Back
[01:40:14] <toner> RifRaf: yeah, I've done some with ze laser
[01:40:22] <toner> I used GearTrax, it worked quite nicely
[02:15:15] <mrsun> hmm cutting depth for CCGT inserts is specced to 0.25 - 8mm ... but for finnishing passes then or very fine corrections of like 0.02mm or something is it possible to do with those? :)
[02:17:07] <mrsun> or am i reading it wrong ...
[02:17:32] <mrsun> no cutting depth of 0.25 - 8mm
[02:19:21] <archivist> because of the way they cut one should set the last fine cut to be on size
[02:20:50] <archivist> often that very small finish cut will just drag and produce a poor finish, main cause is the work hardened skin from a previous cut
[02:24:08] <mrsun> hmm need to learn that then :P
[02:24:42] <mrsun> never got stuff so close even if i measure .. feed in etc feels like the lathes lines arent fully where they should be sometimes =)
[02:26:14] <mrsun> ofc i could check that quite easy :P
[02:27:18] <archivist> another thing to bite you that is not a lathe fault but is related to cut depth, rake and lubrication is built up edge
[02:28:54] <archivist> and a built up edge acts as a cutting edge of some random size this means you dont get what you think you should and causes a poor finish
[02:29:14] <mrsun> mm
[02:29:48] <mrsun> as i do quite a bit in alu i guess buying CCGT inserts wouldnt be a bad idea
[02:29:53] <mrsun> :P
[02:30:27] <DJ9DJ> moin
[02:31:02] <archivist> mrsun, this shows it in slow motion
http://www.youtube.com/watch?v=mRuSYQ5Npek
[02:31:20] <mrsun> yeah seen those =)
[02:31:24] <mrsun> quite beautiful =)
[02:32:35] <archivist> you can see the built up edge breaking off and going into the work and the size change
[02:38:02] <archivist> the lump at 4:20 pushing the metal forward is an example of how big in percentage terms to the tool that the built up edge can get to
[02:52:32] <mrsun> mm
[04:43:57] <Loetmichel> mornin'
[07:03:19] <halo_cast> Hello everybody
[07:05:23] <halo_cast> Is there an advantage to use Mesa cards instead of the Paralell port?
[07:07:08] <archivist> yes
[07:07:54] <halo_cast> which is ....?
[07:08:55] <archivist> list too large for irc :) but better control and more io
[07:09:00] <jthornton> everything
[07:09:21] <cncbasher> theirs no comparison .. mesa wins
[07:09:29] <archivist> there is one disadvantage...price
[07:09:41] <jthornton> encoder feedback for spindle, tons of I/O
[07:09:50] <jthornton> hard to beat free
[07:10:51] <halo_cast> hmmm ... what is a recommended "standard card?
[07:11:15] <halo_cast> for a 3 axis plasma cutter
[07:11:23] <jthornton> stepper or servo
[07:11:29] <halo_cast> stepper
[07:11:45] <jthornton> 5i25 + 7i76
[07:11:56] <jthornton> get the combo with the cable
[07:12:01] <cncbasher> 5i25 + addon card 7i76
[07:12:11] <cncbasher> arh Jt's beat me to it
[07:12:20] <jthornton> lol
[07:12:40] <cncbasher> and dont forget the torch height controler card
[07:12:50] <jthornton> THCAD
[07:14:47] <halo_cast> I already found the thcad. Do I understand correctly that the THCAD must be in the plasma housing? and the 5i25 in the PC? What about interference in this case?
[07:15:30] <jthornton> you can put the THCAD anywhere you like
[07:17:51] <halo_cast> Yes but interference on the line to the pc? There is a plasma cutter next to ...
[07:17:52] <cncbasher> i'd keep the thcad at the plasma cutter and use a sheilded multicore cable to interface , and you should be fine
[07:18:42] <cncbasher> iv'e not had any problems with any i have fitted
[07:21:55] <halo_cast> ic
[07:22:14] <halo_cast> the 5i25 cannot be used directly for the spepper output?
[07:22:21] <jthornton> me neither
[07:22:57] <cncbasher> no you need to add stepper drivers
[07:23:33] <cncbasher> all the mesa card does is give you step and direction and io lines
[07:24:12] <archivist> I think he is referring to why he needs 7i76
[07:24:30] <cncbasher> so add stepper or servo drivers and your good to go , with the approprate power supply of course
[07:24:44] <halo_cast> I understood that the 5i25 provides the outputs. what does the 7i76 do?
[07:24:45] <jthornton> and a field power supply
[07:25:20] <cncbasher> 5i25 is at the pc side , 7i76 is machine interface
[07:25:37] <jthornton> the 7i76 provides all the I/O and connections for drivers and encoders and spindle and and
[07:25:52] <archivist> and optos?
[07:27:24] <halo_cast> And why is this better than the paralell port?
[07:31:31] <jthornton> cncbasher, did you see the G code quick reference thing?
[07:32:25] <jthornton> http://gnipsel.com/linuxcnc/reference/page1.html
[07:32:44] <cncbasher> yea seen the gcode guick reference , great work i can see me using it to repair the old brain
[07:32:51] <archivist> jthornton, needs a way to select alphabetically
[07:33:14] <cncbasher> and an exit button
[07:33:31] <archivist> when that guy pasted code last night I tried to find g80...went to main site
[08:24:15] <JT-Shop> you mean like this
http://linuxcnc.org/docs/html/gcode/gcode.html
[08:25:21] <archivist> yes I had to go there to see what g80 was
[08:25:59] <archivist> one more tab with an index :)
[08:26:14] <JT-Shop> http://imagebin.org/241812
[08:27:02] <archivist> dead servo ?
[08:39:49] <L84Supper> canned cycles, before boxed cycles and even before bucket/bag-o-cycles
[08:50:18] <JT-Shop> http://imagebin.org/241816
[08:50:49] <JT-Shop> archivist: good idea on the index
[08:52:02] <JT-Shop> it's the DC motor gearbox for my ATC, it runs all the time in one direction
[08:52:04] <cncbasher> i'm guessing the other 2 wires is a brake
[08:52:09] <JT-Shop> broke yesterday
[08:52:18] <JT-Shop> that's my guess too
[08:53:16] <cncbasher> if the motor goes up to near where the relays are , i'd guess theirs one stuck
[08:53:17] <JT-Shop> except that the coupling can be turned by hand
[08:53:33] <JT-Shop> so maybe a gear shift?
[08:53:35] <cncbasher> tacho
[08:53:57] <JT-Shop> could be
[08:54:29] <cncbasher> cant be anything else other than either a tacho or brake in that configuration
[08:54:31] <JT-Shop> the motor wires go to three different connectors on the lower right side of the board
[08:54:47] <cncbasher> does the toolchanger run both fwd and back in normal use
[08:55:20] <JT-Shop> yes the motor turns a screw that moves the assb in and out
[08:55:49] <JT-Shop> about 2 or 3 turns lock to lock
[08:55:56] <cncbasher> ok then the dc wires must go to a changeover relay to reverse direction
[08:56:53] <JT-Shop> that makes sense
[08:56:55] <cncbasher> so either a relays stuck or non operational
[08:57:31] <archivist> I would be trying a power supply on the motor :)
[08:58:11] <cncbasher> if the motor runs then the motors ok , it's a switching problem
[08:58:17] <JT-Shop> Brook Crompton AC motor
[08:58:23] <JT-Shop> from UK
[08:58:30] <JT-Shop> tag is hard to read
[08:58:47] <archivist> with that name I was about to say English
[08:58:48] <JT-Shop> 240v 40w
[08:58:59] <JT-Shop> yea that too
[08:59:13] <archivist> is it a capacitor run motor
[09:01:55] <JT-Shop> the end is capped off, might have something under it
[09:02:22] <cncbasher> either a tacho or cam switch
[09:03:28] <JT-Shop> here is the gear head
http://www.maraindustrial.com/cart/motors/accessories/oriental-motor-5gk5k-gear-head.html
[09:03:31] <cncbasher> if it's ran of a seperate driver board then i'd go for a tacho
[09:03:36] <JT-Shop> made in japan
[09:04:03] <JT-Shop> must have been labeled by Brook Crompton
[09:10:11] <JT-Shop> no luck on the motor
[09:15:51] <JT-Shop> the wiring diagram
http://imagebin.org/241818
[09:16:03] <JT-Shop> looks like a fwd and rev coil
[09:16:23] <JT-Shop> and a brake
[09:18:03] <archivist> c100 the capacitor ?
[09:18:20] <archivist> and c101
[09:18:57] <cncbasher> the white will be common
[09:19:12] <cncbasher> and the blk / grey fwd or reverse
[09:19:30] <JT-Shop> the brake must be in the back of the motor
[09:20:00] <cncbasher> so if the motors running forwar all the time , then either on board relay jammed or mosfet gone
[09:20:24] <archivist> those two capacitors on the panel picture do they look bulged
[09:20:45] <archivist> far right
[09:20:45] <cncbasher> i'ld guess the blk(fwd) is forward energised coil
[09:22:04] <JT-Shop> in photo of the board there are two big black things on the right that say CAP on them
[09:22:39] <archivist> yes are they c100/101 and is one looking sick
[09:23:57] <JT-Shop> the bottom one is connected to wires 29 and 31
[09:25:38] <cncbasher> is their a mechanical switch on the tool arm operating a cam style switch by any chance
[09:26:26] <JT-Shop> no, just two proxes one for extend and one for retract
[09:27:29] <JT-Shop> I suspect one of the solder in relays on the right side of the board at this point as well as the brake seems to be broken
[09:27:39] <JT-Shop> gotta run for now
[09:27:40] <JT-Shop> thanks
[09:27:42] <cncbasher> ok if they operate ok back to the panel , then it can only be a changeover relay or
[09:27:51] <cncbasher> mosfet driver
[09:28:01] <archivist> does the motor get hot and hum, is the writing in the middle of that capacitor brown ish/burnt
[09:28:01] <cncbasher> i'd go for relay
[09:28:38] <archivist> it that cap is open it wont run
[09:28:38] <cncbasher> i believe jt says the motor runs ,,, just dont stop
[09:28:59] <archivist> shorted contacts
[09:29:29] <JT-Shop> be back after while time to take the wife to the big city
[09:29:36] <cncbasher> have fun
[09:29:56] <archivist> I probably wont be here, mains is about to drop
[09:45:21] <pcw_home> might be a couple triacs for direction switching
[09:48:35] <pcw_home> Just looked at the picture, look like a relay (if its the stuff in the bottom right)
[09:54:51] <archivist> runs one direction is also a symptom of the limit proxy not seeing it at some end
[10:35:16] <r00t4rd3d> another of my doings:
http://imgur.com/a/f9Zsh
[10:41:01] <DJ9DJ> wow, nice
[10:42:10] <DJ9DJ> what kind of wood is that, r00t4rd3d?
[10:42:30] <DJ9DJ> oak tree?
[10:51:38] <IchGuckLive> hi all B)
[10:51:38] <DJ9DJ> hi livegucker
[10:51:54] <IchGuckLive> biathlon oberhof Germany 1,3
[11:25:27] <mikegg> I've been playing around with the hal component for an arduino Uno. I've got the digital / analog output working. But the inputs don't seem to work. Anyone else fool around with this?
[11:30:35] <halo_cast> kikeegg with HAL or Arduino?
[11:31:41] <mikegg> well, both I guess. I think it's something to do with an argument to the python script for the number of outputs
[11:32:54] <mikegg> I'm not much of a programmer, takes time to sift through the python script, and the arduino code... the hal file is pretty straightforward anyway
[11:33:21] <halo_cast> I am currently starting with the arduino stuff
[11:33:35] <halo_cast> how do u read the inputs?
[11:33:39] <halo_cast> on arduino?
[11:34:15] <mikegg> http://emergent.unpythonic.net/01198594294
[11:34:51] <mikegg> the arduino reads them and passes em through serial via a python script into HAL
[11:35:01] <mikegg> *possibly
[11:39:18] <mikegg> the third bullet at the top of this page:
http://emergent.unpythonic.net/01198594294
[11:39:50] <mikegg> can anyone familiar with python clarify how to choose the partition when component is loaded?
[11:44:56] <IchGuckLive> mikegg: for arduino into python hal ?
[11:46:39] <IchGuckLive> i made this ->
http://www.youtube.com/watch?v=s5_4S24dW8w
[11:50:10] <Tom_itx> where did you get the case?
[11:50:26] <IchGuckLive> REichelt
[11:53:00] <IchGuckLive> Tom_itx:
http://www.reichelt.de/BOPLA/Herstellersuche/103/index.html?ACTION=3;ARTICLE=73243;SEARCH=BOPLA%20BS%20800
[11:53:30] <Tom_itx> nice one
[11:54:57] <mikegg> IchGuckLive: that's really slick. is the source code in the wiki somewhere?
[11:55:21] <Tom_itx> mine's not as fancy
[12:03:57] <IchGuckLive> since arduinu gone from 022 it is not working on the new chips
[12:04:38] <IchGuckLive> Tom_itx: Bopla is world wide available
[12:05:44] <mikegg> I'm using Arduino 1.0.2
[12:05:58] <mikegg> the outputs work..
[12:06:09] <IchGuckLive> mikegg: it is not working with the new arduino interface they changed the serial comunication so the com is braking up at 5min max
[12:06:21] <mikegg> ahh
[12:06:38] <IchGuckLive> and the reestablish is not working as in python for some reason on
[12:07:16] <IchGuckLive> also the i2c intern used to comunicate and store the pcf8574 chips has changed
[12:07:50] <IchGuckLive> this chips dont forget and inform the arduiono if somthin has bean pressed
[12:08:03] <IchGuckLive> a miss but it is all on wirerless gone
[12:08:54] <mikegg> well that's a bummer
[12:09:37] <mikegg> I was going to use the arduino for an analog in. I guess I will try a V-f converter like PCW suggested
[12:09:59] <skunkworks> I have been using the arduino for analog in. No issues yet..
[12:10:28] <mikegg> which model? Uno?
[12:10:37] <skunkworks> uno
[12:10:58] <Tom_itx> 8 or 16u2?
[12:11:01] <mikegg> can I take a peek at your code?
[12:11:25] <skunkworks> https://www.youtube.com/watch?v=h-CdFd2Zakc&list=UUHk52YjGT8HryRYmJKSl-lg&index=3
[12:11:49] <skunkworks> Tom_itx: I don't know..
[12:13:03] <skunkworks> mikegg: I don't have the hal files here.. But I used these directions and interface
[12:13:05] <skunkworks> http://emergent.unpythonic.net/01198594294
[12:13:38] <mikegg> yeah, that's what I was going off too
[12:14:01] <skunkworks> my arduino didn't work until I updated the firmware on it.
[12:14:11] <skunkworks> (it is in the link)
[12:14:24] <skunkworks> the serial connection was flakey
[12:14:30] <mikegg> cool, lemme check that out
[12:19:07] <mikegg> skunkworks: did you use 0022 ?
[12:27:53] <mikegg> not sure that firmware update is applicable to mine. I just got an Uno R3. trying to see what version the firmware is...
[12:30:04] <skunkworks> I don't know...
[12:30:20] <skunkworks> it has been a while.
[12:32:27] <skunkworks> I seem to remember the newer version required a small change to the sketch
[12:33:45] <skunkworks> something to do with BYTE
[12:35:04] <skunkworks> I think this was it..
http://answers.ros.org/question/38761/rospde-compile-error-byte-not-declared-in-this-scope/
[12:35:13] <mikegg> i wonder why the analog / digital read steps are outside of the while(serial.avaiable()) loop...
[12:36:36] <mikegg> i'm not getting any errors or anything. the sketch compiles / uploads fine. the vcp pops up no problem, and the outputs work ok. just the inputs are.. dead
[12:36:51] <skunkworks> oh
[12:37:08] <skunkworks> that is odd
[12:38:05] <mikegg> from the sketch, it looks like if serial.available() was always true, it would never get to the analog/digitalRead steps..
[12:40:07] <IchGuckLive> here is the layout that is in the pendand
http://www.pictureupload.de/originals/pictures/220212193424_all_boards.jpg
[12:40:24] <IchGuckLive> all milled with a 0.8mm bit
[12:55:09] <IchGuckLive> mikegg: here is my arduino code
http://pastebin.com/pSdvuLMd
[12:58:08] <r00t4rd3d> DJ9DJ, yeah oak.
[12:58:33] <DJ9DJ> very nice
[12:59:12] <r00t4rd3d> I just got a nice slab of purpleheart too :D
[13:16:38] <Loetmichel> what is purplehart?
[13:17:45] <Loetmichel> i would like to make something from ebony some time
[13:18:26] <r00t4rd3d> http://i.imgur.com/mdgIK.jpg
[13:19:34] <Loetmichel> h, i see
[13:23:03] <DJ9DJ> violinholz oder so
[13:23:39] <DJ9DJ> ups: Violettholz (D); Amarante (F); Purpleheart (GB); Purperheart (NL)
[13:24:37] <Loetmichel> amaranth
[13:24:48] <Loetmichel> is the name i know.
[13:24:51] <Loetmichel> nice stuff
[13:26:39] <IchGuckLive> im off have a nice suturday where ever you are !
[13:27:45] <r00t4rd3d> its extremely hard.
[13:27:49] <r00t4rd3d> and heavy
[13:30:02] <Loetmichel> i will buy a pices of "drowned" hardwood eventually
[13:30:45] <Loetmichel> the kind that has been under water for at least a century
[13:30:55] <Loetmichel> and is mire stone than wood
[13:30:57] <Loetmichel> more
[13:31:34] <Loetmichel> i think that would make a nice jewellery chest for my wife ;-)
[13:52:28] <ReadError> how much was that piece r00t4rd3d?
[13:55:58] <r00t4rd3d> $38
[13:59:39] <AR_> so i ordered those cheap supported linear bearing rails from china a couple weeks ago
[13:59:45] <AR_> came fine via FedEx
[13:59:59] <AR_> now i ordered a pair again, this time they sent them through DHL
[14:00:12] <AR_> DHL calls me up and says the bearings are classified under anti-dumping
[14:00:20] <AR_> and need all this bullshit paperwork
[14:00:22] <AR_> wtf
[14:00:41] <AR_> also, they sent the packing slip and the seller has my wrong address on it..........
[14:05:34] <r00t4rd3d> call dhl and start screaming at the top of your lungs
[14:06:50] <AR_> lol
[14:07:36] <AR_> they want a copy of 1040 IRS income tax return, social security number, power of atourney paperwork, etc
[14:07:38] <AR_> wtf
[14:08:37] <r00t4rd3d> tell them to just send them back then
[14:08:53] <r00t4rd3d> that will change their minds quickly
[14:10:31] <r00t4rd3d> also tell them they are novelty bearings
[14:12:04] <r00t4rd3d> for a display and they dont really contain any bearing material
[14:13:07] <r00t4rd3d> did it say what they were declared as on your packing slip ?
[14:49:41] <AR_> lol r00t4rd3d
[14:49:53] <AR_> they just put "bearing" on the description
[14:49:54] <AR_> .......
[14:50:14] <AR_> i looked at what the other seller put for the ones that came through with no problem, they were declared as "metal stick"
[14:50:15] <AR_> lol
[14:50:28] <AR_> so yeah i'm just going to say send em back
[14:53:31] <L84Supper> http://news.xinhuanet.com/english/business/2012-08/01/c_131751840.htm
[14:53:40] <L84Supper> U.S. to keep antidumping duties on tapered roller bearings from China
[14:53:54] <DJ9DJ> gn8
[14:55:20] <AR_> It is just these though, L84Supper
[14:55:21] <AR_> http://www.ebay.com/itm/150957554073?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649
[14:55:29] <AR_> do these require duties paid?
[14:56:00] <L84Supper> those don't look like tapered roller bearings to me
[14:56:37] <AR_> no they are just ball bearing slides
[14:56:44] <L84Supper> http://www.biztrademarket.com/User/172409/bb/tapered_roller_bearings_rkc.jpg
[14:57:23] <AR_> so the only bearings that require a fee are tapered roller bearings?
[14:57:37] <AR_> regardless i dont feel like sending all this paperwork to DHL
[14:57:38] <L84Supper> trying to find the actual ruling
[14:58:06] <L84Supper> http://www.usitc.gov/trade_remedy/731_ad_701_cvd/investigations/2012/tapered_roller_bearings_from_china/reviewphase.htm
[14:59:13] <L84Supper> tapered Roller Bearings and Parts Thereof, Finished or Unfinished, From the People’s Republic of China,
[15:00:08] <L84Supper> looks like DHL is flagging them or the US Customs and Border patrol decided that your non tapered bearings are going to be treated like tapered bearings
[15:01:05] <L84Supper> up top you, you could let them know that they are not tapered roller bearings or just send them back
[15:01:42] <AR_> ok
[15:01:43] <L84Supper> lately I've been seeing people error more on the side of dumb and childish
[15:16:38] <ReadError> http://www.youtube.com/watch?v=9qSEfcIfYbw
[15:16:43] <ReadError> thats pretty awesome
[18:04:03] <andypugh> It's quite unusually quiet tonight
[18:09:18] <adb> 'xmas in Russia ?
[18:11:36] <Tom_itx> stir up the pot
[18:12:08] <JT-Shop> andypugh: do you know what the C100 does in this reversing AC motor?
http://imagebin.org/241818
[18:13:07] <Tom_itx> starter?
[18:14:32] <andypugh> I can imagine that live to either 29 or 31 runs the motor one way or the other, with the cap making the other phase lag, to give a sense of rotation. Like a capacitor-run motor, but with symmetrical windings.
[18:15:00] <JT-Shop> ok, that even makes some sense to me
[18:15:17] <andypugh> Black is marked FD andRV on the lower motor.
[18:15:25] <andypugh> So it fits.
[18:15:42] <JT-Shop> it is a fwd/rev motor for sure
[18:15:55] <andypugh> Odd way to do it, if the machine has three-phase
[18:16:53] <JT-Shop> it's my ATC which crapped out yesterday
[18:17:59] <andypugh> Ah.
[18:18:32] <andypugh> If the motor just buzzes in both directions, then C100 looks like a likely culprit.
[18:18:50] <JT-Shop> still not sure of the root problem but the relay appears to have welded the contacts so the motor runs all the time now
[18:19:35] <JT-Shop> I suspect the gearbox is borked and wonder what started the symptom
[18:20:18] <JT-Shop> if the in position prox died that could be the root of the problem
[18:21:18] <JT-Shop> we seem to be getting some strange first posts on the forum
[18:21:33] <andypugh> I found that my new mill is a lot stiffer then the old one, but that it also has a lot more play in the vertical slide.
[18:21:54] <andypugh> I had the knee off, and I think I need to re-adjust the gibbs.
[18:23:46] <JT-Shop> what do you think of this post
http://linuxcnc.org/index.php/english/forum/30-cnc-machines/7716-from-anilam-to-emc2?start=18
[18:29:58] <andypugh> Well, the original thread starter went with Mach3. I am not sure if he ever got it going.
[18:39:09] <JT-Shop> goodnight all
[18:47:30] <pcw_home> JT-Shop: other than jumping on an old thread I'm sure the guys legit
[18:49:37] <pcw_home> He emailed me previously about whether I had any direct knowledge
[18:49:39] <pcw_home> of Anilam Crusader retrofits and I suggested asking on the forum
[18:51:16] <skunkworks> andypugh: how is the hobbing?
[18:52:25] <andypugh> Actually hobbing is some weeks away. Today was mainly spent stripping, cleaning, and assembling properly the rotary table.
[18:53:24] <skunkworks> awww - I am living vicariously through you.... ;)
[18:53:25] <andypugh> Tomorrow might see me attaching a motor so see if the ones I have have sufficient grunt.
[18:59:07] <ReadError> http://brainspl.at/4th-axis.jpg
[18:59:17] <ReadError> thats the 4th axis one of the guys in the reprap chan made
[19:01:00] <ReadError> 485oz stepper on it
[19:04:23] <ReadError> gives a resolution of 0.1125degrees/step
[19:04:29] <ReadError> on 1/16th stepping
[19:13:09] <skunkworks> I think andy needs a little more umph than direct drive
[19:14:10] <skunkworks> ReadError: you don't get much more accuracy than about 1/2 step...
[19:15:27] <skunkworks> so he is lucky if he will get .9 deg resolution
[19:15:34] <ReadError> oh
[19:16:15] <skunkworks> microstepping makes things smoother - less chance of resonence - not more resolution...
[19:16:28] <skunkworks> resonance
[19:17:08] <ReadError> still a pretty novel concept
[19:17:44] <ReadError> so...i was planning on building a cnc router
[19:17:59] <ReadError> do most consumer off the shelf routers have much run out?
[19:18:03] <ReadError> or not much to speak of
[19:20:48] <skunkworks> this is what andy is wanting to do - but bigger. (this is actually andy...)
[19:20:50] <skunkworks> https://www.youtube.com/watch?v=ZhICrb0Tbn4
[19:21:37] <skunkworks> the couple we have are around .007...
[19:22:20] <ReadError> andy sounds very sophisticated
[19:22:26] <ReadError> he makes us americans sound dumb :(
[19:26:28] <skunkworks> we don't need andy for that...
[19:30:31] <lwizardl> hello
[19:30:52] <skunkworks> lwizardl: hello
[19:32:01] <lwizardl> I'm looking for some open source and good CAD software. I'm looking to do layouts of cardboard boxes.
[19:32:58] <ReadError> openscad ?
[19:34:24] <lwizardl> that looks like it will work
[19:48:29] <andypugh> I currently have two CNC (stepper) rotary axes, this is the slow/high torque one:
http://youtu.be/DQlGAGt_zcM
[19:48:51] <andypugh> (the fast one was in the video skunkworks linked to).
[19:51:26] <andypugh> The new mill is going to have a servo-driven one of these:
http://homews.co.uk/Lrgsales50.jpg
[19:53:25] <andypugh> It has already been done (
http://divisionmaster.co.uk/examples.html ) but I hope to find a more compact way to attach the motor.
[20:15:19] <skunkworks> cool
[20:35:17] <L33TG33KG34R> hello guys
[20:35:29] <L33TG33KG34R> I am very confused about hardware required to run linux
[20:35:39] <L33TG33KG34R> well linuxcnc in particular
[20:44:44] <andypugh> L33TG33KG34R: You need "A computer"
[20:45:07] <andypugh> What do you have?
[20:45:09] <jdh> and some sort of IO
[20:45:23] <L33TG33KG34R> well I plan on using mesa cards and hardware
[20:45:32] <andypugh> Well, you can run LinuxCNC without any IO, though it would be frustrating.
[20:45:49] <L33TG33KG34R> I looked at the hardware requirements online and it said that its buggy if ram is greater than 1 GB
[20:45:58] <L33TG33KG34R> I don't have any boards at home that work
[20:46:21] <L33TG33KG34R> I went online to find something useable and everything costs as much as a brand new build
[20:46:45] <L33TG33KG34R> I don't know what to do :( I found a CPU/Mobo/RAM combo for under $150
[20:46:45] <jdh> what do you have that doesnt' work?
[20:47:10] <L33TG33KG34R> none of my boards post because they failed earlier for reasons I don't want to get into
[20:47:15] <jdh> I use 2 & 4GB in mine (atom d525)
[20:47:38] <L33TG33KG34R> yeah the atom board alone costs as much as the new proc/mobo combo I can get locally
[20:47:39] <L33TG33KG34R> new
[20:47:40] <jdh> $90 mobo, $15 ram, $35 case. I used old hard drives
[20:47:53] <andypugh> The LiveCD install is buggy with more than (I think) 3.5GB. The workaround is to disconnect from the internet during install (or temporarily remove some RAM)
[20:48:15] <L33TG33KG34R> well I won't be using a live cd install instead it will be off USB
[20:48:27] <cradek> andypugh: I thought I fixed that a long time ago
[20:48:32] <L33TG33KG34R> so is it alright to use 2/4GB ram and new hardware?
[20:48:41] <andypugh> Sounds like it.
[20:49:05] <jdh> I installed the livecd from a month or so ago with 4GB. NO problems
[20:49:06] <andypugh> Note that the Atom boards include the CPU, as it is soldered on.
[20:49:25] <andypugh> Sory cradek, I am recycling bad info.
[20:50:04] <cradek> andypugh: well it's impossible to know the past with certainty. I sympathize.
[20:50:27] <cradek> andypugh: I'm 90% sure I fixed that and ... some other problems ... with the cd
[20:50:43] <andypugh> L33TG33KG34R: We like the Atom boards because they run fanless on low-power PSUs and have good latency.
[20:51:24] <andypugh> Are you planning a stepper or servo system?
[20:51:42] <L33TG33KG34R> stepper
[20:51:50] <L33TG33KG34R> with maybe encoder feedback in the future
[20:52:57] <andypugh> Feel free to carry on studying, but you are likely to end up with a D525MW (or the currrent equivalent) fitted with the Mesa 5i25 and 7i76 "plug-n-go" kit.
[20:53:21] <L33TG33KG34R> hmm...
[20:53:24] <r00t4rd3d> http://i.imgur.com/MERil.jpg
[20:54:44] <andypugh> That's someting like $80 for the PC, $200 for the Mesa cards. Add $25 for a PicoPSU and $30 for an 8GB flash SSD (planty for a CNC controller) and £15 for a stick ot ram. Then make chips.
[20:55:52] <andypugh> r00t4rd3d: Skillfull tinwork, but I am not sure I like the result.
[21:01:21] <archivist> andypugh, you took a new vertex to bits? did it have shims and adjustment for the worm to gear location?
[21:02:17] <andypugh> It might not actually be a true Vertex. It is badged "Axminster" though the design appears the same.
[21:03:02] <andypugh> The worm-to-gear is adjusted with an eccentric. The worm has a ball pressed in the end, whch bears on an adjusting screw to remove worm axial float.
[21:03:22] <archivist> yes there are vertex ripoff about my B axis is a ripoff but the A is real
[21:03:53] <andypugh> Mine seems to have a nastier version of the rear bearing / clamp than the pictures from Vertex.
[21:03:53] <andypugh> ]
[21:04:18] <archivist> vertex publish a real spec for angular error, others ....
[21:05:05] <andypugh> I think it will do for my purposes though.
[21:05:20] <archivist> and I have had a few surprises with angular error :)
[21:06:59] <archivist> was there any adjustment to centralise the gear to worm axis
[21:07:21] <andypugh> Fore and aft, you mean?
[21:09:52] <archivist> actually thinking about the eccentric which mine has too, that must put the worm axis off the correct plane
[21:10:09] <archivist> except if lucky
[21:10:46] <andypugh> I assume it is a plain gear, rather than enveloping.
[21:11:57] <archivist> for plain after it had worn a bit that could be worse
[21:13:01] <archivist> I must get the colimator and test mine like I tested some a while ago
[21:14:27] <archivist> some "quality" stuff had over a 1/5 degree peak to peak error
[21:14:41] <archivist> 1/2 degree
[21:16:43] <t12> todays discovery was that the lathe coolant pump was wired backwards
[21:16:48] <t12> that or the entire lathe is wired backwards
[21:20:52] <L33TG33KG34R> this motherboard
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813131878 with this processor
http://www.newegg.ca/Product/Product.aspx?Item=N82E16819116408 and this ram
http://www.newegg.ca/Product/Product.aspx?Item=N82E16820139996
[21:22:55] <L33TG33KG34R> costs $10 more than
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813153240
[21:23:22] <L33TG33KG34R> er, sorry
[21:23:29] <L33TG33KG34R> costs $10 less than
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813153242
[21:27:35] <andypugh> http://www.newegg.ca/Product/Product.aspx?Item=N82E16813121596 includes the CPU and has a parallel port (so you could add a 7i43 if you really ran out of IO. Not that that is likely when you can have 216 IO lines from a single 5i25 card.
[21:28:19] <andypugh> You seem to be missing the point that the Jetway has the CPU pre-installed.
[21:29:11] <andypugh> Ah, no, you may not be. I am getting confused. It happens a lot.
[21:29:31] <archivist> age... /me ducks
[21:30:13] <andypugh> Time for me to sleep, I think.
[21:30:17] <L33TG33KG34R> I understand that these atom boards have embedded CPUs on them but atom is going the way of the dodo bird... I don't want to be stuck down the road with obsolete hardware.
[21:31:30] <skunkworks> heh
[21:31:34] <skunkworks> sorry
[21:31:55] <L33TG33KG34R> please speak your mind. this is the internet...
[21:32:29] <skunkworks> if you buy something that has a repalceable cpu - by the time you think it is obsolite - it will be.
[21:32:43] <skunkworks> obsolete
[21:34:20] <skunkworks> as in - you won't be able to buy a cpu that will perform on even close with what the now current motherboards/cpu combos
[21:34:33] <L33TG33KG34R> well at least I will be able to find some cpu's floating around. even if it is obsolete. I just don't like getting an already dying product. 1155 is still going strong. not to mention theres also AMD to look at.
[21:35:47] <L33TG33KG34R> I understand that linuxcnc needs almost nothing in hardware to work, compared to typical desktop boards. dishing out $80 for low performance hardware doesn't make sense.
[21:37:39] <skunkworks> eh - I could go either way...
[21:38:21] <L33TG33KG34R> even the amd APUs fell short
[21:38:33] <L33TG33KG34R> well the fusion series any way
[21:38:54] <L33TG33KG34R> or else I would run custom hardware, like an arm board to run linuxcnc
[21:39:26] <L33TG33KG34R> its really easy to design one not to mention the 1000s that are online just waiting to have linux installed on them
[21:40:34] <skunkworks> sure. Just an fyi - the latest amd processors I have played with have pretty crappy latency
[21:40:47] <skunkworks> like 30-40us
[21:41:11] <L33TG33KG34R> what do you mean?
[21:43:07] <skunkworks> let me see if I can find an example
[21:44:33] <skunkworks> this processor
http://www.newegg.com/Product/Product.aspx?Item=N82E16819106014
[21:44:58] <skunkworks> this motherboard
http://www.newegg.com/Product/Product.aspx?Item=N82E16813131871
[21:45:15] <skunkworks> I bought an older version also - same issue - poor latency
[21:45:45] <skunkworks> I don't know if it is because of the apu - but adding a video card didn't help
[21:47:29] <skunkworks> or just happened to be that combo sucks. Though I have had good luck with amd and asus for linuxcnc up to this point
[21:47:49] <ReadError> AMD is crap
[21:47:55] <ReadError> plain and simple ;)
[21:48:10] <L33TG33KG34R> ReadError: I used to think that but not true
[21:48:14] <ReadError> they have not had a good cpu since the athlon xp
[21:48:18] <ReadError> true story.
[21:49:07] <L33TG33KG34R> yes comparing processors per core, intel has better performance vs amd
[21:49:36] <L33TG33KG34R> but they are truely cheap
[21:49:45] <L33TG33KG34R> great for budget builds and such
[21:49:55] <ReadError> yea but cheap stuff is not always good
[21:50:07] <ReadError> i would rather spend the extra 50 and get something good
[21:50:15] <L33TG33KG34R> yes but can be fore its application
[21:50:19] <L33TG33KG34R> for*
[21:50:38] <L33TG33KG34R> htpcs are one of them if you are trying to go extremely cheap
[21:50:41] <ReadError> i would rather not induce that into my household.
[21:51:14] <L33TG33KG34R> I wouldn't either but for linuxcnc, I would think the reason its slower is because of its cache
[21:51:22] <ReadError> i have erm, 6 intel boxes
[21:51:26] <ReadError> and 2 intel laptops
[21:51:29] <ReadError> amd, not even once
[21:51:42] <ReadError> but like i said, it was solid during the athlon xp days for a bit
[21:51:47] <ReadError> but since then, nogusta.
[21:52:21] <L33TG33KG34R> skunkworks: can you find the processors you used previously before you had the problem with latencies?
[21:52:42] <skunkworks> phenom - does that sound right>
[21:52:45] <skunkworks> ?
[21:52:55] <L33TG33KG34R> yes, but which phenom?
[21:53:02] <skunkworks> hmm
[21:53:28] <L33TG33KG34R> dual cores? quads?
[21:53:32] <L33TG33KG34R> tri? hex?
[21:54:50] <ReadError> who the heck makes a tricore
[21:54:53] <ReadError> thats so dumb
[21:54:59] <ReadError> (AMD)
[21:55:51] <L33TG33KG34R> they were quadcore silicon with one core partially or entirely non functioning. instead of discarding them, they chose to lock down the bad core and sell them at a cheaper price. great way to reduce waste.
[21:56:34] <L33TG33KG34R> its like the clearance bin for open/ed box in stores
[21:56:51] <ReadError> yea but why would i want to buy broken hardware
[21:57:17] <L33TG33KG34R> well, its not like they advertised it as quadcore. so technically not broken.
[21:57:42] <L33TG33KG34R> the broken part is isolated from the working parts
[21:57:52] <skunkworks> I had a bunch - don't think the number of cores really effected the latency
[21:57:55] <ReadError> but why would i buy that when i can get a 2500K cheap
[21:58:47] <L33TG33KG34R> it takes a great deal of effort to manufacture wafers of silicon. why waste stuff that works? they even dropped the price on them
[21:58:54] <skunkworks> here we go....
http://www.electronicsam.com/images/KandT/testing/asus%20m2n68-am%20plus%20motherboard%20with%20amd%20phenom%20x4%209600b%20%28quad%20core%202.3ghz%29.jpg
[21:59:01] <ReadError> and im fairly certain that price:performance ratio
[21:59:04] <ReadError> intel has them beat
[21:59:32] <L33TG33KG34R> yeah... but still amd has their niche and they have it locked down pretty tight
[22:00:00] <skunkworks> I think that is actually what is in the K&T
[22:00:12] <L33TG33KG34R> the cheapest intel proc is ~100 - 120 USD
[22:00:21] <L33TG33KG34R> K&T?
[22:02:02] <skunkworks> https://www.youtube.com/watch?v=39q6kvrSBSk
[22:02:28] <skunkworks> http://www.electronicsam.com/images/KandT/oldkandt.JPG
[22:03:24] <ReadError> http://www.cpubenchmark.net/common_cpus.html#cpuvalue
[22:03:35] <ReadError> amd does not even place in the top 5
[22:03:44] <ReadError> well, its 5th
[22:04:04] <L33TG33KG34R> lol
[22:04:22] <L33TG33KG34R> that machine is HUGE!
[22:04:36] <skunkworks> yes :)
[22:04:39] <ReadError> maybe not even
[22:04:42] <ReadError> lower is better
[22:05:45] <L33TG33KG34R> is the spindle hydraulic too?
[22:07:52] <skunkworks> not anymore
[22:10:29] <skunkworks> ReadError: I don't think lower is better - if it is performance/price - lower the price - higher the number.
[22:11:02] <ReadError> when i add my cnc router
[22:11:17] <ReadError> i was wondering if i should get a mesa card or a PCI-e parport card
[22:11:25] <ReadError> i dont want to swap the cable each time
[22:11:30] <L33TG33KG34R> mesa has pci-e parport card
[22:11:48] <ReadError> yea, but a straight pci-e card is much cheaper
[22:11:59] <ReadError> i have a d525 so my slots are somewhat limited
[22:12:02] <L33TG33KG34R> hope that its not buffered
[22:12:49] <ReadError> i need to decide also, if i should go with a dwalt/OTC router
[22:12:59] <ReadError> or spend the 2-300$ on a nicer one
[22:14:57] <skunkworks> 5i25 alone acts like 2 printer ports...
[22:15:19] <skunkworks> not to mention what you can get in daughter boards....
[22:15:34] <ReadError> how much are those? 200?
[22:15:41] <skunkworks> 89 iirc
[22:15:41] <ReadError> i have never found a good price for them
[22:17:08] <skunkworks> 5I25-SP PCI Anything I/O – 34/ I/O bits LX9 Spartan6 FPGA Standard profile $89
[22:17:52] <ReadError> got the link?
[22:18:22] <skunkworks> http://mesanet.com/
[22:18:28] <skunkworks> left side - price list
[22:22:06] <skunkworks> http://mesanet.com/fpgacardinfo.html
[22:22:09] <skunkworks> cool card
[22:27:10] <skunkworks> I am using 2 of their big brothers...
http://electronicsam.com/images/KandT/conversion/computermesa.JPG
[22:27:35] <ReadError> for 1 machine? or multiple
[22:30:00] <skunkworks> one machine
[22:30:45] <skunkworks> http://electronicsam.com/images/KandT/conversion/panel/electricalbox.jpg
[22:32:28] <ReadError> holy i/o
[22:32:43] <L33TG33KG34R> hehehe... this is nothing compared to where I work
[22:33:23] <L33TG33KG34R> but still, impressive
[22:35:06] <ReadError> i estimate zapping myself 4 times building that, and possibly 2 smaller fires
[22:36:23] <skunkworks> the 2 mesa cards setup as they are gave me 10 axis (+/-10v and encoder counters) and 96 i/o. I pretty much used it all up.
[22:42:31] <ReadError> holy crap
[22:42:41] <ReadError> how????
[22:43:52] <skunkworks> how what?
[22:44:58] <skunkworks> did you see that video? 4 axis - 16 speed gearbox - 60 tool chain, changer, pallets..... lots of i/o needed
[22:54:06] <L33TG33KG34R> what did you use for the main board and processor?
[22:58:04] <skunkworks2> http://www.electronicsam.com/images/KandT/testing/asus%20m2n68-am%20plus%20motherboard%20with%20amd%20phenom%20x4%209600b%20%28quad%20core%202.3ghz%29.jpg
[22:59:11] <skunkworks2> I am pretty sure that is the motherboard and processor
[23:02:15] <L33TG33KG34R> how does that give me mobo and proc?
[23:02:43] <L33TG33KG34R> ah
[23:02:44] <L33TG33KG34R> nvm
[23:02:48] <L33TG33KG34R> sorry
[23:04:12] <skunkworks2> heh
[23:04:49] <skunkworks2> just have to read between the &
[23:16:00] <Gene45> Skunkworks: Hats of to you
[23:20:44] <skunkworks2> thanks - linuxcnc has worked awesome
[23:30:13] <L33TG33KG34R> well, this is a sight for sore eyes
[23:30:14] <L33TG33KG34R> http://www.memoryexpress.com/Products/MX41064
[23:47:27] <Gene45> Skunkworks2: How did you select servos and power supply?
[23:55:05] <frysteev> hey sill question for usb parrell port adapters
[23:55:18] <frysteev> i know they are useless for actual control,
[23:55:43] <frysteev> but are they good enough for something not mission critical , looking turning coolant on and off?
