#linuxcnc | Logs for 2012-02-23

[02:05:41] <DJ9DJ> moin
[03:37:18] <pingufan> Hello, can somebody, please, explain me how I can solve this wrong display? http://www.hantsch.co.at/_temp/Snap033.jpeg
[03:37:52] <pingufan> Watch the vertical black stripe and this yellow colored area in the realtime display.
[03:41:54] <pingufan> I have this since I installed EMC2 on this system. (VIA C7 1000 MHz with OnBoard Unicrome Graphics).
[03:43:38] <pingufan> If somebody has an idea, please let me know. As all the rest (and also the desktop when AXIS is not opened) is perfectly right, this must be an EMC issue.
[03:51:30] <mhaberler> one doesnt follow from the other - looks like an OpenGL problem. see http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TroubleShooting 5.2
[04:09:59] <pingufan> So I can actually do nothing against that - right?
[04:10:39] <pingufan> (except of fixing openGL
[04:13:03] <pingufan> Can I get my 8.04 version switching power really off at shutdown? It always ends with "system halted" but dows not power off.
[04:16:14] <pingufan> Is there a G-Code to mirror the X coordinate? I have to cut out two identical parts, but because they have to be symmetric, I would have to enter now everything mirrored for the second part. Not really funny.
[04:17:14] <pingufan> If i could move to the mirroring point and define it as new X=0, it would be much easier.
[04:24:00] <pingufan> Can somebody, please, tell me how to define i.e. Coordinate system G55 to have +X towards left and -X towards right?
[04:33:59] <pingufan> Nobody here who can help with coordinate system, please?
[04:43:13] <pingufan> m42: Can you, please, help me with Coordinate system of emc2 ? I want that X axis works mirrored.
[04:43:46] <pjm> pingufan what u mean? u want X to go the other way than wat its currently doing?
[04:44:41] <pingufan> I have written a G-Code job (manually). Now I need exactly the same part again, but mirrored.
[04:45:05] <pjm> ah yes i see, probably is a way of doing that
[04:45:40] <pingufan> But how. :/
[04:46:07] <pingufan> I search the docs, find blurry infos about G54..G59, but nothing in detail.
[04:46:30] <pingufan> I am also not really sure that this are the right codes for that.
[04:46:32] <m42> pingufan: and if you invert your XAxis ?
[04:46:40] <pingufan> How?
[04:46:49] <pingufan> (I am a novice)
[04:47:08] <pingufan> I do not want to reconfigure my machine.
[04:47:14] <m42> application \ cnc \ stepconfwizard
[04:47:19] <m42> and modify
[04:47:30] <m42> it's quick and not dangerous for your config
[04:47:45] <pingufan> Oh, oh. No. That becomes a problem, as homing and everything else will be hurt by that.
[04:48:02] <archivist> re use of stepconf is not advisable if any hand editing of the configuration has been done
[04:48:15] <pingufan> I can't believe that this cannot be done by reconfiguring the user coordinate system.
[04:48:20] <archivist> m42 it IS dangerous
[04:48:33] <m42> ok
[04:48:45] <archivist> shouldnt be I know
[04:49:01] <m42> i never modify my stepconf manually :S
[04:49:11] <pingufan> archivist: any idea how to do this trivial issue?
[04:49:15] <m42> so that's why i said that
[04:49:45] <pingufan> And I am happy that it works on my mill. Therefore I also don't change it again.
[04:49:48] <archivist> m42 stepconf does not read its output file
[04:50:14] <m42> pingufan: and you can't modify your Gcode?
[04:51:11] <pingufan> This is hand-written code which moves along a big polygon with curves, edges, ... I'd have to re-calculate every point!
[04:51:12] <archivist> pingufan, if I write some gcode for multiple uses I code in a flag/variable so it changes as needed
[04:51:35] <m42> +1
[04:51:56] <pingufan> You surely are more experienced than me.
[04:52:13] <pingufan> But that does not really help now.
[04:52:13] <archivist> the calculation becomes internal to the code eg O subroutines with a direction parameter
[04:53:10] <pingufan> Aha. And what happens with tool radius compensation, arcs (g2 <-->g3) etc?
[04:53:56] <pingufan> Anyway. There surely is a command which mirrors the X axis.
[04:54:13] <archivist> often I dont use tool compensation on the mill because of the work I do
[04:54:24] <pingufan> This is what I am looking for.
[04:54:55] <pingufan> I use it almost always. When hand-coding G code, it makes life much easier.
[04:55:15] <pingufan> And hand coding is the only way to really learn G-codes.
[04:55:19] <pjm> i guess emc2 doesnt have something liek the g101 command which the haas has for mirror image
[04:55:25] <archivist> mirror is on the wish list http://wiki.linuxcnc.org/cgi-bin/wiki.pl?EmcFeatures
[04:58:23] <pingufan> Ok. So a simple "mirrir", maybe. But can't I overwrite a variable to let X move in opposite direction, and later set it back? Surely not elegant, but would help for now.
[05:00:39] <automata> pingufan: have you looked at grecode?
[05:01:04] <pingufan> No. I never heared from that.
[05:01:30] <pingufan> Anyway. I must get it done. So I will recalculate all points. :(
[05:01:34] <automata> http://code.google.com/p/grecode/
[05:02:23] <automata> try that first...
[05:03:00] <automata> xflip and then shift
[05:03:07] <automata> that may do it...
[05:04:08] <pingufan> Linux or windows program?
[05:04:54] <automata> it is a c program with no dependencies...
[05:05:30] <automata> probably just libc
[05:06:01] <automata> and source is available so a complie will be just a matter of running make (ofcourse with a C compiler installed).
[05:06:01] <pingufan> Can I comment out multiple lines of gcode ?
[05:06:23] <automata> yes. use ( )
[05:06:35] <pingufan> Across multiple lines?
[05:06:38] <automata> yes
[05:06:46] <pingufan> ( in line 3 and ) in line 20 ?
[05:06:50] <automata> try it... it is easy....
[05:07:13] <automata> i mean it is a simple matter to try the multi-line comment...
[05:07:47] <pingufan> I get an error (in AXIS), that the comment is not closed.
[05:08:19] <pingufan> I thought there is a special "comment marker" for that.
[05:14:55] <pingufan> Does a goto and a label command exist? So I could skip code.
[05:15:44] <pjm> u could just use some variable in X moves, then move the datum and X offset i would think
[05:19:09] <jthornton> http://linuxcnc.org/docview/2.5/html/gcode/overview.html#_comments_a_id_sec_comments_a
[05:20:35] <jthornton> pingufan: what are you trying to do?
[05:23:49] <pingufan> I try to mill parts mirrored. Meanwhile I have recalculated all coordinates and enter them.
[05:31:59] <jthornton> I mean when you want to skip lines of code... do you know about the block delete character the slash?
[05:32:16] <pingufan> What I will need in next future, is a perfectly working 2.5D G-Code generation out from AutoCad R14. I have to engrave much text, this is not funny with hand-coding. Is REALIZE the best way?
[05:34:19] <jthornton> I don't have a clue what REALIZE is...
[06:04:48] <_abc_> Has anyone got info on Chinese cnc pendant-controllers like the one in the link http://signtide.en.made-in-china.com/product/DenxpOMchVkP/China-CNC-Controller-DSP-Control-Board-System-for-Motor-Motion.html ? I am looking for the cnc manual (command set supported). Thanks for any info. And it MIGHT end up a linuxcnc machine later at some point (especially if I do not find the manual)
[06:07:37] <IchGuckLive> joe9: this is a better mill for you
[06:07:54] <IchGuckLive> joe9: http://s.glbimg.com/jo/g1/f/original/2012/02/22/620x388.jpg
[06:08:26] <IchGuckLive> shit happens -> joe9: http://www.ebay.de/itm/CNC-3020-Router-Engraver-Drilling-Milling-machine-Neu-/280680085334?_trksid=p5197.m185&_trkparms=algo%3DSIC.NPJS%26itu%3DI%252BUA%26otn%3D12%26pmod%3D120833940263%252B120834810495%252B120813745732%26po%3DLVI%26ps%3D63%26clkid%3D5325942910028628560
[06:29:01] <_abc_> What type of controller is this? http://www.aliexpress.com/fm-store/908764/211369879-506600498/M012-Free-Shipping-3-axis-PCI-motion-control-card-router-engraver-Motion-Control-Board.html
[06:29:20] <_abc_> Is it like an optically separated IO?
[06:34:25] <archivist> _abc_, http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Emc2HardwareDesign
[06:35:26] <_abc_> okay, ?
[06:35:48] <_abc_> So it's a special parallel port?
[06:35:53] <archivist> and why boards like that dont get used
[06:35:58] <_abc_> oh
[06:36:08] <_abc_> That does look like a parallel though.
[06:36:11] <_abc_> Okay I get the point
[06:37:02] <_abc_> Still, 200usec output rate for 16 bits means 2500Hz on the motors which is not too shabby.
[06:37:13] <archivist> getting that linuxcnc has the trajectory planner in realtime IN the pc is the first bit
[06:37:40] <_abc_> Sure, but how does that come into play here?
[06:37:55] <_abc_> Also a USB serial thing will go to 1MBps, 10 times faster than the 115200 serial
[06:38:14] <archivist> usb has crap latency
[06:38:31] <_abc_> Not on serial cdc when properly done, the machine can buffer a bit
[06:38:45] <_abc_> most moves are done open loop in stepper actuated machines
[06:38:55] <archivist> buffers upset realtime feedback
[06:39:08] <_abc_> Who uses realtime feedback at full speed?
[06:39:35] <archivist> anyone with servo
[06:39:50] <_abc_> In fact, the tiny machines I have to do with only have feedback from the end switches, and that only when the machine zeros at the beginning, under operator manual control.
[06:40:18] <_abc_> archivist: what servo? you mean like geckos and other servo drives?
[06:43:16] <archivist> or pwm driven servo amps with encoder feedback
[06:44:40] <archivist> an example http://wiki.linuxcnc.org/cgi-bin/wiki.pl?PWM_Servo_Amplifiers
[08:14:19] <joe9> I have a drill press without a laser sight. I am wondering if there is a smart way/technique to align the drill press to the hole.
[08:29:54] <jthornton> you have a hole now you wish to enlarge?
[08:31:01] <joe9> jthornton: I am reading up on using a drill press. I want to drill 0.5cm (diameter) holes (depth = 1mm ) in a pcb
[08:31:02] <Tom_itx> joe9, http://www.travers.com/skulist.asp?RequestData=CA_Search&navPath=All+Products%2F%2F%2F%2FUserSearch1%3Dedge+finder&q=block+id+37205+and+class+level3+id+29042&minPrice=$23.59
[08:31:05] <Tom_itx> or a dowel pin
[08:31:48] <Tom_itx> oh heck, pcbs don't need that great of accuracy
[08:32:18] <joe9> Tom_itx: It does not have to be very accurate, but I do not want to drill through the pcb either.
[08:32:29] <joe9> I just want a small ditch kind of hole.
[08:32:34] <joe9> Does that make sense?
[08:32:43] <Tom_itx> slot?
[08:32:48] <joe9> yes.
[08:33:14] <Tom_itx> for a locating dimple or something?
[08:33:24] <joe9> reading up on these instructions: http://codepad.org/mdPjr0Ez
[08:34:43] <joe9> i am not sure if the locating dimple is 1mm deep.
[08:35:10] <Tom_itx> that pretty much describes what i just said about the dowel pin
[08:35:36] <joe9> oh, ok. dowel pin == "center-pointed drill rod"?
[08:35:48] <Tom_itx> whatever works
[08:36:13] <joe9> Tom_itx: sorry, I do not know the terminology. did not mean to frustrate you.
[08:36:22] <Tom_itx> you're not
[08:37:13] <Tom_itx> sometimes things require different methods depending on the task
[08:39:13] <joe9> i agree. I understand that I might have to "dimple with sharp scriber joint" + chuck-up a piece of accurately center-pointed drill rod and align the dimple + replace drill rod with drill bit.
[08:39:24] <joe9> do you think I can do something differently?
[08:39:38] <joe9> Tom_itx: I just got those instructions from reading up on the internet.
[08:41:48] <Tom_itx> when i drilled these, i punched the center then just used the drill bit to align into the dimple the punch made: http://tom-itx.dyndns.org:81/~webpage/temp/tiny/tinyTPI2.jpg
[08:41:52] <Tom_itx> worked fine for me
[08:42:34] <joe9> what did you use to punch the center?
[08:44:49] <Tom_itx> http://www.generaltools.com/79--PROFESSIONAL-AUTOMATIC-CENTER-PUNCH_p_34.html
[08:47:09] <joe9> Tom_itx: thanks. on a different note, for those through holes, did you use a clamp to hold the work piece? If you used a clamp, did you use parallels below the piece? Do you remove the parallels while drilling and after clamping?
[08:47:30] <Tom_itx> nope
[08:47:33] <Tom_itx> just drilled em
[08:47:47] <joe9> oh, ok.
[08:47:58] <Tom_itx> with a scrap board under it
[08:49:34] <joe9> ok, thanks.
[08:49:45] <joe9> btw, do you use harbor freight tools?
[08:49:54] <joe9> or, are they a waste of time?
[08:50:33] <joe9> I am looking to buy a center punch, machinist's transfer punch (to align the drill press) and parallels.
[08:51:52] <joe9> I am planning on doing this http://www.woodworkersjournal.com/Main/Articles/Drill_Press_Maintenance_Skill_Builder_Power_Tool_T_8549.aspx to align my drill press.
[08:52:29] <joe9> I am thinking of using the machinist's transfer punch to ensure runout is not bad and to ensure that the table is perpendicular to the drill.
[08:52:57] <joe9> I have a used drill press and I see that the quill seems to be moving a little (a tiny bit), hence the alignment.
[08:55:18] <joe9> or, do you think a machinist's tranfer punch is too much tool for just aligning? Do you recommend a different instrument, perhaps?
[09:01:41] <joe9> Tom_itx: any thoughts, please?
[09:11:39] <Tom_itx> i tend to buy decent tools
[09:11:46] <Tom_itx> gotta go now
[09:11:58] <joe9> ok, thanks. where do you normally buy tools?
[09:32:32] <seb_kuzminsky> this is relevant to my interests: http://www.pagunblog.com/2011/11/17/cnc-machining-an-ar-15-lower/
[09:35:35] <IchGuckLive> seb_kuzminsky: do you got already a mashine or do you search fore one
[09:36:15] <GoSebGo> I have a bridgeport
[09:37:00] <IchGuckLive> is this th same person seb_kuzminsky GoSebGo
[09:38:04] <archivist> yes
[09:38:53] <IchGuckLive> cnc controled
[09:42:12] <IchGuckLive> good communication O.O
[09:44:23] <IchGuckLive> seb_kuzminsky: 3 or 4 axis controlld
[09:45:25] <IchGuckLive> Pa6.6(66) is better to mill then Pa6
[09:46:16] <skunkworks> IchGuckLive: thanks again for the files. I hope to goof around with them soon.
[09:46:19] <JT-Shop> I'm with you on that one seb_kuzminsky
[09:46:52] <IchGuckLive> skunkworks: arduino0022 keep in minde 1.0 is differend
[09:47:23] <skunkworks> ok
[09:47:42] <IchGuckLive> JT-Shop: diddend you also sell cncn mini kids ?
[09:51:04] <JT-Shop> just regular kids
[09:51:46] <IchGuckLive> joe wars alking yesterday from atlante its your region
[09:52:31] <JT-Shop> same country but I'm in the middle and he is on the east coast
[09:53:09] <IchGuckLive> in the USA its around the corner
[09:53:30] <JT-Shop> I've proven beyond a shadow of a doubt that a 6-32 tap will work in a 2.5mm hole but a M3 screw still will not fit... anyone seen my 3mm tap
[09:53:31] <IchGuckLive> here in europ it woudt be a world tour
[10:01:02] <Jymmm> 6/32 == .1875 3mm ==.1181
[10:01:36] <Jymmm> .0984 == 2.5mm
[10:07:11] <IchGuckLive> off time by till 1900MEZ
[10:10:51] <joe9> I am missing the wedge key to remove the chuck from a craftsman drill press. Any suggestions on what else I can use instead?
[10:15:38] <JT-Shop> two screwdrivers
[10:16:21] <joe9> oh, really. let me try that.
[10:16:23] <joe9> thanks.
[10:16:51] <Jymmm> and a chipmuunk
[10:16:52] <Tom_itx> jt, do you watch redjacket?
[10:17:09] <Tom_itx> they revived a russian mortar on it last night
[10:17:26] <Jymmm> you mean sons of guns?
[10:17:30] <Tom_itx> that one
[10:19:44] <Jymmm> it was pretty cool
[10:22:07] <Jymmm> I have no idea what the mortar's payload was, but it looked liquidy, napalm maybe?
[10:22:54] <Jymmm> and it was a HUGE impact, never realized how big of an area one could cover/wipeout.
[10:49:23] <Jymmm> Tom_itx: http://gunnuts.net/2011/12/23/sons-of-guns-loses-ffl/
[10:54:02] <MattyMatty> hi all. still getting this error when trying to jog in stepconf axis test .http://paste.org/45858
[10:54:22] <MattyMatty> http://paste.org/45858
[10:56:24] <MattyMatty> also the latency test won't start and emc2 crashes
[11:00:55] <MattyMatty> http://paste.org/45860 emc2 crash
[11:03:07] <MattMatty> is the linuxcnc repo already installed on the 10.04 cd?
[11:03:54] <MattMatty> if so, then it's all updated
[11:05:41] <skunkworks___> MattyMatty: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TroubleShooting#emc2_doesn_t_run_missing_lapic
[11:07:33] <MattMatty> cheers, I'll work through the whole page :)
[11:19:40] <MattMatty> sweet emc2 runs. cheers skunkworks
[11:30:12] <Cylly> re @ home
[11:30:34] <Jymmm> re @ Earth
[11:30:46] <MattMatty> I'l take this thinkpad to test on my mill before I go any further tracking down the latency and rt error issues
[11:31:57] <MattMatty> I've always had RT errors on my mill's pc too, but it hasn't stopped me working. it's all steppers on the mill
[11:35:53] <MattMatty> joe9 is your chuck a Jacobs chuck screwed into a taper shank? if so I had luck by unscrewing the chuck then screwing it back on over a ring spanner. that gave an even pull
[11:36:40] <joe9> MattMatty: no idea. will check that. thanks for the tip.
[11:42:06] <pingufan> hello. I finished my Tooltip sensor right now. It works very precisely with a small hysteresis and is connected to parallel port Pin 11. When not pressed, Pin11 is pulled to ground.
[11:42:34] <Jymmm> and a resistor?
[11:42:43] <Jymmm> add
[11:44:04] <pingufan> As far as I remember, parallel inputs have the pullup resistor built in. I have an optocoupler between my mill and the PC.
[11:44:16] <pingufan> This perfectly works with home switches.
[11:45:04] <Jymmm> Not all paraports are built the same.
[11:45:37] <pingufan> How do I add the limit switch to EMC2? I have an already configured mill.
[11:45:51] <pingufan> ^^ Well, then this parport is standard. ;)
[11:46:06] <Jymmm> there is no standard.
[11:46:57] <pingufan> Anyway - it behaves identical to an IBM AT and Olivetti M24 (>10 years ago)
[11:47:36] <Jymmm> Is it so hard to just try a resistor and see if that resolves your issue?
[11:48:33] <pingufan> I have no issue, I ask about how to (re-)configure EMC2 to support this switch at all.
[11:48:51] <pingufan> Is this done in stepconf?
[11:48:58] <cradek> Jymmm: wtf are you on about?
[11:49:01] <Jymmm> Yes, I believe os.
[11:49:16] <cradek> pingufan: you'll want to hook that pin to motion.probe-input
[11:49:18] <Jymmm> cradek: I misunderstood what he was inquiring about.
[11:49:25] <pingufan> Indeed.
[11:49:26] <cradek> make sure it goes TRUE when the switch is poked
[11:49:49] <cradek> perhaps you can do it in stepconf if you haven't made any custom changes outside of stepconf
[11:49:58] <Jymmm> pingufan: Well, you didn't actually ask a question initially, just made a statement.
[11:50:29] <pingufan> I am sure that the pin is pulled to ground when the switch is NOT pressed down.
[11:50:49] <cradek> pingufan: you get in and in-not. just pick the right one.
[11:51:12] <pingufan> Where? In stepconf?
[11:52:09] <skunkworks> MattMatty: Great! Lots of times notebooks don't work well for realtime systems
[11:52:19] <skunkworks> Though have you tried the SMI patch?
[11:52:28] <skunkworks> (seems to be an older system?)
[11:52:41] <MattMatty> T40. quite old :)
[11:52:50] <MattMatty> only the old ones have parports
[11:54:53] <MattMatty> T30 actually. I have a pile of vintage thinkpads I forget which one I'm using
[11:55:16] <Jymmm> vinatage, hahahaha,
[11:55:23] <pingufan> cradek: I am now in stepconf, on the page where I configure the pin assignment. There I have currently configured pin 10 as "All Reference Switches", as my limit switches (for homing the machine) are in parallel and connected there.
[11:55:48] <pingufan> Pin 11 is for the tool sensor. What shall I choose for it?
[11:55:54] <MattMatty> 8 years old is vintage for a lappy, although I have older ones
[11:56:13] <MattMatty> a P166 toshiba and a 486 one too :)
[11:56:26] <Jymmm> MattMatty: Then the rocks in my backyard are rare antiques.
[11:56:29] <cradek> "Probe In"
[11:57:21] <MattMatty> my thinkpads are all made in scotland. that makes them classics from the age of british manufacturing
[11:57:34] <pingufan> That is not available. I have Digital Input 0/1/2/3, All Reference Switches, All Limit Switches,...
[11:58:33] <cradek> hm, it's there in my stepconf. If probe support isn't in your version, you'll have to write one line of hal manually
[11:59:02] <MattMatty> on the subject of british computers, has anyone considered a raspberry pi as a controller?
[11:59:09] <pingufan> This is an older emc2 (running on 8.04).
[11:59:17] <cradek> raspberry pie sounds good
[12:00:50] <pingufan> cradek: what shall I select?
[12:01:01] <cradek> hm, it's there in my stepconf. If probe support isn't in your version, you'll have to write one line of hal manually
[12:01:08] <pingufan> Digital Input?
[12:01:30] <MattMatty> they're not on sale yet, but they're neat little boards with ~20 IO, arm SoC and a useful sounding gpu
[12:01:42] <pingufan> I think I can do that. If you tell me, what I have to do.
[12:01:54] <cradek> bbl, lunch
[12:02:09] <pingufan> :)
[12:02:28] <pingufan> Ok. So I close stepconf for now.
[12:02:58] <MattMatty> damned fruit-delinieated computers. it's so hard to get a hungry person interested :)
[12:09:28] <IchGuckLive> hi all O.O
[12:25:28] <pingufan> cradek: Please, tell me when you are back from lunch. I wait.
[12:26:53] <archivist> pingufan, in general people dont do one to one support in irc, just ask the question
[12:28:58] <IchGuckLive> pingufan: agree to archivist
[12:57:51] <pingufan> archivist: You are welcome to help me, too. ;)
[13:03:17] <IchGuckLive> yust usk the problem you got
[13:12:06] <pingufan> The problem I currently have is: My EMC2 (2.3.5) doesn't show me a "Probe In" type in stepconf for my tooltip-sensor. Cradek said that a line has manually to be added to HAL to get the switch operational.
[13:13:15] <pingufan> I neither know _what_ has to be manually added, nor _how_ to do that correctly.
[13:14:44] <IchGuckLive> witch pin woudt you like to use
[13:16:07] <pingufan> Pin 11 is for this sensor. When the button is NOT pressed, this pin is pulled to ground. (I haven't looked if Pin 11 is an inverting input or not, I tell you what is applied from external.)
[13:17:10] <IchGuckLive> Parport i guess o.o
[13:17:20] <pingufan> Yes. Parport.
[13:17:40] <pingufan> Better than RS-232. ;)
[13:17:54] <IchGuckLive> ok
[13:19:57] <IchGuckLive> so go to your mashine config folder
[13:21:07] <IchGuckLive> open the .hal of you mashine
[13:21:47] <IchGuckLive> and at-> net probe parport.0.pin-11-in => motion.probe-input
[13:22:10] <IchGuckLive> by the way how many axes
[13:22:37] <pingufan> I have 3 axes (x,y,z), it is an ordinary mill.
[13:23:25] <IchGuckLive> at it at the last line
[13:24:56] <IchGuckLive> safe it and you done :D
[13:25:36] <pingufan> Does it matter WHERE I insert this line?
[13:25:46] <IchGuckLive> at the end
[13:26:04] <IchGuckLive> please read this -> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Preventing_Probe_Crashes_Using_HAL
[13:27:27] <pingufan> Can I see/test somehow what emc sees on this pin?
[13:27:47] <IchGuckLive> halmeter
[13:27:55] <pingufan> I assume that I must not use stepconf from now on any more. Right?
[13:28:04] <IchGuckLive> yes
[13:28:28] <IchGuckLive> it is not that act to get this in
[13:28:38] <pingufan> Where do I fnd halmeter in my PC?
[13:28:47] <IchGuckLive> axis
[13:29:21] <IchGuckLive> mashine -> halmeter
[13:30:03] <IchGuckLive> choose pins go down to parport.0.11
[13:30:08] <IchGuckLive> hit return
[13:30:37] <jdhnc> he doesn't seem to understand the meaning of 'E-Stop'
[13:31:04] <IchGuckLive> E stands for emergency
[13:32:40] <IchGuckLive> pingufan: if you hit then the button it shoudt change from false to true
[13:33:27] <IchGuckLive> also check signal motion.probe-input
[13:33:48] <IchGuckLive> this shoudt also change
[13:34:13] <IchGuckLive> so im off its late here by till tomorrow
[13:34:18] <IchGuckLive> GN8
[13:36:15] <pingufan> Ok. When I choose parport.0.pin-11.in I get this: Button not pressed = FALSE Button pressed = TRUE.
[13:41:38] <pingufan> Yes. This happens, when I press it. Is this the right direction? Pressed = TRUE ?
[13:42:23] <pingufan> cradek, are you back from lunch?
[13:51:19] <pingufan> cradek: IchGuckeLive was so kind to tell me how to edit the hal file. I verified then with "halmeter" that "parport.0.pin-11-in => motion.probe-input" becomes TRUE, when I press the tooltip-probe-button.
[13:51:52] <pingufan> What is the next step for making automatic tool-length correction?
[13:52:04] <archivist> read some docs
[13:53:21] <pingufan> I love such answers.
[13:55:22] <archivist> well I dont have time at the moment to read them for you
[13:56:28] <pingufan> I can read by myself, but I do not know what to search for. Where is described what I have to do?
[14:00:33] <jthornton> pingufan: you "connect" the parallel port pin to motion.probe-in
[14:00:41] <jthornton> in your HAL file
[14:01:29] <pingufan> Is this another line in the HAL file, or what I already edited?
[14:01:48] <pingufan> Excuse me, but I am a novice.
[14:01:58] <jthornton> I assume you will need to add a line
[14:03:42] <pingufan> Sorry, I stuck. Can you explain me, please?
[14:04:09] <jthornton> net probe-in motion.probe-input <= parport.0.pin-10-in
[14:04:26] <jthornton> but change the pin number to suit your configuration
[14:05:40] <pingufan> Which pin shall I take? The normal one or the inverted one? The normal one becomes TRUE when I press the button.
[14:07:52] <jthornton> http://linuxcnc.org/docview/2.5/html/man/man9/motion.9.html
[14:08:14] <jthornton> find motion.probe-input
[14:08:47] <jthornton> then decide
[14:09:19] <pingufan> Ok. So I need the non inverted. Thank you.
[14:09:43] <jthornton> alex_joni: used to do the same for me... force me to learn
[14:10:19] <raynerd> guys, to run a small cnc router 13"x13" using a TB6560 3 axis driver and some small Y163 1A 5v stepper, will a 12v 5A power supply be enough?
[14:10:33] <pingufan> That's ok, but when I have no idea about how all things belong together, this is a very hard way.
[14:10:57] <jthornton> yes, that is why I give you a hint
[14:11:31] <jthornton> raynerd: 12v will barely run anything
[14:11:57] <raynerd> well it is the original supply that came with this TEP machine..but I`ve changed the driver
[14:12:01] <jthornton> but if your not in any hurry to get anywhere you might live with it
[14:12:03] <raynerd> the motors are original
[14:12:21] <raynerd> in what way will it be limiting?
[14:12:30] <raynerd> slow, torque...?
[14:12:38] <jthornton> speed, torque, acceleration
[14:12:41] <raynerd> ok
[14:12:48] <raynerd> but might be enough just to get it running
[14:12:55] <pingufan> So this "connection" connects this logical pin11 to the G38 code?
[14:12:57] <jthornton> yea, it might work for that
[14:13:03] <jthornton> pingufan: yep
[14:13:30] <jthornton> raynerd: gotta start somewhere if you have it all ready give it a whirl
[14:13:46] <jthornton> have you seen the wiki page on steppers and voltage etc
[14:14:22] <raynerd> yea
[14:14:35] <raynerd> sorry... yes..I have it so just wanted to check I could at least try
[14:14:40] <raynerd> wiki page?
[14:14:44] <pingufan> And how is the whole process of using the tool sensor now? I assume I will have to probe the first tool, to touch off then the zero point of "Z", ....
[14:14:44] * jthornton goes to prepare the ingredients for tomorrows brew session
[14:14:47] <jthornton> yea
[14:15:04] <pingufan> You brew?
[14:15:06] <pingufan> Beer?
[14:15:15] <jthornton> Configuring LinuxCNC - how to make it fit your needs
[14:15:22] <jthornton> http://wiki.linuxcnc.org/cgi-bin/wiki.pl
[14:15:53] <jthornton> pingufan: yes, tomorrow is 10 gallons of Hefeweizen
[14:16:17] <pingufan> I also have a self-built home-brewery.
[14:16:32] <pingufan> But I have not sufficient time.
[14:17:16] <pingufan> Therefore I buy currently ready made wort and save the time of brewing
[14:24:29] <jthornton> http://gnipsel.com/beer/beer.html
[14:26:22] <pingufan> Here is mine: http://brewing.hantsch.co.at
[14:26:56] <pingufan> This is in early stage. Testing. Therefore everything looks a bit naked.
[14:27:10] <pingufan> Meanwhile I am some steps further.
[14:29:16] <pingufan> jthornton: I find nothing that describes how to configure and USE the tool sensor.
[14:32:33] <skunkworks> pingufan: you have a sample gcode program that explains it http://git.linuxcnc.org/gitweb?p=emc2.git;a=blob;f=nc_files/tool-length-probe.ngc;h=8b41233d2afe19ef6b9f17706108d0eb356f066b;hb=HEAD
[14:34:00] <pingufan> skunkworks: tHANK YOU VERY MUCH. This brings light into the darkness.
[14:37:43] <pingufan> Only a question to G30: It makes a linear 3D move to the preset position, or can I define that it first moves to maximum Z and then moves to X/Y ?
[14:38:10] <skunkworks> 3d move.
[14:38:57] <raynerd> Wahooo...all my axis move, but they seem a little juddery!
[14:38:57] <pingufan> Background is: First get as far away as possible from workpiece, then do horizontal move.
[14:40:03] <raynerd> Guys, what is the rule as far as setting the velocity and acceleration on each axis go?
[14:42:04] <raynerd> I currently have it set to 15mm/s for both ... not convinced this is ideal for the sound it is making
[14:43:10] <raynerd> is it just trial and error?
[14:43:42] <pingufan> Are you using steppers?
[14:44:11] <raynerd> yes, sorry. 1A 5v Y163 model steppers
[14:44:50] <raynerd> actually since posting...I have it at 30mm/s velocity and 25mm/s accel. Seems to be running much better
[14:45:39] <jthornton> http://linuxcnc.org/docview/html/common_Stepper_Diagnostics.html#r1_3
[14:46:34] <raynerd> ahh OK.
[14:46:45] <pingufan> If you get wrong positioning, you possibly are accelerating too fast or moving too fast at all.
[14:47:04] <raynerd> I have only just gone through the step config wizzard so haven`t even seen emc2 running yet
[14:47:10] <raynerd> but I`ll try and give that a whirl!
[14:47:51] <pingufan> jthornton: When I add this to my HAL file, AXIS aborts when loading. net probe-in motion.probe-input <= parport.0.pin-11-in
[14:48:55] <pingufan> What is wrong with it?
[14:49:41] <pingufan> This line here works fine and stands right before the problematic one: probe parport.0.pin-11-in => motion.probe-input
[14:56:16] <pingufan> skunkworks: Can you help me with configuring, please?
[14:57:40] <raynerd> Driving me nuts, i`m on the home "splash" emc2 demo and it won`t run. Error Linear move on line 22 would exceed joint 2`s negative limit
[14:57:49] <raynerd> anyone any ideas what I`m doing wrong
[14:58:39] <pingufan> Did you home the machine, and then also touch off all axes?
[14:59:05] <raynerd> oopps.. I`m homed, didn`t touch off!
[14:59:12] <archivist> you have a rectangle showing machine limits with the axis overhanging
[15:00:03] <pingufan> I wonder what is wrong with this line: net probe parport.0.pin-11-in => motion.probe-input
[15:00:15] <pingufan> It causes Axix to crash on startup.
[15:00:27] <raynerd> Ok, I touched off but still nothing
[15:00:45] <archivist> raynerd, does it now fit
[15:00:51] <raynerd> Sorry Archivist, just read your post. I`m sorry, I don`t understand
[15:01:37] <archivist> look in the right hand window does the splash graphic fit the machine limits rectangle
[15:02:06] <jthornton> pingufan: how many times to you have motion.probe-input in your hal file?
[15:02:11] <jthornton> you can only have it once...
[15:03:04] <raynerd> I can`t actually see my machine limits, but I set the table lengths stupidly high
[15:03:07] <pingufan> I manually added this two lines.
[15:03:13] <pingufan> net probe parport.0.pin-11-in => motion.probe-input
[15:03:38] <raynerd> ahh, yes! It does fit in the machine limits box. The box is so big I can`t see the edges without panning
[15:03:42] <pingufan> net probe-in motion.probe-input <= parport.0.pin-11-in
[15:04:02] <jthornton> pingufan: pick any one
[15:04:31] <pingufan> Ok. So they actually do the same and are only in different notation?
[15:05:43] <raynerd> HURRAY!
[15:05:55] <raynerd> cheers Archivist. was the Z axis..
[15:06:04] <pingufan> What was wrong?
[15:06:15] <pingufan> :D
[15:06:50] * jthornton is off to dump weevil infested malt
[15:07:11] <raynerd> OK, nice one! Better stop that now. My motors sound not too good when running in emc2 itself, but fine in step config wizard
[15:07:26] <andypugh> That's odd.
[15:07:42] <andypugh> Even at the same speed?
[15:08:01] <andypugh> pcw_home: Any ideas? https://picasaweb.google.com/lh/photo/yIRAC2T2DcwIgpJorwPa0NMTjNZETYmyPJy0liipFm0?feat=directlink
[15:09:14] <pingufan> raynerd: How are your steppers driven? Full steps, half-steps, microsteps?
[15:09:43] <mrsun> http://a5.sphotos.ak.fbcdn.net/hphotos-ak-ash4/s320x320/395873_10150635627608648_682103647_8764934_1435727691_n.jpg a bit better with the wiring int he cnc computer =)
[15:09:48] <pingufan> In case that you use full-steps, they surely will sound very rough.
[15:11:00] <pingufan> Therefore I use gecko drives. My mill is very silent now. It makes lots of microsteps.
[15:11:35] <raynerd> Full steps at present as that is what my driver is set to currently
[15:12:01] <raynerd> I have the option of also using 1/16 steps, would this be better?
[15:12:38] <mrsun> i use 1/8 now and find that quite good for my mill =)
[15:14:13] <raynerd> u know in emc2 then... (please bear in mind I`m not with my computer this very second, so this is from memory)...if I set my driver to 1/8 step, in the config wizard, for each axis it says steps per rev and something about the type of step
[15:14:27] <pingufan> The more steps, the better. But you will then have to send more steps for the same move, therefore you possibly will be slower.
[15:14:36] <raynerd> obviously I`d put 8 in the type of step (or whatever the wording is), what would I put in the steps per rev?
[15:15:09] <archivist> I use 1/2 step
[15:15:52] <raynerd> yes, so why is everyone using different? My limited knowledge tells me that the lower fraction, the more acurrate it is
[15:15:57] <raynerd> accurate
[15:16:11] <pingufan> Ok, guys. I have to eat something and then go to bed. Have a nice day and thanks for all your help. Now I see light at the end of the tunnel. :D
[15:16:32] <DJ9DJ> gn8 pingu
[15:16:38] <andypugh> raynerd: steps per rev is almost always 200
[15:16:45] <mazafaka> go back to us from that weird light
[15:17:00] <archivist> raynerd, http://www.micromo.com/microstepping-myths-and-realities.aspx
[15:17:05] <mazafaka> or to the U.S. (they live better)
[15:17:47] <mazafaka> Dr. Banana: http://science.slashdot.org/story/12/02/23/161225/interrupted-sleep-might-be-the-best-kind?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+Slashdot%2Fslashdot+%28Slashdot%29
[15:18:28] <mrsun> raynerd, "Drive microstepping" is set to what you have, if its 1/8 you put 8 there
[15:18:34] <mrsun> 1/16 you put 16 etc
[15:18:49] <mrsun> motor steps per revolution is the steps without microstepping
[15:18:55] <mrsun> so for a 1.8 degree stepper its 200
[15:18:59] <mrsun> etc
[15:19:12] <raynerd> ok, mine are 1.8 so steps per rev is 200
[15:19:30] <raynerd> so now I just need to decide my microstepping, set it on the driver and then in emc2 to match
[15:20:47] <raynerd> ahh, I see, put microstepping requires more pulses for a step...so speed is reduced!
[15:20:52] <raynerd> but accuracy gained
[15:21:10] <archivist> accuracy NOT gained see article I linked
[15:22:04] <archivist> position is only guaranteed at full steps
[15:22:07] <mrsun> raynerd, speed should not be reduced realy =)
[15:22:12] <andypugh> Smoothness should improve, but anything more that 8x is a bit excessive.
[15:22:19] <mrsun> microstepping gets rid of resonance etc so everything goes smoother
[15:22:29] <mrsun> but high microstepping rate with high friction == low accuracy =)
[15:22:34] <archivist> compromise
[15:22:38] <mrsun> + you lose some torque
[15:22:42] <mrsun> if im not mistaking
[15:22:45] <mrsun> at higher microstepping
[15:22:48] <archivist> yup
[15:23:08] <archivist> now you may see why I stick at 1/2
[15:23:39] <andypugh> I think I settled on 4x
[15:23:47] <mrsun> 1/8 works very well for me, quite enough power (cant hold the table back how hard i pull) and no stalling of the steppers =)
[15:24:07] <mrsun> tho before i scraped the ways of the mill the freakin thing was so weak i could cry
[15:25:11] <archivist> time to sudo make me a sandwich
[15:26:53] <mazafaka> how nice :)
[15:31:07] <syyl> reminds me to look for the new xkcd comic...
[15:36:36] <andypugh> syyl: It's thursday?
[15:37:17] <DJ9DJ> indeed
[15:37:19] <syyl> wednesday... ;)
[15:37:32] * DJ9DJ looks at syyl
[15:38:01] <syyl> not so close.
[15:38:31] <DJ9DJ> hrhr
[15:38:40] <DJ9DJ> riggly diggly!
[15:39:28] <syyl> ramtamtam!
[15:39:35] <DJ9DJ> yeah
[15:44:41] <archivist> http://i.imgur.com/8fLL8.jpg
[15:45:03] <DJ9DJ> lol
[15:46:24] <Loetmichel> *pfrrrrröööt*
[15:46:26] <Loetmichel> :=
[15:46:27] <Loetmichel> )
[15:47:54] <DJ9DJ> eat more beans to win!
[16:31:05] <DJ9DJ> gn8
[16:31:54] <raynerd> guys, on my driver I have a dip switch to set "decay mode settings" - fast, 25%, 50%, slow are my options...I don`t even know what this is in reference two
[16:31:57] <raynerd> what should it be set to?
[16:32:46] <Loetmichel> raynerd: soinds like the infamous cinese TB6560 boards ;-)
[16:33:01] <archivist> may be the automatic current reduction when it thinks its stopped...phail
[16:34:26] <Loetmichel> archivist: no, its (IIRC) the decay mode for the field. optimum results can vary from motor to motor. start at fast and hear for the sound of the steppers. if they are sounding heathy in ALL speeds you have fond fthe optilam setting
[16:35:20] <raynerd> ok cheers!
[16:35:25] <raynerd> I`ll give it a bash now
[16:35:35] <raynerd> I`ve gone for 1/2 microstepping.
[16:35:51] <raynerd> really don`t know what is best now after reading that document you linked me to Archivist!?
[16:36:11] <Loetmichel> i.e: fast: send the self inducted current at "switch off" via diodes to the PSU-> fast; resistors( or partially closed fets) to bleed the field in 25% and 50% or short circuit the winding after switch off (slow)
[16:41:16] <archivist> fast seems default in the datasheet
[16:42:32] <raynerd> Right OK, really getting somewhere now!
[16:43:54] <raynerd> Motors sound fine on fast, only problem I can see is that only slower speeds I`m getting some groaning, but think it is coming from the slides and lead screw/nut area!!! which is worrying. when it is going at a good constant speed it is fine
[16:44:51] <archivist> you may need to adjust acceleration rate too
[16:45:22] <archivist> too fast and it will stall
[16:45:44] <raynerd> I played around in the setup wizard to get a value below the speed at which it stalled
[16:46:01] <raynerd> I think I`m currently set to 5mm/s velocity and 4mm/s accel.
[16:46:14] <archivist> I added load to the mating for that test
[16:46:21] <archivist> machine
[16:46:55] <raynerd> good point
[16:46:56] <raynerd> I`ve not
[16:47:32] <raynerd> Archivist, I did read your link but I must admit, I`m none the wiser! - would you go for 1/2 step it lower?
[16:47:37] <raynerd> it = or
[16:47:50] <archivist> ecen though I loaded the machine I later had to slow it some more
[16:47:55] <archivist> even
[16:49:08] <archivist> that link basicly gives reasons for and against micro stepping, chose one that is best for your machine
[16:51:45] <raynerd> Is there any reason the "jog" with the arrow keys moves the axis really slowly...it certainly doesn`t go at the velocity set in the setup and during running
[16:53:06] <archivist> look on the left axis panel there is a jog slider adjust to suit
[16:53:49] <archivist> slow is safe
[16:53:59] <archivist> or safer
[16:54:32] <raynerd> ahhh! sorry!! thanks
[17:02:47] <syyl> press shift while joging...
[17:03:10] <Valen> syyl: whats that do?
[17:03:10] <syyl> gives you rapid movement
[17:05:41] <Valen> nifty
[17:13:49] <Loetmichel> raynerd: 5mm/s? so slow?
[17:14:44] <Loetmichel> my little one wiht 1A steppers and TR10*3 screws is going with about 40mm/s G0... without stall ;-)
[17:15:32] <Loetmichel> what was the english term for it... acme thread?
[17:15:52] <archivist> acme yes
[17:15:52] <andypugh> Yes,
[17:16:23] <Loetmichel> for this: __/^^\__/^^\__
[17:16:54] <Loetmichel> opposed to the normal metric thread like this: /\/\/\
[17:18:24] <jdhnc> ITYM they are trapezoidal instead of normal imperial threads.
[17:19:09] <archivist> hmm cannot draw square thread in ascii
[17:20:13] <jdhnc> _|~|_|~|_ approximately square
[17:24:09] <Loetmichel> tr10*3 means trapezoidal 10mm diamater 3mm per turn
[17:24:29] <Loetmichel> so i think acme fits ;)
[17:30:11] <Loetmichel> hmm, the only square thread i've seen latela is the on on my big(ger) lahte driving the sled
[17:30:16] <Loetmichel> lathe
[17:31:04] <archivist> I had to turn one many years ago for a steam valve
[17:34:41] <Loetmichel> right, one can find them on old water taps, too
[17:56:08] <JT-Shop> http://gnipsel.com/cannon/images/carriage07.jpg
[17:56:41] <Tom_itx> that looks nice
[17:56:51] <JT-Shop> thanks, getting there
[17:56:56] <andypugh> I don't think Trapezoidal and Acme are exactly the same thing.
[17:57:02] <Tom_itx> you gonna 'age' finish it?
[17:57:22] <A2Sheds> http://g.co/maps/wkhja I love when google streetview captures scenes in my hometown
[17:57:25] <JT-Shop> clear Olympic Maximum oil finish
[17:57:29] <andypugh> I have always fancied making a wheel.
[17:57:41] <Tom_itx> so did cave men
[17:58:08] <JT-Shop> actually I purchased the wheels, could have made them better myself but pressed for time and they are cheap
[17:58:23] <JT-Shop> made everything else in the photo
[17:58:41] <Tom_itx> do you take these to meets or shows somewhere?
[17:58:48] <Tom_itx> or just play with the neighbors
[17:59:14] <JT-Shop> mostly just play with the other children
[17:59:50] <JT-Shop> there is a thing in Greenville every spring, we might go if I have mine finished by then
[18:01:16] <JT-Shop> last years http://www.pre1840rendezvous.com/oldgreenville.html
[18:02:17] <JT-Shop> some other events in the area http://www.pre1840rendezvous.com/comingevents.html
[18:10:00] <andypugh> Artillery wheels were rather stronger than normal, so you might still end up making your own. I guess it might depend on whether you shoot roundhot or canister :-)
[18:11:31] <andypugh> Annoying Wikipedia page, but I don't know enough to fix it: http://en.wikipedia.org/wiki/Galloper_gun
[18:13:38] <JT-Shop> yes, the artillery wheels had two steel bands on each side like mine and we will shoot round balls and golf balls and aluminum foil balls
[18:14:14] <andypugh> They used "Artillery wheels" on vehicles until the 1930s.
[18:15:16] <JT-Shop> http://www.customwagons.com/store/index.php?app=ccp0&ns=prodshow&ref=1033&sid=e713t4w9r3au6h77d69iqug7ccm733xh
[18:17:23] <andypugh> <chortle> "Cannon and Limper" http://www.customwagons.com/cgi-bin/pro/emAlbum.cgi?c=show_image;p=Real%20Cannons;i=16;img=hope11A.jpg
[18:17:33] <andypugh> I guess if it ran over your foot...
[18:17:59] <JT-Shop> LOL at limper
[18:18:49] <JT-Shop> limber is the correct term
[18:20:07] <andypugh> Aye, I know that. As I know that the galloper gun was used in the Napoleonic wars, not just in the US.
[18:20:35] <andypugh> Hi Evertick
[18:21:10] <Valen> andypugh: to fix a wiki page click edit in the top right ;->
[18:21:22] <Evertick> Hello.
[18:21:24] <Evertick> I could use some advice. I've converted a metal lathe to cnc and use linuxcnc
[18:21:33] <Valen> gold star
[18:21:36] <andypugh> Valen: I do that a fair bit, but in this case I don't know enough about the subject.
[18:21:40] <Valen> ahh
[18:21:59] <Evertick> I'd like to use blender, but havent found any information on how to create gcode for a lathe
[18:22:10] <andypugh> Evertick: A fair few of us have done the same. I hope you started with a better lathe than I did.
[18:22:17] <Evertick> I've only found info on cnc routers and such
[18:22:36] <JT-Shop> Evertick: I use ngcgui for 95% of my lathe ops
[18:22:36] <andypugh> I think Blender is way too complicated for a lathe.
[18:22:48] <Evertick> I already have a router build from buildyourcnc.com
[18:22:53] <Evertick> I'm guessing it is.
[18:23:14] <JT-Shop> now that I have almost unlimited I/O on the plasma I can hook up the start/stop button :)
[18:23:29] <andypugh> I have some simple library routines and a GUI for generic lathe ops, and hand-code for the complicated stuff.
[18:24:02] <andypugh> JT-Shop: ngcgui is probably better than my approach.
[18:24:19] <andypugh> JT-Shop: You have a handy link?
[18:24:21] <Valen> blender isn't that good as far as CAD
[18:24:41] <Evertick> Cool. Would you use ngcgui to actually draft the item?
[18:27:40] <Evertick> What app would you use to design the part?
[18:33:20] <GoSebGo> Evertick: for 2d stuff i've used freecad & pycam
[18:36:16] <Evertick> GoSebGo: I've used pycam before. I also use Qcad. Just have never done a design for a lathe. Only stuff for my router.
[18:36:34] <JT-Shop> andypugh: yea, just a sec
[18:36:48] <Evertick> So I guess the challenge is getting something from *cad into linuxcnc
[18:36:59] <JT-Shop> ngcgui http://linuxcnc.org/index.php/english/component/kunena/?func=showcat&catid=40
[18:37:47] <JT-Shop> Evertick: are you talking profiles or simple shapes?
[18:38:17] <Evertick> I'll start simple, like a chess pawn.
[18:38:43] <Evertick> I'm guessing that would be a profile.
[18:38:45] <JT-Shop> ok that is a profile shape and yea you need to get that into g-code
[18:39:28] <JT-Shop> there is some free dxf to g code stuff out there but I can recall what it is
[18:39:29] <Evertick> JT-Shop: bingo.
[18:40:39] <Evertick> JT-Shop: Does that still fall under the topic of ngcgui?
[18:40:57] <andypugh> I tend to draft it in CAD, dimension to the starts and ends of arcs (and centres to start, to get the I, K parameters) and write the profile by hand. Or just machine like a manual lathe that automatically does the next cut, giving a target X and Z, cut depth, feed and surface speed.
[18:40:58] <JT-Shop> for OD, face, ID, taper, thread, tap kind of things ngcgui is fastest and will even do a profile
[18:41:24] <JT-Shop> what Andy does I tend to do as well for the lathe
[18:41:57] <andypugh> Evertick: This is my system, to get the idea. NGCgui is probably slicker. http://www.bodgesoc.org/lathe/lathe.html
[18:42:09] <JT-Shop> quicker than fiddling with some cam software
[18:42:43] <Evertick> Ok I think I have a rough idea.
[18:43:07] <JT-Shop> Evertick: are you new to a cnc lathe?
[18:44:09] <Evertick> JT-Shop: Yes I am. I had an old manual one that I decided to convert with steppermotors and drivers.
[18:44:41] <JT-Shop> a tip for you then when you view an arc or a tool offset lay on the floor and look up :)
[18:44:57] <JT-Shop> for front tooling
[18:45:07] <Evertick> Ok thanks
[18:45:46] <JT-Shop> get ngcgui installed with some of the subroutines for simple ops and that will get you up and going
[18:46:06] <Evertick> Cool. Thanks. I'll give it a shot
[18:46:07] <JT-Shop> like Andy said profiles are usually easy to lay out in cad and get the offsets
[18:46:42] <JT-Shop> when I do profiles I usually run in radius mode, for most other turning ops in diameter mode
[18:49:04] <JT-Shop> andypugh: your a short timer now
[18:49:25] <andypugh> Indeed.
[18:50:02] <JT-Shop> will you have video and photos of the journey?
[18:52:22] <JT-Shop> I see most of the vessels are in port now
[19:10:41] <andypugh> There will be photos and video, but none from me.
[19:11:21] <andypugh> Have you seen the Virtual Race? My boat in that flew straight through the finish, and grounded.
[19:11:46] <andypugh> No idea what was going on there, but it has left me approximately last out of 40,000
[19:17:56] <andypugh> Loetmichel: I assume you are German, but not Fremder?
[19:40:53] <danimal_laptop> hi John
[19:40:58] <danimal_laptop> hi andypugh
[19:41:20] <danimal_laptop> JT-Shop: im in the market for a plasma
[19:42:20] <andypugh> Hi danimal_laptop , but I am logging imminently
[21:29:55] <A2Sheds> danimal_laptop: hi, how is the shop going with all the new improvements?
[22:35:43] <Jymmm> http://images.piccsy.com/cache/images/picc-11kvellbv-204087-504-251.jpg
[23:09:17] <|n0b0dy|> sharing is caring