymmm> oh well
[08:42:46] <Tom_L> i liked P.F. Changs the time we went
[08:42:54] <BeachBumPete> never tried it
[08:43:05] <_methods> yeah i love pf changs too
[08:43:10] <Tom_L> give it a go once
[08:43:11] <BeachBumPete> Still on the hunt for a 9' Christmas tree for a good price...
[08:43:41] <Jymmm> BeachBumPete: Take your pick, we have about 50,000 to choose from
[08:44:06] <BeachBumPete> There is a PF Changs in the Gardens Mall apparently down south in Jupiter
[08:44:16] <Tom_L> clear one from under some power lines.
[08:44:25] <Tom_L> they're doomed anyway
[08:44:52] <BeachBumPete> what are you talking about?
[08:45:07] <Jymmm> Tom_L: PG&E does a hellva a job on keep trees away from power lines. They just took down a ~70ft one
[08:45:24] <Jymmm> it was dead
[08:45:47] <Tom_L> ahh, pete probably doesn't have many evergreens down south
[08:45:59] <BeachBumPete> eh no
[08:46:26] <Jymmm> It's a trip when they fall that tall of a tree
[08:46:32] <BeachBumPete> whats funny is we lived in Tennessee and Norcal for years and I never went out and FOUND a Tree for our Christmas tree...
[08:47:45] <Jymmm> I got a xmas tree in the front yard... a few of them in fact, all 60ft+ =)
[08:48:10] <Jymmm> tallest being 63.5 ft
[08:48:18] <Jymmm> as of a couple months ago
[08:48:24] <BeachBumPete> well get your azz out there and decorate them!!
[08:48:45] <Jymmm> Short and short-stack do that
[08:48:49] <Jymmm> shorty*
[08:49:11] <BeachBumPete> just need those tree climbin spurs hehe
[08:49:21] <Jymmm> Shorty = squirrles, short-stack = baby squirrels
[08:49:47] <Jymmm> And I'm waiting for Harry and Perry to decorate the rest (the wild peacocks)
[08:49:50] <BeachBumPete> you have trained squirrels that decorate the trees in your yard....JEALOUS!!
[08:50:11] <Jymmm> BeachBumPete: Hell, they come and knock on the front door to be fed
[08:50:36] <BeachBumPete> WOW
[08:50:57] <BeachBumPete> PF Changs is that like a mongolian BBQ or one of those cook it in front of you digs?
[08:52:26] <Jymmm> The wild peacocks are kinda cool though. They just hang around, I'll step outside the garage not realizing they'r ethere, the spook me, I spook them, and then we're all good. lol
[08:52:33] <malcom2073_> PF Changs is a frozen chinese food brand?
[08:52:45] <malcom2073_> Oh, apparently it's also a restaraunt
[08:52:53] <BeachBumPete> we have wild peacocks around here too
[08:53:03] <Jymmm> malcom2073_: started as restraunt, then frozen food branding.
[08:53:09] <malcom2073_> Cool
[08:53:29] <BeachBumPete> in fact there was one in an open field near the storage unit I had and it made the most strange sounds
[08:53:47] <Jymmm> BeachBumPete: They like cat food, and it's good from them too. But dont leave a bowl out or they'll become dependant
[08:54:25] <Jymmm> oh THEY ARE LOUD. But they make some cools sounds too.
[08:54:35] <BeachBumPete> my next door neighbor feeds a family of racoons and they are pretty cool I have seen some of them in my yard at night
[08:54:50] <BeachBumPete> we have some beautiful sand hill cranes around too
[08:55:30] <BeachBumPete> and they have these all white spoon bill looking birds all around too but the peacocks are really neat
[08:55:43] <BeachBumPete> they are REALLY LOUD
[08:58:32] <Jymmm> It KILLS me that every thinks that an "L" configuration is a "rocket stove"
[08:58:40] <Jymmm> everyone*
[08:59:16] <Jymmm> They all miss that one key component of it
[09:00:20] <SpeedEvil> Jymmm: the de Laval nozzle?
[09:01:10] <Jymmm> SpeedEvil: No, the super heated intake air, or at least the channel for it.
[09:02:04] <Jymmm> which removes the mositure in the air before hitting the firebox, promoting higher combutstion.
[09:02:46] <Jymmm> SpeedEvil: As much of a pyro that I am, rocket motors are something that I've been avoiding
[09:03:29] <Jymmm> SpeedEvil: Even saw how to make my own, and yeah, NOOOOOOOOOOOOOOO
[09:04:15] <SpeedEvil> Somewhere I have a diagram of a promising idea for a valveless pumped biprop.
[09:04:35] <Jymmm> SpeedEvil: That would be like me having a flame thrower on my back and there are 100 open contianers of fireworks fresh from china in front of me =)
[09:05:44] <Jymmm> I dont even play with the safety fuse I have
[09:07:10] <Jymmm> I took one inch of safety fuse, put a brick on it, ignited it, and that lil sucker shot off like a rocket 40ft across the yard =)
[09:07:10] <Kucharsky> Hi there
[09:07:13] <Kucharsky> !
[09:08:19] <jthornton> dang finally figured out where I was on the logbot... I need to set up git
[09:12:36] <Kucharsky> Guys I'm consFew months back I bought leadshine stepper drivers with motors for cnc plotter project that’s in my mind for some time now. Recently I noticed that teknic clearpath servos are gaining on popularity. I’m considering dropping steppers in favour of clearpath SD servos. Does anybody have an opinion on clearpath ? Is Step + direction controlling mode only way to control them from linux cnc
[09:13:11] <Kucharsky> sorry I goofed
[09:13:59] <Kucharsky> You get the picture ;) Are clearpath integrated servomotors are good value for money?
[09:14:46] <sync> they are ok, the low input voltage is a joke tho
[09:17:52] <nubcake> hi
[09:22:24] <Kucharsky> Wondering if those mothors would be ok to be retrofitted into something bigger, like old scrapped cnc mill .
[09:25:25] <archivist> I dont think the interface is well suited
[09:28:40] <archivist> use real servos with a proper drive on a big machine so you can tune it properly
[09:29:49] <Tom_L> jthornton, you need git just for that??
[09:35:42] <Kucharsky> archivist: are there any servo solution with comparable pricerange?
[09:35:55] <Kucharsky> archivist: that you know of?
[09:35:57] <Tom_L> servos tend to cost more than steppers
[09:36:17] <malcom2073_> And are typically more difficult to set up for someone unfamiliar with them
[09:39:07] <archivist> I take a dim view of step dir "servos" with no feedback for linuxcnc
[09:43:09] <sync> use stmbl then ;)
[09:43:17] <Kucharsky> Do you guys know of any servo systems that could be bought of the shelf for decent price?
[09:43:26] <Kucharsky> sync: what is stmbl
[09:43:27] <Kucharsky> ?
[09:43:44] <malcom2073_> Define decent?
[09:43:55] <malcom2073_> You're gonna spend 2-3x on servos as you would on steppers. They're 2x3 better though
[09:43:56] <Tom_L> sub 10k range
[09:44:10] <malcom2073_> Arguable more better
[09:44:37] <sync> Kucharsky: the servo drive I codevelop
[09:44:39] <Kucharsky> clearpath from teknic would be around 400$ for axis
[09:44:52] <malcom2073_> That's pretty cheap for servos
[09:45:05] <sync> archivist: it is pretty cool to get drivefaults and not loose homing in linuxcnc, that is very handy
[09:45:57] <Kucharsky> clearpaths are not servos I think they are markedted as servos since they work in closed loop internally
[09:46:08] <Kucharsky> and they have the ability to report failures
[09:46:13] <Kucharsky> at positioning
[09:46:14] <malcom2073_> Servo-ish then
[09:46:21] <Kucharsky> exactly
[09:47:00] <malcom2073_> I set up my knee mill with steppers for less than $700. Equivalant servos would cost me $2-3k, probably more?
[09:47:24] <Kucharsky> sync: I'm looking at https://github.com/rene-dev/stmbl
[09:48:11] <sync> yes
[09:51:04] <IchGucksLive> hi is the macys parade today ?
[09:51:24] <Kucharsky> sync: wow this project is amazing, I'm wondering if this controller would be suitable for 2 servos i have laying around. Servos by siemens. Let me check spec for them.
[09:52:10] <Kucharsky> sync: do you know what is the cos of hardware to get stmbl working?
[09:52:26] <Kucharsky> sync: sorry I ment cost
[09:52:56] <sync> I think in the old hardware is around 100€ or a bit more, the new one will be a bit cheaper
[09:53:31] <malcom2073_> ooo cheaper is good!
[09:53:32] <rene-dev> Kucharsky: less than $100 if you build it yourself.
[09:54:16] <rene-dev> It probably works with your servos
[09:54:33] <malcom2073_> Oh nice, it's really come down in price!!
[09:54:57] <sync> it is not like it was more expensive before
[09:55:38] <malcom2073_> I specced out the parts list (took a while to find US distributor equivalants), and it was significantly more expensive than the EU distributors heh
[09:55:59] <malcom2073_> This was like a year ago or so, I dunno whenever I talked to you about it
[09:58:15] <rene-dev> New version is significantly cheaper
[09:58:25] <malcom2073_> Ooo, and there's an octopart BOM now
[09:58:27] <malcom2073_> That's fantastic
[09:58:44] <rene-dev> Yes, but that's the old version
[09:59:09] <malcom2073_> Still cheaper than what I had found before, I may have to revisit stmbl for my big router, awesome
[09:59:30] <Kucharsky> can you Guys have a look at my servos to see if it'' work with stmbl ? https://github.com/rene-dev/stmbl
[09:59:34] <Kucharsky> sorry wrong link
[09:59:50] <Kucharsky> I ment this link http://imgur.com/a/gSDv0
[10:00:10] <Kucharsky> motor use resolvers to encode position
[10:01:23] <rene-dev> Yes
[10:01:35] <Kucharsky> I'have them lying around in garage, bought them with 2 rexroth linear tables
[10:01:42] <rene-dev> Works perfectly, I have a similar one
[10:01:49] <Kucharsky> NO WAY!
[10:02:28] <rene-dev> The list of supported feedback systems is long
[10:03:48] <rene-dev> Resolver motors pretty much always work no problem
[10:06:38] <Kucharsky> rene-dev: I see you're main contributor, how the hell did you manage to write such an complicate piece of low level software?
[10:07:58] <IchGucksLive> THANksgiving Day parade LIVE http://www.earthcam.com/events/thanksgiving/2016/?cam=midtown4k_thanksgiving2016
[10:10:08] <rene-dev> Kucharsky: 2 main contributors for the software... its not really complicated once you figure out how stuff works :D
[10:10:41] <Kucharsky> rene-dev: are there any similar open source project to yours?
[10:10:55] <Kucharsky> rene-dev: never seen one before
[10:11:59] <rene-dev> None that support industrial voltages or feedback systems. None that have a properly implemented control loop.
[10:12:05] <archivist> you can run a servo off the parallel port, all the code is also in linuxcnc
[10:13:27] <rene-dev> So how do you connect a resolver to parport? How abs encoders? Sin/cos interpolation for linear glass scales?
[10:14:05] <rene-dev> High speed encoders?
[10:14:44] <archivist> it was a proof of concept https://emergent.unpythonic.net/projects/01142347802
[10:15:03] <rene-dev> And Linuxcnc has no cascaded pid for motor control
[10:16:18] <rene-dev> Yes, that is a dc servo
[10:18:19] <Kucharsky> rene-dev: where to start? No many resources on web on how to use this controller.
[10:18:24] <pcw_home> actually many people have set up cascaded PID in LinuxCNC
[10:19:05] <pcw_home> (though LinuxCNC on most hardware is probably too slow for current loops)
[10:19:25] <rene-dev> Kucharsky: we have a wiki and a irc channel
[10:20:02] <Kucharsky> https://github.com/rene-dev/stmbl/wiki
[10:20:03] <Kucharsky> ?
[10:21:45] <rene-dev> pcw_home: and no one seems to do it... and not really feasible.
[10:22:32] <rene-dev> Kucharsky: yes... starting point is getting some hardware
[10:25:13] <rene-dev> pcw_home: problem is you have to make sure the inner loops can reach their target within the cycle of the outer loop. Usually the inner ones are just quicker. We do it by dynamic clamping.
[10:25:40] <IchGucksLive> Macys live https://www.youtube.com/watch?v=J1i4n11Ia9g
[10:30:41] <pcw_home> rene-dev: not true at all many people have implemented cascaded velocity/position loops
[10:30:50] <pcw_home> (in linuxcnc)
[10:31:18] <Roguish> pcw_home: what do you mean 'cascaded' ??
[10:32:01] <pcw_home> position loop PID output to velocity loop PID setpoint
[10:32:21] <Roguish> i've a setup with yaskawa motor/drive and linuxcnc. yaskawa has its internal pid. then linuxcnc has its pid
[10:33:05] <Roguish> it's a 'double' loop, i call it. maybe what you are speaking of. very interesting to tune....
[10:34:01] <rene-dev> pcw_home: any example? I have never seen one.
[10:34:02] <pcw_home> most motor controls have three cascaded PID loops (current, velocity, position)
[10:34:28] <Roguish> ah, ok. now i got ya.
[10:34:33] <pcw_home> Theres at leas a couple on the LinuxCNC forum
[10:35:05] <rene-dev> Roguish: yes, that is the classic way of doing it. Its actually very easy to tune, if it's configured correctly.
[10:36:11] <Roguish> rene-dev: i found that tuning the yaskawa 'soft' made tuning lcnc much easier. yaskawa in velocity mode.
[10:36:32] <pcw_home> we also had a customer doing current control with linuxCNC (100 KHz servo thread)
[10:37:08] <Kucharsky> rene-dev: can your controller be bought assembled? Not in separate parts?
[10:37:09] <rene-dev> pcw_home: with what hardware?
[10:37:30] <pcw_home> our FPGA card, custom A-D
[10:37:53] <pcw_home> (no video on PC host)
[10:38:22] <rene-dev> Kucharsky: not at the moment. Maybe we do a production run later, there is a lot of interest.
[10:39:26] <rene-dev> pcw_home: but what is the point. Its much easier just to put everything on a $3 micro controller
[10:40:30] <rene-dev> pcw_home: and you still need the power stuff. Feedback input. Galvanic isolation. We just put everything one one board.
[10:40:54] <pcw_home> It depends on how much processing you need
[10:43:14] <rene-dev> We have a 72 mhz 32bit cortex m4 with fpu just to output pwm and current coltrol. Internal opamps and adc. Isolated from that is the servo cpu, same one, but 168mhz. 192k ram, 1mb flash. Should be enough :)
[10:43:32] <pcw_home> depends on what you are doing
[10:44:10] <pcw_home> Plenty for motor control, not for all power control tasks
[10:44:29] <rene-dev> We already do a lot of stuff, and struggle to get past 50% load
[10:44:31] <IchGucksLive> live from street https://www.youtube.com/watch?v=JmdIHrvFAZg
[10:44:54] <rene-dev> What control tasks?
[10:46:15] <rene-dev> Servo amp just gets target position and velocity from linuxcnc via smartserial
[10:47:02] <pcw_home> LinuxCNC/HAL can be used for other than motion control tasks
[10:47:56] <Jymmm> pcw_home: Can LinuxCNC/HAL cook a turkey?
[10:48:21] <IchGucksLive> the timer will do the job
[10:48:37] <Jymmm> IchGucksLive: I said COOK, not notify =)
[10:48:39] <IchGucksLive> and pwm controls the oven temp
[10:48:56] <Jymmm> IchGucksLive: a 99¢ watch cna do that =)
[10:49:04] <IchGucksLive> with robot you can get a nice crust
[10:49:36] <Jymmm> and carnival ride too... https://www.youtube.com/watch?v=wOCmoYU6h1Q
[10:49:40] <IchGucksLive> BYE the way is there a cheep part for getting part temp to lcnc
[10:50:05] <IchGucksLive> im looking at 300deg C
[10:50:41] <IchGucksLive> to parport of cause
[10:50:45] <IchGucksLive> mabe pwm in
[10:50:56] <rene-dev> pcw_home: we also have a similar hal implementation on the drive. You can reconfigure it at runtime to do stuff with i/o or add a trajectory planner to the input...
[10:52:19] <rene-dev> pcw_home: someone used our hal on a io board to control gripper, force-torque sensor and other io on robot arm
[10:52:26] <pcw_home> Yeah I think its neat that you set up the drive in a HAL like way
[10:55:22] <rene-dev> We already had requests from automation engineers, because you can do a lot of simple tasks just with the drive, no plc required
[10:55:32] <pcw_home> Have you considered a high performance version? (using MOSFETs)
[10:55:34] <pcw_home> how fast can you run you control loop?
[10:55:58] <IchGucksLive> rene-dev: the xtradrives can do that also
[10:58:05] <rene-dev> Current loop is synced to pwm, so 20khz. But something like 200khz is no problem
[10:59:10] <rene-dev> pcw_home: 20A 400V works for most retrofits... we have a 400v mosfet Prototype from sync
[10:59:30] <pcw_home> 20 KHz is a bout it for IGBTs
[11:01:09] <rene-dev> Yes, we use a igbt module. Sorry, its 15 khz at the moment.
[11:01:21] <rene-dev> IchGucksLive: ah, interesting
[11:02:04] <rene-dev> But yes, sync is experimenting with mosfets.
[11:02:12] <pcw_home> Yeah I think our 8I20 default is 12 KHz and 20 KHz max (but lower current limit at 20 KHz)
[11:04:09] <rene-dev> Yes, what I don't like about the 8i20 is the size, and you need an extra board for the feedback. And the default position to current pid setup is not ideal, and a bit limited.
[11:06:54] <rene-dev> But 12-20 is good
[11:13:45] <pcw_home> Yeah the feedback works best if linuxcnc runs a servo loop at 4 KHz or better (or linear motors that have low commutation rates)
[11:13:46] <pcw_home> We could improve the commutation quite a bit by supplying velocity for lookahead
[11:16:15] <pcw_home> I notice OnSemi has some new 30 to 75 Amp IGBT modules that are pretty cheap for the 50 and 75 A range
[11:18:19] <IchGucksLive> im off to home HAPPY thanksgiving
[11:22:00] <rene-dev> That's no problem when everything just runs at 5kz
[11:23:36] <rene-dev> The velocity is also used. Proper velocity feed forward. Rarely seen that done properly on a linuxcnc setup :D
[11:24:12] <rene-dev> We use iram256 module, not too bad
[11:24:52] <pcw_home> velocity feed forward is always used for velocity mode drives (at least if tuned correctly)
[11:25:29] <rene-dev> I know, but many people dont tune correctly
[11:25:37] <pcw_home> also used for stepgen of course
[11:26:01] <rene-dev> Yes, I hope so
[11:26:19] <pcw_home> One thing LInuxCNC lacks is proper FF2
[11:27:09] <pcw_home> ( you really want the trajectory planner to supply velocity _and_ accel for the next segment )
[11:28:14] <pcw_home> but it does not, so the best you can do is D/DT in HAL (so you are always a sample time late)
[11:28:53] <pcw_home> (with FF2)
[11:29:03] <rene-dev> Yes
[11:30:09] <pcw_home> This show up clearly on the stepgen (which is basically an perfect velocity mode servo)
[11:30:53] <pcw_home> I mean nearly perfect ( few PPM timebase differences etc )
[11:42:11] <rene-dev> Yes. Stepgen is just p and ff1?
[11:48:04] <pcw_home> Yep
[11:48:24] <pcw_home> (well a little FF2 if you are fussy)
[11:48:56] <pcw_home> (FF2 = seconds from position read to velocity write)
[11:52:59] <pcw_home> P is normally1/ thread_period (which means errors are fixed in the next period)
[11:56:41] <pcw_home> For the hardware stepgen, I added the DPLL to remove position sampling jitter you needed to bound acceleration (or bound PID maxerror)
[11:56:55] <pcw_home> before I added
[12:01:18] <rene-dev> Hmm, I never had issues with the jitter
[12:04:49] <pcw_home> step motors are fussier about the pulse stream than a servo (that just low pass filters phase noise)
[12:09:11] <pcw_home> imagine generating a pulse stream for a stepper with a hardware stepgen (a DCO basically) and moving at a high speed
[12:09:13] <pcw_home> what happens if you have 50 usec jitter on the servo thread so the position sample is 50 usec late?
[12:09:14] <pcw_home> The PID makes a large (incorrect) adjustment to the DCO frequency
[12:10:17] <FloppyDisk525> That's not a bad lathe to convert in the sfbay area: http://sfbay.craigslist.org/sby/tls/5887706844.html
[12:13:01] <CaptHindsight> looks good
[12:13:12] <CaptHindsight> don't pass that up
[12:17:56] <CaptHindsight> http://rockford.craigslist.org/tls/5879127032.html Warner Swasey 1AC automatic lathe - $1250
[12:19:55] <pcw_home> rene-dev: stepgen with no sampling time correction and with correction:
[12:19:57] <pcw_home> http://freeby.mesanet.com/nodpll.png
[12:19:58] <pcw_home> http://freeby.mesanet.com/dpll.png
[12:20:00] <pcw_home> notice PID error scale
[12:22:03] <rene-dev> Uuh
[12:23:33] <rene-dev> So how does that work exactly?
[12:23:52] <IchGucksLive> hi
[12:27:33] <pcw_home> In this case it uses a DPLL to synchronize to LinuxCNCs servo thread, filtering out the sample time jitter
[12:29:23] <pcw_home> in hal:
[12:29:24] <pcw_home> setp hm2_[HOSTMOT2](BOARD).0.dpll.01.timer-us -50
[12:29:26] <pcw_home> setp hm2_[HOSTMOT2](BOARD).0.stepgen.timer-number 1
[12:30:19] <pcw_home> so the stepgen position is sampled 50 usec before the nominal servo thread time
[12:34:17] <pcw_home> you have to have some lead time (50 usec in this case) because of baseline wander in scheduling latency do to CPU load (video playing etc)
[12:43:11] <rene-dev> pcw_home: ah, thanks. I see if that is something we need at siome point.
[12:45:04] <IchGucksLive> English tutorial on Resculping from Picture English as some requierd here https://youtu.be/7UqUJpeKtKQ
[12:59:34] <tiwake> well
[12:59:40] <tiwake> time to go stuff my face senseless
[13:09:33] <IchGucksLive> in germany jesterday about 100.000 birds where crushed cause of birdflue
[13:09:45] <IchGucksLive> i the US they eat them all today ;-)
[13:14:33] <IchGucksLive> im off Gn8
[13:24:11] <Cromaglious_> not I we're having IKEA meat balls
[13:37:22] <Vq> mmm, swedish meatballs :)
[13:46:43] <Jymmm> Cromaglious_: Are those next to the ikea cfl lightbulbs?
[14:00:19] <sync> pcw_home: 20kHz is clearly not the limit for IGBTs
[14:00:41] <Jymmm> nor ICBM's
[14:04:17] <pcw_home> No but standard ones are not of much use at 20KHz or above due to switching losses
[14:04:43] <sync> I use standard ones at over 100kHz with no issues
[14:04:59] <pcw_home> Then you are not looking closely at your losses
[14:05:06] <sync> I am
[14:05:15] <pcw_home> whatever
[14:05:16] <sync> efficency is similar to mosfets in that application
[14:05:55] <pcw_home> switching losses are much higher in IGBTS
[14:06:22] <pcw_home> look at the uJoules/transition
[14:06:28] <pcw_home> terrible
[14:06:42] <pcw_home> way too slow
[14:11:10] <sync> I use fgh60n60smd, they perform as well as a variety of fets I tested in that application
[14:15:59] <pcw_home> At 600V its true that IGBTs are much better on a cost basis
[14:21:04] <pcw_home> but even the one you show would have 170W switching losses at 400V 60A
[14:21:42] <pcw_home> (at 100 KHz)
[16:01:41] <andypugh> Anybody here speak Python?
[16:02:12] <andypugh> I am trying to pass a dict from one function to another.
[16:02:13] <andypugh> http://www.pastebin.ca/3741982
[16:05:12] <Jymmm> andypugh: I'd ask in #python
[16:05:25] <andypugh> Why didn’t I think of that?
[16:07:26] <Jymmm> andypugh: It's all that foggy mist you breath ;)
[16:09:01] <Jymmm> andypugh: ...it clouds the mind
[16:11:15] <FinboySlick> andypugh: You still got a video of that custom electric motor you had made?
[16:19:03] <JT-Shop> andypugh: http://paste.ubuntu.com/23529442/
[16:20:10] <andypugh> FinboySlick: I haven’t ever had a custom motor made. I have made one myself, though.
[16:22:25] <FinboySlick> Bad grammar ;) I meant 'have made'.
[16:34:33] <JT-Shop> andypugh: http://paste.ubuntu.com/23529505/
[16:36:13] <andypugh> JT-Shop: Yes, but now try printing “args” in the function, not the global “words”
[16:37:24] <andypugh> FinboySlick: https://www.youtube.com/watch?v=WEI6TBwydqs
[16:39:50] <JT-Shop> carp
[16:40:21] <FinboySlick> andypugh: Thanks :)
[16:50:01] <JT-Shop> andypugh: sorry wasn't paying attention http://paste.ubuntu.com/23529576/
[16:50:14] <JT-Shop> you want key word args I think
[16:51:44] <andypugh> JT-Shop: Someone on #python sptted the problem. I add a “self” do “def” out of habit, and that wasn’t right in the case of G7x
[17:27:11] <Deejay> gn8
[18:25:13] <enleth> pcw_home: what was the reason behind isolating the spindle interface on 7i76E separately from both the logic and field I/O sections?
[18:26:17] <pcw_home> VFDs tend to be fairly nasty noise sources
[18:27:33] <pcw_home> in this case we needed a separate PWM --> analog output circuit anyway so its just a matter of adding the PWM OPTO
[18:44:41] <enleth> pcw_home: ok, I was wondering if that's something "mundane" like this, or catering for weirdo piece of shit VFDs like the one I pulled from the ORAC, with all the control lines (including the analog input) having their ground directly connected to 230VAC
[18:46:39] <enleth> I suppose this would, in fact, work with that VFD as a side effect if I hadn't thrown it away, but I sure hope there weren't too many more designs like this
[18:47:00] <enleth> and with luck, there are none newly made nowadays
[18:47:07] <pcw_home> Yeah it can also be used for some of the cheap KBIC type PWM/SCR controllers that are not isolated but its not really recommended
[18:49:42] <pcw_home> ( its isolated to 2500V RMS but really not safe to have line voltage so close and mixable with LV signaling )
[18:50:34] <enleth> well, duh
[18:52:24] <Duc_mobile> enleth: how is the bridgeport conversion going
[18:52:34] <enleth> anyway, after 3 weeks of waiting for the bloody customs agency to clear the package, I've received my 7i76E today which means I can continue to work on ORAC while I wait for some more Bridgeport parts to arrive
[18:52:50] <enleth> Duc_mobile: I'm halfway through tuning the new servo drives properly
[18:53:17] <enleth> Duc_mobile: and likewise through the new servo power supply
[18:53:20] <Duc_mobile> enleth: which step servo units did you end up with?
[18:53:54] <enleth> Duc_mobile: I'm using the original SEM brushed DC servos with some used Copley Controls 423 drives off ebay, and 7i77
[18:54:24] <enleth> the drives cost me $15 a piece, got 4 of them so I have one spare just in case
[18:54:58] <Duc_mobile> thats not bad
[18:55:00] <Duc_mobile> cheap
[18:55:26] <Wolf_> wish the ones i’m going with were that cheap
[18:55:31] <enleth> they are a bit oldschool as far as configuration goes, with an internal DIP header that accepts discrete resistors and capacitors
[18:56:03] <enleth> but the tuning procedure is in fact rather straightforward if you have a scope and a signal generator (or a scope with a signal generator)
[18:56:12] <t12> |win 12
[18:56:30] <t12> er
[18:57:05] <Duc_mobile> was that a QC spindle or cat 40 on that thing
[18:57:11] <enleth> QC30
[18:57:57] <enleth> I'm still not entirely decided on whether I should have this spindle bored through to accept ISO30/NMTB30 with a drawbar
[18:58:45] <Duc_mobile> I would like to since Ive had tooling come loose a few times
[18:58:47] <enleth> but for now it's QC30
[18:59:06] <enleth> are you using a proper wrench or just tightening the locknut by hand?
[18:59:15] <enleth> it seems to be important
[18:59:24] <Duc_mobile> using the factory wrench
[18:59:37] <enleth> well, that's a bit weird
[19:00:07] <Duc_mobile> I think I need to have all the tool holders ground to use the same distance from the taper to shoulder
[19:00:14] <enleth> have you ever taken the locknut's internal mechanism out to examine it?
[19:00:34] <enleth> it is somewhat prone to cracking if it's worn out
[19:01:13] <enleth> if it's cracked near one of the lower relief pockets, it won't maintain a proper grip, but you can't see the crack until you pull it out of the nut
[19:01:36] <Duc_mobile> I have once before but I probably should do it again and see if it is still adjusted correct
[19:02:27] <enleth> Duc_mobile: is it the original old nut?
[19:02:42] <Duc_mobile> yes it is. I do have a spare nut also
[19:03:33] <enleth> disassemble both and compare the wear of the current one's mechanism to the spare
[19:04:12] <Duc_mobile> Ill do that this weekend since all projects are on hold since I may have a new house soon
[19:04:13] <enleth> it might also be that the pull ears on the old one are bent and/or worn away into a slope on their upper surface
[19:04:29] <enleth> that's also something you can't see without taking it apart
[19:05:04] <Duc_mobile> Ill need to fix it since I have about 55 tools for the thing
[19:05:26] <enleth> and the last thing - before you install the spare, take full measurements of the pull ring or whatever the internal part should be called
[19:05:41] <enleth> first, you might need another spare someday and they're kind of hard to come by
[19:06:51] <Duc_mobile> mine is probably worn since I think it is original to the machine
[19:07:10] <enleth> I mean the spare one
[19:07:17] <enleth> or are both used?
[19:07:51] <enleth> second, I'd be extremely grateful for a dimensioned sketch of that part because I might have ground mine down for a wrong shoulder thickness, which means I'd need another one as soon as I buy actual proper QC30 tooling instead of modified chinese ISO30
[19:07:54] <Duc_mobile> http://www.thebionicuniverse.com/wp-content/uploads/2015/04/QC-Locknut-Instructions.pdf
[19:09:24] <enleth> I mean the internal rotating part that the locking ears and extraction ears are on - it's quite complex
[19:09:26] <Duc_mobile> both are used
[19:09:31] <enleth> oh well
[19:09:38] <Duc_mobile> I can get some measurements worked up from it
[19:10:10] <enleth> I had one made from the original part I had, but it was so badly worn out that some of the dimensions used to make a new one had to be guesswork basically
[19:10:39] <enleth> and I had to do some grinding to get it right with the chinese ISO30 faceplate
[19:10:54] <enleth> s/faceplate/flange/
[19:10:57] <enleth> which I think is thinner than a proper QC30 flange
[19:11:11] <enleth> so it might not grip a proper holder anymore
[19:11:19] <Duc_mobile> the spare one has some of the blackoxide removed but doesnt look worn out
[19:11:46] <Duc_mobile> never mind there is a small step in the ears
[19:12:36] <Duc_mobile> wonder if I can tig weld them up before regrinding
[19:12:42] <enleth> does it look like there's some of the original thickness left on them?
[19:12:56] <enleth> I wouldn't bet on that - this is spring steel
[19:13:05] <enleth> welding it might do funny things to its springiness
[19:13:39] <enleth> like causing it to crack very soon
[19:14:30] <Duc_mobile> there is a small step of original material
[19:14:45] <Duc_mobile> and dam one just sold on ebay for 15 dollars
[19:14:57] <enleth> anyway, if you cared to take enough measurements of that part to model it properly, it could come in handy for us, or someone else out there
[19:15:17] <enleth> I could volunteer to actually model it from a freehand drawing
[19:16:39] <Duc_mobile> Ill try to knock it out this weekend.
[19:17:51] <enleth> it's simple enough that any job shop worth its name should be able to produce it, but you *have* to know the original dimensions well, especially the relative distances of ear surfaces and the outer faces of the whole part
[19:18:50] <enleth> worst case, it won't extract the tools properly
[19:19:05] <enleth> I mean, that's the worst case you want, instead of not gripping properly
[19:19:26] <enleth> mine doesn't, so I keep a wooden mallet close to the mill
[19:20:13] <Duc_mobile> same with me. small tap on the side of the tool holer to get it to pop out
[19:20:39] <enleth> for some reason, the drill arbor pops out by itself most of the time, while the ER16 holder doesn't
[19:20:41] <Duc_mobile> http://www.ebay.com/itm/Quick-Change-Tool-Holder-R8-Master-Bridgeport-Fits-NMTB-30-Erickson-Collets-/322329180492?hash=item4b0c50f94c:g:RFMAAOSwLF1YA-Bj
[19:20:54] <enleth> even though they are identical as far as the flange goes
[19:21:15] <Duc_mobile> but its the distance from the seating in taper to the flange
[19:21:37] <enleth> and thickness of the extractor ears
[19:22:13] <enleth> $250 is probably more than Kennametal wants for a new one
[19:22:32] <enleth> I mean, a new locknut, not an adapter
[19:22:36] <Duc_mobile> yea
[19:23:18] <Duc_mobile> when you took about your QC nut which screws were in the unit
[19:24:12] <enleth> AFAIR there were more than 3 for some reason
[19:24:52] <enleth> all of them with the hex holes stripped round
[19:25:06] <enleth> I had to use a dremel to cut a slot for a flathead screwdriver in all of them
[19:25:14] <Duc_mobile> mine have 3 holes. but spare has 2 long and 1 short
[19:26:46] <enleth> I'm pretty sure I've got two short screws in to indicate the position of the keys when fully open, and one long to actually keep the thing together
[19:27:55] <Duc_mobile> there is a internal snap ring also
[19:28:09] <enleth> easy to remove with a small flat screwdriver
[19:28:58] <Duc_mobile> yep I will need to grind one of the screws
[19:30:21] <enleth> BTW, this screw configuration I described is actually easier to use than what they suggest in the manual - if you see a single screw with only holes adjacent, it's definitely the key; if you see two adjacent screws, grab the locknut and feel the back side to see which of the two has the third one on the opposite side, that's the other indicating one
[19:30:44] <enleth> might have something to do with the fact that the cadmium plating was completely gone on mine
[19:31:08] <Duc_mobile> mine are on equally spaced apart
[19:31:48] <enleth> you should still be able to install the indicating screws opposite each other
[19:33:02] <Duc_mobile> found a picture similar to mine
[19:33:04] <Duc_mobile> http://www.practicalmachinist.com/vb/attachments/f38/11009d1237159629-please-identify-bridgeport-spindle-p3100028.jpg
[19:34:42] <enleth> oh, OK, so there are only 3 holes
[19:34:50] <enleth> there are 6 on mine
[19:35:22] <enleth> which seems to allow for more finely tuned adjustment of the open position stop screw
[19:35:31] <Duc_mobile> ah
[19:36:36] <Duc_mobile> do you have 2 screws that protrude into the ring area?
[19:36:47] <enleth> no, just one
[19:36:59] <enleth> the other two indicate key positions only
[19:37:22] <enleth> it doesn't seem to be necessary to have two long screws installed anyway
[19:38:05] <enleth> they don't do anything other than stopping the outer ring against a key when you loosen it up by hand
[19:38:09] <Duc_mobile> ah
[19:38:22] <Duc_mobile> Ill need to look on my mill to see whats on there
[19:38:37] <enleth> it's not like they need to hold up against any real torque
[19:41:00] <Duc_mobile> pulling the one off the mill now
[19:43:11] <Duc_mobile> I really need that bigger shop
[19:45:51] <Mario_> Hello dears.
[19:46:08] <Mario_> Any G code expert ?
[19:46:10] <Mario_> #!/bin/bash voltage=$1 feedrate=$2 halcmd setp thc.voltage $voltage halcmd setp thc.feedrate $feedrate exit 0
[19:46:26] <Duc_mobile> uh/
[19:46:27] <Mario_> What is the symbol "$" used for ?
[19:47:25] <Duc_mobile> not sure
[19:47:30] <archivist> you are asking about a shell script problem not gcode
[19:47:49] <Duc_mobile> enleth: well my keys were well worn and need to flipped for one. and one was loose
[19:47:55] <archivist> g code has no knowledge of bash
[19:47:57] <Mario_> This is a sheel script.
[19:48:06] <Mario_> Sorry,
[19:48:45] <Mario_> So Its defining a M code using shell script.
[19:48:57] <archivist> but the answer is parameters to the script
[19:49:17] <Mario_> Its on page 295 of linux docs: http://linuxcnc.org/docs/2.7/pdf/LinuxCNC_Documentation.pdf
[19:50:09] <Mario_> Ok. The "$" symbol is used by the shell script to transfer the parameters to voltage and feddrate.
[19:50:41] <Mario_> Oh man. More 450 pages i will finish reading this linux cnc doc.
[19:51:39] <Mario_> Thanks a lot archvist. I appreciate your help.
[19:52:50] <Wolf_> I have tired reading that manual…
[19:53:53] <Wolf_> my cnc machine doesn’t seem to care that I gave up
[19:54:13] <archivist> reference manual, find what you need to know rather than a full read
[19:55:24] <Mario_> I didnt know nothing about gcode, HAL, UI's so i tough reading the docs will be worthy.
[19:55:31] <Wolf_> IIRC a lot of things are kinda over explained in there
[19:55:49] <Wolf_> from a just wanting to run the machine aspect
[19:57:17] <Mario_> I have to make a hotwire to cut styrene blocks. Did anybody did it before ?
[19:57:41] <archivist> yes
[19:57:41] <Wolf_> probably
[19:57:43] <Mario_> I could speed up this machine if i find a sample configs file somewhere.
[19:58:22] <Duc_mobile> enleth: the wear looks similar between the two units
[19:58:45] <archivist> Mario_, google linuxcnc hot wire
[19:58:50] <Mario_> I nedd to cut styrene blocks to fit inside shipping containers walls as insulation.
[19:59:52] <archivist> look at the 4 axis config for hot wire
[20:00:43] <Mario_> Hot wire are 4 axis not just 2 axis ?
[20:02:05] <archivist> to get odd shapes you might want 4
[20:02:25] <archivist> angular cuts
[20:03:01] <Mario_> Right.
[20:04:46] <Mario_> Ths a lot archivist. I will take a rest now. HAve a good night everybody