Back
[00:37:41] <hatch789> does anyone here play with the "precise build" on buildbot? I am trying to install it on my 11.10 Ubuntu and it shows the package as broken??
[00:38:01] <hatch789> linuxcnc-sim is what I'm selecting
[01:02:30] <hatch789> hey does anyone know if there is a version of linuxcnc that will run on Ubuntu 11.10?
[03:06:15] <DJ9DJ> moin
[03:54:56] <alex4nder> yoh
[04:19:31] <Mjolinor> http://www.digital-kaos.co.uk/forums/f4/can-you-pass-british-citizenship-exam-259483/
[04:19:54] <Mjolinor> gobsmacked me, 10 out of 24, rejected and I am English
[04:53:23] <cncbasher> yea but thats the point , we are not supposed to pass it , we get kicked out next
[04:58:18] <psha[work]> Mjolinor: it should be performed in the form of competition - for each candidate random citizen is selected
[04:58:34] <Mjolinor> :)
[04:58:35] <psha[work]> then they both pass test and who get's better result wins citizenship
[04:58:53] <psha[work]> looser is deported to somali
[04:58:59] <cncbasher> the one carrying the hand grenade wins
[04:59:12] <Mjolinor> I rekcon it shoudlnt be performed art all, stick the stuff fomr the channel tunnel in a great big wall round the south of england and filt eh channel tunnel with all the non britsh people that are here already
[04:59:16] <Mjolinor> sorted tehn :)
[04:59:33] <cncbasher> their was a reason were an island ...
[04:59:39] <cncbasher> then they build a tunnel
[04:59:58] <Mjolinor> or better still, dig a channel across the country say aorund birmingham, shove the bit south of that across to france
[05:00:22] <cncbasher> i'll bring me konatsu
[05:10:01] <theorbtwo> BTW, you don't just need your offical trival test for citizenship, but also for your indefinate leave to remain. That caught me out at one point -- couldn't do the test in time, needed to get another two-year extension.
[05:10:48] <theorbtwo> I do find it rather silly that your reaction to failing the test is to say that we should get rid of all the foreigners, rather then saying "oh, I guess it's harder to stay here legally then I thought".
[05:11:09] <theorbtwo> (I'm an American expat living in the UK, for six years now.)
[05:11:18] <Mjolinor> my reaction to failing isnt to get rid opf all the foreigners :)
[05:11:41] <Mjolinor> I had that belief before I took the test :)
[05:12:26] <Loetmichel> mornin'
[05:13:28] <theorbtwo> Mjolinor: Do you think I should be thrown out of the country?
[05:13:46] <Mjolinor> I have no idea
[05:13:58] <Mjolinor> not knowing you makes htat hard to pass judgement on
[05:14:14] <theorbtwo> But I'm a foreigner, and you said that they should all be thrown out.
[05:14:21] <Mjolinor> ok, then yes
[05:15:50] <Mjolinor> it is a whole lkot more complicated than that but for sure I don't think that quiz asks the right questions at all
[05:16:11] <theorbtwo> I'm pretty sure it doesn't, but I have no idea what the right questions are.
[05:16:19] <Mjolinor> but I dont know if htat quiz is representative fo the sort of things you have to know to get citizenship
[05:17:13] <theorbtwo> Mjolinor: It is, though the fee is not to be sneezed at. I actually think the fee should be shorter, but the times should be much longer.
[05:17:26] <Mjolinor> the fee?
[05:18:37] <theorbtwo> £836 for me, to apply for citizenship.
[05:18:55] <Mjolinor> :0
[05:18:59] <Mjolinor> thats a lot
[05:19:33] <theorbtwo> Yeah. That's why I haven't yet.
[05:19:50] <cncbasher> welcome to rip off britian
[05:19:55] <Mjolinor> :)
[05:20:19] <cncbasher> thats legal extortion
[05:21:14] <theorbtwo> Yeah. Luckily, I'm not at the point in the process where it's still "pay me more money, or it's closing time for you -- don't care where you go, but you can't stay here".
[05:21:17] <Mjolinor> well we have to pay the illegals soemhow
[05:22:18] <theorbtwo> Yeah, I have no experince with how easy it is to stay in the country illegally.
[05:22:37] <theorbtwo> Knock wood, I'll be able to keep saying that.
[05:22:41] <Mjolinor> where are you from?
[05:22:49] <cncbasher> easier than you trying to do it correctly and legal thats for sure
[05:23:08] <Mjolinor> it does seem to be sort of random and totally unfair
[05:23:19] <Mjolinor> who gets exported that is
[05:24:04] <theorbtwo> I'm American. Deporting me would be easy -- the UK and EU don't like to admit that the US tortures people.
[05:24:17] <Mjolinor> you can stay :)
[05:24:25] <cncbasher> you passed
[05:24:33] <Mjolinor> anyone that realises how much america sucks and leaves is welcome in my book :)
[05:26:36] <theorbtwo> It makes sense to me. The UK doesn't condone torture, and won't send people to someplace they are likely to be tortured. It's a principled stand. It has some unfortunate consequences, but I respect the UK for saying that it's worth it to not ship people off to be tortured.
[05:29:04] <Mjolinor> i dont understand how the US htinks that if anyone in the world broke their laws whether they are in the states or not, they ahve a right to have them taken their for trial
[05:29:42] <Mjolinor> if the US dont want their citzens to see web sites that are not in the US then the US government should stop it, taking the people that made the website cannot ever be morally correct
[05:45:07] <archivist> I beat you Mjolinor ....still fail though 11/24 one better guess
[05:46:14] <cncbasher> arh well thats 2 less
[05:47:02] <cncbasher> i'll be joining too i recon
[06:06:23] <Mjolinor> :)
[06:06:30] <Mjolinor> my son did it, he got 10 too
[06:06:53] <cncbasher> ok so thats 4 of us dead and burried
[06:07:30] <Mjolinor> its hard to know what questions should be asked really
[06:07:44] <Mjolinor> certainly I dont see european parliement members to be relevant
[06:08:11] <Mjolinor> It shoud be things like when was hte normal conqwuest, fire of london, guy fawkes and stuuf I think
[06:08:36] <Mjolinor> what % of UK population live in scotland
[06:08:40] <Mjolinor> who the hell knows that?
[06:14:58] <cncbasher> easy questions are do you carry grenades or mortar shells outside of your day to day job
[06:15:16] <cncbasher> if yes then youv'e passed
[06:15:38] <cncbasher> any questions on the uk etc is a fail
[06:17:58] <Mjolinor> shoudl be one mark off for knowing how long you have to work before drawing handouts
[06:19:10] <cncbasher> ther'd all pass thats one
[06:19:24] <cncbasher> genuine citizens would fail
[07:25:40] <jthornton> arggg time to start the 6 month process of realigning my internal clock :/
[07:29:23] <Loetmichel> hihi, and in 6 month the same ;-)
[07:38:18] <Mjolinor> anyone familiar with Siemens 420 VFDs?
[07:38:55] <Mjolinor> thought i had is sussed after much readonig yesterday but apparenlty not
[07:39:33] <Mjolinor> I set the max freq to 120 Hz (50 is default) but the PWM still only gets 50 Hz at 10 volts so I am missing somehting and as there are maybe 1000 pages in the documents it is not easy to find
[08:31:43] <jthornton> input scale?
[09:07:37] <Mjolinor> nah, its the inverter, pwm input goes as it shoudl 0 -10 volts but the inverter is only making 50Hz at 10 volts.
[09:07:49] <Mjolinor> I htink I may have found it but ther eare so many parameters you can set in these things
[10:13:31] <mazafaka> end mills have diameter which differs from zero, that's why today's my detail is being welded up. Otherwise, it would be a perfect detail.
[10:37:03] <ssi> jthornton: still no luck... the control wants to come on now, but the inrush current when I hit control on trips the overcurrent on the VFD :(
[11:36:33] <archivist> gene77, just checked the rotary configs on the 5 axis limits are set min -99999 max 99999 only makes sense on the a axis though :)
[11:38:40] <tehDarkAura> hmmm thinking about using aluminum 5/16th rods to guide x axis
[11:38:54] <tehDarkAura> is that stuff strong enough you think?
[11:39:10] <archivist> are you milling paper or something stronger
[11:39:26] <tehDarkAura> copper clad board and wood is the goal
[11:40:24] <archivist> 5/16 seems awfully slim unless bolted every inch to something solid
[11:40:24] <tehDarkAura> maybe i can get away with 5/16 stainless pipe...
[11:40:33] <tehDarkAura> ahh okay
[11:41:11] <ssi> at least use thompson shaft :P
[11:41:32] <tehDarkAura> need it to hang supported only by the end for 36"
[11:41:46] <tehDarkAura> ohh im not sure what that is -- /me goes to google it
[11:41:49] <ssi> then use 3/4" thompson shaft
[11:41:50] <ssi> heh
[11:42:01] <archivist> think of the energy in a spinning shaft where that energy gets released, bending your rails or cutting the wood
[11:42:10] <ssi> http://www.mcmaster.com/#linear-shafts/=gm942v
[11:42:30] <tehDarkAura> cool thanks :)
[11:42:39] <tehDarkAura> appreciate it people :)
[11:42:55] <ssi> np
[11:43:59] <archivist> my small lathe has about a 20mm spindle, way to small for the overhang the provided chuck has
[11:49:23] <alex4nder> hey
[12:46:14] <Tugge> What program have you been usign for PCB-making?
[12:49:23] <mrsun> theres something strange going on :( Y axis behaves like the X axis when it comes to the missing 0.05mm and getting it back the next step
[13:22:18] <tom3p> yesterday freenode was blocked by OpenDNS ( "in response to Conficker Virus" )
[13:26:42] <tehDarkAura> altium designer is by far my favorite pcb design software
[13:26:56] <tom3p> i want to make my old linuxcnc 8.04 into a VM so i have Apt360 et al available.
[13:26:56] <tom3p> does anyone have experience installing VMwarer sever on Ubuntu 10.04?
[13:27:31] <tehDarkAura> ive used a lot of the free ones and they are okay too -- but if you can find a copy of altium its amazing -- even the auto route alone is well worth it
[13:28:27] <IchGuckLive> tehDarkAura: id you decine good then autorout is good on all systems
[13:29:10] <IchGuckLive> tehDarkAura: does altium has linuxcnc code generation module inside
[13:30:26] <tehDarkAura> it does gerbers easily extra step i guess
[13:31:57] <IchGuckLive> ok but gerber does only the standards you need to translate in Eagle you can go parallel and se at the reload button the produced code then you can move around with the numbers to get better routlines
[13:32:04] <Tugge> I meant to ask that what do you use to generate g-code from gerbers? I asked wrong question :D
[13:32:39] <IchGuckLive> Tugge: we got a gerber translater in our Wikipedia simple g-code generators
[13:33:09] <IchGuckLive> as i programmed it asome time ago B)
[13:33:53] <IchGuckLive> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Converting_Tools
[13:35:17] <archivist> Tugge, some/a lot use pcb2gcode
[13:35:26] <Tugge> IchGuckLive: Thanks a lot :) I will try it tomorrow. This looks quite neat software.
[13:36:04] <IchGuckLive> Tugge: its in python so you can do what ever you want to do ad your need
[13:36:14] <flyback> I love how cnc shops now
[13:36:20] <flyback> save their millings
[13:36:24] <IchGuckLive> Tugge: look at the youtube Howto use it video
[13:36:26] <flyback> cause it's worth money :)
[13:36:54] <flyback> hell even iron milling is. cement, chemistry companies buy it if it's not worth remlt
[13:36:57] <flyback> remelt
[13:37:31] <IchGuckLive> flyback: on cutters ?
[13:37:35] <flyback> yeah
[13:37:39] <IchGuckLive> or rest pices
[13:37:42] <flyback> grinders, etc
[13:37:55] <flyback> the metal scrap you have from making stuff
[13:38:18] <IchGuckLive> Raw material is worth alot at this times
[13:38:27] <flyback> ye[
[13:38:46] <IchGuckLive> and the chips even more if sorted
[13:39:12] <flyback> and like I said, the stuff not worth reclaiming for remelt and a foundry, sometimes chemistry, etc companies use it
[13:39:19] <IchGuckLive> so clean comeplete bevor brass and make a Hospital clean bevor copper
[13:39:32] <flyback> yep
[13:39:43] <flyback> copper is neat its anti bacteria properties
[13:40:41] <IchGuckLive> agree my company had 7t in stock from years ago there wars a real money race to sell it
[13:41:51] <IchGuckLive> my traslater gives me no return EDM with copper no wire cut how is this named in english ???
[13:42:42] <flyback> not sure but I think I get the idea :)
[13:43:19] <IchGuckLive> elektrodes from 1980-1990 for molding beeng connected ad gave a output to 200.000Euros
[13:43:43] <tom3p> EDM with copper no wire cut? drahterodieren? wire edm? senkerodieren = sink edm
[13:44:01] <IchGuckLive> senkerodieren = sink edm B)
[13:44:25] <flyback> hell i've seen some companies that recycle plastic, cloth and carpet cuttings
[13:44:29] <flyback> back into usuable product
[13:44:32] <flyback> pretty slick
[13:45:21] <IchGuckLive> we 2 the wire channel are only 0,3mm new material the rest of the 2-3mm wall is shit
[13:52:08] <IchGuckLive> isssi: nice to see you
[13:52:22] <isssi> hi
[13:53:17] <IchGuckLive> how is the wether in bulgaria
[13:54:22] <Mjolinor> heeks is confusing me again
[13:54:36] <IchGuckLive> Mjolinor: can i help ?
[13:54:40] <Mjolinor> http://homepage.ntlworld.com/dknpower/3.png
[13:54:42] <Mjolinor> maybe
[13:54:58] <Mjolinor> I got what I wanted eventually but the solution seems odd
[13:55:20] <Mjolinor> The 3 in that screen shot needed to stand above the circle
[13:55:30] <IchGuckLive> this is awallins workingfild
[13:55:49] <Mjolinor> in order to get the G code to do that I had to subtract the 3 form the circle and make a 3 chaped hole in the circle
[13:56:23] <IchGuckLive> whi dident you extrude to a real 3D
[13:56:24] <Mjolinor> it seems to be backawards somewhat
[13:56:38] <Mjolinor> i just followed your video
[13:56:48] <isssi> today is hot
[13:56:50] <IchGuckLive> i see 2D no 3D
[13:57:06] <IchGuckLive> isssi: nice
[13:57:34] <ssi> isssi: omg it's like the weird palindromic parallel universe me! D:
[13:58:08] <IchGuckLive> Mjolinor:
http://foengarage.de/fiat_logo_heeks.jpg
[13:58:24] <Mjolinor> http://homepage.ntlworld.com/dknpower/3_2.png
[13:58:29] <Mjolinor> that is with it visible
[13:58:43] <Mjolinor> it is "not visible" in the first one
[13:59:19] <flyback> I bet there's a class of cnc machines out there for recycling materials
[13:59:24] <Mjolinor> hmm, thats a bit more advanced than where I am :)
[13:59:42] <flyback> milling old train wheels into metal bowls or something :P
[14:00:09] <IchGuckLive> Mjolinor: as i understand you woudt do the 3 standing out on top of the circle
[14:00:21] <Mjolinor> I did that
[14:00:30] <Mjolinor> but when I made the gcode it erased the 3
[14:00:44] <Mjolinor> as it is now it works fine and cuts what I wanted it to cut
[14:00:58] <IchGuckLive> no you need to extrude the circle to minus and the 3 to plus and then bool them together
[14:01:03] <Mjolinor> I couldnt get the gcode to go round the 3 when the 3 stood above the circle
[14:01:13] <Mjolinor> I did that
[14:01:32] <IchGuckLive> to go around the 3 use path
[14:01:46] <Mjolinor> well I extruded the circle to -2 and the 3 to -0.5 then joined htem together
[14:01:59] <Mjolinor> Ill try again, it is a lot of steps to remmeber the first time
[14:03:47] <Mjolinor> what is "path"?
[14:04:23] <IchGuckLive> minus on mines will extrude into itch other not ontop
[14:05:32] <IchGuckLive> Mjolinor:
http://www.anderswallin.net/
[14:06:06] <Mjolinor> also of note. If I have the pocket and click on it then edit what I want in the properties box at the bottom left then all is good
[14:06:24] <Mjolinor> if I double click on properties to it opens the dialog box then heeks will crash
[14:06:48] <Mjolinor> if I double click on pocket to it opens the dialog box then heeks will crash
[14:06:56] <Mjolinor> igcnore the first of thsoe two :)
[14:07:29] <Mjolinor> is that fiat sign one that you did?
[14:07:36] <IchGuckLive> i guess this is what you want to see for output
http://www.anderswallin.net/wp-content/uploads/2012/01/exterior_emc2_offset-300x205.png
[14:08:49] <IchGuckLive> https://github.com/aewallin/linuxcnc-scripts
[14:09:05] <IchGuckLive> use this
[14:09:47] <isssi> just finish the new stepper board , it works fantastic , bether than any other
[14:10:23] <IchGuckLive> isssi: witch drivers ? microstepping ß
[14:11:28] <isssi> yep ,
http://imagebin.org/202908
[14:12:06] <isssi> analog spindle , 6 stepper , touch probe , 6 limits , 6 zeros , 12 outputs and 20 inputs.....
[14:13:44] <IchGuckLive> oh a mesa %i25clone
[14:14:02] <IchGuckLive> isssi: or just breakout board
[14:14:04] <isssi> no , modified 7I43
[14:14:57] <isssi> its all i need to control the machine + gecko drivers.
[14:15:04] <isssi> simple to connect and use
[14:15:18] <IchGuckLive> a i see
[14:15:27] <IchGuckLive> nice work
[14:15:35] <isssi> yep , works great
[14:17:17] <deuplonicus> hmmmm..... so if anyone remembers I was trying to figure out if my second gantry motor is losing steps. I made a test run g code program to run the gantry back and forth. It runs flawlessly! only g20, g0 and g4 commands are used. However in another gcode we made for cutting out hardstops it loses steps on the gantry's left motor whenever doing long distances (over 2 inches)
[14:17:52] <deuplonicus> I was looking at the gcode, assuming the problem lies there. G64 is one of the first several codes called, and I'm under heavy suspicion this is my problem.
[14:20:05] <ssi> isssi: what's the green plastic bit that is holding the boards?
[14:24:13] <isssi> the plastic housing is from pfoenix company in germany
[14:24:21] <isssi> \verry usefull
[14:24:40] <ssi> looks useful
[14:25:04] <isssi> yep
[14:28:46] <issssi> sweeet, lots of taken up bs nicks
[14:30:02] <isssi> ooppps , its issy
[14:31:16] <member02> why would one motor lose steps on a gantry, but none of the other motors lose steps?
[14:31:34] <Tom_itx> friction
[14:32:22] <mikegg> when was the last time you sacrificed a chicken?
[14:32:24] <isssy> when lose steps during acceleration or at planar move?
[14:33:20] <member02> it isn't friction, we have the bearings loosened and it still happens
[14:33:28] <member02> mikegg, good point!
[14:33:38] <ssi> maybe your gantry is too heavy for your motors?
[14:33:41] <ssi> maybe that one motor is bad?
[14:33:51] <ssi> some times if they've been disassembled they'll lose a bunch of their rated torque
[14:34:13] <member02> Well.... when I sdid the stepper diagnostics test with simple looped gcode it runs fine
[14:34:41] <ssi> the moves that lose steps have more accel than the test?
[14:35:02] <member02> they shouldn't, it loses most of it's step when in the middle of a long pass
[14:35:20] <ssi> maybe your gantry is binding in the travel?
[14:37:05] <member02> we thought so too, but then we loosened and double checked the squaring and the table, and ran it loose on the top bearings only. it seemed fine. I think it is gcode, because in the stepper test progrm it runs to top speed back and forth alll day without error
[14:39:06] <member02> would removing a g64 command cause bad things to happen?
[14:46:35] <JT-Shop> you can't remove it, only choose one
[14:48:44] <member02> well, I removed it already and ran it, seemed the same, still missing steps on that one motor
[14:49:58] <deuplonicus> We've checked a bunch of stuff, the stepper diagnostics test to look for missing steps runs perfectly fine
[14:50:14] <deuplonicus> so now we're just trying to figure out why we're losing steps on other gcode
[14:52:52] <deuplonicus> my test code runs fine without losing steps for a hundred loops, but not this gcode from cambam, this is how it starts:
[14:53:02] <deuplonicus> ( T0 : 0.245 ) G20 G90 G64 G40 G0 Z0.125 ( T0 : 0.245 ) T0 M6 ( Engrave1 ) G17 M3 S1000
[14:53:17] <deuplonicus> then from there it's all G0 and G1 commands
[14:54:23] <ssi> JT-Shop: still no luck on powering this beast up... when the control comes online, it trips the VFD's overcurrent
[14:54:46] <ssi> I've got the schematics spread out on the walls around me, and I think I can pull the control off and power it off its own 220
[14:55:31] <ssi> it looks like L1/L2 comes off and supplies 220/1ph to the control, and L2/L3 comes off and supplies 220/1ph to the stepdown xformer
[14:56:12] <tom3p> if ok manually after rest of machine is up, try a delay-on timer?, give it 10 seconds of other inrush to settle
[14:56:12] <archivist> deuplonicus, check speeds you can get up to when testing, dont over specify top speed and acceleration
[14:56:50] <ssi> tom3p: I don't know what I would delay exactly... I think the control itself just has a massive amount of inrush
[14:56:57] <ssi> tom3p: it's not even starting any motors at that point
[14:56:59] <tom3p> delay connection of load
[14:57:23] <ssi> there's nothing to delay
[14:57:28] <ssi> I bring the VFD up to freq
[14:57:34] <ssi> then I throw the master disconnect on the machine
[14:57:39] <ssi> that makes the fans in the control come up, and estop light on
[14:57:44] <ssi> then I hit "control on"
[14:57:46] <ssi> and at that point, it trips
[14:58:08] <ssi> also this VFD just might be too small
[14:58:18] <ssi> it's a 3hp VFD, and nameplate says FLA is 13.5, which is around 4hp
[14:58:31] <ssi> but I'd think it wouldn't draw 4hp with the 2hp spindle motor not running!
[14:58:34] <deuplonicus> archivist, what do you mean?
[14:59:01] <tom3p> clamp on current probe show anything?
[14:59:09] <tom3p> ssi ^^^
[14:59:34] <archivist> deuplonicus, run the spiral.ngc
[15:00:07] <ssi> no, it's too fast
[15:00:18] <ssi> it's showing like 4A (!!!) with just the control fans running in estop mode
[15:00:32] <pcw_> I dont think in general that transformers. electronics, etc like being powered by a 5 KHz squarewave carrier
[15:00:32] <ssi> and when I hit control on, it trips faster than the probe can register
[15:00:33] <pcw_> modulated with 60 Hz
[15:00:49] <ssi> pcw_: yeah it ate the noise filters pretty quickly
[15:00:57] <deuplonicus> ok, i'll run spiral
[15:01:13] <ssi> but like I said, based on the schemes it looks like the ONLY place that 3ph is carried throughout is to the spindle and coolant motors
[15:01:53] <ssi> so I guess I'll rewire it so the main disconnect accepts 1ph, and then the VFDs replace the contactors for the motors
[15:02:31] <deuplonicus> is there anything specific we should look for besides missed steps?
[15:04:00] <ssi> suppose there's anything significant to the fact that the control power is taken from L1/L2, and the local 1ph is taken from L2/L3?
[15:04:11] <archivist> deuplonicus, missed steps can be because you are accelerating faster that the motors/hardware allows or going too fast where the torque is lower
[15:04:19] <ssi> I guess since they're floated at both ends with transformers, it probably is irrelevant
[15:05:37] <pcw_> I doubt it makes any difference (except in a 3 phase setup where they may have tried to balance the loads)
[15:06:22] <ssi> ah yea, balance likely is the reason
[15:06:30] <ssi> ok good, I should be able to make that happen
[15:06:38] <ssi> I guess I should order up a second small VFD to run the coolant pump
[15:06:47] <ssi> it's called out as 1/8 hp
[15:07:23] <ssi> whoa
[15:07:26] <ssi> http://dealerselectric.com/item.asp?cID=28&scID=165&PID=1007
[15:07:26] <ssi> $33 really?
[15:08:12] <djdelorie> sounds about right. 2-3 HP VFDs can be gotten for around $100 - $150
[15:08:34] <ssi> yea I know, but most of them are $100 minimum, even the tiny ones
[15:08:41] <ssi> I just might have to try the $33 one
[15:08:42] <ssi> hahaha
[15:10:08] <deuplonicus> ok, i'm almost ready to run spiral.ngc
[15:14:49] <archivist> deuplonicus, and for a giggle try tort.ngc
[15:16:10] <deuplonicus> well, I paused the spiral because it sounds like its missing a ton of steps
[15:16:55] <pcw_> a big thanks to Seb and the buildbot!
[15:16:57] <archivist> slow your acceleration and top speed limits and retry
[15:16:57] <pcw_> emc2-sim runs fine on 12.04
[15:19:36] <ssi> nice
[15:19:44] <pcw_> there's interaction between top speed and maxaccel as well
[15:19:46] <pcw_> (the higher the top speed, the less headroom is left for acceleration)
[15:19:54] <deuplonicus> we've already tried slowing down the max veloccity and accelleration, it skips even more steps when we run any gccode really slow
[15:20:30] <archivist> deuplonicus, what drivers are you using
[15:21:42] <archivist> some rubbish drivers drop the current and are too slow to restart when a step comes along
[15:24:31] <deuplonicus> We are using the Hobby CNC Pro controller, we know of the current idle issue. But I was under the impression that only happens when it starts up from rest, and you can hear it go into current reduction mode after 10 seconds of standing still.
[15:25:22] <archivist> define "rest" that is just running slow
[15:25:31] <deuplonicus> "rest is non moving
[15:25:36] <archivist> nope
[15:25:40] <pcw_> it only goes idle when all axis are idle for 10 seconds?
[15:25:47] <ssi> when I started building my gantry machine, I was using a hobby cnc board... I had a lot of problems with it, and eventually just swapped it for a G540... MUCH happier with the gecko
[15:26:05] <deuplonicus> we cannot change driver boards, simply they cost too much
[15:26:42] <pcw_> (pcw resists SA remark about what your time is worth)
[15:27:15] <deuplonicus> pcw, if time was the issue, we would've bought Mach 3 licensing, instead of using this awesome program
[15:27:44] <djdelorie> if time were an issue, you'd hire someone with a cnc machine...
[15:27:45] <pcw_> Doubt that that would help step drive issues
[15:27:53] <archivist> you would still have the hardware issue
[15:28:04] <deuplonicus> we have an idea to see if it's the hardware vs the code
[15:28:13] <deuplonicus> swap motor connections on the gantry
[15:28:43] * djdelorie sees a market for a "hardware stepper simulator" that keeps track of step/direction counts and tells you where the stepper SHOULD be
[15:29:00] <ssi> they're called encoders :)
[15:29:04] <deuplonicus> or at least to rule out if it is mechanical
[15:29:27] <djdelorie> no, it's an anti-encoder. Given the pulses, tell you the position. Encoders are given the position, and tell you the pulses.
[15:29:43] <deuplonicus> so should I try the Hobby CNC board mod a lot of people have done to turn off the current idle reduction?
[15:29:50] <archivist> yes
[15:29:53] <ssi> a device that you feed pulses and it tells you a position is a stepper
[15:29:54] <djdelorie> I mean something you connect to the same inputs as your stepper drive, with a digital display that says "count=NNN"
[15:29:56] <ssi> the encoder verifies the stepper :P
[15:30:17] <ssi> you should be able to get halui to tell you how many counts there have been
[15:30:19] <archivist> and mechanics
[15:30:19] <djdelorie> so you can tell if the problem is the stepper, or the software...
[15:31:07] <deuplonicus> djdelorie, that would be a nice gadget
[15:31:15] <archivist> steppers stall though so must never be allowed to miss a step
[15:33:59] <pcw_> Just an updown counter. But for debugging step drives is would have to have the same setup/hold/pulsewidth/pulsegap timing as the drive to be of much use
[15:34:22] <deuplonicus> the only reason I don't think it is the current idle reduction causing missed steps, is because the missed steps are sporadic and happen during constant motion
[15:34:45] <pcw_> Whats your latency
[15:35:18] <pcw_> ?
[15:35:39] <deuplonicus> max jitter was 10,000
[15:35:44] <deuplonicus> ns
[15:36:36] <deuplonicus> well, currentl max jitter is 8,000ns, I got 10,000 after letting it sit for hours on end
[15:36:50] <pcw_> random missed steps suggest:
[15:36:52] <pcw_> latency problems
[15:36:54] <pcw_> electrical problems (noise)
[15:36:56] <pcw_> violation of step drive pulse width ,setup or hold timings
[15:40:13] <deuplonicus> awesome, we've have grounded sheilded wiring everywhere. Latency seems appropriate according to everything I've read, So should I read the motor controller's IC's datasheet and verify the second three?
[15:40:18] <mrsun> any idea of my problem? im backlash compensating, tho on both X and Y when i go positive i get exactly 1mm each time i press +, then i turn, first step is 0.95mm, next is 1.05mm, next is 0.95mm next is 1.05 ... every freakin time, anywhere i try on the screw =)
[15:40:44] <pcw_> one drive working and another not also suggests
[15:40:46] <pcw_> a bad drive or marginal timing or marginal electrical drive
[15:40:52] <ssi> mrsun: what happens with the backlash comp turned off?
[15:41:02] <mrsun> ssi, hmm, havent tried realy .. good idea =)
[15:41:09] <mrsun> but its still a astrange phenomena
[15:41:26] <mrsun> gonna check tomorrow after work =)
[15:42:35] <pcw_> Yes I would verify the step timing
[15:43:10] <deuplonicus> If I switch my wires on the gantry, and the problem persists on the same motor, wouldn't that be a motor issue then? (it occurs when there is no tension on the bearins so we are pretty certain it is not mechanical)
[15:46:58] <pcw_> I would sure suspect the motor in that case (you could also measure the torque of both motors and check the they are the same) This is more of a pain to do if you have idle current reduction
[15:48:31] <deuplonicus> awesome, well, we must leave the machine (I really wish it was in my apartment), and work on it again tomorrow! Thanks for the help guys!
[15:54:53] <ssi> aw crap I knew I forgot something in my mesa order the other day :(
[15:57:17] <JT-Shop> I hate when that happens
[16:23:50] <pcw_> the sim axis.ini (inch) in 2.5 has a +Z limit of .0001 is this intentional? (it wont run the axis demo)
[16:39:27] * flyback sets up his new soldering iron on his 38th bday to test out, whoooohooo
[16:41:18] <TekniQue> damn you're old
[16:42:30] * Mjolinor at 56 thinks the opposite
[16:43:15] <TekniQue> a friend of mine died one week ago, at 56
[16:43:22] <flyback> sorry :/
[16:43:26] <flyback> <flyback>
http://www.sparkfun.com/products/10707
[16:43:26] <flyback> <d-tech> acidblue: any chance your adsl ip passthrough setting might be at blame?
[16:43:26] <flyback> <flyback> I love my aoyue 968 station and it's still the best iron I
[16:43:58] <flyback> but it being 35w it can't do thru hole, just not enough wattage
[16:43:58] <flyback> <flyback> aoyue even realized this and released the 968+ with 70w iron
[16:46:17] <Mjolinor> I have a 968 and the itron is excellent
[16:48:35] <flyback> 968 or 968+/A
[16:50:53] <Mjolinor> 968
[16:51:40] <Mjolinor> my workbench has 2 wellers, one pace adn the 968 and I think the 968 is best for general use
[16:53:22] <flyback> oh it's a good iron it just can't do thru hole
[16:53:32] <flyback> I crank it all the way up and sit on a joint for 10 mins
[16:53:33] <flyback> nothing
[17:05:28] <DJ9DJ> gn8
[17:33:56] <Roguish> hey all...simple question i hope. just did a git pull to get latest, good compile and all.... when starting linuxcnc, the 'my configurations' is empty, but all my configurations are still there.
[17:34:14] <Roguish> how does the 'my configurations' get populated????
[17:42:10] <cradek> are your configurations in ~/emc2 or ~/linuxcnc?
[17:43:23] <Roguish> ah, emc2
[17:45:09] <Roguish> got it. thanks cradek!!!!!
[17:45:16] <cradek> welcome
[17:45:19] <cradek> it was nothing :-)
[17:46:00] <Tom_itx> has the live cd been updated to reflect the name changes?
[17:46:08] <cradek> no
[17:46:33] <cradek> it's possible we may have a new cd after 2.5 is released, but I'm not doing it, so I can't promise
[17:48:33] <flyback> http://www.surplustraders.net/
[18:14:59] * flyback goes outside to grill his bday steak, bbl
[18:58:23] <Kirk_Wallace> Hello. I need to add an RTS feature that is in my demo source as an #ifdef. How and where is an #ifdef defined? Thanks.
[19:00:51] <SWPadnos> Real Time Strategy?
[19:01:39] <Kirk_Wallace> ?
[19:01:53] <SWPadnos> RTS is a game genre :) what do you mean by it?
[19:02:11] <SWPadnos> (like Starcraft, Warcraft, etc)
[19:02:32] <djdelorie> Kirk: the answer is "usually in a Makefile or *.h file", regardless of what RTS means...
[19:03:34] <SWPadnos> true, that is the right answer, though it may not address the real question
[19:03:50] <Kirk_Wallace> Oops. No Ready To Send in my Modbus source. It says "ifdef RTS_ENABLE" plus code to set it up. I need to figure out how to get this code compiled with RTS enabled.
[19:04:02] <SWPadnos> (which is probably something more like "how do I enable this feature")
[19:04:05] <SWPadnos> ah
[19:04:09] <djdelorie> ./configure --help ?
[19:04:29] <SWPadnos> modbus stuff has no ./configure entries, unless someone added it after I wrote the module ... :)
[19:04:44] <djdelorie> then look for CFLAGS in Makeifle, or some global .h file. I'm just guessing here...
[19:05:26] <SWPadnos> kirk_wallace, one easy way to do it is to add #define RTS_ENABLE to the modbus.h file, somewhere near the top probably (search for it, they may have an #undef in the source)
[19:05:52] <Kirk_Wallace> It's for FreeModbus running on an AVR.
[19:05:56] <SWPadnos> the other ways require you to change the Makefile in some way
[19:05:58] <SWPadnos> oh
[19:06:05] <SWPadnos> the other end of Mobus :)
[19:06:09] <SWPadnos> +d
[19:07:15] <SWPadnos> do you have a makefile, or do you just avr-gcc "some stuff" ?
[19:07:43] <Kirk_Wallace> I have a Makefile. I was used to "define" "something" "is something else" but define RTS_ENABLE should do it?
[19:08:12] <SWPadnos> yes. somewhere in the Makefile, add "-DRTS_ENABLE" to the cc options
[19:08:39] <djdelorie> for "ifdef" it doesn't matter what you define it to. If you don't give a value, the default is '1', but just "being defined" is enough
[19:08:40] <SWPadnos> without the quotes, of course (though I think that may not matter in this case)
[19:09:22] <Kirk_Wallace> Okay, great, I'll give it a try. Thanks. Be back soon.
[19:09:33] <SWPadnos> see you. I'm out too
[19:11:34] <Kirk_Wallace> djdelorie: Thanks. I'm going to try the CFLAG thing.
[19:12:02] <jdhnc> Do you use some random IO pin for RTS/CTS?
[19:14:14] <Kirk_Wallace> No, it is defined in the ifdef section as PB0.
[19:14:57] <jdhnc> so, semi-random :)
[19:17:01] <Kirk_Wallace> I added -DRTS_ENABLE to -> CFLAGS += -Wall -gstabs -DF_CPU=10000000UL -DRTS_ENABLE -Os -Wall -Wstrict-prototypes
[19:17:42] <Kirk_Wallace> But make says there is nothing to do, so I guess I'll touch the source file.
[19:18:14] <djdelorie> or rm the *.o
[19:22:46] <jdhnc> or jumper out the handshaking on the other end
[19:26:37] <Kirk_Wallace> I'm using half-duplex and need to switch the 485 driver. The compile completed without errors but the comm. test still doesn't work. I might have to actually do some work and get out some tools.
[19:54:52] <Nick001> On a lathe - how do i get 2 offsets for the same tool-like m6 t7 g43 h7 and m6 t7 g43 h10 - I need 2 different x offsets for t7
[19:57:14] <Nick001> using these 2 g codes gives an error
[20:01:38] <Kirk_Wallace> It has been a while since I've thought about this, but on my lathe I have two or three tools at each turret position. I recall selecting a tool to get the turret position then set a separate tool offset, but the way the tool table works might have changed since then.
[20:16:00] <Nick001> I have the same hardinge you have - just need 2 offsets to turn 2 dia's and control each one. 1 offset won't do it. I'm changing the 2nd dia in the program right now
[20:17:55] <Nick001> I'm using 2.4.6 right now.
[20:31:42] <Kirk_Wallace> I briefly looked over the table info on the wiki and its completely different than what I have on my lathe. Sorry I'm not much help. I'll need to go through this when I upgrade.
[20:36:55] <Nick001> Thanks for the effert BTW - when you go to 2.5 - the turret comp won't write to the files like it does in 2.4.6
[22:04:49] <ssi> made progress!
[22:04:52] <ssi> got the axes jogging
[22:28:26] <tehDarkAura> hey that cnc position display seems pretty simple -- anyone make their own?
[22:29:18] <tehDarkAura> would like to grab a pinout and have an avr display to an lcd rather then the 8 segment led things
