#linuxcnc | Logs for 2013-03-25

Back
[01:23:24] <LeelooMinai> http://i.imgur.com/4dWrzWP.jpg <- I am trying to figure out what screws are expected to be used with this part. the holes are 6.6mm... M6? THough 0.6mm extra seems like a loose fit to me - is it supposed to be like that?
[02:12:46] <Loetmichel> mornin'
[02:52:20] * LeelooMinai ponders what size of screw for general CNC frame assembly. M6 M8? It's a 50cm by 50cm footprint fixed gantry frame
[02:52:34] <LeelooMinai> what size of screw to use*
[02:53:13] <LeelooMinai> Not sure if there are any rules for that
[03:00:16] <mrsun> LeelooMinai, isnt it for adjustability ?
[03:00:22] <mrsun> once you torque them down it shouldnt move :P
[03:00:39] <LeelooMinai> You mean my previous question?
[03:00:47] <mrsun> the hole size M6 6.6mm
[03:00:49] <mrsun> :P
[03:00:49] <LeelooMinai> Well, not sure... that's why I asked.
[03:01:07] <mrsun> as its all in mm i guess its metric so
[03:01:12] <mrsun> M6 would be closest :P
[03:01:18] <mrsun> else 1/4"
[03:01:22] <mrsun> 6.35mm
[03:01:24] <mrsun> closer but
[03:01:31] <mrsun> imperial :P
[03:01:40] <LeelooMinai> Well, I know that much:)
[03:01:57] * LeelooMinai looks around a bit worried
[03:02:07] <LeelooMinai> What is this channel about anyways? :)
[03:02:18] <mrsun> everything! :P
[03:02:40] <LeelooMinai> A lot of people, but it's always dead when I look in here - weird.
[03:02:45] <mrsun> hehe =)
[03:03:03] <mrsun> what timezone? :P
[03:03:04] * archivist denies being dead
[03:03:25] <LeelooMinai> I eam Canada east coast - it's 3:50 am here
[03:03:29] <mrsun> archivist is one of those who doesnt seem sleep
[03:03:41] <mrsun> almost always when you ask something he will answer .. whatever time ;P
[03:04:26] <LeelooMinai> I am building a CNC and was searching for mythical channel with a lot of people talking all the time about practical aspects of it... I guess I was naive:p
[03:05:09] <mrsun> LeelooMinai, this is it ... its not mythical but alot of people who knows a hell of alot =)
[03:05:36] <mrsun> except me then .. im just absorbing all others knowledge :P
[03:06:06] <LeelooMinai> Maybe I am few years late or something - everyone talked about everythink already and they just sit with gray beards and muse about deep things
[03:06:29] <archivist> hmm beard, yes
[03:06:41] <mrsun> talking about shopdogsam ? ;P
[03:07:44] <DJ9DJ> moin
[03:08:55] * LeelooMinai sighs
[03:09:09] <LeelooMinai> Oh well - I just hope I will not build some mechanical abomination
[03:10:09] <archivist> I did that already :)
[03:10:41] <LeelooMinai> What? Forgot one of the axis? :p
[03:10:47] <archivist> http://www.archivist.info/cnc/stage5/P2180014.JPG
[03:11:14] * LeelooMinai blinks
[03:11:51] <archivist> http://www.archivist.info/cnc/stage7/
[03:12:28] <archivist> made from "available parts"
[03:12:49] <LeelooMinai> I was just going to ask if you robbed some Japanese robotic car assembly plant:)
[03:13:37] <LeelooMinai> It looks very... random
[03:15:11] <archivist> the column is from a lathe, the spindle is its headstock
[03:15:56] <LeelooMinai> Hmm... I think I have a different plan:)
[03:16:04] <archivist> xy from a tool setters jig
[03:17:22] <LeelooMinai> I am sure it's practical, but it looks like some mad-man creation:)
[03:19:12] <archivist> designed for small gear like parts http://gears.archivist.info/
[03:19:44] <mrsun> haha i love that mill archivist ... :P
[03:21:20] <LeelooMinai> Such a monster machine, for such tiny gears...
[03:22:19] <archivist> machines do need to be solid and rigid for metal cutting
[03:22:41] <LeelooMinai> Right, I know... but I think you took it to some new level:)
[03:23:33] <archivist> the column before I added the angle to it was far too flexible
[03:25:09] <LeelooMinai> You mean that huge yellow thing in the middle was flexible?
[03:28:33] <archivist> very
[03:30:07] <archivist> it was very poor in resisting twist
[03:30:31] <archivist> because it was just a channel section
[03:31:16] <LeelooMinai> Aha, well, that motor there seems kind of big
[03:32:31] <Valen> ok at some point in the past I machined up a soldermask
[03:32:47] <Valen> I had a gerber from kicad
[03:32:47] <archivist> the motor is 1/2hp
[03:33:13] <Valen> and using only open source tools I cut a solder stencil out of an aluminium can
[03:33:22] <Valen> any ideas how the hell i did that?
[03:33:35] <Valen> pcb2gcode with a negative offset is not the answer it seems
[03:34:02] <LeelooMinai> Valen: You ask others how you did that? :)
[03:40:02] <archivist> Valen, in your irc logs ?
[03:40:24] <Valen> could be, i know i boasted about it but i'm not sure if i detailed it lol
[03:41:32] <archivist> problem is how long ago too
[03:42:53] <archivist> my bot logs #kicad and in here to the same table :)
[06:41:43] <jthornton> if I have 4 inputs, one for each position is there a component that will output 1 - 4 based on the inputs?
[06:53:22] <archivist> what should happen if two pins are set
[06:53:48] <jthornton> in this setup that is not possible
[06:54:03] <jthornton> there are 4 proxes for the turret and one target
[06:54:16] <jthornton> so as it rotates only one at a time can be on
[06:57:01] <jthornton> maybe I'm trying to solve the problem the wrong way too
[06:59:01] <archivist> dont you already have ladder logic for turning the turret
[06:59:35] <jthornton> yes, but mine has an absolute encoder and my logic is based on that
[07:00:20] <jthornton> I see a way to do this in ladder now
[07:00:21] <archivist> so the encoder already has your number in some form
[07:00:32] <jthornton> yes
[07:00:48] <archivist> dunno what you are really trying to do
[07:01:58] <jthornton> lift and spin turret until input matches tool request number then down and lock
[07:05:49] <jthornton> http://imagebin.org/251550
[07:05:57] <jthornton> I think this will do what I want
[07:07:05] <archivist> just need to compare the right pin with request rather than make a number
[07:55:38] <cncbasher> JT:see the component Arceye did for my toolchanger
[07:55:53] <cncbasher> and the modified iocontrol
[07:56:02] <jthornton> does yours have a prox for each station?
[07:56:11] <cncbasher> grey encoder
[07:56:27] <jthornton> ok this one has a prox for each station
[07:56:32] <cncbasher> 3 optos
[07:56:40] <jthornton> so I think it is simple to do in ladder
[07:56:45] <cncbasher> easy to modify i would have thought
[07:57:32] <cncbasher> it also feeds back to axis via iocontrol so real position feedback based on encoder
[08:55:26] <JT-Shop> thanks for the files, rather clever that ArcEye
[08:55:41] <JT-Shop> this one is very simple to do now that I think about it
[08:56:28] <jdh> you must 'know' more dw33bs than I do.
[08:56:37] <jdh> <wrong channel, probably>
[08:56:52] <JT-Shop> sounds plausable
[08:57:13] <alex_joni> JT-Shop: what ArcEye?
[08:57:24] <JT-Shop> on the forum
[08:57:27] <alex_joni> ah
[08:57:34] <JT-Shop> hi Alex
[08:57:43] <alex_joni> I see now it's not related to welding arc's ;)
[08:58:06] <JT-Shop> he may be I don't know :)
[08:58:13] <alex_joni> heh
[08:58:26] <alex_joni> sounds like a THC of some sorts :D
[08:58:42] <alex_joni> all well on your end?
[08:58:57] <JT-Shop> yes
[08:59:00] <JT-Shop> how about you?
[09:00:10] <alex_joni> yup, all's well
[09:00:39] <JT-Shop> did the user map die with the upgrade or other technical issue?
[09:02:29] <alex_joni> yup
[09:02:38] <alex_joni> the map component was for joomla 1.0
[09:02:49] <alex_joni> couldn't find a worthy replacement
[09:05:26] <JT-Shop> thanks, I was just wondering
[09:17:15] <cradek> it was sure nice back when it worked.
[09:18:51] <cradek> my My Profile has my latitude and longitude. does it use it for anything?
[09:19:31] <cradek> the User List is ridiculous and useless
[09:20:48] <cradek> "show users within N miles from me" would be useful
[09:33:07] <JT-Shop> I wrote a php mysql that generated a kml file that can be loaded with google maps
[09:34:46] <JT-Shop> http://maps.google.com/maps?q=http://gnipsel.com/spyder/test4.kml
[09:56:07] <alex_joni> JT-Shop: now all is left is to integrate that with CB
[09:57:42] * jthornton wonders how you do that?
[10:10:13] <JT-Shop> I did read that if you have more than 200 data points in the kml file the load time starts to slow down
[10:10:38] <alex_joni> JT-Shop: this looks promising: http://tech.reumer.net/Google-Maps/Documentation-of-plugin-Googlemap/integration-with-cb.html
[10:12:37] <alex_joni> but I can't look at it in depth right now
[10:12:56] <alex_joni> and we don't want to start hacking linuxcnc.org again, since cradek just made it secure a short while ago ;)
[10:14:49] <JT-Shop> looks good
[10:15:06] <JT-Shop> alex_joni: have you ever ran the database sync?
[10:15:19] <JT-Shop> it reports a zillion records out of sync
[10:28:16] <alex_joni> JT-Shop: which one?
[10:31:27] <alex_joni> CB Tools: Check Users Database?
[10:34:17] <JT-Shop> yes
[10:36:16] <alex_joni> I did run the sync a long time ago
[10:36:21] <alex_joni> but it shouldn't really matter
[10:36:32] <alex_joni> as we handle all logins/stuff from CB directly
[10:36:39] <alex_joni> so the CB database is what matters
[10:37:03] <JT-Shop> ok I didn't know that's why I asked while you were here
[10:46:03] <alex_joni> I'm always here :D
[10:50:31] <pcw_home> sounds kinda spooky, like the gospel of St T
[10:50:33] <pcw_home> split a piece of wood and I am there lift a stone and I am there
[11:06:45] <jdh> St T uses screen(1) ?
[11:21:19] <cradek> R2E4_: http://www.mattshaver.com/v2e3/index.htm
[11:21:34] <cradek> very early retrofit of a machine similar to ours
[11:22:04] <cradek> if I understand right, this was the first EMC retrofit outside NIST
[11:25:09] <L84Supper> wow, so old the pictures of it on the internets are fading
[11:25:36] <cradek> those are likely polaroids placed on a color scanner
[11:26:20] <cradek> there were a few years where we all used scanners before we had digital cameras :-)
[11:26:58] <L84Supper> I found an old one of mine in a box the other day
[11:27:14] <cradek> maybe not - each of those images has 652 as its largest dimension. might have been an actual camera?
[11:27:29] <cradek> I still regularly use a scanner
[11:28:02] <L84Supper> we have to for docs that require an actual signature
[11:28:28] <L84Supper> cradek: what year was that retrofit?
[11:28:32] <cradek> I use mine mostly in place of a photocopier
[11:28:47] <cradek> L84Supper: probably around 99? it was before my time.
[11:29:20] <cradek> (mshaver sometimes shows up here)
[11:29:30] <jdh> I've had to fax documents the last few weeks. It's amazingly annoying
[11:30:19] <cradek> I've lost that capability (from home) in the last two years - no landline anymore
[11:31:19] <jdh> I have a semi-landline. All the house phones got to a cell box thing
[11:32:53] <pcw_home> just need an acoustic coupler on your fax machine...
[11:33:52] <cradek> pcw_home: remember those terrible cordless phones? I used to use one on my acoustic modem. weird overlap of technologies.
[11:34:26] <pcw_home> That is funny
[11:34:55] <cradek> it wasn't round, so it didn't seal in the cups. very annoying.
[11:35:28] <cradek> oh I miss those old round handsets that you could actually hold with a shoulder comfortably
[11:35:49] <jdh> my first one was a big blocky thing with square speaker/mic and a long pull out antenna
[11:36:39] <cradek> yep that's the era. they worked so badly, but we thought they were cool. ha, just like today's phones!
[11:36:56] <pcw_home> I still have a couple big blocky Motorola cell phones
[11:37:36] <cradek> do you still have analog service in your area? that actually worked pretty well (for the couple hours your battery lasted...)
[11:37:59] <jdh> hasn't all AMPS been gone for years?
[11:38:20] <alex_joni> cradek: "A badly scanned picture showing the front of the mill with the keyboard and monitor supported by a monitor arm."
[11:38:31] <pcw_home> I think its gone here (I have not even tried turning them on for years)
[11:39:46] <alex_joni> cradek: I got one of these for christmas http://www.geek.com/wp-content/uploads/2009/04/yubztalk.png
[11:39:57] <L84Supper> I'd like to make a smart phone with a rotary dial :) just for fun and old time sake
[11:40:31] <alex_joni> it's pretty fun to walk down the street talking into that
[11:41:07] <L84Supper> make it look like a princess phone
[11:42:26] <cradek> alex_joni: that's silly, but it would be useful if you were forced to use a cell phone to do actual work at a desk
[11:42:41] <alex_joni> I use a handsfree for that
[11:43:04] <cradek> those are obnoxious to the caller and also everyone around you :-/
[11:43:10] <cradek> I hate them
[11:43:50] <alex_joni> why is that?
[11:44:29] <L84Supper> some of those people aren't actually on hands-free phones but are schizophrenic
[11:44:29] <cradek> they sound echoey, and most don't let you both talk at once
[11:44:39] <alex_joni> mine's this one: http://www.sonymobile.com/global-en/products/accessories/mw600/
[11:44:40] <cradek> and everyone around you hears the conversation
[11:45:00] <alex_joni> probably we're talking about different things then
[11:45:11] <cradek> oh you mean a headset type thingy
[11:45:31] <cradek> sorry, handsfree to me meant speakerphone
[11:45:33] <alex_joni> bluetooth thingie
[11:45:58] <alex_joni> ah, I don't get why they still allow those functions (speakerphone)
[11:46:04] <alex_joni> it's not really usable anyways
[11:46:28] <L84Supper> I asked about this other day, what application would you build on for 5-axis CAM? PyCAM or??
[11:46:37] <alex_joni> depends on the budget
[11:46:51] <cradek> what do you mean build?
[11:47:09] <archivist> L84Supper, did you see that OOM crash mentioned on the mailing list
[11:47:22] <L84Supper> cradek: expand it to support 5-axis
[11:47:55] <L84Supper> archivist: nope
[11:48:48] <cradek> I have no idea...
[11:50:44] <archivist> L84Supper, the thread was [Emc-users] potrace output to scalable gcode. Exists?
[11:51:29] <archivist> http://osdir.com/ml/emc-users-enhanced-machine-controller/2013-03/msg00329.html
[11:51:39] <archivist> 11 hours
[11:53:35] <alex_joni> L84Supper: ask in #cam
[11:56:50] <L84Supper> http://code.google.com/p/opencamlib/ heeksCNC used those libs
[12:03:38] <alex_joni> this is interesting: http://sourceforge.net/apps/mediawiki/free-cad/index.php?title=6-Axis_Robot
[12:31:34] <c60> morning all, anyone had any experience hooking up a pmdx-107 vfd spindle controller with linuxcnc, I'm using a pmdx-126 too.
[12:32:15] <c60> I've managed to get it working in test mode, but it only briefly runs when I hit the stop button in linuxcnc, doesn't start when I hit start
[12:35:24] <DJ9DJ> re
[13:03:57] <IchGuckLive> hi all B)
[13:06:41] <L84Supper> alex_joni: thanks I hadn't noticed the whole section on robots before http://sourceforge.net/apps/mediawiki/free-cad/index.php?title=Robot_Workbench
[13:32:39] <tjb1> Might be a Haas Service Engineer :D
[13:34:22] <IchGuckLive> O.O
[13:35:09] <IchGuckLive> tjb1: i worked trou a truckload of sheeds with the powermax 45
[13:35:23] <IchGuckLive> used 25 nozzels
[13:35:33] <tjb1> Wow
[13:35:45] <Loetmichel> hmmm, has anyone this PC in use for linuxCNC? have some issues with jitter and low performance when usinf vnc to supervise the mill.. -> http://www.pollin.de/shop/dt/MTM5ODkyOTk-/Computer_und_Zubehoer/Hardware/Komplett_Rechner_Barebones/Desktop_PC_FUJITSU_SIEMENS_ESPRIMO_E5600_XP_COA.html
[13:36:03] <tjb1> IchGuckLive: Do you have height control?
[13:36:22] <Loetmichel> ... would it be a wise idea to get an Athlon 64 x2 3800 cpu for it?
[13:36:34] <Loetmichel> or better invest in a new bigger PC?
[13:36:43] <IchGuckLive> Loetmichel: i use P5915
[13:37:37] <IchGuckLive> http://www.ebay.de/itm/Fujitsu-Esprimo-P5915-Intel-Celeron-440-2-0-GHz-2-GB-RAM-160-GB-HDD-/350696742364?pt=DE_Technik_Computer_Peripherieger%C3%A4te_PC_Systeme&hash=item51a7279ddc
[13:37:52] <IchGuckLive> 55Euroas
[13:38:28] <IchGuckLive> tjb1: no i dont i use M200 to get height regulated at some places
[13:38:45] <tjb1> are you setting pierce height?
[13:39:10] <IchGuckLive> tjb1: i start plasma at Z3.8 as in the datasheet F500 to Z1.3 at P1 sec
[13:39:22] <tjb1> What is m200?
[13:39:25] <Loetmichel> IchGuckLive: that should help me in which way to determine if the jitter problems will be gone with a bigger dualcore cpu on mine?
[13:39:46] <IchGuckLive> Loetmichel: i got 16k of jitter
[13:40:12] <IchGuckLive> tjb1: m200 is my Zero Z G92
[13:40:53] <tjb1> Loetmichel: The Intel D525MW does pretty good with linuxcnc
[13:41:01] <IchGuckLive> detecting on G38 set the parameter G92 in my case Z-3
[13:42:18] <tjb1> Have you spoken with Jim Colt?
[13:42:26] <tjb1> That seems like a bunch of nozzles
[13:42:51] <IchGuckLive> 120 sheeds 240 sqm of material
[13:44:02] <IchGuckLive> tjb1: did you experiment with nozzel 1mm
[13:44:14] <IchGuckLive> the industrial ones have 2.2
[13:44:26] <tjb1> nozzle 1mm?
[13:44:38] <IchGuckLive> for pricise small cut
[13:44:54] <tjb1> height or hole diameter?
[13:45:09] <IchGuckLive> hole diameter
[13:45:31] <IchGuckLive> Loetmichel: querry ?
[13:45:33] <Connor> tjb1: the Intel D525MW is our of production...
[13:45:58] <tjb1> ? You can still buy it
[13:46:10] <Connor> Where? Only place I found was Ebay.
[13:46:16] <tjb1> IchGuckLive: I didnt know you could buy different sized nozzles
[13:46:25] <IchGuckLive> yes
[13:46:54] <IchGuckLive> but i got a cnc lathe with 2 turrets B)
[13:47:19] <tjb1> Amazon has them Connor
[13:47:31] <Loetmichel> tjb1: i have the system already running
[13:47:43] <Connor> Looks like all the stock got dumped to Ebay and non mainstream distributors..
[13:47:43] <tjb1> Well not really "Amazon" but they arent much more than I paid for it
[13:47:55] <Loetmichel> just a little much jitter (7k++) and some performanve issues with webcam and vnc ;-)
[13:47:57] <tjb1> Did they say why they killed it?
[13:48:01] <IchGuckLive> 525 in germany at reichelt.de
[13:48:15] <Connor> Not really..
[13:48:26] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=14133
[13:48:33] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=14136
[13:48:38] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=14139
[13:48:43] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=14142
[13:49:31] <Connor> Loetmichel: Where did you get the extrusion of the table top ?
[13:49:52] <Loetmichel> Connor: chinese machine
[13:50:01] <Connor> Oh, you ddn't make that one ?
[13:50:12] <Connor> I'm looking for that type of table top to replace my MDF one...
[13:50:16] <Loetmichel> complete with spindle , 4th axis and inverter
[13:50:47] <tjb1> That your thumb Loetmichel ?
[13:51:25] <Loetmichel> http://www.ebay.de/itm/321061312512 <- this machine
[13:51:33] <Loetmichel> tjb1: hmm?
[13:51:44] <tjb1> http://www.cyrom.org/palbum/main.php?g2_itemId=11136
[13:51:54] <Loetmichel> yeah
[13:52:01] <Loetmichel> little accident with the bandsaw
[13:52:12] <Loetmichel> nothing serious
[13:54:50] <Loetmichel> this thumb has seen worse... 3,0mm 2 flute TC mill bit, 24krpm... one second not "fully aware" -> http://www.cyrom.org/palbum/main.php?g2_itemId=2958
[13:55:50] <Loetmichel> nice 3.0mm hole through the nail, thumb, bone and out ...
[13:56:00] <Loetmichel> didnt even hurt ;-)
[13:56:04] <Loetmichel> (at first ;-)
[13:56:24] <tjb1> your thumb trying to kill itself?
[13:57:05] <Loetmichel> tjb1: left thumb, simply wromg time wrong place... every time ;-)
[14:03:13] <cradek> eeeee
[14:06:54] <Loetmichel> sorry, cradek, should have flagged that as "gory" :-)
[14:08:09] <cradek> eh it wasn't gory at all.
[14:08:40] <cradek> hope it was a fast drill cycle so it pulled back out...
[14:09:24] <Loetmichel> ... i tend to be a little careless with myself... one time my apprentice came into the workshop. sees me, draws his cellphone: *KLICK* " just for the book: 'how to NOT do it!'" ;-)
[14:09:53] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=12244
[14:11:02] <Loetmichel> ... he then helped me to extiguish the trashcan (below my knees) where i hat thrown 2 sheets of alcohol-soaked towels ;)
[14:14:14] <Loetmichel> no, it wasnt drill cycle, it was my owb lazyness.
[14:14:19] <Loetmichel> own
[14:14:52] <L84Supper> Loetmichel: do you work with any of the color laser printers for Tempest?
[14:14:54] <Loetmichel> machin was in par position, i attempted to change the workpiece and was too lazy to stop the spindle
[14:15:08] <Loetmichel> l49: yes
[14:15:22] <Loetmichel> L84Supper: yes
[14:15:27] <Loetmichel> sole dell, some HP
[14:15:30] <Loetmichel> some
[14:15:54] <L84Supper> Loetmichel: any pref for which is more dependable?
[14:16:27] <Loetmichel> i am afraid: when we get them on the BSI-list they are already EOL ;-)
[14:16:56] <L84Supper> Loetmichel: are any LED vs laser?
[14:17:04] <Loetmichel> last reliable and easy to shield model i had in my grasp wa a HP cp3525
[14:17:13] <Loetmichel> was
[14:17:38] <Loetmichel> i had no led printer in my fingers.. (knowingly)
[14:18:26] <archivist> OKI did a LED iirc, used a grin lens
[14:19:20] <L84Supper> OKI, Kyocera, Fuji and Xerox have the high density heads I'm interested in
[14:20:48] <archivist> we borrowed the lens for another printer project :)
[14:38:55] <c60> hrm, anyone have ideas how to setup pwm/spindle control?
[14:41:13] <r00t4rd3d> with what would be a good starter
[14:44:52] <r00t4rd3d> c60, what controller do you have?
[14:45:11] <L84Supper> c60: let us know what hardware you are using (boards, motors etc etc)
[14:45:31] <r00t4rd3d> he has a pcb and some flux
[14:45:42] <r00t4rd3d> couple zip ties
[14:45:49] <r00t4rd3d> bring it all together
[14:46:27] <c60> I'm using a pmdx-126 motherboard and a pmdx-107 for the pwm- speed control, the 107 has a test function which works and changes the speed, so wiring is hooked up right. If I turn the spindle on via linuxcnc nothing happens, but when you hit stop it turns on briefly and stops,
[14:46:53] <tjtr33> steves_logging, hmmm how to notify steve stallings when only his logger is here? well... i read you made an EDM 'pulser' is it available for purchase/open source?
[14:46:55] <c60> I'm just trying to force the spindle-pwm to be positive using absolute but not having much success
[14:46:59] <tjtr33> speaking of pmdx :)
[14:47:19] <r00t4rd3d> steve responds to email
[14:48:47] <tjtr33> r00t4rd3d, all i find is ~Steve@wsip...blahblah blah you got address?
[14:49:08] <r00t4rd3d> http://linuxcnc.org/docs/html/examples/spindle.html
[14:49:47] <r00t4rd3d> i think i got it from the domain register info
[14:50:01] <tjtr33> k thx
[14:50:25] <r00t4rd3d> c60, googling linuxcnc hal spindle control will probably help you if no one is around
[14:51:09] <r00t4rd3d> http://linuxcnc.org/lucid/emc2/index.php/italian/forum/24-hal-components/3721-spindle-control
[14:51:23] <r00t4rd3d> c60, did you set up a hal component for spindle control?
[14:51:37] <r00t4rd3d> i >think< thats what you need to do
[14:51:55] <c60> r00t4rd3d: I have that in my .hal from using stepconf setup,
[14:53:10] <c60> the thing that's stumping me is if I turn the spindle on, and make sure it's not at 0 rpm, nothing happens until I hit stop then it starts and stops just as quickly
[14:53:37] <r00t4rd3d> something in your config isnt right
[14:53:55] <r00t4rd3d> have you looked through the ini and hal files ?
[14:54:32] <c60> r00t4rd3d: ya something's not right, I've been looking through the hal file, havn't really checked the ini yet
[14:55:21] <c60> I can hear the relay click on and off when I hit the spindle on and off buttons
[14:55:42] <c60> and again, when I hit stop it turns on briefly and stops again.
[14:56:27] <c60> I kept reading about the pwm needing to be forced absolute, but i'm not sure how old that info is, and I am having problems implementing it
[14:57:04] <r00t4rd3d> cradek pcw_home archivist
[14:57:51] <r00t4rd3d> waiting for pcw to come alive is probably your best bet
[14:58:35] <c60> I might have found something, brb
[14:58:37] <tjtr33> does halmeter of halscope show the signal act as you expect?
[15:00:24] <r00t4rd3d> i need to do something with hal so I can learn that shit
[15:01:09] <r00t4rd3d> open file, jog, start
[15:01:20] <r00t4rd3d> thats my linuxcnc routine
[15:02:01] <r00t4rd3d> my spindle control is my thumb
[15:02:59] <chopper791> Hello everyone, I have a quick question (I hope). When my Z axis makes a direction change the traverse of X and Y almost come to a stop. When Z axis is not doing anything the X and Y will move along in rapid speed no problem.
[15:03:09] <chopper791> Any clues to what this could be
[15:03:29] <r00t4rd3d> are all your motors the same?
[15:03:43] <r00t4rd3d> drawing the same power
[15:04:01] <chopper791> Yes
[15:04:36] <r00t4rd3d> are they new?
[15:04:53] <chopper791> Yes all brand new
[15:06:36] <chopper791> Is there a mode that I could have set EMC2 in to make it slow down like this? I know that if I use a G64 the speeds will slow down to keep within parameters
[15:06:49] <chopper791> This happens even without G64
[15:08:37] <r00t4rd3d> turn down your acceleration and velocity
[15:09:20] <chopper791> Ok.... How will that change anything(Just for my knowledge)?
[15:09:29] <r00t4rd3d> just what i am reading
[15:09:35] <r00t4rd3d> http://www.cnczone.com/forums/linuxcnc_formerly_emc2/149811-x-axis_stalling_linuxcnc.html
[15:09:51] <r00t4rd3d> post #3
[15:09:55] <r00t4rd3d> from cncbasher
[15:09:56] <r00t4rd3d> lol
[15:10:44] <r00t4rd3d> cut your accel and vel numbers in half and see if it still does it
[15:10:51] <chopper791> This poster is loosing steps. I am remaining dead nuts with no step lose issues.
[15:11:23] <r00t4rd3d> how big is your power supply?
[15:12:03] <chopper791> If I run the X and Z no problem , Y and Z no problem, X and Y then the taverse slows when Z is added to the mix.
[15:12:10] <chopper791> Power supply is 40v 10a
[15:12:48] <chopper791> The drivers will consume no more then 7.2amps
[15:13:23] <chopper791> total
[15:13:28] <chopper791> combined
[15:14:07] <r00t4rd3d> did you try turning down the speeds/
[15:14:08] <r00t4rd3d> ?
[15:15:24] <r00t4rd3d> ive only ever had weird motor issues when I had my numbers set too high
[15:15:46] <chopper791> I will give it a try
[15:16:32] <r00t4rd3d> rerun stepconf and open your existing config and just try cutting the acceleration and velocity numbers in half and see if it still stalls
[15:17:12] <cpresser> if you run something like "G64 P0.1; G0 X0 Y0 Z0; G0 X100 Y100 Z-10; G0 X200 Y200 Z0;" does it stall at x,y=100?
[15:20:29] <r00t4rd3d> finally someone who can help him more then me and he is afk :/
[15:20:44] <c60> That didn't work, hal scope looks to be reporting all positive pwm that changes as I alter spindle speed,
[15:21:05] <c60> should halui.spindle.start change ?
[15:22:18] <c60> motion.spindle-on changes to true, but halui.spindle.start stays false
[15:24:47] <tjtr33> start simple, what does halui.spindle.is-on show?
[15:27:08] <c60> It changes to true
[15:28:07] <tjtr33> next monitor halui.spindle.start and halui.spindle.stop while you ask for it to be on, then off
[15:29:40] <c60> ok, those do nothing.
[15:29:44] <c60> they stay at false
[15:30:40] <tjtr33> do you have them dependent on other signals in your hal? are they connect to physical or gui btns?
[15:31:48] <c60> Not that I know of, I have a spindle gui in the manual control interface with a stop a direction button and +/- buttons.
[15:33:58] <tjtr33> is your config a supplied config? maybe you posted it already ( or very close to a supplied config )?
[15:34:57] <c60> I did post it, It's pretty close to a supplied config, generated from stepconf, the only thing I added was to do a touchoff and a joystick control, which was awhile ago and works fine
[15:35:48] <tjtr33> something like net spindle-enable motion.spindle-on => parport.0.pin-14-out ? for the enable
[15:36:10] <tjtr33> and does halmeter/scope AND and a meter agree?
[15:37:27] <tjtr33> in the above the amplifier's enable would be physically wired to the pport pin 14.
[15:37:30] <c60> net spindle-cw => partport.0.pin-14-out
[15:37:35] <c60> is what's in mine
[15:38:03] <tjtr33> what about an enable? is you spindle amp enabled by a pin?
[15:38:29] <c60> spindle-on => parport.0.pin-17.out
[15:38:36] <tjtr33> (you could have it hot wired but that'd be scarey)
[15:38:57] <tjtr33> sorry on != enable enable sez 'ready' and on sez 'go'
[15:39:59] <c60> There's also a net spindle-on <= motion.spindle-on => pwmgen.0.enable
[15:40:29] <c60> got an idea, hold on.
[15:40:42] <tjtr33> yes that enables the pulse stream that creates differnt speeds, but amp enable is another, ok
[15:42:26] <c60> I have a spindle-on => parport.0.pin-17-out
[15:42:45] <tjtr33> and hows it look on halmeter/scope?
[15:42:48] <c60> a spindle-pwm => parport.pin-16-out
[15:42:58] <c60> and a spindle-cw => partport.pin-14-out
[15:43:22] <tjtr33> # connect "spindle on" motion controller pin to a physical pin
[15:43:23] <tjtr33> net spindle-on motion.spindle-on => parport.0.pin-09-out
[15:43:38] <tjtr33> ^^ example cfg supplied with stepper
[15:44:40] <tjtr33> take a look in the stepper config ( i'm assuming, sorry, but you seem to be missing some bits )
[15:45:15] <L84Supper> tjtr33: http://www.ebay.com/itm/ELOX-EDM-Model-12-3816-Ram-Sinker-w-Wizard-150-Controller-Astra-200D-Elox-Power-/110803455519?pt=LH_DefaultDomain_0&hash=item19cc668e1f
[15:45:26] <tjtr33> you said "motion.spindle-on changes to true" but not what its connected to
[15:45:33] <L84Supper> are those Elox units worth keeping alive?
[15:46:48] <tjtr33> L84Supper, gawd an astra, they're beasts, but run forever, thats actually a split supply, runs 2 machines, easy to find operators
[15:47:32] <tjtr33> has a tiny o-scope on each half ("" watch for grass to grow") very lo rez
[15:48:06] <L84Supper> 70's tech
[15:49:15] <tjtr33> interesting fill technique, air pressure pushed into base raises N lbs of oil into tank.
[15:49:49] <c60> does "net spindle-on <= motion.spindle-on => pwmgen.0.enable" mean put the motion.spindle-on value into both pwmgen and spindle-on?
[15:49:53] <L84Supper> I run into lots of Elox units around here
[15:50:13] <L84Supper> also Mitsubishi http://www.ebay.com/itm/Mitsubishi-Wire-EDM-FX-1-Machine-/221204161572?pt=LH_DefaultDomain_0&hash=item3380cba424 or similar
[15:50:22] <c60> and with the net spindle-on => partport.0.pin-17-out end up taking the value from spindle-on and outputting it on pin 17?
[15:52:17] <tjtr33> c60 well, yes but i'd think you'd want net "somePhysPin <= motion.spindle-on => pwmgen.0.enable" and connect somePhysPin to the enable on yuur amp
[15:52:34] <tjtr33> hmm, you added to it...
[15:53:15] <tjtr33> yes, i'd agree, and p17 should be wired to the amp enable
[15:54:12] <c60> how does halui.spindle.start relate to spindle-on
[15:55:09] <tjtr33> after the device is enabled then you can ask it to run. ( arm weapon, then trigger it (
[15:57:35] <tjtr33> if i say 'after the device is ON then you can ask it to run' you can see why ON is a bad term, and enabled is proper, ( but on is used :(
[15:58:29] <c60> this pmdx-107 is using 3 pins, 14- for direction, 16 - for pwm, and 17 can be used as a charge pump signal dip switch controled, which orignally I had it turned off, its now on, but made no difference.
[15:59:16] <c60> I changed hal to have net spindle-on => parport.0.pin-17-out but it made no difference either
[16:00:08] <c60> the part that is still throwing me for a loop is if the rpm is >100 then when I hit stop, it turns on briefly and stops, so it's like it's getting the right signals when I hit stop,
[16:01:34] <andypugh> c60: spindle-on is a "Signal" as is the first thing after _every_ net command. You can actually choose any name you want for a signal.
[16:01:47] <c60> also I hear the relay's clicking on/off when I hit start/stop but halui.spindle.start is always false
[16:01:51] <tjtr33> aint no enable on that thing. are the jumpers suited for your vfd & does the relay respond?
[16:02:33] <andypugh> c60: Sounds like you need to invert the parport "enable" pin.
[16:02:34] <c60> wireing works fine, it has a button to put just the device in test mode and the vfd spins up my spindle all good.
[16:03:46] <c60> the board can also function where it looks at the pwm, so no pwm, then off, and if there is a pwm, then on.
[16:04:30] <tjtr33> have you changed anything since this conversation began? i understand the standalone debug test is ok
[16:05:16] <c60> not really, well I change something, it doesn't work, i change it back.
[16:05:42] <c60> hmm, something in this manual about it needed a pwm freq between 25hz and 1000hz to be considered valid
[16:06:05] <c60> a 0% or 100% is considered a stop command.
[16:07:06] <andypugh> Ah, in that case perhaps the scales are messed up and 1 rpm = full scale = off.
[16:07:07] <tjtr33> ok, try a midrange PWM, then theres just 2 signals in that pdf, (3rd optional) do pin 14 and 16 respond as you expect?
[16:07:22] <andypugh> what is the pwmgen scale set to?
[16:08:14] <c60> 14 and 16 respond as expected, pwmgen scale is 20000, I have freq at 1000 right now, and offset is 0, dither-pwm is true
[16:09:19] <JT-Shop> I have a 24vdc input and output and need to count the input to some preset value like 100 then turn on the output for a second and reset the counter. The transition may be as fast as every 1/2 second so slow
[16:09:20] <tjtr33> " it needed a pwm freq between 25hz and 1000hz to be considered valid" try 500
[16:09:41] <JT-Shop> is there any single cheap chip that can do this?
[16:09:59] <c60> the freq was 50, but I just changed it, still no dice.
[16:10:01] <tjtr33> pic/avr
[16:10:13] <c60> I just tried 500 too, same thing,
[16:10:13] <andypugh> JT-Shop: Pity it's 100, not 128, or you could use a binary counter.
[16:10:37] <JT-Shop> I have to be able to set the count after calibration... it's for a tipping bucket rain gauge
[16:11:21] <tjtr33> c60 ok, when you say p16 acts as expected AND 500 doesnt work, does halscope show 500hz on p16?
[16:12:21] <JT-Shop> http://cq.cx/ladder.pl
[16:12:24] <andypugh> JT-Shop: http://docs-europe.electrocomponents.com/webdocs/0ca0/0900766b80ca0e3a.pdf
[16:12:54] <tjtr33> c60 and does a voltmeter agree ( 2.5V or 5V )
[16:13:19] <JT-Shop> andypugh: thanks
[16:13:50] <tjtr33> arduino is cheap to some ppl
[16:14:28] <andypugh> That chip is 20p. But by the time you have put in on a PCB and wrapped it in logic, the Arduino might be cheaper.
[16:15:17] <tjtr33> and can always be used for something else when you find the simpler/cheaper solution
[16:15:37] <JT-Shop> I have an Arduino to test with
[16:16:00] <c60> How do I know if the halscope is showing 500hz, and how can I get it to pause the graph, it keeps appearing and disappearing,
[16:16:35] <tjtr33> JT-Shop, it has a delay_ms() built in, but not 24V tolerance w/o opto or such
[16:17:00] <tjtr33> c60, theres a pause btn in upper right
[16:17:08] <tjtr33> maybe its called stop,
[16:17:13] <Tom_itx> JT-Shop, what you makin?
[16:17:23] <JT-Shop> tipping bucket rain gauge
[16:17:23] <Tom_itx> oh the selector for your turret?
[16:17:58] <tjtr33> wow like the japanese bamboo tumbler
[16:18:31] <JT-Shop> dunno
[16:20:42] <c60> tjtr33: Upper right where? nothing labeled as such
[16:21:35] <c60> I measured the voltage coming out of the pmdx107 which should be 0-10v depending on the pwm, and it matches based on what I set spindle speed at,
[16:21:35] <tjtr33> c60 hold on gotta fire up lcnc * ha;scope
[16:22:33] <tjtr33> c60 thats enuf, if the vom follows the spindle pot 0-10v
[16:23:50] <tjtr33> and it was top right, box labeled "run mode' , a checkbutton labled 'stop'
[16:24:46] <c60> hmm, well I gotta go, I'll have to pick this up tomorrow.
[16:25:29] <tjtr33> ok, check p14 if p16 seems ok, hi or low not open
[16:26:42] <c60> yup, p14 follows hi, then low when stoped
[16:27:55] <tjtr33> then checkk againt table 1 in section 2.4
[16:28:05] <tjtr33> best o luck!
[16:28:05] <andypugh> C60 did you try inverting the enable pin as I suggested some time ago?
[16:28:40] <andypugh> If it's an opto-enable then it is quite likely to be pull-down for on.
[16:30:11] <c60> andypugh: ya I tried inverting, same deal,
[16:30:42] <andypugh> That is strange then.
[16:30:52] <andypugh> Is the VFD common to breakout 0V?
[16:30:55] <c60> I tried changing dip switch #3 as per section 2.4 but didn't change behavior
[16:31:41] <andypugh> (Bear in mind I have no idea what hardware you are talking about)
[16:32:33] <c60> andypugh: that's helpful, the wiring is correct since it works from the board's test mode,
[16:34:34] <tjtr33> the wiring that is correct is the wiring that does not use the control side wires( linuxcnc side). in test mode it ignores the control side wiring
[16:34:47] <c60> ok, I'll try tomorrows,
[16:35:07] <c60> tjtr33: yes that's true, but all the wiring is done by the pmdx-126 motherboard, it's all prewired.
[16:35:55] <tjtr33> ok, is some of that wiring spindle specific?
[16:36:26] <tjtr33> from linuxcnc to the pmdx-126 mobo?
[16:36:38] <c60> wierd, if I play with the spindle override, when I cross to zero, the spindle turns on momentarily and then it stops
[16:36:54] <c60> and I can hear the relay's clicking
[16:37:13] <c60> so I'm guessing it has to do with the pwm signal,
[16:37:20] <c60> fresh brain tomorrow will help,
[16:37:23] <c60> thanks for all your helps
[16:37:34] <tjtr33> you can scope or meter a pwm
[16:37:56] <c60> looking does not a see-er make ;)
[16:38:10] <tjtr33> its a tail its a trunk
[16:45:43] <tjtr33> I'm awaiting 6 ac servos, the prints show UVW (power) and SD+ SD- for encoder.
[16:45:43] <tjtr33> I guess i'll circumvent the serial format puzzle and dig inside. Are Hall sensors the norm?
[16:52:07] <tjtr33> i hope there enough room to drag the encoder and hall wires out, its a 6dof puma , and space for even a few wires will be tricky
[17:09:20] <R2E4_away> Hi guys
[17:09:30] <tjb1> hello
[17:09:39] <R2E4_away> Anyone seen skunkworks around?
[17:10:43] <DJ9DJ> gn8
[17:22:25] <kwallace> gene77: I saw your name on a message when I searched for RTAI MIDI. Are you using Linux with MIDI? I'm thinking about getting a MIDI keyboard, but I want to work out some details first.
[17:37:48] <tjtr33> http://www.copleycontrols.com/motion/pdf/encoder-guide.pdf lists different serial data formats for encoders
[18:23:24] <andypugh> tjtr33: Disappointingly light on actual data formats, it seems?
[18:23:59] <tjtr33> yes, but hope springs eternal
[18:26:25] <tjtr33> i'm now trying to find where the hall effects would wire in. the 8i20 doesnt want them, seems theyre just gpio presented to bldc comp.
[18:26:25] <tjtr33> the 8i20 doesnt want them, seems theyre just gpio presented to bldc comp.
[18:26:37] <tjtr33> sorry
[18:29:06] <tjtr33> andypugh, though they do say the names of the protocols, like 'absolute a' <-- a wonderful thing to google for ;)
[18:30:38] <r00t4rd3d> So I just made this guy a sign, "Als House, Als Rules" and I put Mickey Mouse on both sides giving the middle finger.
[18:30:59] <r00t4rd3d> in oak
[18:31:17] <tjtr33> add "Als feet are sane " if he's German
[18:31:35] <r00t4rd3d> i dont think he is german
[18:31:43] <r00t4rd3d> and the sign is done
[18:33:04] <andypugh> tjtr33: Yes, the 8i20 just does what it is told, it's a truly dumb drive.
[18:33:11] <andypugh> All the magic happens in HAL.
[18:33:41] <Tom_itx> voo doo magic
[18:33:46] <andypugh> It's working well for me though.
[18:33:47] <tjtr33> hey thx for all work you've put into this ac drive bldc 8i20 stuff
[18:34:43] <andypugh> Have you seen http://youtu.be/haO5ZJq-Y3c ? That's an 8i20 and resolvers.
[18:34:47] <tjtr33> it occurs to me i might not have to thread wires thru the joints, i could go outside ( borg )
[18:34:56] <tjtr33> oh thx, will look
[18:35:49] <tjtr33> sounds great, i remember the b4 vid
[18:36:07] <tjtr33> sounds great hell, sound pro
[19:16:40] <JT-Shop> well at least after all this work on the BlueWing the CB radio seems to work :)
[19:25:07] <tjtr33> ti sensorless bldc http://www.youtube.com/watch?v=szgVUfyX8JM&feature=endscreen&NR=1 interesting, the V from whats like a current sense tap on each phase
[19:45:44] <andypugh> Interesting, but only applicable to trapezoidal commutation, and I don't think it will actually give any torque at zero speed.
[19:53:04] <tjtr33> dunno, i have yards to learn about this stuff. lead on McDuff!
[19:54:06] <tjtr33> and the IPAnema is cool, like a tethered/damped NIST robocrane
[19:56:48] <r00t4rd3d> My machine runs 1000% times smoother with the new z axis i built
[19:57:52] <r00t4rd3d> my depths are money across 30inches
[20:28:55] <Tom_itx> andypugh, you think i should give up on that motor as a spindle motor?
[20:29:41] <andypugh> Probably.
[20:29:57] <Tom_itx> it might be good for something...
[20:30:02] <Tom_itx> i'll toss it on the pile
[20:51:17] <cradek> andypugh: ooh, air cylinder plus block & tackle sounds really smart
[20:52:20] <cradek> as does bicycle chain and sprocket
[20:52:48] <cradek> although the apparent endpoint isn't fixed in space then...
[21:14:18] <andypugh> I am thinking of a pair of vertical-axis sprockets. The geometry does vary slightly as it swings round. admittedly.
[21:14:54] <Connor> andypugh: What are you building ?
[21:17:36] <andypugh> Me? Nothing. A guy on the mailing list wants to CNC a magnetic profile-following cutting torch.