Back
[01:54:34] <DJ9DJ> moin
[07:36:44] <PetefromTn> Hello fellas
[07:38:47] <jthornton> morning Pete
[07:39:04] <PetefromTn> Mornin' JT
[07:47:46] <mr_new_> anyone here familar with xilinx tools?
[07:50:09] <PetefromTn> nope sorry...
[07:54:49] <mr_new_> not so easy to compile an mesa firmware
[07:55:45] <R2E4> cradek: you here?
[08:20:01] <mr_new_> as i hate errors while compiling -.-
[08:20:51] <R2E4> I've never compiled anything without errors.....
[08:20:57] <mr_new_> specialy when i dont know an ide^^
[08:21:09] <AR_> usually i compile without errors
[08:21:15] <mr_new_> R2E4: errors arent the problem, but not knowing the ide
[08:21:24] <mr_new_> and xilinx ide is not easy i think
[08:21:57] <AR_> loll
[08:22:02] <AR_> it sucks
[08:22:13] <R2E4> I didnt mean it errors out and aborts the compile, if you look while it is crunching, you will most proably see a few errors.
[08:22:26] <R2E4> I allways do.
[08:23:16] <AR_> xilinx errors are usually ver vague
[08:23:21] <AR_> very*
[08:23:56] <AR_> "connection may cause high skew"
[08:24:02] <mr_new_> it cant find a library function
[08:24:11] <AR_> what function
[08:24:12] <mr_new_> but if i add, i get more errors
[08:24:24] <mr_new_> sslbp
[08:25:03] <AR_> ask azonenberg in ##electronics
[08:25:16] <R2E4> BP boss 9 with Aspire is not good. Bridgeport post does not work with Aspire. It is for the EZTrack BP.
[08:25:21] <R2E4> BAAAA!!!!
[08:25:23] <AR_> he is the xilinx queen
[08:25:33] <mr_new_> AR_: its in the mesa firmware
[08:25:37] <AR_> oh
[08:41:46] <r00t4rd3d> R2E4, im sure it does.
[08:42:04] <r00t4rd3d> what post processor are you using?
[09:06:20] <xxoxx> can linuxcnc drive an EDM?
[09:06:33] <AR_> why not
[09:06:36] <xxoxx> I've got a die sinker
[09:06:56] <xxoxx> trying to figure out how to convert it to CNC operation
[09:07:01] <AR_> there are X Y Z axis but it doesnt have to be in a mill or lathe configuration
[09:07:18] <AR_> you could drive a robot with it
[09:07:20] <xxoxx> EDM the tool gets eaten away
[09:07:27] <xxoxx> so Z zero changes
[09:08:08] <xxoxx> wonder how to handle this
[09:08:28] <AR_> keep subtracting from your Z moves
[09:08:42] <xxoxx> right. that's what I was thinking
[09:08:52] <xxoxx> just wondering if this been done in LinuxCNC
[09:09:03] <xxoxx> or I'd have to add that myself
[09:09:27] <xxoxx> it would be cool to cut steel or brass mold that way
[09:09:36] <AR_> yeah
[09:10:37] <JT-Shop> edm site:linuxcnc.org
[09:11:00] <r00t4rd3d> thats helpful
[09:11:03] <r00t4rd3d> :D
[09:11:36] <xxoxx> i will 'dig' around ... hehehe
[09:12:02] <xxoxx> trying to convert a SIEG SU-1 to CNC
[09:16:31] <jdh> cool
[09:17:26] <jdh> like this:
http://www.grizzly.com/products/Mini-Horizontal-Vertical-Mill/G0727 ?
[09:19:29] <xxoxx> yeah. same chasis, different motor
[09:19:37] <xxoxx> SU-1 is the brushless version
[09:19:54] <xxoxx> I figure this would be quite suitable for CNC job
[09:20:01] <xxoxx> all axis being down below
[09:29:21] <jdh> looks like not much Z space in vertical mode
[09:30:30] <xxoxx> nope
[09:30:37] <xxoxx> but I only build small stuff
[09:30:51] <xxoxx> mostly aluminum molds
[09:30:56] <xxoxx> 5MM cutting depth
[09:31:35] <xxoxx> also do horizontal milling for gears
[09:32:58] <xxoxx> SU1
[09:33:19] <xxoxx> SU1, SC2, C0, and a steel bodied CNC router/mill
[09:33:42] <xxoxx> SU1, SC2, C0, would like to all eventually convert to CNC
[09:35:33] <xxoxx> thinking about getting a Bolton AT125
[09:35:47] <jdh> I have a c0 ( I think... 7x lathe?)
[09:35:54] <xxoxx> yeah. baby lathe
[09:35:58] <jdh> SC2
[09:36:31] <xxoxx> SC2 is the mini lathe, with torquey brushless motor
[09:41:59] <r00t4rd3d> my lathe:
http://imgur.com/a/MYeQ0
[09:42:11] <xxoxx> SU-1 spindle runout about 0.002mm, pretty surprising
[09:42:23] <AR_> that is horrible
[09:42:41] <xxoxx> ?
[09:42:53] <xxoxx> 0.002mm, not inch
[09:43:28] <xxoxx> r00t, a DIY nano lathe ?
[09:44:03] <r00t4rd3d> holds square stock that 1.5, 1.0, .75 inch
[09:44:09] <r00t4rd3d> 6.5 inches long
[09:44:24] <r00t4rd3d> for my drill press
[09:45:31] <xxoxx> ok. nice
[10:58:24] <Loetmichel> re @ home
[11:13:47] <ve7it> Jymmm, ping...
[11:25:18] <L84Supper> http://www.youtube.com/watch?v=rpBnurznFio RC plane launched from balloon at 33km (20 miles)
[11:25:19] <Tecan> (rpBnurznFio) "Space Glider - FPV to Space and Back!" by "David Windestål" is "Tech" - Length: 0:06:01
[11:33:10] <jdh> I hope that is a bot
[11:35:18] <AR_> i am not a bot
[11:35:41] <jdh> prove it.
[11:36:06] <AR_> ggggggggggggggggggg
[11:36:27] <L84Supper> Tecan comments on any youtube video posted
[11:36:43] <jdh> because he is bored?
[11:37:23] <AR_> yeah
[11:37:25] <AR_> he has no life
[11:37:54] <jdh> me either, but I don't comment on some youtube videos
[11:38:17] <L84Supper> http://www.youtube.com/watch?v=xZOiNdkJ8SU
[11:38:18] <Tecan> (xZOiNdkJ8SU) "5-Axis Machining, 5-Axis Mill" by "MDAPrecision" is "Tech" - Length: 0:06:43
[11:38:30] <L84Supper> a bot
[11:39:21] <L84Supper> http://www.youtube.com/watch?v=vf5VdQzuSjQ
[11:39:22] <Tecan> (vf5VdQzuSjQ) "Home made 5axis CNC mill.wmv" by "suke koro" is "Tech" - Length: 0:06:00
[11:39:48] <L84Supper> not sure who runs it
[11:40:05] <jdh> https://www.youtube.com/watch?v=QbRlACyP7Aw
[11:40:32] <jdh> sure, he doesn't like mine.
[11:40:37] <L84Supper> https://www.youtube.com/watch?v=QbRlACyP7Aw
[11:40:49] <jdh> http://www.youtube.com/watch?v=QbRlACyP7Aw
[11:40:50] <Tecan> (QbRlACyP7Aw) "Rocky Horror tunnel at Madison Blue Spring" by "Bob Bowling" is "Sports" - Length: 0:07:46
[11:40:54] <jdh> he's insecure
[11:40:58] <L84Supper> http vs https
[11:41:19] <L84Supper> https://www.youtube.com/watch?
[11:41:30] <L84Supper> http://www.youtube.com/watch?v=QbRlACyP7Aw
[11:43:15] <L84Supper> Tecan: !help
[11:44:52] <L84Supper> http://www.youtube.com/watch?v=
[11:53:41] <Tecan> haha
[11:54:11] <Tecan> feedback welcome, if you dont like colors or the function it can be changed
[11:54:29] <Tecan> theres a channel ignore option for the YT bot
[11:54:50] <Tecan> but i am really bored alot :P
[11:55:43] <AR_> there are no colors
[11:55:57] <jdh> not in my vt screen
[11:56:30] <AR_> colors are filtered in this channel
[11:56:31] <AR_> mode +c
[11:59:16] <JT-Shop> so after a week of putzing around with the backhoe it finally starts... it rained last night so now it's too muddy to dig :(
[11:59:27] <jdh> got lucky!
[11:59:57] <JT-Shop> no, I finally figured out I put the gasket in the fuel filter incorrectly...
[12:08:11] <pcw_home> gasket was installed in fuel economy mode?
[12:09:27] <AR_> upside down
[12:12:52] <IchGuckLive> hi all B)
[12:12:52] <L84Supper> anyone have a source for conveyor belts ~6' long, 1' wide, with really high repeatability? +- 0.002 (50um)
[12:14:07] <IchGuckLive> L84Supper: USA
[12:14:13] <L84Supper> anywhere
[12:14:16] <jthornton> yea if you put it in just right it blocks off the passage to the exit LOL
[12:16:00] <IchGuckLive> L84Supper: only flat or some shape also strait or in curve
[12:16:14] <L84Supper> IchGuckLive: just straight
[12:16:21] <L84Supper> and flat
[12:16:25] <IchGuckLive> for a CNC control
[12:16:38] <IchGuckLive> or axis transporting
[12:17:05] <L84Supper> servo controlled at a constant rate 1gram loads 0.5m/sec
[12:17:42] <AR_> build one
[12:17:46] <IchGuckLive> so why not a 4layer timing belt it messures that 0.002
[12:17:47] <L84Supper> IchGuckLive: we don't need the servo included, just loking for the belts, guides rollers as an assembly
[12:18:05] <L84Supper> AR_: don't have time to build everything :)
[12:18:08] <IchGuckLive> at 25,4mm widt
[12:18:13] <IchGuckLive> 1"
[12:18:15] <AR_> rolling pins
[12:18:26] <IchGuckLive> or one foot
[12:18:28] <AR_> screw to 2x4
[12:18:31] <AR_> ????
[12:18:32] <AR_> profit
[12:18:55] <L84Supper> AR_: we already tried that and got infested with termites
[12:19:04] <AR_> lol
[12:20:12] <IchGuckLive> L84Supper:
http://www.asconveyorsystems.co.uk/belt-conveyors.html
[12:20:16] <L84Supper> vacuum would be nice as well
[12:20:30] <IchGuckLive> http://www.asconveyorsystems.co.uk/Vacuum-Conveyors.html
[12:20:44] <L84Supper> heh, yes made by Dorner, near us
[12:21:17] <L84Supper> they seem to make belts for everyone
[12:21:39] <IchGuckLive> as they are within the top 10
[12:22:15] <AR_> litecoin client looks just like bitcoin client
[12:22:18] <AR_> hm
[12:22:20] <AR_> conspiracy
[12:23:00] <IchGuckLive> L84Supper: 800 USD
[12:23:31] <L84Supper> 330mm width
[12:23:34] <IchGuckLive> 225M040600A0906 50" X 4"
[12:23:45] <L84Supper> 2m length
[12:24:16] <IchGuckLive> Dorner 2200 Conveyor 24" x 48" x 46" 650
[12:25:29] <L84Supper> a Dorner system is ~$12K with everything to spec, not sure if they can hold +- 50um in Y axis even if we have a laser mic on the edge
[12:25:35] <IchGuckLive> Dorner Conveyor Series 2200 6' ft X 8" Wide dorner motor 1000USD
[12:26:55] <IchGuckLive> i see the replacements are so expensive as it is worth to get a NEW with Percents
[12:27:23] <IchGuckLive> Nice company staregie
[12:30:55] <IchGuckLive> L84Supper: youve seen the Faulhaber system
[12:46:52] <L84Supper> IchGuckLive: Faulhaber for a conveyor? no
[12:47:12] <IchGuckLive> no the hole system
[12:49:32] <L84Supper> IchGuckLive: have link, I don't see it on their website
[12:51:04] <IchGuckLive> http://www.faulhaber.com/sprache2/n426848/n.html
[12:51:59] <IchGuckLive> 6feet is your length how high woudt you go 80mm 3"
[12:52:46] <IchGuckLive> i woudt go for a pricice Timingbelt system 2 lines and mound a normal transport belt on it
[12:54:34] <IchGuckLive> do you need full rotation or only baxk and forth
[12:55:38] <IchGuckLive> the HDT 14M woudt fit your need
[12:57:17] <sadara> Two questions: 1: Has anyone here designed/built a linear drive? 2:What type of motor (what would it be called) if you took a car alternator and applied power to both the stator and the rotor coils? It's not a PMSM, because they is no permanent magnets, and because the rotor is energised directly, it not a coreless either. I'm trying to research it a bit, but need to know what it is called
[12:58:47] <IchGuckLive> sadara: welcome to our ICQ
[12:58:48] <pcw_home> its a synchronous 3 phase motor
[12:59:07] <IchGuckLive> why liniar wotory better to control
[12:59:38] <WalterN> L84Supper: well, the laser was at my door, but I wasent awake to sign for it... so I'll probably get it tomorrow
[13:00:11] <sadara> IchGuckLive: linear is faster, lower inertia, and just as easy to control
[13:00:13] <IchGuckLive> WalterN: did you find a powersupply
[13:00:37] <pcw_home> the alternator is drivable like a PMSM as long as you supply field power
[13:00:43] <sadara> pcw_home: Just syncronous, PMSM is Permanent magnet sync, is there an acronym or something?
[13:01:19] <WalterN> IchGuckLive: kinda... one was $600 for just the driver (its not a complete power supply...)
[13:02:15] <pcw_home> Just synchronous (lots of BIG AC motors used to be synchronous not sure if they are common anymore)
[13:02:20] <WalterN> I dont think I want to pay that much for 5 pieces of electronics
[13:02:31] <sadara> found it, DC-excited sync motor
[13:04:24] <sadara> Thanks pcw. (I'm figuring a linear drive is just a flat motor, and you don't need slip rings to energise the "rotor")
[13:05:14] <pcw_home> They have one advantage compared to PMSM in that you can adjust the
[13:05:16] <pcw_home> field for different torque/speed characteristics
[13:05:18] <Connor> pcw_home: On pete's encoder + prox sensor issue. I have some optos that will work.. I just need clarification on what to do with the index side of the opto. Do I tie the index to the collector (with resistor in series and then tie that to the mesa input ?
[13:05:43] <Connor> I.E. index acts as if it's suppling the Vcc to the opto...
[13:06:20] <L84Supper> IchGuckLive: continuous motion in one direction, 300mm/sec
[13:06:41] <pcw_home> index --> collector --> emitter --> pulldown --> 7I77 input
[13:08:20] <Connor> okay. the resistor on the pulldown is tied to grown.. simple schematic would be helpfull.
[13:08:34] <pcw_home> the fly in that particular ointment is that the 7I77 encoder input already has a pullup in TTL mode so the pulldown needs to be quite stiff (if you had a active low index it would be better)
[13:09:32] <pcw_home> If you use a differential encoder you would have both index polarities available
[13:09:40] <L84Supper> IchGuckLive: length is 2 meters, width is 330mm
[13:10:10] <Connor> We will be using a differential encoder.. but only in single ended mode to start off with unless we have issues..
[13:10:35] <pcw_home> then the active low index --> opto emitter --> opto collector --> 7I77 index input
[13:10:42] <L84Supper> IchGuckLive: the concern with a timing belt is the vibration it will introduce
[13:10:43] <Connor> I have a slew of PC817 opto's
[13:11:30] <IchGuckLive> L84Supper: agree on that
[13:11:58] <pcw_home> (using the active low index signal eliminate the pulldown and dueling pullups/pulldowns)
[13:12:35] <Connor> pcw_home: okay, so at that point the active_low index acts as a syncing ground.
[13:13:04] <Connor> would we use the index or index not input on the 7i77 ?
[13:13:29] <pcw_home> you need to use the TTL input (I forget which one)
[13:14:07] <Connor> That should be the regular one I would think. not the NOT ..
[13:14:24] <IchGuckLive> L84Supper: so staying with vaccum Dorner Conveyor Series 2200
[13:14:49] <L84Supper> IchGuckLive: so far, but looking for options
[13:15:45] <L84Supper> if we 3D print parts directly onto a conveyor we will also want the same accuracy, <50um
[13:18:06] <IchGuckLive> L84Supper: why not robot at M30 gets the part of the table
[13:18:07] <AR_> why
[13:18:31] <IchGuckLive> L84Supper: and Robot restarts the programm
[13:18:59] <L84Supper> IchGuckLive: the current application is sheets of foil/film/paper
[13:19:11] <IchGuckLive> ah ok
[13:19:14] <Connor> pcw_home: looks like pin 23 is IDX5 and ping 24 is /IDX5
[13:19:38] <L84Supper> but a 3D printer might use a belt to carry parts through different deposition systems for different types of layers/materials
[13:19:45] <IchGuckLive> Dorner 2100 Series 930132 Conveyor Belt 10" X 84" this is the best offer i get
[13:20:12] <L84Supper> IchGuckLive: yeah, talked to their engineers again earlier today
[13:20:14] <pcw_home> 7I77 manual should say which in to use when in TTL mode
[13:21:02] <IchGuckLive> i woudt go for a pallet changer
[13:21:16] <Connor> Only thing I found in manul is on page 53
[13:21:20] <IchGuckLive> this gives you a better then <50µm
[13:21:26] <Connor> INPUT TTL MODE THRESHOLD 1.4 1.8 Volts
[13:21:35] <Connor> that's the ONLY TTL I found in the manual
[13:22:21] <Connor> When used with single ended encoders, the ENCA+, ENCB+ and IDX+ signals are wired to the encoder and the ENCA-,ENCB-, and IDX- terminal left unconnected.
[13:22:32] <IchGuckLive> im off by
[13:24:51] <sadara> L84Supper: I've seen linear motors do 80m/sec
[13:25:03] <sadara> (that is meters btw, not mm)
[13:26:47] <L84Supper> sadara: yeah, we are only going 300mm/sec tops
[13:26:59] <L84Supper> but it has to be a loop, no back and forth
[13:29:46] <sadara> I had a borked promtional video once of a engine block tied down on a linear drive x/y table been machined. You could see the promo was for the fast table, as the machine did a lot of unnessecery tool changes etc, and lots and lots of rapids. then one of the tie down broke and the motor went through a wall.
[13:30:33] <Jymmm> ve7it: pong
[13:31:37] <L84Supper> sadara: quite a missile
[13:32:12] <L84Supper> sadara: there are 5gram sheets
[13:32:14] <ve7it> Jymmm, hey... I think I need a faster computer... compiling a distribution image for an imbedded linux box I am working on takes 1.5 days
[13:32:32] <L84Supper> sadara: so the worst case is probably a paper cut :)
[13:33:17] <sadara> idk, 5 gram sheet of tungesten at 80m/sec --->
[13:33:25] <sadara> whatcha cutting?
[13:33:34] <ve7it> Jymmm, maybe a good year to upgrade... there has been some income.... any ideas on good i7 proc/mb/memory combos
[13:33:38] <Jymmm> ve7it: You could stop watching 42 pr0n all at the same time too, that might speed something up! And what takes 1.5days to compile?
[13:35:05] <ve7it> basically a cd linux distribution for an arm processor
[13:35:16] <Jymmm> ve7it: Either hit CL for a deal, or what pretty much do is it Freiday's Fry's ad for Laptop De Jour
[13:36:00] <Jymmm> ve7it: Dell E6xxx seem promising
[13:36:11] <ve7it> it seems a lot of new m/b has the windows 8 bios crap
[13:36:27] <Jymmm> bios crap?
[13:36:33] <ve7it> eufi?
[13:36:52] <ve7it> It wont let you install linux!
[13:37:18] <L84Supper> yeah, more crappy BIOS
[13:37:43] <Jymmm> ve7it: take a look, see what you think
http://www.dell.com/us/business/p/latitude-e6430/pd
[13:37:58] <pcw_home> Ha the was a newspaper article today that blamed low PC sales on move to mobile _and_consumer dislike of Windows 8
[13:38:04] <Jymmm> ve7it: I saw one on CL for $400
[13:38:46] <Connor> pcw_home: You see my above statements on the TTL input info in the manual ?
[13:39:04] <Jymmm> ve7it:
http://www.ubuntu.com/certification/hardware/201202-10552/
[13:40:13] <pcw_home> Yes, so you use the + inputs for TTL
[13:41:03] <Jymmm> ve7it: Max ram 16GB
[13:41:42] <ve7it> Jymmm, dont really want a laptop.... a fast desktop for a compile engine is better... but fast gammer systems concentrate on fast video... VGA video is all I care about
[13:42:11] <Jymmm> ve7it: budget?
[13:42:37] <ve7it> Jymmm, i DONT NEED A $500 video card... budget is <$1000
[13:44:36] <carper64_lb_> only thing that intrests me is that a graphics card is compatable for real view use with solidworks
[13:45:02] <Connor> pcw_home: okay.. just want to be sure. I do have one more question.. then I'll shut up.. when using the active low.. is that a negative voltage, or just the inverse of the other side ?
[13:45:24] <Jymmm> ve7it: I thought this was pretty cool myself....
http://www.newegg.com/Product/Product.aspx?Item=N82E16856101117 video
http://www.youtube.com/watch?v=l00FV8bG0D4
[13:45:24] <Tecan> (l00FV8bG0D4) "Shuttle SH67H3 Barebones Mini PC Unboxing & First Look Linus Tech Tips" by "LinusTechTips" is "Tech" - Length: 0:09:41
[13:45:49] <ve7it> Jymmm, was thinking of i7 quad processor, 8 or 16gb of 2400Mhz DDR3 ram, an IVY bridge motherboard... not sure if a solid state drive would benefit compiling or not... maybe mounting /usr/bin in a ram drive would be better.
[13:46:17] <Jymmm> ve7it: It would, but you said < $1000
[13:47:09] <pcw_home> active high index is ~0V when no index and ~5V at index
[13:47:11] <pcw_home> active low index is ~5V when no index and ~0V at index
[13:47:24] <Jymmm> ve7it: SuperMicro mobo's could do the quad CPU's and big ram
[13:49:23] <Connor> okay. got it. Should be simple enough..
[13:51:20] <Jymmm> ve7it: I kinda stay away from big klunky boxes anymore
[13:52:04] <carper64_lb_> i find the atom board runs linuxcnc eay
[13:52:08] <carper64_lb_> easy
[14:01:21] <Jymmm> ve7it: Shuttle + i5-2500K + 16gb + 64gb ssd + 1TB hdd == $662 USD
[14:04:06] <Jymmm> ve7it: add $150 for i7-3770K (3.5GHz)
[14:05:17] <ve7it> Jymmm, too many choices... newegg has 17 pages of intel motherboards
[14:05:35] <Jymmm> ve7it: intel brand mobo?
[14:06:07] <Jymmm> ve7it: I have a love/hate relationship with intel brand mobo's
[14:07:02] <ve7it> intel proccessor...
http://www.newegg.com/Product/Product.aspx?Item=N82E16813157330 seems like it might be a good board
[14:07:06] <Jymmm> ve7it: some of the gigabyte mobo's have aoverclock buton on them
[14:07:31] <sadara> what new systems are getting low latency these days?
[14:07:57] <Jymmm> ve7it: you want a mid tower system?
[14:07:59] <sadara> (and have 3 pci slots :)
[14:09:01] <ve7it> Jymmm, doesnt really matter... fast disk, fast memory bandwidth and a quad processor probably win.... I will have to cruise toms hardware for a while
[14:09:02] <Jymmm> ve7it: try to get one with intel gigabit ethernet, relatek nic is kludgy
[14:09:43] <Jymmm> ve7it: 4 i7 cpus alone will cost $1200
[14:10:14] <ve7it> fast network is not an issue..... talking to the internet is throttled through a $29 wireless router and then a cable modem
[14:10:25] <Jymmm> ve7it: bbefore ram, ssd, or mobo
[14:10:33] <ve7it> Jymmm, just thinking of 1 i7!
[14:10:36] <Jymmm> ve7it: it's BUGGY
[14:10:58] <Jymmm> ve7it: you said quad processor twice... that's 4 cpus
[14:11:10] <ve7it> quad core
[14:11:10] <Jymmm> ve7it: did you mean quad core?
[14:11:16] <Jymmm> ahhhhhhhhhhhhhhhhhhhh
[14:11:20] <Jymmm> * WHACK*
[14:11:31] <Jymmm> ve7it: No HF for you!
[14:12:03] <Jymmm> ve7it: We'ree keeping you on 11m from now on!
[14:12:34] <ve7it> Jymmm,
http://www.newegg.com/Product/Product.aspx?Item=N82E16819116501 really eats up the budget!
[14:13:38] <ve7it> nice at 77w.... and includes pretty hot graphics, so a separate video card is not needed
[14:16:31] <Jymmm> ve7it: that cpu, the shuttle Z77, 16gb, 64 ssd == $890 USD
[14:19:25] <Jymmm> Bummer, the shittle uses realtak nic, crap.
[14:19:51] <Jymmm> I need a good VM box
[14:22:23] <PetefromTn> hello!!
[14:29:21] <L84Supper> ok so a 1m/sec linear motor will actually work instead of a belt
[14:29:46] <AR_> yes
[14:30:30] <L84Supper> they just speced a belt since nobody ever tried a linear motor
[14:30:39] <L84Supper> figures
[14:31:03] <AR_> inferiors
[14:31:54] <L84Supper> but the price of a 6 foot linear motor is about that of a precision conveyor belt
[14:34:37] <L84Supper> 121,000 meters of travel per week, how many weeks until the bearings are shot?
[14:36:30] <L84Supper> need more durability specs
[14:39:08] <L84Supper> that is 4.5 million inches per week
[14:40:45] <AR_> maglev
[14:40:47] <L84Supper> thats workable without an aire bearing
[14:40:53] <L84Supper> air even
[14:41:32] <L84Supper> yeah mag or air bearing is best but the cost might be too high and the lead times are >3 months
[14:55:25] <sadara> L84Supper: I'm working to bring linear motors to the masses !
[14:56:22] <sadara> I'm trying to figure out a good design, I'll be prototyping it over the comming weeks
[14:56:50] <sadara> what I don't get is why they are so damn expencive
[14:56:53] * skunkworks_ hands sadara a bunch of HD maganets...
[14:57:28] <sadara> skunkworks_: I'm thinking I'll go magnet free, as ND mags are $$$
[14:57:46] <sadara> copper magnet wire is only $$
[14:58:00] <skunkworks_> your going to use electromaganets?
[15:00:23] <sadara> I havn't desided, you can do it one of two ways, either using "electromagnets" for your "rotor", or using to motors back to back, like an inrunner running inside an outrunner ( or even three motors, "rotor" coils, "stator" coils, them more "rotor" coils, this way you seem not to need a linear bearing at all)
[15:01:51] <sadara> or with permanent magnets as you said, but I'm trying to keep this low cost and high performance, and it's not that difficult by the look of it, but I can say that as I havn't prototyped anything yet
[15:03:27] <L84Supper> sadara: what srange of sped and accell
[15:03:45] <L84Supper> sadara: what range of sped and accelleration?
[15:04:32] <sadara> I'm not exactly sure yet, but the motors are scaleable, just add more laminations (the same as making a stepper longer by adding laminations
[15:05:18] <sadara> power wise, anywhere you to about 3.6 kw
[15:05:27] <sadara> s/you/up
[15:05:55] <PCW> there are linear stepmotors (just unrolled and rotor/stator swap)
[15:06:38] <sadara> yeah, I was looking for one, but I couldn't find any priceing, so I'm assuming they are out of financial reach for most ppl
[15:07:29] <PCW> and lnear induction motors
[15:08:32] <sadara> It's my understanding that linear induction motors are tricky to do right
[15:08:36] <PCW> then there are the MHD liquid metal motors/pumps
[15:09:20] <L84Supper> I need to hit ~2m/second for a distance of 2 meters in one direction
[15:09:48] <JT-Shop> black powder
[15:10:01] <sadara> ^^ beat me to it
[15:10:09] <JT-Shop> lol
[15:10:22] <L84Supper> 1.5m/sec min
[15:10:56] <sadara> what mass?
[15:11:33] <L84Supper> 5 gram load + whatever the mass of the carriage itself is
[15:12:00] <sadara> easy, compressed air
[15:12:05] <sadara> do you need precision?
[15:12:24] <L84Supper> it's a piece of paper
[15:12:27] <L84Supper> heh
[15:12:43] <L84Supper> need to move a sheet through a printer
[15:12:48] <sadara> how are you grasping the paper?
[15:13:08] <L84Supper> vacuum table ~1Kg
[15:13:22] <sadara> rollers seem the obvious choise, what is the reason for not using them?
[15:13:34] <L84Supper> roller or belt?
[15:13:47] <sadara> rollers
[15:13:51] <L84Supper> 50um paper thickness
[15:14:01] <L84Supper> sheet not web
[15:14:28] <sadara> hrmmm.... what about electrostatics?
[15:14:44] <L84Supper> foil paper, conductive
[15:14:56] <sadara> on a roll?
[15:15:05] <L84Supper> sheet not roll/web
[15:15:18] <sadara> charge the sheet up
[15:15:22] <L84Supper> 12" x 16"
[15:15:29] <L84Supper> can't
[15:15:43] <sadara> keep it in position using electrostatic charge of same polarity
[15:15:59] <sadara> L84Supper: Whay not?
[15:16:10] <L84Supper> not under my control
[15:16:17] <L84Supper> we get what they give us
[15:16:24] <L84Supper> automated high speed line
[15:16:32] <sadara> does it have to stay exactly flat?
[15:16:38] <L84Supper> yes
[15:16:57] * sadara is thinking realy large diameter roller, like 2-3m
[15:17:44] <L84Supper> belt vs roller
[15:17:51] <L84Supper> stainless vacuum belt
[15:18:06] <sadara> if it can't be bent slightly, does it have to travel in an exactly staight line?
[15:18:18] <sadara> L84Supper: forget the belt for a moment
[15:19:13] <L84Supper> yes in X and Y
[15:19:42] <L84Supper> straight in X with as little Y jitter as possible
[15:20:02] <sadara> so y can't be curved?
[15:20:08] <L84Supper> no
[15:20:11] <sadara> I'm thinking of this as a printer in a production line. sheet of paper comes in, get printed, goes out, whatever happens in the printer is up to you?
[15:29:21] <L84Supper> this belt is the printer
[15:30:17] <L84Supper> sadara: whwn you said roller I thought roller conveyor not a drum
[15:31:24] <L84Supper> a drum that size would be hard to hold to spec
[15:31:28] <sadara> if you can vaccuum the paper onto a rapidly spinning large diameter drum, have the print head directly above the drum, and then airjet the paper off the drum
[15:31:59] <L84Supper> we don't load or unload we just expect some other machine to load and unload
[15:32:21] <L84Supper> we just start when the paper goes down on one end and move it to the other end
[15:32:34] <sadara> what tolerance? (I would be more concerned with keeping the paper square to x axis on the drum)
[15:32:45] <L84Supper> we have no idea what puts it down or pick it up
[15:33:19] <sadara> I would need more information to help with this
[15:33:38] <L84Supper> we compensate for squarness with a 4 camera system that measure reference points then we adjust for skew
[15:34:13] <sadara> if you can adjust for skew, can't you adjust for off-axis distortion?
[15:34:39] <L84Supper> I'd just a like a conveyor that does not jitter in the Y (side to side) since the x is constant speed
[15:35:07] <sadara> yeah, a large diameter drum emulates that
[15:35:16] <L84Supper> yes, but the Y axis movement is only a concern between when the camera inspects it and 20 inches later when the printer puts the ink down
[15:36:18] <sadara> put the camera 20 inches back along the circumferance of the drum
[15:36:26] <L84Supper> the camera might see oe thing and if there is distortion before the ink goes down that is the only problem
[15:37:06] <sadara> if you do it right, the paper would be held firmly on the drum, either unisng electrostatics or vaccuum
[15:37:11] <L84Supper> giant 2 meter flyweel
[15:38:02] <L84Supper> thats not going to fit in the floor space available
[15:38:34] <L84Supper> not many places can machine a 2m dia drum
[15:38:46] <sadara> I may be wrong, but I'm working it out to 20 rpm
[15:39:09] <sadara> go to 100rpm and make drum smaller
[15:39:57] <L84Supper> lets just sat drum won't work since we are told that the belt has to be flat
[15:40:17] <L84Supper> we have no control over what comes before or after this machine
[15:40:23] <sadara> so your trying to accelerate a stainless belt to 2m a sec in 2m?
[15:40:25] <L84Supper> it would make it easier if we did
[15:40:32] <L84Supper> no
[15:40:44] <L84Supper> constant velocity belt
[15:40:52] <sadara> cv @ 2m/s?
[15:41:17] <L84Supper> i was just running the numbers if a linear motor was used instead of a belt
[15:41:31] <sadara> ahhh, I'm following you now
[15:41:47] <sadara> how many sheetsper min?
[15:41:51] <L84Supper> no the belt would only travel at 0.3m /sec
[15:42:09] <L84Supper> 6 second cycle time between sheets
[15:42:22] <L84Supper> 6 foot long belt (2m)
[15:42:37] <sadara> linear motor will do that really easily
[15:42:57] <sadara> your looking for .2g
[15:43:08] <L84Supper> it's funny how on IRC if something can be missunderstood it most certainly will be :)
[15:43:20] <sadara> true
[15:43:31] <sadara> COTS linear motors go up to serveral 100G
[15:46:51] <sadara> even a cheap(ish) 1G will easily meet your requirements
[15:48:59] <L84Supper> 5 gram paper, 1-2kg vacuum table is whats moving
[15:49:46] <L84Supper> now the problem is how to connect vacuum line that whips around +-1m at 2G's
[15:49:54] <sadara> 1-2kg moved by 60 newtons?
[15:50:37] <sadara> thats 60m/s/s
[15:51:01] <sadara> how far?
[15:51:43] <sadara> (btw, 60newton is about the smallest standard linear drive)
[15:51:46] <L84Supper> it has to slow down to <0.5m per second for 1.4 seconds and the travel is 2 meters
[15:52:16] <L84Supper> fast. slow, fast for one direction and then fast all the way back
[15:52:22] <sadara> easy
[15:52:49] <L84Supper> yeah, people here though 2G's of accel would require gunpowder
[15:53:13] <sadara> http://intellidrives.com/Linear-actuator-index.htm
[15:53:16] <L84Supper> they probably never worked with linear motors
[15:53:29] <sadara> LSS-120 seems right
[15:53:41] <sadara> your peak force can be with a dc brake
[15:53:43] <L84Supper> hiwin, parker, aerotec, there are several
[15:54:12] <L84Supper> it's more lead times with linear motors 2 meters long
[15:54:21] <L84Supper> 3-4 months for some vendors
[15:55:20] <sadara> 150n/m acceleration, brake at 450N/m, precison travel through the printer, accerate again, then brake, I need to know the exact distances to work out the timing
[15:55:31] <sadara> L84Supper: 2m is off the shelf
[15:55:56] <sadara> the linear induction drive (read = more $$$) are cut to length on same day
[15:56:39] <L84Supper> yeah, they delivered fast before
[16:10:37] <L84Supper> sadara: linear motor/actuator is 0.5x the cost of a conveyor belt
[16:12:54] <sadara> L84Supper: that was one explancive conveyor
[16:13:06] <sadara> L84Supper: Whare did you get a price?
[16:17:13] <DJ9DJ> gn8
[16:17:26] <sadara> gnate?
[16:18:13] <DJ9DJ> good night!
[16:20:25] <sadara> L84Supper: I'm going to get some sleep, it's 5am here, and I have wrok ing the morning
[16:27:17] <brummer> not sure when to jump in with questions but I have a shizuoka cnc mill with bandit control i intend to convert to linuxcnc. i believe it has brushed dc servos though I need to verify. Mesa boards seem to be the most recommended type here.
[16:32:13] <L84Supper> yes
[16:32:48] <PCW> Nah they're carp
[16:32:53] <PCW> crap
[16:33:21] <L84Supper> for the halibut?
[16:34:20] <PCW> just for the halibut
[16:36:33] <L84Supper> http://www.intellidrives.com/Rotary_Tables_Selection-si-59.html nice servo direct drive rotary tables
[16:37:53] <L84Supper> PCW: have an amp for something this size
http://www.intellidrives.com/product_info_1.php?cPath=&products_id=46
[16:38:56] <L84Supper> 14A peak, 5.1A continuous hree phase brushless Y winding (standard)
[16:44:31] <PCW> 8I20 maybe
[16:51:09] <L84Supper> http://sensing.honeywell.com/asdx%20series%20ultra-low%20pressure%20silicon%20pressure%20sensors anyone ever see a board with several of these on it to monitor low pressures in a system?
[17:02:23] <L84Supper> PCW: do have an I2C IO card?
[17:02:29] <L84Supper> do you
[17:03:13] <L84Supper> or use GPIO and modify the fpga?
[17:04:15] <L84Supper> nevermind I just found sensors with SPI
[17:20:50] <andypugh> So, if I turn the current up on my brushless motor to the level it seems to need, instead of going faster, it squeals horribly.
[17:21:42] <andypugh> I can't decide if this means that I am saturating the iron, or something else.
[17:21:55] <JT-Shop> yikes
[17:22:09] <pcw1204> If it squeals its probably means you are current limiting on the 7I39
[17:22:52] <andypugh> pcw1204: This is at 0.4 durt-cycle on the 7i39
[17:23:00] <andypugh> (duty cycle)
[17:23:50] <andypugh> But, it only has 1.5 ohms winding resistance, I suspect it basically isn't "inductive" enough.
[17:24:19] <pcw1204> duty cycle determines voltage so 40% of full voltage can easily be too much current unless spinning
[17:24:55] <andypugh> Right, which fits with my thoughts that I need to re-wind it with thinner wire.
[17:25:23] <pcw1204> if its inductance you might try a high PWM rate
[17:25:38] <andypugh> It's already at 40khz
[17:26:21] <FinboySlick> pcw_home: got another quick one for you if you have a moment.
[17:26:38] <pcw1204> depending on your magnetic circuit you may have pretty low indcutance
[17:27:13] <andypugh> I think that is the problem. Hence wanting to try thinner wire.
[17:27:29] <FinboySlick> Some of the circuitry I am testing is fairly wide and ISE complains that it ran out of IOBs since the chip doesn't have enough external pins to match. Obviously this is all going to be internal logic and not connected to pins, is there a way to tell ISE to just not care about this?
[17:28:14] <pcw1204> chose a bigger chip?
[17:28:29] <pcw1204> or wire it all to a big mux
[17:30:18] <andypugh> A sort-of similar motor I have comes out at 5 ohms DC resistance. Measuring inductance is a bit beyond my means at the moment.
[17:30:34] <FinboySlick> pcw_home: Could I wire the pins to registers?
[17:32:22] <pcw1204> yes as long as they can be read externally
[17:32:39] <pcw1204> (otherwise it will all get optimised away)
[17:33:11] <FinboySlick> OK.
[17:35:50] <pcw1204> remember inductance is proportional to turns squared
[17:37:57] <andypugh> Current wire is .45mm / 26 gauge. I am trying to decide between 30 gauge (half the area, twice the turns, 4x the DC resistance, 4x the inductance) and .28mm (32 gauge) where the factor is 2.6 / 6.6
[17:39:49] <pcw1204> I suspect 30 would be better
[17:39:50] <andypugh> I suppose I could try both, it's only really really tedious to re-wind. :-)
[17:40:14] <pcw1204> but thats just a guess
[17:40:18] <andypugh> Yes, I suspect I would get into current-density problems with the 32.
[17:41:06] <andypugh> Though 32-gauge delta might be right.
[17:41:15] <pcw1204> can you measure how much torque you are getting with a DC supply for a sanity check?
[17:41:46] <pcw1204> (or measure the BEMF)
[17:42:09] <andypugh> Measuring bemf sounds easier.
[17:42:42] <pcw1204> do you have a lab supply?
[17:42:56] <andypugh> I used to design dynos for testing small electric motors, but that was a few jobs ago ;-)
[17:43:16] <andypugh> The DC supply is easy. measuring the torque less to.
[17:43:40] <pcw1204> BEMF when spun will give you an idea of the correct number of turns
[17:52:47] <andypugh> about 15V at the fastest I can flick it.
[17:52:53] <andypugh> (on the scope)
[17:53:31] <andypugh> So I will try the 30 gauge, keeping the option of switching to delta connection.
[17:58:34] <andypugh> (The wires are now too short to reach the PCB anyway, so I need to do some reconstruction, I might as well experiment with different winding)
[18:02:12] <PetefromTn_> Hey guys, anyone feel like helping me walk thru wiring up the logic on my Hitachi WJ200 spindle drive?
[18:07:10] <JT-Shop> yikes!
[18:07:33] <PetefromTn_> That bad huh?
[18:08:35] <JT-Shop> well I know nothing about a Hitachi drive lol
[18:08:43] <JT-Shop> is it just on and off?
[18:09:07] * JT-Shop is glad I have an Automation Direct GS2 drive that speaks to LinuxCNC
[18:11:02] <PetefromTn_> Nah I have the 5i25/7i77 combo and I am wanting to hookup the +/-10v analog control signal and also figure out how to wire this 24volt relay that switches on the spindle motor fan to the run signal on the VFD so it only comes on when the spindle is runnin.
[18:11:36] <JT-Shop> I use motion.hmmm let me find that pin
[18:12:21] <PetefromTn_> LOL....
[18:12:57] <JT-Shop> maybe it is halui.program.is-running
[18:13:04] <JT-Shop> let me check the BP config
[18:13:44] <PetefromTn_> No I mean the VFD ouputs a signal which is hooked to an external relay that can be used to switch on and off the relay to power the fan...
[18:15:48] <PetefromTn_> It is on pin 12 on the Hitachi VFD and you can configure the output to switch the relay..
[18:30:55] <andypugh> Well, that is me committed to rewinding my motor. I have a bare stator and a pile of copper.
[18:31:09] <JT-Shop> the washing machine motor?
[18:31:50] <andypugh> No, this one:
https://plus.google.com/photos/108164504656404380542/albums/5747722155741347649/5685034302549414130?banner=pwa
[18:32:18] <JT-Shop> that's one you made right?
[18:32:40] <andypugh> PetefromTn_: Does the spindle drive take +/- 10V, or 0-10V + direction?
[18:33:12] <andypugh> JT-Shop: Yes, so rewinding it is just a case of re-doing something I did before.
[18:33:51] <skunkworks> andypugh: is that why it wasn't running as well as you thought it should before? (open wire)
[18:34:05] <PetefromTn_> andypugh: Lemme check a sec.
[18:34:48] <andypugh> skunkworks: I think that the whole motor is a bad match for the 7i39. Not enough impedance for the 7i39 to drive into.
[18:35:09] <skunkworks> smaller wire?
[18:35:31] <andypugh> That's the plan. Smaller wire, more inductance.
[18:35:39] <skunkworks> neat
[18:35:52] <andypugh> More tedious to wind..
[18:36:18] <skunkworks> heh
[18:36:54] <PetefromTn_> here is a link to the manual PDF..... I am trying to figure out the logic right now. Currently it is wired up and working under front panel control.
http://www.driveswarehouse.com/documentation/Hitachi/WJ200R.pdf
[18:38:30] <PetefromTn_> on page 24 of the manual it says the analog voltage input specs.. It does not actually say it is +/-. At least I don't think it does.
[18:40:16] <andypugh> Yes, 0 to 10V between O and L.
[18:40:58] <PetefromTn_> okay not H and L then...
[18:41:02] <Tom_itx> what size wire did you use to begin with?
[18:41:08] <Tom_itx> and what are you going to?
[18:41:15] <andypugh> No, H is the reference voltage out of the drive.
[18:42:01] <PetefromTn_> Where and how would you get direction input then?
[18:42:35] <andypugh> The FW and RV terminals.
[18:42:45] <andypugh> (page 25)
[18:43:08] <andypugh> For extra fun it seems that you can set up the terminals in a number of ways in software.
[18:43:52] <PetefromTn_> Not seeing the fw and rv...
[18:44:15] <andypugh> Table in Page 25.
[18:44:49] <andypugh> It looks like you can configure any of the logic inputs to be any of the functions in that table.
[18:45:19] <PetefromTn_> Are you sure it is page 25 cause I don't see it...
[18:45:29] <JT-Shop> http://www.spyder.brp.com/compare-models?categoryCode=RT
[18:45:40] <andypugh> It is the page that says 25 at the bottom.
[18:46:02] <Tom_itx> gotta love pdf files
[18:46:32] <Tom_itx> JT-Shop, you gettin a new toy?
[18:46:42] <PetefromTn_> Oh okay not the PDF page now I see it..LOL
[18:47:53] <JT-Shop> making one Tom_itx
[18:48:44] <PetefromTn_> SO what you are saying is that there are some terminals pins that are configurable and I can hookup an output from linuxCNC to two of them one for FWD and one for REV. What pins are the configurable outputs?
[18:49:21] <andypugh> Page 27 explains, I think.
[18:49:45] <andypugh> I am looking at the 7i77 manual to try to figure out which wiring scheme to use.
[18:50:40] <JT-Shop> hmmm wrong link lol
[18:50:48] <PetefromTn_> Thanks man...I am looking at page 23 and it seems that pins 1,2, and L are the intelligent input terminals and perhaps that is how I should hookup.
[18:51:13] <andypugh> 7i77 are sourcing type, so you need to follow the scheme on the bottom of page 29, ]
[18:51:39] <andypugh> Pins 1 to 7 are all "intelligent inputs"
[18:51:52] <andypugh> L is the GND I think.
[18:52:09] <JT-Shop> http://www.youtube.com/watch?v=JynwVKXzc9s
[18:52:10] <Tecan> (JynwVKXzc9s) "Supertramp Live 2011: Bloody Well Right [Full HD]" by "ComeFaceTheMusic" is "Music" - Length: 0:07:07
[18:52:30] <andypugh> Page 27 is the one you need to read carefuly.
[18:54:29] <PetefromTn_> Okay so it looks like a command output from the 7i77 pins output go to the intelligent input pins say 1 and 2 and then where would you hook the ground from L.. Also remove the jumper right?
[18:56:20] <PetefromTn_> I think I get it Andy, thanks man....
[18:57:03] <andypugh> A few more things.
[18:57:24] <PetefromTn_> Now I just need to figure out how to switch this external relay for the fan...
[18:57:42] <andypugh> The motion.spindle-speed-out pin is bipolar, it's a negative number for reverse.
[18:58:25] <andypugh> So, I suggest passing that through an ABS hal function.
http://www.linuxcnc.org/docs/html/man/man9/abs.9.html
[18:58:40] <PetefromTn_> are you talking about the 7i77 or the vfd?
[18:59:08] <andypugh> Then wire the abs.0.is-positive and abs.0.is-negative pins to the fwd/reverse outputs.
[18:59:23] <andypugh> Neither, I am talking about HAL.
[18:59:59] <PetefromTn_> I am sorry for my lack of understanding, hal controls pinouts on the 7i77, correct?
[19:00:11] <andypugh> Not really.
[19:00:33] <andypugh> HAL controls what the pins on the 7i77 are used for.
[19:02:15] <PetefromTn_> How would you wire the pins for fwd and rev? I originally thought it would take a +/- 10v and the polarity would control the direction but that will not work now so I am not sure if there is a way to use the analog output from the 7i77 to control both the frequency/speed as well as the direction or do I need to configure a couple output pins from the 7i77.
[19:02:56] <andypugh> If you just connect motion.spindle-speed-out to hm2_5i25.0.7i77.0.0.analogout-03 (or whatever the pin is called) then anticklockwise rotation commands will output a negative voltage on the analog output, and that will rather confuse the VFD>
[19:03:26] <andypugh> I think you need to use a couple of output pins for FWD/REV
[19:04:39] <PetefromTn_> Okay so we are definitely going to tie up a couple field outputs then....
[19:05:50] <andypugh> It looks like it.
[19:06:21] <andypugh> Are you in danger of running out?
[19:07:15] <PetefromTn_> No not really but I have not even really started on the toolchanger yet so I am trying to keep them spared as much as possible...
[19:08:19] <andypugh> You can always hook up another 48 IO lines by adding a smart-serial card.
[19:08:40] <PetefromTn_> Yeah I know but I am trying to make this work with just the one card..
[19:09:05] <Valen> still cold there andypugh?
[19:09:14] <andypugh> I am just saying that it isn't an impossible situation of you run out of outputs.
[19:09:23] <andypugh> Warming up here.
[19:09:28] <PetefromTn_> Yeah I understand..
[19:09:36] <Valen> saw your + photos, heh cold
[19:10:18] <andypugh> That's actually my parent's village, we got rather less snow where I live.
[19:10:31] <Valen> ahh
[19:10:43] <andypugh> Only 200 miles away.
[19:11:30] <PetefromTn_> So basically configure the outputs for fwd and rev and the O and L for the freq then. What do you think about hooking up the run signal output relay for the fan? I have a 24 volt relay that is hooked to the 220v leads hooked to the fan leads and I want to configure it to come on with the run signal whenever the motor is spinning.
[19:14:28] <PetefromTn_> I currently have it hooked up to the pin 12 and the CM2 terminals. Right now without the computer control and just using the panel to run the motor the fan comes on and stays on as soon as power is on to the vfd... Cannot seem to figure out how to make it switch with the run signal only when the motor is spinning.
[19:15:27] <andypugh> You can configure an output terminal to be RUN (Page 36).
[19:16:54] <andypugh> Note what they say about the diode on the relay though.
[19:18:18] <PetefromTn_> Oh okay that may work better actually. I am gonna research these ideas here man thank you so much. I am gonna sign off now so I can go out there and play with it. Appreciate you.
[19:18:39] <andypugh> Time for me to sign off and sleep, too.
[19:19:04] <andypugh> Good luck. DOn't rush it.
[19:21:57] <PetefromTn_> thanks dude goodnight.
[19:22:50] <andypugh> Night all
[19:59:45] <Valen> hmm
http://www.ebay.com.au/itm/CNC-Milling-Machine-Bridgeport-series-2-cnc-mill-textron-/140947073081?pt=LH_DefaultDomain_15&hash=item20d119a839
[20:37:43] <PetefromTn_> Wow that is an old looking machine. It also has that massive longish millhead rail mounting. Dunno if all the series II have that but I don't think so. Kinda reminds me of a tracer mill.
[20:48:38] <skunkworks> steppers
[20:49:54] <L84Supper> https://www.youtube.com/watch?v=72Ig_Dk9pbA&feature=youtube_gdata_player I called them earlier, and it doesn't play music when it spins :(
[20:49:56] <PetefromTn_> Surely not but it does say that in the ad...
[20:51:27] <PetefromTn_> What's am RGV stage?
[20:51:47] <PetefromTn_> Looks EXTREMELY WELL built
[20:51:50] <L84Supper> 2 axis rotary stage
[20:52:10] <L84Supper> RGV is just a model number/name
[20:52:56] <L84Supper> https://www.youtube.com/watch?v=nlAfy_ha57M High Speed Gimbal
[20:56:52] <Tom_itx> what's the application?
[20:58:09] <Valen> PetefromTn_: which one?
[20:58:26] <PetefromTn_> WHich one what?
[20:58:34] <L84Supper> heh
[20:58:52] <L84Supper> the gimbals are for a 3D printer
[20:59:15] <Valen> PetefromTn_: Looks EXTREMELY WELL built
[20:59:54] <PetefromTn_> First one...
[21:00:05] <PetefromTn_> THis looks very nice too...
https://www.youtube.com/watch?NR=1&v=dO6pUaRlZTI&feature=endscreen
[21:00:43] <PetefromTn_> The Hoyt Bow I mean...
[21:00:56] <L84Supper> what bow?
[21:01:41] <PetefromTn_> ROFL...
[21:01:52] <PetefromTn_> Anyone into Archery?
[21:04:50] <r00t4rd3d> anyone know what a fromag keyseater is?
[21:05:56] <jdh> soemthign that cuts keyways?
[21:06:04] <r00t4rd3d> http://www.reddit.com/r/CNC/comments/1c68mm/hi_im_currently_operating_a_fromag_keyseater/
[21:06:13] <r00t4rd3d> i guess
[21:06:40] <jdh> they make dedicated machines for that?
[21:06:46] <r00t4rd3d> i guess
[21:10:46] <r00t4rd3d> http://imgur.com/a/JDTXM#0
[21:10:56] <r00t4rd3d> i like his vaccum table thing
[21:11:10] <r00t4rd3d> http://imgur.com/a/JDTXM/layout/blog
[21:36:24] <PetefromTn> hello..
[21:40:04] <PetefromTn> Okay guys I managed to get the shielded cable run from my 7i77 to my Hitachi wj200 VFD as per Andy's suggestions. Now I need to configure the two outputs from field I/O to set the FWD and REV signals in the VFD. Looking thru signals I see spindle CW and CCW signals. How can I configure these signals to pins 18 and 19 on the 7i77? Trying to figure this one out now...
[21:43:03] <Connor> PetefromTn: Let me look to see what the Pnpconf program does with that..
[21:44:23] <PetefromTn> Okay man but I thought this was a custom setup IE no encoder feedback and such...
[21:44:56] <Connor> Well... I need to look at the base config first to make sure we have no conflicts.
[21:47:04] <PetefromTn> Okay but why would we have conflicts...
[21:47:55] <Connor> Well. Because when we ran it.. I think you assigned the encoder and output stuff already.. I just need to double check.
[21:49:07] <PetefromTn> will this work?
[21:50:06] <PetefromTn> 4. Spindle Direction
[21:50:07] <PetefromTn> If you have direction control of your spindle the HAL pins motion.spindle-forward and motion.spindle-reverse are controlled by M3 and M4. Spindle speed Sn must be set to a positive non-zero value for M3/M4 to turn on spindle motion.
[21:50:07] <PetefromTn> To link these pins to a parallel port pin, put something like the following in your .hal file making sure you pick the pin that is connected to your control device.
[21:50:07] <PetefromTn> net spindle-fwd motion.spindle-forward => parport.0.pin-18-out
[21:50:07] <PetefromTn> net spindle-rev motion.spindle-reverse => parport.0.pin-19-out
[21:50:18] <Connor> No.
[21:50:26] <Connor> because parport isn't the mesa card.
[21:50:40] <PetefromTn> I mean with mesa information, that is right out of the manual silly..
[21:51:21] <PetefromTn> something like 5i25.7177.0.pin18 etc...
[21:54:19] <Connor> to be honest.. I'm not sure.
[21:54:31] <Connor> My spindle only uses a single output.
[21:55:13] <Connor> That sounds right.
[21:55:23] <PetefromTn> really how? You gotta have speed and direction and usually grounds for each..
[21:57:23] <PetefromTn> I am also gonna want to setup a spindle drive enable signal unless that is the analog out too typically.
[21:57:33] <Connor> Sorry. Use 1 output for on/off. 1 output for CW/CCW and then 1 output for PWM which converts to 0-10v
[21:58:36] <Connor> In your case. forward and reverse on separate pins... is the other accepted way. no need for on/off.. because of both are off.. drive is off.
[21:59:25] <PetefromTn> Well that is fine I would like to find a way to only use one pin for direction but I am not sure how to do that on the wj200...
[21:59:44] <Connor> I don't think you can.
[21:59:59] <PetefromTn> I dunno I need to read the manual some more...
[22:00:18] <Connor> it's either 1 pin for on/off + 1 pin for dir + 1 pin for analog out.. or 1 pin for cw + 1 pin for ccw + 1 pin for analog out.
[22:01:27] <PetefromTn> Looking at Lee's Hal file I notice that there are no pins designated in any of the spindle lines of code. Not really sure what it looks like but I would expect to see a pin number somewhere. There is spindle Cw anc CCW in the halui area of the file tho...
[22:01:57] <Connor> His may use -10 to +10
[22:03:35] <PetefromTn> I suppose...
[22:04:27] <Connor> So, looking at the manual... Looks like FW is pin 1, RV is pin2 ?
[22:04:37] <Connor> (on the VFD not the mesa)
[22:05:15] <PetefromTn> Well actually the intelligent inputs 1-7 are configurable so in essence yes..
[22:05:45] <Connor> okay. so, yea.. your going 2 burn 2 I/O pins.. 1 for CW and 1 for CCW
[22:06:16] <Connor> What *I* would do.. is go ahead and config the hal file.. and leave the VFD unconnected..
[22:06:36] <Connor> set the spindle speed and direction.. and then use the hal configuration to verify the pins are active..
[22:06:42] <PetefromTn> yeah page 27 of the manual shows what any of the configurable pins can be set to...just input the correct code.
[22:06:45] <Connor> and then take a volt meter and check the output voltage.
[22:07:22] <PetefromTn> Config the hal file is what I don't understand LOL..
[22:07:42] <Connor> okay. let me look at the one I copied from your macihne..
[22:09:58] <Connor> Okay.. You have two lines in your .hal file..
[22:10:00] <Connor> net spindle-cw <= motion.spindle-forward
[22:10:00] <Connor> net spindle-ccw <= motion.spindle-reverse
[22:10:22] <Connor> I think they need to read something like this..
[22:11:27] <PetefromTn> If you look at the manual for the VFD page 27 there is an intelligent input setting called FWD/REV and it says three wire interface...trying to figure out what the hell that means.
[22:12:32] <Connor> net spindle-cw motion.spindle-forward => hm2_5i25.0.7i77.0.pin18
[22:12:32] <Connor> net spindle-ccw motion.spindle-reverse => hm2_5i25.0.7i77.0.pin19
[22:12:44] <Connor> note: I'm not sure of the name of the pin.. you'll have to verify that..
[22:12:57] <PetefromTn> Okay that looks like what I expected are you sure that will work?
[22:13:36] <Connor> Well.. I'm saying.. try that.. with the VFD disconnected from the mesa.. and check the output signals first.
[22:13:48] <Connor> using hal config and a volt meter. :)
[22:14:01] <Connor> 3 wire interface is, CW, CCW, Analog
[22:14:09] <Connor> I'm guessing.
[22:16:03] <PetefromTn> Well if ONE pin input can be setup for fwd and rev then no actually..
[22:18:25] <Connor> Still have to have a on/off or enable.. I don't think they trust the 0-10v analog to be 0v for the OFF position.
[22:19:09] <PetefromTn> dunno...reading.
[22:20:07] <L84Supper> http://emc2arduino.wordpress.com/ Emc2Arduino
[22:21:57] <Connor> time to eat.. back in a bit.
[22:22:14] <PetefromTn> enjoy...
[22:31:10] <PetefromTn> Connor when you get back here can you email me that copy of my hal and INI files so I don't have to try to take it out of the machine with my phone LOL....thanks dude.
[22:32:52] <L84Supper> http://forums.reprap.org/list.php?155 I get a headache just reading through these posts
[22:39:21] <mr_new> good morning everyone
[22:39:48] <PetefromTn> whaddya mean ya get a headache? Morning new.
[22:39:53] <FinboySlick> mr_new: Good morning.
[22:43:19] <mr_new> hm i need a credit card -.-
[22:43:24] <L84Supper> PetefromTn: misguided expert opinions
[22:43:59] <PetefromTn> L84Supper: ah I see what you mean just reading thru a couple posts...
[22:46:36] <L84Supper> http://www.repetier.com/ was their solution vs build on Linuxcnc
[22:48:13] <L84Supper> http://reprap.org/wiki/Reprap_host_software is another
[22:48:50] <L84Supper> or
http://smoothieware.org/
[23:11:50] <zultron> L84Supper, I'm working on putting together an open-source CAD/CAM/CNC workflow.
[23:13:13] <zultron> LinuxCNC obvi. I have PyCAM 0.5.1 packages under review for Fedora, and about to fix tool setting persistence for 0.6.
[23:14:03] <zultron> FreeCAD, which package I co-maintain in RPMFusion (OpenCASCADE licensing issues keep it out of Fedora).
[23:14:51] <zultron> FreeCAD has a promising CAM module in early stages of development. Some knowledgeable folks behind it.
[23:16:43] <zultron> They're creating a new FreeCAD "CAM workbench". They'll lift the post-processor from HeeksCNC and have various sources for TPGs.
[23:17:39] <PetefromTn> Never used freecad, how does Sheetcam or Cambam work for linuxCNC?
[23:36:19] <L84Supper> zultron: our current interest is in CAM for additive manufacturing
[23:36:54] <zultron> That's great. PyCAM needs a little TLC.
[23:37:20] <zultron> Have you been talking with Lars?
[23:37:30] <L84Supper> it's either start from PyCAM or Heeks
[23:37:57] <zultron> You mean HeeksCNC running with HeeksCAD?
[23:38:48] <L84Supper> not so much any CAD
[23:38:52] <zultron> Or using parts for a new project?
[23:39:12] <zultron> Right, but HeeksCNC was a plugin for HeeksCAD, iiuc.
[23:39:54] <L84Supper> haven't decided yet, use parts of the others and start over or just build from where they left off
[23:40:18] <zultron> A common workflow is to design the part in FreeCAD, bring an STL model to HeeksCAD with the HeeksCNC plugin (or else PyCAM), and then g-code to LinuxCNC of course.
[23:40:48] <L84Supper> serious commercial users will have NX, Mastercam or similar
[23:40:56] <zultron> Yup.
[23:41:06] <zultron> Is that your market?
[23:41:20] <L84Supper> but for prosumer maybe something open source will be agood fit
[23:42:05] <L84Supper> yeah, at first it's just OEM's, later there will be low cost consumer type printers
[23:42:43] <L84Supper> right now there isn't even a commercial CAM tool for additive manufacturing
[23:43:02] <L84Supper> it's the wild frontier
[23:43:06] <zultron> Ha, didn't know that. Sounds like an opportunity.
[23:43:29] <L84Supper> up until now there weren't any 3d printers that were worthy
[23:43:37] <zultron> And you're not interested in CAD?
[23:44:00] <zultron> What're they lacking that would make them worthy?
[23:44:02] <L84Supper> they all were just glacially slow pro machines that just sliced STL's for 2.5D printing
[23:44:39] <L84Supper> well it would be nice if FreeCAD could do what NX does :)
[23:44:44] <zultron> Heh, that's true. Voxelizing a model seems an obvious way to do it.
[23:44:49] <L84Supper> or even solidworks
[23:45:41] <L84Supper> but there are enough CAD tools around
[23:45:55] <zultron> Another reason for PyCAM & FreeCAD is the whole chain can be tied together with python.
[23:46:02] <L84Supper> CAM is the real missing piece since the printers add materials vs cut/remove
[23:46:10] <L84Supper> and the tools are different
[23:46:38] <L84Supper> nozzles, printheads, lasers, DLP projections etc vs rotating cutting tools
[23:46:48] <zultron> Right.
[23:47:34] <L84Supper> so it's part printer driver and part cnc machine
[23:47:42] <roh> well.. seems the tools for both arent what they could be. (both additive and substractive)
[23:48:08] <zultron> I think PyCAM's as far ahead as any of them. Lars said it would be easy to add 5-axis (currently 3 only), so maybe that means it's flexible enough to do additive, too.
[23:48:10] <roh> atleast the open ones or affordable
[23:48:49] <L84Supper> 5-axis would be a good start
[23:49:23] <L84Supper> if we add tools for 3d printing then the closed commercial vendors would be copying PyCAM soon
[23:49:39] <zultron> 5 axis would put it clearly ahead of any other free/libre project I've seen.
[23:49:40] <L84Supper> they haven't even considered it yet
[23:49:58] <roh> i'd be happy with 3 axes
[23:50:18] <L84Supper> Siemens is really interested in seeing what the g-code and tools end up looking like
[23:50:33] <zultron> Sounds like additive would put it clearly ahead of any commercial solution.
[23:50:33] <t12> http://www.todayandtomorrow.net/wp-content/uploads/2009/05/axes.jpg
[23:51:11] <L84Supper> yeah, people tend to find the 3-axis PyCAM toolpath generation slower than simulation or actual cutting times on a machine
[23:51:33] <zultron> Which version were they using?
[23:51:37] <L84Supper> zultron: yes, it would
[23:52:12] <L84Supper> might generate some good press, like linux being ahead of M$ with new hardware support
[23:52:34] <zultron> 0.5.1 was horrible. Some cowboys rode through town a year or two ago and fixed up the data structures and redid the numerical with numpy. Supposed to be much better now.
[23:52:48] <zultron> But I can't stand using it without tool setting persistence.
[23:53:00] <zultron> So that's my first job. :)
[23:53:27] <zultron> Fixing what the last cowboy broke, heh
[23:54:10] <L84Supper> there are lots of STL models floating around already, people need to be able to generate tool paths
[23:54:39] <L84Supper> and the repcrap is just that
[23:55:31] <L84Supper> they are so self absorbed that their software will stay with the toy GGG printers
[23:56:03] <L84Supper> but real printers need CAM that is fast and works well
[23:58:06] <L84Supper> https://github.com/Aeva/voxelpress she started working on some of this, but it's focused on GGG
[23:59:39] <L84Supper> http://voxelpress.org/rasterization_demo.html
