#emc | Logs for 2010-10-14

[00:04:00] <theorb> theorb is now known as theorbtwo
[03:00:42] <ChanServ> [#emc] "This is the #emc channel - talk related to the Enhanced Machine Controller and general machining. Website: http://www.linuxcnc.org/, wiki at http://wiki.linuxcnc.org/"
[04:52:59] <flyback> http://cgi.ebay.com/High-performance-CNC-Stepper-Motor-Driver-Up-tp-5-6A-PK-/380133899642?pt=LH_DefaultDomain_0&hash=item5881befd7a
[05:04:13] <elmo39> $99 to control ONE stepper motor?
[05:04:39] <flyback> no I think it's several
[05:04:48] <elmo39> doesn't look like it
[05:04:55] <flyback> oh then it's lame
[05:04:59] <flyback> IT'S *CANUCKED*
[05:05:00] <flyback> IT'S *CANUCKED*
[05:05:39] <elmo39> interesting line: "It can offer servo-like performance."
[05:05:45] <elmo39> damn Canucks
[09:00:40] <pjm> pjm is now known as Guest87973
[13:42:43] <JT-Hardinge> G83 the Z is the final depth of the hole and R is the start position for the feed move?
[13:45:28] <elmo42> yes, start and retraction depth.
[13:45:49] <elmo42> it will pull out to the R value every Q depth of cut
[13:45:59] <elmo42> until it reaches the Z final depth
[13:46:59] <JT-Hardinge> ok, I was confused after reading the manual
[13:48:25] <pepsi> what kind of cereal does jesus like?
[13:48:29] <pepsi> holy-o's :(
[13:52:01] <JT-Hardinge> well, that worked nicely
[13:52:33] <JT-Hardinge> now to figure out how to do a back side chamfer before parting off...
[14:14:05] <skunkworks> cradek: how is the 4th?
[14:15:11] <elmo42> JT-Hardinge: very carefully?
[14:16:41] <cradek> skunkworks: I didn't work on it last night - still the same state as when I made the silly video
[14:18:14] <skunkworks> you made a video?
[14:18:18] <skunkworks> * skunkworks goes looking
[14:18:36] <cradek> http://www.youtube.com/watch?v=kxgSmr01UQg
[14:19:57] <skunkworks> cool!
[14:21:13] <awallin> cradek: how do you handle backlash in the old manual table?
[14:21:39] <cradek> awallin: the worm gear has adjustable engagement - the whole thing is eccentric
[14:21:48] <SWPadnos> aren't we all
[14:22:41] <cradek> awallin: I haven't tested it yet, but it appears to be adjustable so backlash is .001-2" at the outside of the table
[14:22:46] <JT-Hardinge> cradek: you need to turn a thumb guard for your camera :)
[14:23:08] <cradek> JT-Hardinge: yeah, embarassing
[14:25:17] <cradek> wish I had taken pictures of the mechanical work - to make the shaft from the worm long enough, I spliced it with a pressed-in 2" dowel pin
[14:25:35] <cradek> 1/4 x 2"
[14:26:10] <JT-Hardinge> not as embarrassing as this is http://www.youtube.com/watch?v=G4-lUNa0CpY
[14:26:41] <cradek> haha
[14:27:01] <cradek> sproink! there goes an insert and a little screw...
[14:27:24] <cradek> I did that once too - hit the end of the work (aluminum) though - much softer
[14:27:25] <JT-Hardinge> it left a mark too
[14:27:50] <cradek> did it hit the collet?
[14:28:02] <JT-Hardinge> yea
[14:29:36] <skunkworks> I wonder what kind of trouble I can get into with 4 tons of force...
[14:29:59] <skunkworks> (actually it would be twice that for the first second...)
[14:30:19] <JT-Hardinge> depends on how big your tooling is :)
[14:30:31] <skunkworks> I should make some videos of it splitting logs.. ;)
[14:30:43] <skunkworks> dad would probably frown on that.
[14:31:36] <cradek> I've used Jr as an arbor press... presses nice and straight
[14:31:45] <skunkworks> I bet
[14:31:58] <skunkworks> we have a rather large drill press that works great for that also
[14:32:19] <skunkworks> (and a h-frame hydraulic press that we break out for the big jobs)
[14:33:24] <JT-Hardinge> I hate it when I don't get what I expect and don't know for sure why...
[14:34:20] <cradek> yeah like getting the chocolate with coconut in it?
[14:37:05] <JT-Hardinge> kinda like that... I touched off a right hand turning tool to the spindle and added 0.500 (the width of the tool) but when I check it with my material Z0 is not even with the end of the material...
[14:53:16] <Als> Als is now known as AlSMT
[15:10:15] <JT-Hardinge> hmmm, the parting tool is off to...
[15:12:43] <elmo42> well, where did you touch off? to the newly machined face? Need to touch off all tools right in the beginning unless the program is written with touching off to a previously machined face...
[15:13:15] <JT-Hardinge> no, I touch off of the spindle always for tools then take any tool and touch off my g54 offset
[15:23:59] <cradek> JT-Hardinge: are you using master or 2.4?
[15:24:55] <JT-Hardinge> master
[15:25:17] <JT-Hardinge> the tools that I use 0.000 for the offset seem correct
[15:25:35] <cradek> do you have the right tool touch off option set?
[15:25:42] <JT-Hardinge> I'm using tool touch off to fixture
[15:27:22] <JT-Hardinge> hmmm, my parting tool is off the width of the tool
[15:27:54] <JT-Hardinge> when I touched it off I put in 0.125 (the width of the tool) so the right edge would be 0.000
[15:28:04] <JT-Hardinge> and got it backwards :/
[15:28:12] <JT-Hardinge> should have been -0.125
[15:28:23] <cradek> you were starting to worry me...
[15:29:07] <Jymmm> JT-Hardinge started worrying me a LONG time ago
[15:30:07] <JT-Hardinge> I should go and fix the backhoe instead of making parts :)
[15:40:09] <Jymmm> Has anyone used or tinkered around with car airflow sensors or pressure switches by chance?
[15:41:14] <L84Supper> does having to replace broken ones count? :)
[15:41:23] <MarkusBec_> MarkusBec_ is now known as MarkusBec
[15:41:44] <Jymmm> Sure, if you know the pinout, voltage and pressure level that it triggers at =)
[15:42:10] <Jymmm> Or if you know of one that can easily be adapted for other uses =)
[15:45:58] <JT-Hardinge> something is not correct with touch off when you enter a value other than 0.0000 and use touch off to fixture...
[15:47:31] <Jymmm> what if you enter zero, then go back and change it?
[15:47:35] <JT-Hardinge> I have to run to town now... but no matter what I enter other than 0 the Z position of tools that are 0 offset don't match up to tools that have a value in the offset
[15:48:50] <cradek> JT-Hardinge: is g59.3 (fixture coordinate system) zero set to match the reference tool?
[15:49:34] <cradek> when you get back we can go through testing it together - I'm not understanding what you are saying is wrong
[15:50:09] <Jymmm> Why are there g-codes for tool offsets?
[15:50:57] <Jymmm> It would seem to me that would be a config thing, not a per-job thing
[15:50:58] <cradek> you can set and load tool offsets with gcodes
[15:51:31] <Jymmm> But why would you do that? Shouldn't they always be the same between jobs?
[15:51:32] <cradek> it's true that usually you don't have those gcodes in your part program
[15:52:18] <cradek> sort of - but sometimes you want them different
[15:52:28] <Jymmm> simple example?
[15:52:40] <JT-Hardinge> cradek: I'm not sure what you mean about the g59.3 zero
[15:52:45] <cradek> like jt's parting tool - which side of it should be Z=zero?
[15:52:55] <Jymmm> ah
[15:53:14] <Jymmm> cradek: ty
[15:53:34] <cradek> JT-Hardinge: you set up the "touch off fixture" by putting your reference tool against it with no TLO, and setting G59.3 to zero there
[15:55:00] <JT-Hardinge> I didn't do that
[15:55:18] <cradek> then when you use touch off to fixture, you get offsets that compare to that point
[15:55:44] <cradek> I don't know if I'm explaining it adequately
[15:55:46] <JT-Hardinge> I just took each tool up to the spindle face and one dowel diameter away touched off the Z for the tool as 0 except for the parting tool
[15:56:14] <cradek> that should work fine
[15:56:28] <JT-Hardinge> that's what I thought too
[15:56:45] <cradek> who cares where G59.3 is, if you don't care exactly what lengths you get in the tool table
[15:57:01] <cradek> do you have an active g92 offset?
[15:57:13] <JT-Hardinge> no
[15:57:56] <JT-Hardinge> I do have a G54 offset for Z
[15:58:17] <cradek> ok, that's normal and should not affect touch-off-to-fixture
[15:58:56] <JT-Hardinge> I'll do some more testing when I get back from town
[16:00:00] <cradek> hm, I bet g92 should be ignored for touch-off-to-fixture (it is not)
[16:00:07] <cradek> so confusing...
[16:57:02] <Jymmm> So, anyone have any ideas on car airflow/airpressure switches?
[17:38:05] <Jymmm> Hey, at 0:43, it that 1.0v per division??? http://www.youtube.com/watch?v=ijK2LsN8ipw
[17:41:29] <cpresser> yes.. but he is possibly using a 1:10 probe
[17:43:05] <Jymmm> If so, then that would be .1v per division?
[18:02:34] <Fox_Muldr> is there a g-code command which does relative movement to the current position or only absolut movement to 0,0,0?
[18:05:47] <Fox_Muldr> ah i found it. i have to change to incremental distance mode with G91 than i can do relative movements and after that go back to absolute distance mode with G90
[18:10:27] <elmo42> that is the way to do it :)
[18:10:56] <cradek> I never get to answer the easy questions!
[18:12:22] <psha> cradek: it's a crime to spend you on easy questions :)
[18:12:30] <Jymmm> How much wood can a woodchuck chuck, if a woodchuck could chuck wood?
[18:12:32] <cradek> haha
[20:13:41] <DJSprocket> hey i need to know somethin
[20:13:42] <DJSprocket> g
[20:14:12] <micges> ask
[20:14:30] <DJSprocket> i have a vinyl cutter im fitting to EMC2 with an XYZ and an A he A needs to rotate to keep the blade cutting straight with the direction of travel of the cutter
[20:14:36] <DJSprocket> how would i do that?
[20:15:26] <jepler> by programming the A axis explicitly with each XY move
[20:15:46] <DJSprocket> is there a software that does it automagically
[20:16:03] <archivist_attic> this has been discussed a few times, worth scanning logs
[20:16:16] <jepler> here's a link to a previous mailing list discussion about the topic: http://thread.gmane.org/gmane.linux.distributions.emc.devel/1696
[20:16:41] <DJSprocket> how do i copy?
[20:26:31] <DJSprocket> ok i read through that mailing list... no help at all
[20:28:34] <cpresser> DJSprocket: what do you expect? it shows how to modify your gcode
[20:28:49] <DJSprocket> without any explicit help
[20:28:58] <cpresser> yeah, thats open-source :D
[20:29:10] <cpresser> are you familiar with programming?
[20:29:41] <cpresser> if yes, adding this C-Axis-function to 'grecode' my be an option
[20:29:45] <DJSprocket> in c++ a little but not enough to convert the G file into text read it compute the angles and spit out an A co-ordinance
[20:31:40] <cpresser> http://code.google.com/p/grecode << this may be a good start
[20:31:57] <cpresser> it already has lots of features for gcode manipulation..
[20:32:00] <DJSprocket> what if i needed this running yesterday
[20:32:13] <cpresser> then you are fucked up :D
[20:33:04] <cpresser> take a lot of money and buy some software/controll which handles this. as sad as it is.. but there may be no ready-to-use solution
[20:33:31] <DJSprocket> i heard that there was someone who already had this working somewhere
[20:35:41] <cpresser> unfrotunately I only know the german word for such a tool (schleppmesser), but not the english translation.. so I cant help search :/
[20:35:57] <DJSprocket> tangential knife
[20:37:26] <DJSprocket> its not translating if you called me something...
[20:39:52] <DJSprocket> there should be some cam software that automatically puts in C axis somewhere
[20:42:45] <cpresser> I have already seen a lot of such cutters at work. most for simple cartboard
[20:43:01] <cpresser> but they all come with commercial proprietary software..
[20:43:36] <DJSprocket> yeah i dont want proprietary i want free as thats all i can afford. i have everything else i need to run this thing except for the rotary axis
[20:43:47] <DJSprocket> another question about stepper motors
[20:44:40] <DJSprocket> my steppers are rated for 5.1 volts can i safely run them overnight at 12 volts for a production run or would i have to put a resistor in line with the V+ on my driver boards?
[20:45:14] <cpresser> found a tool: http://www.cnc-club.ru/forum/viewtopic.php?t=35
[20:49:00] <DJSprocket> needs inkscape to run is this the right one? http://inkscape.org/download/
[20:49:15] <cpresser> yes. thats inkscape
[20:49:44] <DJSprocket> k thanks that should help. now the stepper question please
[20:49:55] <cpresser> but inkscape ist no CAD/CAM tool.. make sure to check if it messes up your cordinates after generation gcode
[20:50:32] <cpresser> I mean, make sure a one-inch-line in your DXF-File (or whatever) is also a one-inch-line in the generated gcode
[20:50:52] <cpresser> stepper.. I dont know. would have to take a look at the datasheet
[20:52:02] <DJSprocket> ive been trying to track down the datasheets cant find them anywhere
[20:52:44] <DJSprocket> 19 year old steppers
[20:53:46] <DJSprocket> custom windings and the whole 9 yards
[20:54:11] <DJSprocket> but the questions is would the resistors absorb the extra amperage and keep the motors from heating up so much
[20:54:47] <DJSprocket> or would the steppers just pull more current
[20:54:47] <cpresser> depends on your driver-circuit
[20:54:59] <DJSprocket> qkits K-179
[20:55:04] <DJSprocket> ill grab a link
[20:55:11] <cpresser> does it drive voltage or currant?
[20:55:16] <cpresser> *current?
[20:55:43] <DJSprocket> i dont know the difference it just puts pulses to the right winding
[20:55:45] <cpresser> my guess would be voltage.. current-driver seem odd for steppers :)
[20:56:24] <cpresser> its the electrial characteristic of the power-soure.. depending on this, the effect would change
[20:56:26] <DJSprocket> http://store.qkits.com/moreinfo.cfm/QK179
[20:57:16] <cpresser> assuming, it is driving just the voltage, adding a resistor would limit the total current
[20:57:39] <cpresser> there will also be a voltage drop across the resistor.. burning your additional energy
[20:58:14] <DJSprocket> well i do need to run the steppers off 12 volts thats what my PSU will supply without going into protect mode
[20:59:00] <cpresser> my steppers run at 42Volts, but their design is safe for that voltage :)
[20:59:22] <DJSprocket> mine are rated at 1 amp at 5 volts
[20:59:26] <cpresser> you sould mesaure the inner resistance of your stepper coils
[20:59:41] <DJSprocket> i'd have to decrease the amperage to get them to stop over heating right?
[20:59:48] <cpresser> if 1Amp at 5Volts is right, they would have 0.2Ohms
[21:00:29] <DJSprocket> 4.2 ohms to the center tap and 7 oohms across the two outside wires
[21:00:58] <cpresser> ah yes.. 5Ohms,. of course :D
[21:01:06] <cpresser> dammit.. I inverted that :D
[21:01:24] <DJSprocket> no big deal
[21:01:26] <cpresser> so you would add 7Ohms resistors to drop 7Volts at 1Amp
[21:01:44] <cpresser> thus burning 7/12 of your total power.. quite a waste :D
[21:01:45] <DJSprocket> the resistors i have are 20 ohms 95 watts
[21:02:03] <cpresser> 95Watts? how big are they?
[21:02:18] <DJSprocket> 4 incches long 1 inch wide .25 inches thick
[21:02:39] <cpresser> huge :D
[21:02:43] <DJSprocket> yeah
[21:02:48] <DJSprocket> biggest ive had my hands on
[21:02:53] <cpresser> put two in parallel.. it gives you 10Ohms
[21:03:12] <DJSprocket> shouldnt that decreases the resistance in parralel
[21:03:18] <cpresser> thus you would have 15Ohms, 12Volt -> 0.8Amp
[21:03:26] <DJSprocket> oh yeh sorry
[21:03:49] <cpresser> 2x20 in parallel = 10Ohms.. the stepper has 5Ohms in series -> 15Ohms
[21:03:53] <DJSprocket> but you said i need 70 ohms
[21:04:17] <cpresser> 23:01 < cpresser> so you would add 7Ohms resistors to drop 7Volts at 1Amp
[21:04:23] <DJSprocket> to drop to the voltage of 5
[21:04:38] <DJSprocket> oh 7 ohms not 70
[21:05:05] <DJSprocket> can you put a space between the 0 and the O its hard to read
[21:05:10] <rooks> #slug
[21:05:11] <cpresser> 7 Ohms :D
[21:05:16] <DJSprocket> ty
[21:05:37] <cpresser> 7 Ohm + 5 Ohm = 12Ohm -> 1 Amp @ 12 V
[21:05:43] <cpresser> thats the basic Idea
[21:05:55] <DJSprocket> so the resistors would soak .2 A?
[21:06:21] <cpresser> 10 Ohms + 5 Ohm = 16 Ohm -> 0.8 Amp @ 12V
[21:06:35] <cpresser> yes. you would lose about 20% of your power
[21:06:59] <DJSprocket> which would help to alleviate the heat problem?
[21:07:05] <cpresser> just get a 5Volt power supply.. thats easy. plus it doesnt waste energy
[21:07:24] <cpresser> heat problem?
[21:07:36] <DJSprocket> i dont have the money plus i would need a 4 amp at least power supply
[21:07:55] <DJSprocket> yeah the steppers heat up alot at 12 volts
[21:08:43] <DJSprocket> quickly too
[21:08:54] <cpresser> at 12Volts, whithout the resisors your steppers run at 2.4Ampere.. thats 240% of the designed spec :)
[21:09:07] <cpresser> so yes, the resistors would help.
[21:09:21] <DJSprocket> so what would i need to get it down to 100% just the 2 20's in parralel?
[21:09:45] <cpresser> no, you would need a 7 Ohm resistor
[21:10:10] <DJSprocket> i dont have a 7 ohm that wouldnt burn out at the proposed amperage
[21:10:24] <DJSprocket> i guess i should have said below not to
[21:10:49] <cpresser> as said above:
[21:10:55] <cpresser> cpresser> 10 Ohms + 5 Ohm = 16 Ohm -> 0.8 Amp @ 12V
[21:11:02] <DJSprocket> so 10 ohms is fine
[21:11:13] <cpresser> would solve your heat problem, whit little power loss
[21:11:19] <DJSprocket> id rather run them below what they need anyways
[21:11:39] <DJSprocket> would i lose my speed gain by putting the resistors in?
[21:11:55] <cpresser> sorry, cant answer that, since I dont know :)
[21:11:57] <DJSprocket> ok
[21:12:02] <cpresser> just test it :)
[21:12:11] <DJSprocket> i will in a bit
[21:12:29] <cpresser> most likely you will only loose torque, since speed is controlled via the step-rate
[21:12:31] <DJSprocket> gotta go upstairs and do some soldering first
[21:12:52] <DJSprocket> their 100 ounce inch steppers lol
[21:13:38] <cpresser> * cpresser is gone for now. bedtime :)
[22:46:31] <flyback> * flyback bbl, bathroom break
[22:56:10] <JT-Hardinge> TMI
[22:56:55] <cradek> JT-Hardinge: did you figure out the offset problem? I fixed something (that was not affecting you) but otherwise my testing showed that it is working right
[23:03:22] <JT-Hardinge> no I think the problem was PEBKAC or however you spell it, I was correct with my offsets for the parting tool etc. but something else was wrong
[23:04:08] <JT-Hardinge> dunno just went back from the start and set all the tools offset from the spindle face and checked the position in G54 and all was I expected
[23:04:36] <JT-Hardinge> thanks for asking
[23:14:23] <JT-Hardinge> and I just made a nice part, but I should have used cutter comp on the angled parts :/
[23:21:02] <cradek> uh-oh :-)
[23:21:04] <cradek> darn
[23:22:06] <cradek> how do you set X offset? I use the spindle for Z, but don't know how to do X well
[23:22:06] <JT-Hardinge> only 25 thou deep 45 so it really don't matter this time
[23:22:21] <cradek> is it still mounted? a file is good for that...
[23:22:21] <JT-Hardinge> usually a test cut
[23:22:42] <cradek> ah I suppose you don't have any X work offsets, so you can still use 'fixture'
[23:22:54] <cradek> cool...
[23:22:56] <JT-Hardinge> it came out fine just some chamfers on the front and back sides of the part
[23:23:29] <JT-Hardinge> I didn't even think about fixture for X
[23:25:27] <cradek> and I didn't think about lathes when I wrote it...
[23:25:31] <cradek> but it seems fine
[23:26:06] <JT-Hardinge> seems to work here just fine
[23:26:13] <cradek> good, thanks
[23:26:36] <JT-Hardinge> I think I'll make an extra one just for fun now :0
[23:26:39] <JT-Hardinge> :0
[23:26:42] <JT-Hardinge> :0
[23:26:43] <cradek> now where are the M4x0.7 socket setscrews? oh - nowhere - right
[23:26:55] <cradek> grr - off to the hardware store
[23:27:21] <JT-Hardinge> good luck at the hardware store finding that
[23:27:47] <cradek> it's a decent one - usually I'm lucky
[23:27:57] <JT-Hardinge> cool
[23:29:25] <JT-Hardinge> oh, I wrote a comp for an elapsed time meter, works pretty cool
[23:30:12] <JT-Hardinge> 2 min 12 seconds to make that part :)
[23:30:20] <JT-Hardinge> * JT-Hardinge heads inside now