Back
[02:16:25] <DJ9DJ> moin
[02:21:24] <theos> o/
[05:58:42] <jthornton> anyone want some chiggers? you can have mine
[06:00:42] <jthornton> I like this description −V display lots of debugging junk
[06:04:29] <r00t4rd3d> a chinese n*gger?
[06:05:43] <r00t4rd3d> not sure why i starred out the i, really doesnt make it any less offensive
[06:06:59] <r00t4rd3d> you got bugs dude?
[06:07:05] <r00t4rd3d> lol
[06:09:15] <r00t4rd3d> you should have wore a condom
[06:12:39] <jthornton> on my legs?
[06:14:42] <r00t4rd3d> i think its almost fun removing ticks
[06:15:12] <r00t4rd3d> as long as they are on someone else
[06:15:17] <jthornton> at least you can see/feel them crawling up your leg
[06:15:19] <r00t4rd3d> or in i should say
[06:15:53] <theos> ...
[06:17:08] <jthornton> how can I tell what functions I have when I import hal?
[06:17:37] <theos> calmly...
[06:19:48] <r00t4rd3d> does it generate an ini ?
[06:24:05] <jthornton> looks like a pin parameter wrapper to me
[06:34:39] <jthornton> how can I run a stepper x steps then return back to 0 in hal
[06:39:01] <awallin> jthornton: how do you want to trigger that? do you want to repeat that ramp? siggen?
[06:39:58] <jthornton> I want to run a stepper 1000 steps the reverse direction and run 1000 step and repeat
[06:40:15] <jthornton> just using hal
[06:41:55] <awallin> so siggen might have a 'ramp' or triangle-wave output? or you could integrate a square wave to get a triangle-wave
[06:42:33] <jthornton> stepgen will handle the acceleration
[06:42:54] <jthornton> I'll use stepgen in velocity mode
[06:44:00] <jthornton> so I guess I need to flipflop the velocity input
[06:44:07] <jthornton> somehow
[06:47:42] <awallin> if you only care about velocity, not exact position, then just a square-wave from siggen to stepgen velocity input
[06:49:06] <jthornton> well I do care about the end positions
[06:49:42] <awallin> I don't know how stepgen in velocity-mode is used when one also wants to position. a PID-loop around then stepgen?
[06:49:56] <mrsun> just measure the distance you want to move
[06:50:20] <mrsun> G1 X100
[06:50:22] <mrsun> G1 X0
[06:50:26] <mrsun> repeat
[06:50:38] <mrsun> and it will accelerate, move the distance, descelerate, then do the same thing in reverse :P
[06:50:40] <awallin> "hal only" was one requirement.. :)
[06:50:44] <jthornton> say I want to move 1000 steps I can read my position from stepgen.N.counts
[06:50:44] <mrsun> gah
[06:50:46] <mrsun> =)
[06:51:26] <jthornton> so when stepgen.N.counts = 1000 I want to negate the velocity command until stepgen.N.counts = 0
[06:52:10] <jthornton> anyone know how this works?
http://linuxcnc.org/docs/html/man/man9/flipflop.9.html
[06:54:00] <awallin> http://en.wikipedia.org/wiki/Flip-flop_%28electronics%29
[06:54:18] <awallin> I think it just transfers the input to the output on the rising clock-edge
[06:54:40] <jthornton> ok, I don't think that would work then
[06:55:03] <awallin> I would generate that 0-1000 triangle wave with siggen and use a stepgen in position mode.
[06:55:21] <awallin> possibly with some limiters in-between if velocity or acceleration limits are a problem
[06:55:36] <jthornton> let me see if I can figure that out
[07:11:23] <jthornton> triangle with my current settings goes from -1 to 1 then back to -1 is there a way to make it go 0-1 then back to 0
[07:14:35] <jthornton> ahh set offset and amplitude to 1/2 of desired positive
[07:16:20] <jthornton> doesn't seem to be a way to start and stop it
[07:16:39] <Jymmm> jthornton: It's okey to talk to yourself, but when you start answering yourself is another matter.
[07:16:57] <jthornton> doesn't matter to me
[07:17:20] <jthornton> I could go talk to little Ricky but she don't talk much
[07:17:31] <awallin> etiher siggen has a clock-input? or just an update function associated with it that should be run in some thread
[07:17:35] <Jymmm> jthornton: Ok, here try on this shiny new jacket
[07:18:41] <jthornton> only amplitude, frequency, and offset are in pins
[07:19:44] <awallin> I guess it is allways on then. You will have to gate the output if you don't want your motor continuously running
[07:21:02] <Jymmm> jthornton: 1000 steps, reverse, 1000 steps. is that what you want?
[07:22:14] <jthornton> I can't think of how to make sure I start at 0 with freggen
[07:22:20] <jthornton> Jymmm, yes
[07:22:47] <Jymmm> using a counter?
[07:23:38] <jthornton> using stepgen
[07:26:07] <Jymmm> ok, how do you loop in stepgen?
[07:29:49] <Jymmm> can stepgen.N.down be set, or is that a RO parameter?
[07:38:09] <Jymmm> If (steps % 1000){dir * -1; steps==1}
[07:41:27] <jthornton_> up and down are outputs for a up/down stepper driver
[07:42:10] <Jymmm> jthornton_: use modulus 1000 as the trigger
[07:42:42] <Jymmm> and * -1 to flip between post and neg
[07:42:44] <Jymmm> pos
[07:42:49] <jthornton_> trigger for what?
[07:43:14] <Jymmm> if (steo % 1000){blah}
[07:43:21] <jthornton_> I don't see a modulus component
[07:43:22] <Jymmm> % is modulus
[07:43:39] <Jymmm> it's a math operator
[07:43:47] <jthornton> in HAL?
[07:43:57] <Jymmm> + - / * % ^ etc
[07:44:08] <jthornton> in HAL?
[07:44:21] <Jymmm> In general, not sure what hal has
[07:44:38] <jthornton> I'm not doing this in general
[07:44:51] <jthornton> I only have HAL components to work with
[07:45:02] <jthornton> http://linuxcnc.org/docs/html/
[07:45:44] <jthornton> flipflop would do it I think if I can figure out how to trigger it
[07:46:18] * jthornton goes to walk the dog and ponder the problem
[07:50:20] <Jymmm> jthornton: classic ladder has it
[07:50:56] <Jymmm> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?ClassicLadderExamples
[07:51:59] <Jymmm> http://wiki.linuxcnc.org/cgi-bin/wiki.pl/wiki.pl?action=browse&diff=1&id=ClassicLadder
[07:53:33] <Jymmm> http://linuxcnc.org/docs/html/gcode/overview.html#sec:Binary-Operators
[07:57:17] <Loetmichel> mornin'
[08:01:11] <jthornton> Jymmm, I'm not using classicladder
[08:01:34] <jthornton> maybe I just need to write a component to do the job
[08:01:40] <Jymmm> Well, then tell whoever wrote hal to add a modulus operator
[08:06:25] <Loetmichel> sooo, milling the last 100 slot plates for PCs (company work at home)... if there's not another order ... ->
http://bambuser.com/v/2911623
[08:09:25] <Jymmm> your not done yet? eeeeeesh
[08:09:41] <Loetmichel> Jymmm: last 100 of 600
[08:10:51] <Jymmm> I still dont see the purpose for it to begin with. stickers would have been cheaper, easier, faster
[08:11:48] <Jymmm> and easier to read too
[08:12:22] <Jymmm> I hope they are paying $10/each for those.
[08:14:29] <MattyMatt> for 600 making a die to stamp them mighta been easier
[08:15:19] <Loetmichel> Jymmm: no
[08:15:53] <Loetmichel> i am lucky when i can have the time spend here as additional holidays
[08:16:05] <Loetmichel> ... the tools are payed by the company, though
[08:16:24] <MattyMatt> that's a win :) I need my time paying for
[08:16:43] <MattyMatt> "I just gotta cnc my lathe, here's what I need for it"
[08:17:10] <Jymmm> "1) A lathe"
[08:17:30] <Jymmm> "2) a shop for the lathe"
[08:17:31] <MattyMatt> that's almost true. mine barely counts
[08:18:17] <MattyMatt> I stripped the leadnut again last night, I tightened the gibs too tight because I was trying parting
[08:18:43] <MattyMatt> gotta fill it will molten delrin and tap it again
[08:19:26] <Jymmm> If you stripped it AGAIN, use something else other than delrin AGAIN
[08:19:44] <MattyMatt> oh it's lasted OK :) the gibs were very tight
[08:19:58] <Jymmm> k
[08:20:14] <MattyMatt> the original was bronze, and that was stripped when I got it
[08:20:25] <Jymmm> I swear, it jsut sucks finding a good headlamp
[08:20:53] <MattyMatt> eevblog dude does reviews. he caves in his spare time
[08:21:12] <Jymmm> They dont make one is the problem
[08:21:41] <MattyMatt> he was impressed with one
[08:21:59] <Jymmm> they are all impressive, jsut not good
[08:22:48] <Jymmm> BlackDiamond storm is the only IPX8 rated I've found. But it takes FOUR AAA's, WTF?!
[08:23:03] <Jymmm> NOBODY uses AA's (barely)
[08:23:22] <Jymmm> AA's are 2.5 times more than AAA
[08:23:36] <Jymmm> I dont want to carry around 16 AAA spares
[08:23:50] <MattyMatt> why noy 18650 LiIon?
[08:23:55] <Jymmm> There are SINGLE AA lights that output 200+ lumens
[08:24:02] <jdh> for how long?
[08:24:17] <Jymmm> not common enough in a pinch
[08:24:21] <Jymmm> 40-60m
[08:25:15] <Jymmm> I love the dimming feature of the storm, but the battery life sucks
[08:25:35] <Jymmm> and the 5mm leds suck more power than the high output led.
[08:27:24] <Jymmm> I want 1 or 2 AA, dimmable or 3 mode, IPX8. that really doens't seem too much to ask for.
[08:27:57] <MattyMatt> seen eevblog's mod of the priveton tec with brighter lED?
[08:28:20] <Jymmm> well I suspect they're one of the ultralight crowd
[08:28:22] <MattyMatt> ^princeton
[08:28:44] <Jymmm> no I haven't
[08:29:28] <MattyMatt> http://www.youtube.com/watch?v=l2vvsexdguQ
[08:31:23] <MattyMatt> ah 3x AAA
[08:32:10] <awallin> Loetmichel: wow is that video really live?
[08:33:52] <awallin> looks like pci-card brackets ?
[08:49:42] <Loetmichel> right
[08:49:54] <Loetmichel> awallin: it is
[08:50:07] <Loetmichel> and right, pci card brackets
[08:50:15] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=13438
[08:50:43] <Loetmichel> have to get 2 audio cables otu of the PC... and on the NIC was enough room
[08:50:46] <Loetmichel> ... times 600
[09:00:34] <Damianos> What is the cheapest way to acquire to a cnc machine? I have a linux box loaded and ready to go with EMC2 but I need a small CNC machine to get started. The only thing I've found that seems sturdy enough is DIYLILCNC
[09:01:43] <Loetmichel> Damianos: build one yourself?
[09:02:03] <Damianos> Yes, something DIY
[09:02:21] <Damianos> of course if there is something preassembled that is cheaper I'm all ears lol
[09:02:23] <Loetmichel> so DO it?
[09:02:32] <Loetmichel> not that i know of
[09:02:51] <Loetmichel> see what you have laying around, think a bit and start building ;-)
[09:03:41] <Damianos> ok, but I would do a lot better if I had some established plans to follow
[09:04:01] <MattyMatt> no plans demand you buy the parts they specify
[09:04:02] <Damianos> keep in mind that I'm a total newb at this
[09:04:07] <MattyMatt> improvise is cheaper
[09:04:22] <Jymmm> MattyMatt: Just FYI...
http://www.youtube.com/watch?v=0DqHicC6QL4&feature=relmfu
[09:04:23] <Damianos> not if I make many mistakes
[09:04:25] <Jymmm> AAA
[09:04:26] <MattyMatt> look at photos to steal ideas
[09:05:05] <MattyMatt> I stole my Y axis from solsylva, but I made it even cheaper by using drawer slides :)
[09:06:09] <r00t4rd3d> Damianos, microcarve
[09:06:29] <r00t4rd3d> http://www.microcarve.com/mcMV1.html
[09:07:09] <MattyMatt> http://makeyourbot.org/mantis9-1
[09:07:32] <Damianos> PERFECT!!! and they're in GA too! (I live right by the state line)
[09:08:41] <r00t4rd3d> that still needs stuff like motors and electronics
[09:08:49] <r00t4rd3d> but not much
[09:09:37] <r00t4rd3d> and microcarve, John is pretty active and helpful.
[09:09:38] <r00t4rd3d> http://www.cnczone.com/forums/diy-cnc_router_table_machines/109390-my_newest_desktop_machine.html
[09:09:41] <r00t4rd3d> he has a new version out
[09:09:47] <Jymmm> MattyMatt: If I could find a waterproof 18650 battery holder, could take that Fenix AAA head and wire the two together
[09:10:17] <Loetmichel> Damianos: most of the diy plans i have seen have many construction errors, too
[09:10:49] <Loetmichel> my little machine is completely selfbuild, the sith or seventh iteration and STILL i have some errors in it
[09:11:00] <MattyMatt> Jymmm you need a reprap :)
[09:11:03] <Loetmichel> that doesent prevent it from doing some buisiness work
[09:11:24] <MattyMatt> or just mill one
[09:11:28] <Jymmm> MattyMatt: Nah, a lathe
[09:11:50] <MattyMatt> yeah o ring is easier than making a square gasket
[09:12:35] <Damianos> well I've never built anything like this before. I'm a web developer with some design experience so not much hands on. I was hoping to do something as small and as cheap as possible so I can get a little practice. I figured later I can make something bigger and more substantial
[09:12:58] <Jymmm> MattyMatt: I could see doing a "T" shaper design.
[09:13:05] <Jymmm> or L
[09:13:06] <Loetmichel> MattyMatt: i have used simple household square rubber bands as gaskets with good results.
[09:13:33] <Loetmichel> but a o-ring would be more fitting if the gasket is opened a lot
[09:13:58] <MattyMatt> ah jymm you're jibe about the delrin has reminded me. there's an epoxy putty that sets hard into a low friction bearing surface. that's the stuff to pack my leadnut with
[09:14:00] <Jymmm> and I found a place that ONLY sells o-rings, nothing else.
[09:15:37] <Jymmm> MattyMatt: there ya go =) Make sure to scuff up the surface before applying
[09:16:25] <Jymmm> MattyMatt: Actually test on a scrap delrin, may not adhere too well
[09:16:46] <MattyMatt> on a scrap bronze. i'll clean out the delrin if I use it
[09:16:56] <Jymmm> =)
[09:17:48] <MattyMatt> I had to glue in the delrin pug, trying to tap a right hand thread into a lefthand threaded plug had the obvious result
[09:18:15] <Jymmm> a secure fastener =)
[09:19:22] <MattyMatt> it's a baby lathe. I only wanted a temporary M8 leadscrew to survive until I cnc'ed it (estimated 6 months 2 years ago)(
[09:19:49] <Jymmm> dont ya hate when that happens =)
[09:19:54] <MattyMatt> not worth buying an acme tap for
[09:20:05] <MattyMatt> still not
[09:20:06] <Jymmm> buy? make!
[09:20:28] <MattyMatt> no grinder or torch
[09:20:45] <Jymmm> can you afford $15 ?
[09:20:47] <MattyMatt> or spare left hand 16tpi stud
[09:21:18] <MattyMatt> it's an old British lathe, and everything non-metric costs twice as much here now
[09:22:00] <Jymmm> MattyMatt: buy/borrow a cheap $15 angle grinder...
http://www.embeddedtronics.com/acmetap.html
[09:22:33] <MattyMatt> hmmz I got an angle grinder. not very elegant for tool grinding tho
[09:22:47] <Jymmm> doesn't need to be
[09:23:14] <Jymmm> and some cutoff wheels
[09:23:21] <Jymmm> to make the groove
[09:23:24] <Jymmm> s
[09:23:58] <Jymmm> chuck the acme in a drill press and have at it with the grinder
[09:26:20] <MattyMatt> original leadscrew was 3/8x16. I guess if I got some 1/2"x16 I could restore lathe to almost original
[09:26:43] <MattyMatt> no need tho, it's a pig lathe and I'll probably make it metric cnc
[09:26:44] <Jymmm> MattyMatt: Just giving you options is all.
[09:27:08] <MattyMatt> lots of meat on the original bronze nut to tap larger
[09:27:19] <MattyMatt> not enough to change pitch tho
[09:27:21] <Jymmm> MattyMatt: might want to bookmark that acme tap, it comes up every few years
[09:27:58] <MattyMatt> yeah I've seen the trick
[09:29:03] <Jymmm> MattyMatt: WOW, this has some impressives specs.... 27hrs @ 3 lumen
http://www.rei.com/product/821450/fenix-mini-ld01-flashlight
[09:29:26] <MattyMatt> I've got 3m of twin start TR12x6P3 acme I'll be making a tap with any offcuts left over after mill & lathe get the rest
[09:29:33] <Jymmm> wonder if they have a AA version
[09:30:16] <MattyMatt> haven't decided whether to cnc this lathe or make a mini slantbed instead
[09:30:57] <MattyMatt> I decided to make the slatbed until I got the MDF home and realised how soft MDF is
[09:31:18] <MattyMatt> so not even a prototype in MDF thanks
[09:32:12] <MattyMatt> maybe just as a non-running model, but running it would jeopardise the good parts
[09:33:21] <MattyMatt> I'll use the MDF to make moulds for an epoxy-granite one instead :D
[09:34:19] <Jymmm> MattyMatt: Eeeeesh
http://www.fenixlight.com/UploadFiles/201111311168652.jpg
[09:34:42] <Jymmm> 1xAA
[09:35:39] <Jymmm> $42 USD amazon
[09:35:46] <MattyMatt> 4 of them, and duct tape, and you have what you originally asked for :)
[09:35:56] <Jymmm> why 4 ?
[09:36:03] <Jymmm> NSEW?
[09:36:23] <Jymmm> MattyMatt:
http://www.amazon.com/Fenix-Flashlight-Headband-18-22mm-Diameter/dp/B001NZO85O/ref=pd_bxgy_sg_text_b
[09:36:34] <MattyMatt> redundancy
[09:36:58] <MattyMatt> "I'm going up to 4 on you"
[09:37:22] <Jymmm> MattyMatt:
http://www.rei.com/product/830864/fenix-ld10-flashlight-mega-pack
[09:38:28] <MattyMatt> so you're happy you can get a good headlight now?
[09:38:59] <Jymmm> thats just fugly with the added battery case =) and $100
[09:39:09] <MattyMatt> seems expensive for a torch IMO. that's what I paid for my lathe :D
[09:39:27] <Jymmm> Yeah, at least a AA one.
[09:40:04] <jdh> I have a $1300 torch
[09:40:23] <Jymmm> jdh: the OTHEr kind of torch =)
[09:40:42] <jdh> it is an HID lamp
[09:40:54] <Jymmm> oh, eh
[09:41:14] <jdh> and another $800 one, and a bastardized $600 one
[09:41:32] <Jymmm> I'm VERY VERY HAPPY with my cree 18650 one
[09:41:49] <Jymmm> I'm able to drive on the freeway with it at night
[09:42:12] <jdh> I'm fairly sure the highway patrol wouldn't agree with taht.
[09:42:55] <Jymmm> some light is better than zero light
[09:43:57] <Jymmm> AWESOME light
http://www.dealextreme.com/p/ultrafire-th-t60-ha-ii-xm-lt60-5-mode-1200-lumen-white-led-flashlight-with-strap-1-x-18650-57007
[09:44:40] <jdh> http://www.divegearexpress.com/lights/rx10.shtml
[09:44:46] <jdh> ^^ more awesome light
[09:44:51] <Jymmm> I did get these batteries, TRUE 2400mah
http://www.dealextreme.com/p/trustfire-protected-18650-3-7v-true-2400mah-rechargeable-lithium-batteries-2-pack-20392
[09:45:12] <Jymmm> jdh: Nah, you just got ripped off dude!
[09:46:09] <MattyMatt> I got "3000" mA/h ones on ebay cheap. not tested yet tho
[09:46:33] <Jymmm> MattyMatt: "cheap" is the relative term =)
[09:46:52] <MattyMatt> not with my cashflow it ain't. if they work at all they're good
[09:47:06] <Jymmm> MattyMatt: I wasn't talking about price =)
[09:47:34] <MattyMatt> if they burn the lappy, well that was cheap on ebay too
[09:47:48] <Jymmm> heh
[09:48:26] <Jymmm> jdh: what is 900 lux in lumens (equiv)?
[09:49:45] <Jymmm> jdh: this says it's 84 lumen, is that right?
http://www.unitconversion.org/illumination/lux-to-lumens-per-square-foot-conversion.html
[09:50:41] <Jymmm> jdh: and do must items need that heavy duty enclosure when diving?
[09:50:48] <Jymmm> most
[09:50:48] <MattyMatt> surely they'd have to tell you the beam angle etc if they are quoting lux
[09:51:34] <Jymmm> MattyMatt: yeah, kinda why I asked. I was just joking when I said he got ripped off, but if it REALLY was 84 lumen, ouch
[09:52:44] <MattyMatt> I'm still impressed with the blue LED in the end of my ciggy lighters
[09:52:53] <Jymmm> heh
[09:53:10] <MattyMatt> it's exactly where you need it if you were just about to inspect sth with a naked flame
[09:54:15] <MattyMatt> that second you take to consider if your finger will burn is just long enough to remember it's there
[09:55:43] <jthornton> dang go for a hike in the woods and by the time you get back 4 spammers have registered and no one has offered a solution to my puzzle
[09:55:46] <Loetmichel> sooo, 100 pcs engraved, toolchange... 100 pcs "drilling" ;-)
[09:56:37] * jthornton searched the contributed components to make sure I'm not duplicating a wheel
[09:58:20] <jthornton> and some retard wants me to write him a program for his mac to duplicate a python script I wrote ROFLMAO
[10:02:45] <jthornton> http://pastebin.com/dYE5X2QZ
[10:05:35] <Jymmm> MattyMatt: The Fenix LD10 (R5) can use a AA or 14500 is seems...
http://www.youtube.com/watch?v=OaoUxnwIZWA
[10:05:59] <Jymmm> MattyMatt: not listed in the manual though.
[10:06:38] <pcw_home> jthornton: you gave me a lawnmower but it wont mow trees, please fix it
[10:07:29] <jthornton> lol
[10:15:40] <jdh> where did you get 900lux?
[10:17:50] <Jymmm> fiik, missed the 5 in there
[10:18:28] <jdh> I use that one in the ocean, my cave light is brighter and has a better beam angle for signaling
[10:18:57] <Jymmm> jdh: do you really need all that casing?
[10:19:04] <Jymmm> for diving
[10:19:10] <jdh> depends on the kind of diving.
[10:19:28] <Jymmm> example?
[10:19:45] <jdh> 250feet under water, 1 mile back in a cave.
[10:20:11] <Jymmm> due to the pressure? abuse proofness?
[10:20:41] <jdh> a light is "life support equipment"
[10:20:51] <Jymmm> thats why I'm asking.
[10:21:27] <jdh> due to pressure, and any potential abuse.
[10:21:37] <Jymmm> it's delrin, I'd thik aluminum would be better
[10:21:51] <jdh> that light is rated to 500ft, but the cord is not what I would have picked.
[10:22:01] <jdh> delrin is almost ideal for underwater use.
[10:22:24] <Jymmm> ok, I'll bite... why?
[10:22:29] <bpuk> jthornton: if you modify comp to work with s32 (comp_s32.comp) you should be able to use two of them. one to compare if stepgen.N.count = 1000, one to compare if stepgen.N.count = 0. comp_s32.0.out goes to flipflop.0.set, comp_s32.0.out goes to flipflop.0.reset
[10:22:55] <jdh> corrosion resistance, more forgiving mechanical aspects (dropping an aluminum light can ruin it.)
[10:22:55] <jthornton> bpuk, thanks
[10:23:24] <Jymmm> jdh: how so?
[10:23:35] <MattyMatt> delrin hates chlorine, so I'd think seawater would be bad for it
[10:23:53] <jdh> there is no real free chlorine in seawater afaik.
[10:23:54] <MattyMatt> seawater REALLY hates aluminium tho
[10:24:10] <Jymmm> MattyMatt: Well, titanium is best for salt water
[10:24:27] <jdh> aluminum deforms, sealing surfaces get defaced/bent
[10:24:51] <Jymmm> k, and the oxidation of saltwater
[10:25:01] <Jymmm> I forget about that one =)
[10:25:29] <Jymmm> HDPE s pretty good too
[10:25:43] <Jymmm> is
[10:27:04] <MattyMatt> PP is used for boats
[10:28:09] <Jymmm> I like PP, unless alcohol is involved
[10:28:48] <MattyMatt> jdh, it's all ions in solution, so the chlorine would attack the delrin locally afaics
[10:29:12] <jdh> but, it doesn't.
[10:29:54] <bpuk> http://pastebin.com/ZmkreZEu
[10:30:01] <bpuk> haven't tried it - but it should get you fairly close
[10:31:15] <bpuk> with s32 instead of s_32 - forgot it was comp for a second :P
[10:32:35] <bpuk> http://pastebin.com/PNsyZarc
[10:32:44] <bpuk> sorry for the link spam, that's the last one
[10:33:04] <jdh> comp links are much more on topic than dive lights!
[10:35:12] <bpuk> surely that depends on if the dive light housing was machined using linuxcnc?
[10:37:58] <jdh> they are.
[10:39:45] <Jymmm> pcw_home: Damn right jthornton should fix it so it mows trees!
http://www.youtube.com/watch?v=fPQakP3mYSA&feature=plcp
[10:42:10] <bpuk> at the risk of straying on topic again - is anyone familiar enough with G71 cycles on other controllers to answer a couple of questions?
[10:43:43] <Jymmm> bpuk: Not a risk anyone should be willing to consider.
[10:44:27] <bpuk> true, true, everyone needs a treemower
[10:45:05] <Jymmm> Blame jthornton, he's the one giving away treeless cutting mowers!
[10:45:30] <Tom_itx> where will all the bunnies live?
[10:46:00] <Jymmm> The BunnyHut made from all the lumber cut down with the tree cutting mower.
[10:46:22] <Tom_itx> you are making work out of this. leave the forest and let the bunnies live there
[10:46:54] <Jymmm> portable CNC lumber mill!
[10:47:19] <Jymmm> CNC pre-fabbed bunny huts!
[10:48:05] <Tom_itx> Jymmm for your disks, get a bar of 2.5" SS and a good cutoff blade
[10:48:10] <Tom_itx> have a good time with it
[10:48:40] <Tom_itx> predrill the end holes before you face them off
[10:48:56] <Tom_itx> no DIE required
[10:49:31] <Jymmm> Tom_itx: Heh, I think I'll have a rough time finding a blade that can create a 0.020" thick disc =)
[10:49:48] <MattyMatt> wire edm?
[10:49:52] <Jymmm> But, I could use the 2.5 rod to make a punch
[10:50:01] <Jymmm> MattyMatt: waterjet
[10:50:01] <bpuk> plunge sparker? ;)
[10:50:50] <Jymmm> who/where has 2.5 rod tool steel and the tube to fit it?
[10:51:08] <MattyMatt> i've been thinking if an edm slitting saw would work. spinning shim as the electrode
[10:52:10] <MattyMatt> seem like it should, if the disk is belt drive so the solenoid doesn't have to move the motor
[10:54:09] <Jymmm> Man, I only need one and only 4"
http://www.amazon.com/dp/B000H9TJOC/ref=biss_dp_t_asn
[10:55:26] <Jymmm> ah ha! I'll hit the metal scrap yard
[10:56:08] <Tom_itx> it doesn't need to be SS
[10:56:23] <Jymmm> thats O1 tool steel, not SS
[10:56:39] <Tom_itx> cut off 3" and sell the rest on ebay
[10:57:42] <Jymmm> for a punch, would it be better to have the end surfaced flat, or concaved?
[10:59:00] <Jymmm> or even just beveled?
[11:01:09] <Tom_itx> no way to sharpen beveled
[11:01:22] <Jymmm> ah
[11:03:40] <Jymmm> If I used an arbor press, do you think a puck 1" thick 2.5" diam would be enough?
[11:04:23] <Jymmm> I could bolt it to the arbor's ram then.
[11:10:06] <jthornton> wd5jwy, you around?
[11:17:05] <jthornton> http://linuxcnc.org/index.php/english/component/kunena/?func=view&catid=48&id=23317&limit=6&start=18#23383
[11:19:37] <bpuk> that's a more elegant solution :)
[11:20:09] <jthornton> thanks
[11:20:56] <jthornton> I have one more issue with my component if I change the velocity in it doesn't update till it reaches max or min limit
[11:21:29] <jthornton> http://pastebin.com/UrvdtiWz
[11:25:13] <bpuk> http://pastebin.com/LVU6c6GX
[11:25:41] <jthornton> excellent, thanks
[11:33:33] <jthornton> http://linuxcnc.org/index.php/english/component/kunena/?func=view&catid=48&id=23317&limit=6&start=18#23384
[11:42:03] <wd5jwy> Yes, John I'm here
[11:47:32] <wd5jwy> John, I'm reading the updates you did on the thread - spooler.comp
[11:47:56] <wd5jwy> but, I think you are assuming I'm using a lead-screw - but I'm not
[11:48:05] <wd5jwy> unfortunately
[11:48:09] <wd5jwy> I wish it were that easy
[11:49:23] <wd5jwy> the wire feed traverse mechanism has to be driven by a physical CAM
[11:50:02] <wd5jwy> It's possible that it could be done with a lead screw, but the traverse action has to be non-linear
[11:51:04] <wd5jwy> the CAM is actually cartiod-shaped (somewhat like a heart shape) - the traverse shaft is driven off the side of that CAM as it rotates
[11:52:01] <Tom_itx> like the cam on a compound bow?
[11:52:23] <wd5jwy> I dunno - never seen a cam on a compound bow
[11:52:34] <wd5jwy> let me see if I can find some pics on the net
[11:53:15] <Tom_itx> http://en.wikipedia.org/wiki/File:Compound_Bow_Pulley.jpg
[11:54:00] <wd5jwy> OK, yes, but instead of oval-shaped - these CAM would be heart shaped
[11:54:21] <wd5jwy> the traverse shaft is driven off the edge of the CAM as it rotates
[11:55:07] <wd5jwy> the wire feed mechanism is mounted on the traverse shaft and moves left-to-right
[11:55:38] <wd5jwy> this forces the wire onto a coil form in a special way to achieve what they call a "universal winding"
[11:56:19] <wd5jwy> http://www.youtube.com/watch?v=Lt0uHyzr4C0
[11:56:45] <wd5jwy> that's a video of a home-built hand-cranked coil winder that does what I'm trying to do with stepper motors
[11:56:53] <wd5jwy> except
[11:57:12] <wd5jwy> the cams used on that machine are oval instead of heart-shaped
[11:57:31] <wd5jwy> oval cams get close to a true "universal winding" but not perfect
[11:57:58] <wd5jwy> I'll machine the heart-shaped cams on my CNC once I get the physical coil winder built
[12:00:15] <wd5jwy> with oval cams, the wire spends too much time being wound on the outside edge of the coil
[12:00:31] <FinboySlick> wd5jwy: That's great music to wake up to.
[12:00:47] <Jymmm> MattyMatt: FOUND IT!!!
http://www.fenixlight.com/viewnproduct.asp?id=92 - I like his accent...
http://www.youtube.com/watch?v=WuoPz7hm7Co
[12:01:07] <wd5jwy> with heart-shaped cams, the peak at the bottom and dimple at the top of the cam forces the traverse to reverse direction quicker and results in a better coil
[12:03:39] <Tom_itx> why are you winding your own chokes?
[12:04:09] <pcw_home> sounds like a triangle wave would be close
[12:04:40] <pcw_home> (instant reverse)
[12:04:41] <Jymmm> wd5jwy: building new or restoring?
[12:05:32] <wd5jwy> There's no one making these particular kind of chokes any longer - I would be winding them for myself and perhaps sell a few
[12:05:43] <wd5jwy> they are used in old radios - can't buy them anymore
[12:05:54] <Jymmm> heh, restoring =)
[12:06:02] <wd5jwy> I'm trying to build my own coil-winding machine - not restoring
[12:06:15] <wd5jwy> the old machines that did this kind of coil are very, very hard to find
[12:06:21] <wd5jwy> and expensive when you do find them
[12:06:39] <Jymmm> wd5jwy: for the purpose of restoring old rigs =)
[12:06:46] <wd5jwy> YES
[12:06:58] <wd5jwy> old transmitters and receivers
[12:07:14] <pcw_home> so JTs code will do triangle so should be optimum (at least if instant reversal is optimum)
[12:07:21] <wd5jwy> you can buy new machines that will wind this type of coil, but they run 40 to 50K
[12:07:21] <Jymmm> I gave up tube LONG ago
[12:07:36] <Jymmm> unless a linear =)
[12:08:01] <wd5jwy> Yes, John's code would do it, but I would rather use a CAM instead of a screw
[12:08:19] <wd5jwy> If it doesn't glow in the dark, it's not a real radio
[12:08:24] <wd5jwy> :-)
[12:08:30] <Jymmm> Wait, I take that back... I did have a mobile motorola tube at one time
[12:08:43] <pcw_home> That makes it rather awkward to change profiles however
[12:08:57] <wd5jwy> agreed
[12:09:15] <wd5jwy> but, I would most likely only be winding one or two type of coils
[12:09:34] <wd5jwy> but, it's fairly easy to just CNC a new CAM if needed
[12:09:46] <Jymmm> wd5jwy: My TM-D700a glows in the dark... via LED's in the remote head =)
[12:09:55] <wd5jwy> there ya go
[12:10:13] <wd5jwy> I dunno, tubes are just fun
[12:10:28] <wd5jwy> I have solid-state radios too
[12:10:47] <wd5jwy> tubes obviously can't compete with modern gear
[12:10:48] <Jymmm> wd5jwy: Oh no, dont get me wrong, they put out a PERFECT signal, just big, power hungry, etc
[12:11:30] <Jymmm> not feature they can't, but singal wise, hell yes.
[12:11:32] <wd5jwy> but it's just amazing to me how well the old engineers did with such limited tech
[12:11:51] <jdh> the same will be said in 50 years.
[12:11:59] <wd5jwy> I'm sure
[12:12:16] <wd5jwy> the real tech advances are all in cellular these days
[12:12:16] <jdh> not every field can have a john moses browning
[12:12:34] <Jymmm> wd5jwy: But, I'm still waitig for th FCC to make vhf+ all digital to conserve bandwidth
[12:13:02] <wd5jwy> like that's really needed huh
[12:13:11] <wd5jwy> the repeaters around here are all barren
[12:13:34] <Jymmm> wd5jwy: Look at 220, nobody has the rigs (barely)
[12:13:50] <wd5jwy> I know - or 440 for that matter
[12:14:12] <Jymmm> 900, 1200
[12:14:16] <wd5jwy> there's some activity here on 440, but mostly just the AM, FM radio engineers
[12:14:41] <Jymmm> 900 I think we're at risk of losing, since there are no ham rigs on the makret, just modded commercial ones
[12:15:15] <wd5jwy> Ham radio is just having a difficult time competing with the Interwebs
[12:16:02] <Jymmm> Eh, first disaster that takes out a major hub for weeks and it won't be =)
[12:16:16] <wd5jwy> agreed
[12:16:32] <Jymmm> http://users.innercite.com/kj6ko/page8.html
[12:16:42] <wd5jwy> most of the storm-spotters around here are hams
[12:17:05] <wd5jwy> cool site
[12:17:34] <Jymmm> What I love is that everyone thinks/feels they'll be able to rely on their cellphone if SHTF =)
[12:17:55] <wd5jwy> that's true - that's the first thing that goes down
[12:18:07] <wd5jwy> I work in cellular - AT&T System Engineer
[12:18:11] <wd5jwy> the system just can
[12:18:18] <wd5jwy> can't handle the load
[12:18:33] <Jymmm> wd5jwy: OH, then you must know about WPS too
[12:18:45] <wd5jwy> WPS?
[12:18:48] <Jymmm> http://wps.ncs.gov/
[12:19:00] <wd5jwy> ah, yes
[12:19:15] <wd5jwy> and CMAS - wireless emergency service
[12:19:24] <wd5jwy> we're just rolling that out
[12:19:30] <Jymmm> we (civilians) are just so screwed =)
[12:20:01] <wd5jwy> if you want to be able to communicate in an emergency - get a ham license
[12:20:30] <Jymmm> and $5000+ in gear, antenna, feeline, etc =)
[12:20:36] <Jymmm> feedline
[12:20:39] <wd5jwy> true
[12:21:00] <wd5jwy> but, you could just get a technician license and buy a handheld
[12:21:11] <wd5jwy> at least you would be able to talk to someone if you had to
[12:21:51] <Jymmm> Been there, done that 12+ years ago. Issue is I need to get my GENERAL to get HF priv. 10m for simplex distance.
[12:22:18] <wd5jwy> Yeah, 10 meters is an overlooked band - in my opinion
[12:22:22] <wd5jwy> and 6 as well
[12:23:08] <Jymmm> well, at least with HF you can get a few hundred miles. VHF+ 50 or so if you are lucky.
[12:23:29] <Jymmm> I'm thinkin mobile here
[12:24:27] <FinboySlick> Jymmm: when shtf, fuel goes away not much longer after cell.
[12:24:32] <wd5jwy> I've always wondered if ham radio could provide a sort-of backup Internet system if we ever had a really major nation-wide emergency that wiped out the web somehow
[12:24:43] <Jymmm> I had a friend that loved talking form Calif to .au on 7w =)
[12:24:46] <r00t4rd3d> lol
[12:25:04] <Jymmm> wd5jwy: HF version of APRS?
[12:25:20] <Jymmm> wd5jwy: maybe at 2400 baud ;)
[12:25:21] <wd5jwy> yeah, sort of
[12:25:45] <wd5jwy> yes, but still you could get information out - sort of like the old billboard systems
[12:26:13] <Jymmm> Yep, works for me. FIDONET!
[12:26:15] <wd5jwy> obviously, you couldn't stream data
[12:26:28] <Jymmm> store and forward works
[12:26:32] <wd5jwy> but, it would be a way to get information out if needed
[12:26:41] <wd5jwy> if there was no other way
[12:27:56] <Jymmm> wd5jwy: I did a SF to LA run once, gave all my friends a url to findu.com so they could see where I was at and how far out. It worked great
[12:28:20] <wd5jwy> yeah, that's pretty cool - and scary at the same time
[12:28:47] <Jymmm> yeah, the tracking part is scarry. but not that much more than fox hunting
[12:28:59] <wd5jwy> yep
[12:29:37] <Jymmm> Unles you are evil and park under the local TV broadcast towers ;)
[12:31:30] <Jymmm> (and offset about 10Khz being the 'fox')
[12:34:12] <wd5jwy> So, do any of you guys have a good understanding of the Stepgen in Hal?
[12:34:23] <wd5jwy> and how to configure it?
[12:35:39] <Jymmm> I'd say that's a job for Super jthornton!!!
[12:36:34] <wd5jwy> Yeah, it probably is - but I've bothered him so much over the last week, I hate to ping him again
[12:36:57] <wd5jwy> I think I'm wearing out my welcome
[12:37:46] <Jymmm> nah
[12:37:50] <wd5jwy> I'm trying to setup the stepgen in "velocity" mode instead of "step and position" like it's normally used
[12:38:15] <wd5jwy> and the docs are a little confusing on how to do that
[12:38:48] <wd5jwy> in other words, instead of using gcode to tell the stepgen to "go to this position"
[12:39:06] <wd5jwy> I want to just turn on the stepgen and let it run at a set speed
[12:39:18] <wd5jwy> until I tell it to stop (disable)
[12:39:18] <pcw_home> man stepgen
[12:39:24] <skunkworks__> wd5jwy: doable..
[12:39:30] <wd5jwy> yeah, i did that
[12:39:30] <skunkworks__> ^
[12:40:09] <wd5jwy> I hope I'm just doing something wrong in my .hal file - and I'm sure I am
[12:40:15] <IchGuckLive> Hi all
[12:40:17] <wd5jwy> just can't figure out what yet
[12:40:19] <skunkworks__> do you want a gui to control it?
[12:40:26] <wd5jwy> yes
[12:40:30] <wd5jwy> I have a gui built
[12:40:33] <wd5jwy> that's the easy part
[12:40:42] <wd5jwy> I have a Hal Toggle button
[12:40:56] <IchGuckLive> then run a python
[12:41:04] <wd5jwy> I tie that signal to the stepgen.enable pin
[12:41:13] <wd5jwy> I've set up the threads
[12:41:21] <wd5jwy> I've set the stepgen scale
[12:41:50] <wd5jwy> I've tied the stepgen.out to the parport pins, but I get nothing
[12:41:53] <IchGuckLive> wd5jwy: what is the main goal for this button
[12:42:08] <wd5jwy> to enable the stepgen to start making pulses
[12:42:48] <skunkworks__> you also need to set the stepgen.N.velocity-cmd
[12:42:51] <skunkworks__> also
[12:42:52] <wd5jwy> I capture the state of the toggle button in a python file and it does enable the stepgen enable pin
[12:43:06] <wd5jwy> yes, I'm trying to do that with a spinbutton
[12:43:41] <wd5jwy> but the spinbutton value doesn't seem to get passed for some reason
[12:43:53] <skunkworks__> ok - can you run halcmd and look at the pins to see what the pins are doing?
[12:43:59] <skunkworks__> ah
[12:44:02] <wd5jwy> yes
[12:44:22] <wd5jwy> I see that I'm enabling the stepgen enable pin with the toggle button - that works
[12:44:50] <wd5jwy> it's the spinbutton value that I'm trying to pass to the stepgen.N.velocity-cmd - that doesn't seem to be working
[12:45:31] <bpuk> have you tried setting the velocity command manually - i.e. setp stepgen.0.velocity-cmd 100
[12:45:39] <wd5jwy> ah
[12:45:41] <wd5jwy> no
[12:45:42] <wd5jwy> not yet
[12:45:44] <skunkworks__> can you set it manually in halcmd?
[12:45:48] <skunkworks__> heh
[12:45:51] <wd5jwy> suppose that would be easy to try
[12:45:52] <skunkworks__> I type slow
[12:45:57] <bpuk> heh
[12:46:17] <bpuk> it at least tells you if everything else is setup correctly - then you can work on getting the spinbutton working
[12:46:43] <wd5jwy> so if I did that - shouldn't I be able to see that state of the pin change in the "show pin" hal command?
[12:46:51] <wd5jwy> or no
[12:47:48] <bpuk> I'd go with yes, but haven't used halcmd all that much
[12:47:53] <wd5jwy> I also might be routing signal wrong in my hal file as far as getting the signal out the parport, but I can work on that later
[12:48:06] <wd5jwy> let me try setting the velocity-cmd manually
[12:48:45] <wd5jwy> here's the "net" command in the hal file
[12:48:46] <wd5jwy> net p-scale stepgen.0.velocity-cmd <= gladevcp.hal_spinbutton1-f
[12:49:30] <wd5jwy> currently when I look at the value of that pin with "show pin" it is equal to zero
[12:49:37] <wd5jwy> so, let's see what happens
[12:49:56] <skunkworks__> you will need to unhook the glade.hal_spinwhatever pin first
[12:50:17] <skunkworks__> then sets p-scale 100 or whatever
[12:50:51] <wd5jwy> would I do that in the python file or the hal file?
[12:51:02] <wd5jwy> on in glade?
[12:51:04] <skunkworks__> you can do it in halcmd
[12:51:48] <skunkworks__> uunlinkp
[12:51:54] <skunkworks__> -u
[12:52:11] <wd5jwy> ok
[12:53:35] <jthornton> wd5jwy, the example is using stepgen in velocity mode
[12:53:36] <wd5jwy> ok
[12:53:54] <jthornton> I just tossed the spooler.comp in there for fun
[12:54:05] <wd5jwy> yeah, I appreciate that work
[12:54:28] <jthornton> but a variable velocity input would work too based on position and easier than a cam
[12:55:11] <wd5jwy> aren't we already trying to use a velocity input?
[12:55:23] <wd5jwy> net p-scale stepgen.0.velocity-cmd <= gladevcp.hal_spinbutton1-f
[12:56:07] <jthornton> yes, what I meant was if you varied the velocity based on position you could achieve the same motion as the cam
[12:56:17] <wd5jwy> for CNC purposes, isn't the stepgen set up in position mode instead of velocity?
[12:56:26] <wd5jwy> ah
[12:56:27] <jthornton> normally yes
[12:56:45] <wd5jwy> I agree, but the problem is that I need non-linear motion
[12:56:59] <wd5jwy> we're talking sine and cosine functions for screw positioning
[12:57:05] <wd5jwy> I'm not that good at trig
[12:57:32] <wd5jwy> if it was linear, it would be easier to use a lead screw
[12:57:40] <jthornton> me neither but siggen does put out sine and cosine
[12:58:00] <wd5jwy> yes, I know, but I don't know how to develop the equations to get the right motion
[12:58:06] <wd5jwy> I'm stupid
[12:58:23] <wd5jwy> I know it could be done and it is done in commercial coil winders
[12:58:47] <wd5jwy> but, I can cut heart-shaped cams all day long on my cnc
[12:58:49] <wd5jwy> much easier
[12:58:57] <jthornton> go with easy
[12:59:20] <Jymmm> that's what she said
[12:59:31] <wd5jwy> that's what I told my wife - please marry me, I'm stupid
[12:59:36] <wd5jwy> she said no the first time
[12:59:39] <wd5jwy> smart woman
[13:00:02] <Jymmm> first (hundred || thousand) time ?
[13:00:20] <wd5jwy> that and please just "go away"
[13:00:26] <Jymmm> she was just playing hard to get is all ;)
[13:00:28] <wd5jwy> but I didn't
[13:00:34] <wd5jwy> anyway
[13:00:39] <wd5jwy> so, stepgen
[13:00:56] <wd5jwy> I'm gonna try a few things - see if I can just get the motors spinning
[13:01:18] <wd5jwy> yeah, she was hard to get
[13:01:25] <wd5jwy> still don't know why she married me
[13:01:30] <jthornton> you should be able to see how I connected the toggle button to the stepgen enable and duplicate that
[13:01:47] <jthornton> then work up from there
[13:01:47] <wd5jwy> yes, the toggle button works
[13:01:57] <Jymmm> wd5jwy: jepler or cradek when around might be able to give you some pointers on the trig/formulas
[13:02:01] <wd5jwy> I can see the enable pin changing when I hit the button
[13:02:19] <wd5jwy> what's not working is getting the value from the spinbutton into the stepgen
[13:03:25] <wd5jwy> so <skunkworks> suggested trying to set the stepgen.velocity manually to see if I can make it change
[13:03:29] <jthornton> open up my glade project and look for the items in bold for the spinbutton also look for the adjustment1
[13:03:50] <jthornton> you have to have an adjustment or the spinbutton does nothing
[13:04:19] <wd5jwy> oh, ok - so maybe that was the problem
[13:04:26] <jthornton> if you run my sample you can see it change
[13:04:37] <wd5jwy> ok
[13:05:09] <jthornton> in any case I'm just heading out, so update the forum if you run into a snag
[13:05:23] <wd5jwy> ok, will do - thanks again
[13:05:29] <jthornton> welcome
[13:06:59] <Jymmm> wd5jwy: It was the your pocket-protector, slide-rule, and love of vacuums tubes that always attract the hawt chicks ;)
[13:07:08] <pcw_home> Not sure there is any trig involved, if you want linear motion and instant reversal a triangle is correct (and what a cardioid cam approximates)
[13:08:47] <wd5jwy> @ <Jymmm> LOL
[13:09:02] <Jymmm> =)
[13:10:02] <wd5jwy> @<pcw_home> The problem is that I'm trying to achieve non-linear motion - a Cartiod or kind of Heart-shaped pattern
[13:10:22] <wd5jwy> it's needed due to the way the wire needs to be wound on the coil
[13:10:43] <wd5jwy> triangle approaches that but not close enough
[13:11:05] <wd5jwy> the wire size I'm using is close to .0005 in diameter
[13:11:21] <wd5jwy> so I need to get close to .00025 tolerance
[13:11:41] <wd5jwy> which is difficult - and may be impossible with home workshop tools
[13:11:58] <wd5jwy> I know it could be done better with the lead-screw drive, I just can't do the math
[13:12:24] <wd5jwy> but I can cut CAMs on my CNC that will get close enough
[13:14:17] <wd5jwy> here's a good pic
[13:14:18] <wd5jwy> http://www.google.com/imgres?imgurl=http://bp1.blogger.com/_gQvjhDmJj6o/R2T2YLLXmbI/AAAAAAAAALI/ieQ91OrZFng/s400/heart%2Bcam.gif&imgrefurl=http://kh-designandtechnology.blogspot.com/&h=150&w=90&sz=3&tbnid=dFFQRlgsuC3FFM:&tbnh=90&tbnw=54&prev=/search%3Fq%3Dheart-shaped%2Bcams%26tbm%3Disch%26tbo%3Du&zoom=1&q=heart-shaped+cams&usg=__NFKGiqGcFEnMgU32E1eDe63_koA=&docid=xS_7oGfkf_MnmM&sa=X&ei=utgvUMDPKK
[13:14:20] <wd5jwy> nq2AXPy4CQCQ&ved=0CFYQ9QEwAg&dur=15
[13:14:33] <wd5jwy> oops
[13:14:40] <wd5jwy> let me get a shorter url
[13:15:22] <wd5jwy> http://tinyurl.com/8gul3ed
[13:18:40] <pcw_home> Yes but thats just a linear motion cam (= triangle)
[13:19:02] <wd5jwy> yes, it's close, but not quite
[13:19:54] <wd5jwy> the dimple at the top and the peak at the bottom of the cam is critical
[13:20:30] <wd5jwy> it forces the traverse mechanism to change directions quickly but smoothly
[13:20:33] <pcw_home> The books ive seen on col winding say linear cams are used for universal coils
[13:20:40] <wd5jwy> not instantaneously
[13:21:01] <pcw_home> so add an accel limit to the stepgen
[13:21:15] <wd5jwy> I suppose you could do that
[13:21:42] <wd5jwy> that would approach the heart-shaped traverse reversal that I need
[13:21:53] <wd5jwy> interesting thought
[13:22:04] <wd5jwy> and then I wouldn't need to produce cams
[13:22:40] <archivist> I would never create cams with the power of linuxcnc under the hood
[13:22:50] <wd5jwy> yeah, that's true
[13:23:29] <archivist> are you trying to create a wave winder
[13:23:38] <wd5jwy> yes, sort of
[13:24:06] <wd5jwy> a universal-wound coil has certain electrical properties that are needed in HF circuits
[13:24:24] <wd5jwy> that's why they came up with the idea almost 100 years ago
[13:24:35] <wd5jwy> it lowers the self-capacitance of the coil
[13:25:01] <wd5jwy> illuminating self resonance
[13:25:02] <archivist> btw the coil winder out in the garden has a leadscrew and just knocks the reversal mech
[13:25:51] <archivist> wave winder was also for self support and high voltage
[13:25:57] <wd5jwy> true
[13:26:39] <wd5jwy> and the problem with trying to reverse directions quickly is that the wire tends to get pulled across the coil instead of lying down next to the adjacent turn
[13:26:44] <bpuk> the coil winders I use at work (which are winding much larger diameter stuff, typically 3-10 mm) use a friction drive on a smooth rod - which confused the heck out of me at first
[13:27:36] <bpuk> the internal cam on the changeover lever gives the same smooth motion at changeover as a heart shaped cam. It is wierd to see a traverse without a screw though :P
[13:27:37] <archivist> wave winding is a bit special
[13:28:01] <wd5jwy> it is - and damn hard to do correctly
[13:28:15] <wd5jwy> almost impossible with enamel covered wire
[13:28:23] <wd5jwy> requires some type of serving
[13:28:26] <wd5jwy> in the old days
[13:28:29] <wd5jwy> they used cotton
[13:28:36] <wd5jwy> now they use a form of nylon
[13:28:43] <wd5jwy> and that wires is asspensive
[13:29:33] <Jymmm> wd5jwy: Maybe you can get the source code and incorporate that into linuxcnc
http://coil32.narod.ru/index-en.html
[13:30:17] <wd5jwy> yeah, if I can ever get it working
[13:30:40] <archivist> http://www.youtube.com/watch?v=aoIr4VT8d80
[13:30:52] <wd5jwy> I suppose to use a lead-screw traverse drive, I would need a high turns-per-inch lead-screw
[13:31:01] <wd5jwy> and micro-step the motor
[13:32:55] <wd5jwy> Some of the old coil winders like that had as many as 150 gear sets and numerous cams
[13:33:26] <mrsun> http://www.youtube.com/watch?v=g7Sm8XAeUGk&feature=related
[13:33:31] <mrsun> haha i like the cam action counter =)
[13:33:34] <wd5jwy> it's amazing what they were able to do with those
[13:35:30] <archivist> http://info.ee.surrey.ac.uk/Workshop/advice/coils/winders/wave/winder2.jpg
[13:36:31] <Jymmm> mrsun: Heh, I guess whatever works =)
[13:36:39] <archivist> all the gearing can be done in hal just needs "some" setup
[13:40:04] <archivist> I have a handle that I am about to use on the hobbing machine that may be off an AVO wave winder, and have some of the gears
[13:43:48] <pcw_home> http://www.scribd.com/doc/45849334/Coil-Winding
[13:45:34] <wd5jwy> yes, that book discusses the heart-shaped cams and the winding ratios required for universal coils
[13:48:23] <pcw_home> but of course cams and gears are obsolete if you have a motion controller
[13:48:31] <Jymmm> login to dl, argh
[13:49:22] <wd5jwy> My original intent was to use the motion controller to achieve the ratios I needed
[13:49:24] <wd5jwy> but
[13:49:45] <wd5jwy> you guys have convinced me it's better to forget about trying to use cams
[13:49:53] <archivist> :)
[13:50:25] <wd5jwy> by using a lead-screw instead - there's really no limit to what you could do
[13:50:59] <wd5jwy> I think I was worrying too much about the math involved
[13:51:14] <pcw_home> yes any kind of spool wind is possible
[13:51:15] <wd5jwy> easier to adjust code instead of real-world cams
[13:51:20] <archivist> as long as your resolution/accuracy is smaller than your smallest wire size
[13:51:31] <wd5jwy> that's the problem
[13:51:44] <wd5jwy> I'll certainly have to micro-step the traverse motor
[13:51:45] <archivist> not a problem really
[13:51:49] <wd5jwy> but that shouldn't be a problem
[13:52:03] <archivist> you can gear the traverse too
[13:52:15] <wd5jwy> yes, I thought of that - timing belt maybe
[13:52:45] <bpuk-home> use a HTD timing belt to downgear a bit
[13:52:48] <pcw_home> My first real job was winding transformers. The accuracy is not really an issue
[13:52:59] <wd5jwy> I'm going to use 80 20 strut for the machine platform
[13:53:09] <wd5jwy> so I can move things around easily
[13:54:34] <pcw_home> not really an issue because level wind tends to self align and open winds like universal dont care
[13:54:58] <archivist> A friend locally worked at a transformer/coil winding co, hence the reason I have a winder
[13:56:09] <pcw_home> The ones I used had a mechanical counter with a motor stop at 0
[13:57:28] <Jymmm> pcw_home: By motor, you mean 2 or 4 legged?
[13:58:20] <archivist> I know linuxcnc has been used for a couple of winding machines but I dont remember a wave machine done yet
[13:58:28] <pcw_home> I hated doing heavy wire since we didn't have winders that could guide it and you had to guide it by hand and end up with bruised thumbs
[13:58:30] <pcw_home> No jymmm these were not horse powered winders, I'm not quite that old
[13:59:19] <Jymmm> pcw_home: I wans;t implying age, as much as manual crack or animal crank
[13:59:50] <Jymmm> lke the video archivist linked to
[14:00:02] <Jymmm> I thought I saw a crank handle on the side of it
[14:00:11] <pcw_home> No, these were electrical motor driven and very fast for fine wire
[14:00:15] <archivist> the hand crank is only for setting up
[14:00:20] <Jymmm> archivist: ah
[14:01:21] <archivist> later machines would have foot peddle speed control
[14:01:21] <Jymmm> If someone figures out how to auto thread a spool, wind it, and thread the next one automatically, please let me know =)
[14:01:52] <archivist> there are videos on youtube doing just that
[14:02:14] <pcw_home> They must do that for little inductors etc
[14:02:33] <archivist> cotton reels etc
[14:02:43] <Jymmm> I haven't found any videos of it yet.
[14:03:00] <Jymmm> these are hanks, not so much spools
[14:04:10] <archivist> there is one winding about 20 reels per cycle, clever use of a sticky bar to keep the ends between reels iirc
[14:04:36] <Jymmm> Yeah, I linked that a ways back, but it's on a cardboard card spool
[14:04:58] <Jymmm> I'm talking like embroidery floss
[14:04:59] <archivist> so adapt
[14:05:39] <Jymmm> I need to make these essencially
http://cdn.dickblick.com/items/631/00/63100-1009-3ww-l.jpg
[14:05:52] <Jymmm> but can't find a video of them being made.
[14:07:36] <archivist> you dont need a video, just think of granny holding out two hands
[14:07:55] <skunkworks__> wd5jwy: don't rely on microstepping for accuracy
[14:08:19] <wd5jwy> yeah, I was wondering about that
[14:08:38] <wd5jwy> probably have to use gears and timing belt
[14:08:40] <Jymmm> Right, but how do you cut to finish it, eject it (so to speak), thread a new start, and repeat, all hands off?
[14:09:00] <skunkworks__> 1/2 steps is as good as you get
[14:09:09] <archivist> jymm now go back to the other video, merge
[14:09:24] <wd5jwy> ok, good to know - I'm new to stepper motors...........and Linux too
[14:09:40] <Jymmm> archivist: they use adhesive on the cards, there will be no cards and no adhesive.
[14:09:59] <archivist> Jymmm, add small grippers
[14:10:20] <skunkworks__> micostepping just makes things smoother
[14:11:09] <Jymmm> archivist: Sure, but how does the start of the thread stay on the "arms" till it has a chance to have a few winds ?
[14:11:32] <archivist> should I repeat the small gripper part
[14:12:06] <Jymmm> archivist: if the amrs are spinning, hows is the gripper going to follow them?
[14:12:20] <archivist> its ON the arm!
[14:13:11] <Jymmm> ok, then how do you reseat the next coil in the grippers?
[14:14:42] <Jymmm> when you finish a hank, it's ejected, pulling the thread out of the grippers. So how is the beginning of the NEXT coil get threaded in the grippers?
[14:15:25] <Jymmm> err, cut and ejected
[14:20:12] <archivist> http://www.youtube.com/watch?v=RfnxYTXGkyU&feature=related
[14:24:02] <archivist> so simple even ladies can use it
[14:25:55] <Jymmm> Not even close. thats all MANUAL hands on of securing and removal of the hanks.
[14:27:08] <Jymmm> This is close, but uses the spool to secure the end of the winding and the beginning of the next one.
http://www.youtube.com/watch?v=6rFzbTUB_g4&feature=player_detailpage#t=93s
[14:30:42] <jmkasunich> any suggestions for recovering from a locked-up Axis without losing home position?
[14:31:25] <Jymmm> is it just the GUI thats locked up?
[14:32:01] <jmkasunich> I believe so
[14:32:19] <jmkasunich> not refreshing screen or responding to keypresses
[14:32:23] <jmkasunich> other windows work
[14:32:55] <Jymmm> can you query emc in a shell window?
[14:33:07] <jmkasunich> I guess I don't have to - it came back!
[14:33:21] <Jymmm> ie, will it respond to remote commands basically.
[14:33:38] <jmkasunich> after being locked up for at least several minutes, when I walked over there to try something else it was awake again
[14:33:55] <jmkasunich> fscking GUIs :)
[14:34:13] <Jymmm> KEYSTICK FTW!
[14:34:45] <jmkasunich> yeah!
[14:35:28] * jmkasunich goes back to making chips, not to be seen for another year
[14:35:48] <Jymmm> eh, only 4 months.
[14:44:31] <Jymmm> That's a big ass sewing machine.... the material moves 4-6 heads
http://www.youtube.com/watch?v=6KI5Hnv7p5I
[14:49:00] <jmkasunich> more like 13 or 14 heads
[14:57:55] <DJ9DJ> namd
[15:11:29] <skunkworks__> hit and run
[15:32:54] <FinboySlick> Anyone has recommendations for a tiny Atom based board with just USB and LAN? I don't need anything else.
[15:34:58] <Jymmm> does it have to be atom?
[15:35:29] <FinboySlick> The ALIX 3D is perfect... But my USB hardware hates the OHCI controller.
[15:35:44] <Jymmm> i dont knwo what that is
[15:35:44] <FinboySlick> So I need intel UHCI apparently.
[15:36:59] <Jymmm> what are you trying to do? usb NAS ?
[15:38:04] <Jymmm> $100 ok?
[15:38:26] <Jymmm> all inclusive; case ps, wallwart, etc
[15:40:16] <FinboySlick> Jymmm: I need something to run my top-secret audio encoding software and pipe it out a professional grade (ballanced audio) audio card. It fetches its stuff off of network.
[15:40:32] <FinboySlick> It has to be portable.
[15:40:39] <Jymmm> ReadError: Do you have a balanced charger?
[15:40:51] <Jymmm> FinboySlick: Is it linux based?
[15:41:02] <FinboySlick> The audio card is Focusrite Scarlett 2i2
[15:41:11] <FinboySlick> And it's one picky bitch :P
[15:41:19] <FinboySlick> Yeah, it's linux based.
[15:41:43] <Jymmm> FinboySlick: Why not hack a WD LiveTV, they are really cool.
[15:41:49] <Jymmm> tiny
[15:41:59] <Jymmm> hdmi
[15:42:13] <Jymmm> ethernet, wifi, usb
[15:42:14] <FinboySlick> Jymmm: I don't want hdmi or even video.
[15:42:29] <FinboySlick> But does it have an FPU?
[15:42:33] <Jymmm> I know, but it's REALLY small
[15:42:44] <ReadError> yea Jymmm
[15:42:54] <ReadError> mine does 4 batteries at once
[15:42:56] <ReadError> all balanced
[15:43:00] <ReadError> up to 6 cells
[15:43:41] <FinboySlick> I have some marvel based (think GuruPlug) hardware but lack of an FPU just kills my software encoding. Even at 1.2Ghz it's not fast enough.
[15:44:40] <Jymmm> FinboySlick:
http://en.wikipedia.org/wiki/WD_TV
[15:44:53] <FinboySlick> Jymmm: Yeah, was reading that.
[15:45:12] <Jymmm> ReadError: You can't actually USE the batteries while being charged (like a laptop does) can you?
[15:46:05] <Jymmm> FinboySlick: for a media player, I love it
http://www.wdc.com/en/products/products.aspx?id=330
[15:47:10] <FinboySlick> Jymmm: It looks really cool... But I don't have time to port everything to this platform. I really have to stick to x86 if I want to meet the deadline.
[15:47:43] <Jymmm> FinboySlick: Ok, just a suggestion. Have you tried the PicoITX by Via?
[15:48:00] <Jymmm> size of a deck of cards
[15:48:45] <FinboySlick> hmmmm, via is uhci I think, that's an option.
[15:48:57] <Jymmm> http://en.wikipedia.org/wiki/Pico-ITX
[15:49:27] <Jymmm> FinboySlick: your in EU?
[15:49:34] <Jymmm> you're
[15:49:43] <FinboySlick> Jymmm: No, Canada.
[15:50:06] <Jymmm> oh, hmmm. I got one for sale if you're interested =)
[15:51:03] <Jymmm> Comes with warranty that expires upon receipt =)
[15:52:58] <ReadError> nah Jymmm
[15:53:27] <Jymmm> ReadError: k. does your charger get hot?
[15:53:38] <Jymmm> do you rapid charger your batteries?
[15:57:12] <ReadError> nah it stays cool
[15:57:16] <ReadError> has fans inside it
[15:57:20] <ReadError> i charge around 3C
[15:57:48] <Jymmm> ReadError: I want to use 1 or 2 banks of 4x 18650's to make a sort of battery pack, that can be charged via wall, car, solar, or "other" method, and output anywhere from 2 to 16V with the use/help from step up/down converters. But I want the batteries to be as pampered as much as possible.
[15:58:34] <Jymmm> ReadError: i ust have no clue on balanced chargers
[16:00:05] <Jymmm> ReadError: But, I did find some holders
http://s.dealextreme.com/search/18650+holder
[16:06:04] <ReadError> sorry running in and out got some beer brewin
[16:06:41] <ReadError> if you want to go easy on them
[16:06:45] <ReadError> dont exceed 1c
[16:12:35] <DJ9DJ> gn8
[16:20:24] <Jymmm> ReadError: ping me when you get back/sober; whichever comes later
[16:38:56] <joe9> wondering if anyone can help me out. I get this warning in linuxcnc "Program exceeds machine minimum on axis Z" . my program:
http://codepad.org/DZFhsDOC axis Z configuration in my .ini file
http://codepad.org/89kmeTRs
[16:39:19] <joe9> the relevant line that throws out this error is: G00 Z0.20000 ( retract )
[16:39:31] <joe9> I have no idea why I get that warning.
[16:39:46] <joe9> I changed the .ini file to have my limits between 1.00 and -5.5
[16:41:49] <Tom_itx> G91?
[16:42:04] <Tom_itx> does it get switched back?
[16:42:32] <joe9> yes, the next line after that.
[16:42:42] <joe9> http://codepad.org/kKI7sqLs Tom_itx
[16:43:32] <joe9> Tom_itx: thanks for checking the program out. I know it is too much to ask of anyone. but, I am clueless at this point.
[16:43:37] <ReadError> why do you use negative joe9 ?
[16:43:48] <ReadError> i found it much much easier to just use 0 to 5.5
[16:44:00] <ReadError> 0 being the lowest
[16:44:13] <Tom_itx> what line is the error on ?
[16:44:16] <ReadError> but still set to where the actual spindle cant hit the bottom
[16:44:26] <joe9> Tom_itx: line 130
[16:44:33] <joe9> Tom_itx: G00 Z0.20000 ( retract )
[16:44:41] <joe9> if i remove that line, the program runs fine.
[16:44:44] <Tom_itx> you don't set G91 back there though do you?
[16:44:55] <Tom_itx> add a G91 above and G90 below it
[16:44:58] <Tom_itx> see what happens
[16:45:07] <joe9> ok, will do. thanks.
[16:45:08] <Tom_itx> or just the G91
[16:46:09] <Tom_itx> that's what you do on lines 23..25
[16:46:18] <joe9> Tom_itx: adding a G91 and G90 worked without any warning.
[16:46:31] <joe9> http://codepad.org/0jUTcuLO
[16:47:37] <joe9> Tom_itx: the Z code is being generated. I can sed it to add G91 before and G90 after that line.
[16:47:47] <joe9> but, my G55 coordinates should take care of that.
[16:48:03] <joe9> and, sometimes adding a G91 and G90 might not make the program work correctly.
[16:48:06] <Tom_itx> G55 won't change the mode
[16:48:32] <joe9> I set the touchoff of G55
[16:49:02] <joe9> and, I want
http://codepad.org/MWVnfU0c
[16:49:17] <joe9> the Z move to be a movement on the G55 Z axis.
[16:49:21] <joe9> Does that make sense?
[16:49:35] <Tom_itx> set G55 change G91
[16:49:37] <joe9> I want it to go to the 0.2 on G55 Z axis.
[16:49:46] <Tom_itx> set the G55 first
[16:49:50] <joe9> absolute coordinates of G55
[16:49:52] <joe9> I mean.
[16:49:55] <Tom_itx> then execute the G91
[16:50:02] <Tom_itx> mmm
[16:50:28] <Tom_itx> why are you switching back and forth anyway?
[16:51:16] <joe9> in my touchoff code, i do not know the absolute coordinate on Z axis to back up to.
[16:51:29] <joe9> so, I am using the relative mode to go back by 0.1 inches and try again.
[16:51:35] <joe9> it is just for the touchoff code
[16:51:51] <joe9> in the normal code, I want it to go to the absolute Z=0.2 of G55
[16:51:55] <joe9> not relative mode.
[16:51:58] <joe9> does that make sense?
[16:52:16] <Jymmm> it would if I had any.
[16:52:26] <Jymmm> ;)
[16:52:58] <joe9> Tom_itx: there is a lot more program code between line 132 and 134 that I removed for the sake of fixing this issue.
[16:53:25] <Tom_itx> what happens when you switch to G55 then execute G90
[16:53:26] <joe9> after the G55 touch off is set in line 122, I run my normal program code.
[16:53:54] <joe9> Tom_itx: like this, you mean
http://codepad.org/8FaCKJYH
[16:54:14] <Tom_itx> yeah
[16:54:28] <Tom_itx> seems it's getting confused, i dunno
[16:54:38] <Tom_itx> you may even need one after the G54
[16:54:45] <Tom_itx> it may be redundant though
[16:55:00] <Tom_itx> i don't know how linuxcnc reacts to that
[16:55:30] <joe9> Tom_itx: it says the same warning message about "Program exceeds machine minimum on axis Z"
[16:55:45] <Tom_itx> is G90 relative to world?
[16:55:51] <Tom_itx> (machine coordinates)
[16:55:58] <joe9> trying this now:
http://codepad.org/6qJglZvt
[16:56:33] <joe9> tom_itx, same message again.
[16:57:09] <Tom_itx> did you try running it with the G91 like i originally suggested?
[16:57:27] <Tom_itx> to see if it gives you the desired position
[16:58:46] <joe9> Tom_itx: it worked this time, but, it is not a good idea to do so in my program. as the z=0.2 is the safe distance, that the bit should retract to after doing whatever it is doing.
[16:59:10] <joe9> and, it needs to come back to the absolute of Z=0.2 all the time.
[17:00:05] <joe9> i think linuxcnc is getting confused as I am changing G55 Z on the fly.
[17:00:08] <joe9> Is there something that I can do to set the G55 at startup, so, that program limits are complied.
[17:00:09] <Tom_itx> why are there 2 G54 lines in a row on 110 112?
[17:00:11] <joe9> and linuxcnc does not error out.
[17:00:39] <joe9> Tom_itx: Is it wrong to have that?
[17:00:44] <joe9> Tom_itx: I can change it.
[17:01:09] <Tom_itx> not sure you need both
[17:01:21] <Tom_itx> 110 isn't setting it is it?
[17:01:36] <Tom_itx> don't you need a G10 or something to set it?
[17:02:28] <joe9> Tom_itx: removed all the extra g54's, I think.
http://codepad.org/N9Y8UO0K
[17:02:31] <joe9> same message still.
[17:02:41] <Tom_itx> what about line 110?
[17:02:49] <Tom_itx> is it supposed to be saving the position?
[17:02:53] <joe9> Tom_itx: yes, in line 103 I have G10 L2 P2
[17:02:56] <joe9> to set G55.
[17:04:04] <joe9> Tom_itx: line 110 -- just coming back to the values that I had saved earlier in line 58
[17:04:07] <Tom_itx> ( save the current values to return to later )
[17:04:07] <Tom_itx> G0 X #<current_x_position> Y #<current_y_position>
[17:04:15] <Tom_itx> but you're not saving it are you?
[17:04:32] <joe9> Tom_itx: in line 58, I am saving those values.
[17:04:59] <joe9> Tom_itx: the probe comes back to the X and Y position where it was before starting the touch off process.
[17:05:48] <Tom_itx> does the final post work?
[17:06:01] <Tom_itx> with the G55 then the G90?
[17:06:07] <Tom_itx> 129 130
[17:06:19] <joe9> no, it does not.
[17:06:33] <joe9> I get the same message about "exceeding minimum Z"
[17:06:40] <Tom_itx> but the G91 after it does?
[17:07:11] <joe9> yes, I if I put a G91 at line 131, I do not get that message.
[17:07:19] <Tom_itx> can you save the mode to a var so you know what it's set to?
[17:07:26] <Tom_itx> i dunno if you can do that or not
[17:07:39] <joe9> but, why?
[17:07:53] <Tom_itx> ok, try this.. add the G91 before the G55 and leave the G90 after
[17:07:57] <Tom_itx> see what happens
[17:07:59] <joe9> I think linuxcnc is miscalculating the Z minimum distance and throwing that message wrongly.
[17:08:28] <Tom_itx> we're trying to get it back in it's proper frame of mind if it is
[17:09:00] <Tom_itx> does changing modes G90 G91 require a linear move?
[17:09:02] <joe9> http://codepad.org/2CgBC9Wd like this you mean.
[17:09:03] <Tom_itx> i forget
[17:09:12] <joe9> same message still.
[17:10:35] <joe9> oh, really. I can remove the G91 usage if that is the reason for the message later on.
[17:11:03] <Tom_itx> try adding a G1 Z.000001
[17:11:07] <Tom_itx> see what happens
[17:11:20] <Tom_itx> or .00001
[17:11:22] <Tom_itx> whatever
[17:11:53] <joe9> tom_itx, where ?
[17:12:09] <Tom_itx> after the mode change
[17:12:31] <Tom_itx> i'm not sure really.
[17:12:45] <joe9> Tom_itx: I removed all usages of G91 from the program and still the same message.
[17:12:49] <joe9> let me try the G1
[17:13:16] <Tom_itx> add it between the G55 and G91 too just to see
[17:13:39] <Tom_itx> (i'm guessing at this point)
[17:14:13] <joe9> this is with G01 before the G00, same message (btw, I put the G91 back into the program)
[17:14:26] <joe9> tom_itx, ok, let me add the G1 where you suggested.
[17:14:48] <Tom_itx> beyond that i dunno what to tell you
[17:15:20] <joe9> Tom_itx: no luck. thanks for trying.
[17:15:30] <joe9> seems to be a bug.
[17:15:49] <joe9> I can understand linuxcnc not knowing where the Z at that point is. as Z is being set dynamically.
[17:15:50] <Tom_itx> view the status as you single step thru it
[17:16:15] <Tom_itx> whatever vars are in use there etc
[17:16:26] <Tom_itx> maybe you can spot it that way
[17:16:39] <joe9> Do you mean using the "arrow" button?
[17:16:53] <Tom_itx> yeah i think
[17:16:54] <joe9> The first time I click that button, it gives me that message.
[17:16:59] <r00t4rd3d> http://www.liveleak.com/view?i=7c7_1345319885
[17:19:14] <joe9> Tom_itx: could it be something to do with G92?
[17:26:52] <Tom_itx> i don't think i use G92 much or if at all
[17:26:59] <Tom_itx> not sure how it acts
[17:28:34] <Tom_itx> from the description, it could screw you over good though if you don't keep track of it
[17:31:23] <Tom_itx> you're not using G92
[17:41:05] <joe9> Tom_itx: maybe I should be using the tool length stuff instead of setting Z value on touchoff.
[17:41:16] <joe9> will have to research that further. will keep you posted.
[17:41:40] <Tom_itx> i use the tool table on my stuff
[17:43:19] <joe9> Tom_itx: if the tool length is known, then that would work. I do not have the depth rings so I do not know the tool length reliably.
[17:44:04] <Jymmm> ReadError: Whatcha think?
http://www.amazon.com/Thunder-Balance-Charger-Discharger-Software/dp/9269806723
[17:45:55] <joe9> Tom_itx: this is interesting. I added a M06 after the G10 L2 P2 and no message now.
[17:47:24] <Tom_itx> are you changing tools in the code? i didn't read thru it all
[17:47:44] <Tom_itx> but that would signal a tool change and to re read the offsets
[17:48:54] <joe9> no message, but, it gives me this error: emc/task/emctask.cc 389: interp_error: Need tool prepared -Txx- for toolchange
[17:48:57] <joe9> Need tool prepared -Txx- for toolchange
[17:49:09] <joe9> so, I guess that will not work.
[17:50:22] <jdh> doesn't that just mean that it calculated the resultant Z to be outside your limits? (using work offsets, tool height offsets, etc)
[17:51:51] <Tom_itx> jdh did you see the code?
[17:51:58] <jdh> tl;dr
[17:52:05] <ReadError> Jymmm,
http://www.hobbyking.com/hobbyking/store/__11170__Turnigy_4x6S_Lithium_Polymer_Battery_Pack_Charger.html
[17:52:07] <ReadError> thats what i run
[17:52:07] <jdh> and, probably not really relevant.
[17:52:34] <joe9> jdh: this is the code:
http://codepad.org/KwTZeXqR
[17:52:57] <Tom_itx> jdh the issue is around line 129 130 or so
[17:53:15] <joe9> i am trying to do a probe and then run the main program, without linuxcnc complaining about limits
[17:53:39] <joe9> I checked with cradek on #linuxcnc-devel and he seems to think that it would not be possible.
[17:54:00] <joe9> as linuxcnc would have no way of calculating the future Z value when there is a probe.
[17:54:14] <joe9> and, the only option is to use the "Run anyway" button.
[17:54:35] <joe9> this message has saved my ass a few times before, and, hence I want the message/mechanism to work correctly.
[17:55:09] <joe9> the only option at this point is to split the program into 2 programs, one for setting the touchoff values and the other for the actual running of the code using those touchoff values.
[17:58:13] <joe9> ReadError: what is your feedrate when milling pcb's?
[17:58:36] <joe9> mine seems to be doing a good job at 6 inches per minute. I am not sure if that is a bit too much.
[18:15:45] <Jymmm> ReadError: In a battery PACK, I know there are sense wires going back to the charger, but how does the charger balance EACH cell in the pack, when there is only one power cable going to the pack from the charger?
[18:36:04] <jthornton> wd5jwy, around?
[18:39:54] <Tom_itx> jthornton you gettin your front end worked out?
[18:40:16] <jthornton> not yet, been busy
[18:42:30] <jthornton> thinking about the layout mostly now
[18:44:12] <ReadError> Jymmm,
[18:44:15] <ReadError> it has a balance plug
[18:44:21] <ReadError> that detects the voltage of each sell
[18:44:27] <ReadError> and discharges the higher ones
[18:44:30] <ReadError> to balance
[18:44:42] <jthornton> now spammers are creating a random 12 letter.com for their email
[18:45:49] <jthornton> last 5 from the same ip address
[18:45:55] <jthornton> they are so dumb
[19:01:24] <r00t4rd3d> dumb or smart
[19:04:45] <r00t4rd3d> oh, your talking about forum spammers
[19:04:56] <jthornton> dumb cause every registration must be approved or they are wasting there time
[19:05:02] <jthornton> yea
[19:06:03] <r00t4rd3d> use phpBB
[19:06:37] <jthornton> not my choice as to what we use
[19:06:46] <ReadError> http://gawker.com/5935685/topless-femen-activist-takes-down-giant-cross-with-chainsaw-to-protest-pussy-riot-verdict-%5Bnsfw%5D
[19:06:48] <ReadError> nsfw
[19:07:00] <ReadError> lesson: how not to use a chainsaw
[19:07:05] <r00t4rd3d> with phpBB you cant set questions
[19:07:17] <r00t4rd3d> Whats this forum name?
[19:07:29] <r00t4rd3d> they also have anti spam mods
[19:07:51] <r00t4rd3d> who chooses what software to run, chris?
[19:07:52] <jthornton> LinuxCNC
[19:08:01] <jthornton> dunno
[19:08:43] <r00t4rd3d> who owns the website?
[19:09:15] <jthornton> dunno
[19:10:12] <r00t4rd3d> Registrant: Make this info private
[19:10:12] <r00t4rd3d> Henry, Raymond**
[19:10:12] <r00t4rd3d>
[19:10:12] <r00t4rd3d> 358 Swante Aho Road
[19:10:12] <r00t4rd3d> Crystal Falls, MI 49920
[19:10:13] <r00t4rd3d> US
[19:10:32] <r00t4rd3d> who is ray henry?
[19:10:55] <jthornton> a developer that is not around much
[19:11:16] <jthornton> I've not seen him on here in a few years
[19:11:21] <r00t4rd3d> how about steve stallings
[19:11:48] <jthornton> don't know who he is
[19:11:54] <r00t4rd3d> he is the tech contact
[19:12:01] <wd5jwy> Hey John - I'm here
[19:12:36] <wd5jwy> I wasn't able to install the .comp file for some reason
[19:12:55] <jthornton> quick question the ratio between the winder and the guide is for wire spacing?
[19:13:04] <jthornton> you need to install dev
[19:13:37] <wd5jwy> yes, sort of
[19:13:43] <jthornton> let me see I put that in the manual somewhere
[19:14:02] <jthornton> sudo apt-get install linuxcnc-dev
[19:14:08] <jthornton> http://linuxcnc.org/docs/html/hal/comp.html#_installing
[19:14:15] <wd5jwy> ah, ok
[19:14:24] <jthornton> so the wire spacing is even except on the ends?
[19:15:37] <wd5jwy> yes, let me find you a pic - it will tell you a lot
[19:15:44] <jthornton> ok
[19:16:07] <r00t4rd3d> isnt a pic the talks a video?
[19:16:25] <r00t4rd3d> that
[19:16:55] <wd5jwy> http://tinyurl.com/92dahnv
[19:17:16] <r00t4rd3d> makes me really sad the Late Show had to layoff employees and that fat idiot took a 10 million pay cut.
[19:17:32] <jthornton> is that three seperate coils?
[19:17:46] <r00t4rd3d> what is that thing?
[19:17:49] <wd5jwy> no, one continuous wire - but 3 pies
[19:17:52] <wd5jwy> it's a RF choke
[19:18:14] <jthornton> so wind one pie move over wind the second pie?
[19:18:30] <wd5jwy> yes, but moving over is the easy part
[19:18:37] <r00t4rd3d> what do you do with that?
[19:18:53] <wd5jwy> the commercial machines that do this......
[19:19:22] <jthornton> it almost looks like a dwell at the end of the stroke then a constant velocity to the other side
[19:19:25] <wd5jwy> use sine and cosine functions to achieve the proper traverse movement
[19:19:55] <wd5jwy> well, that's just it - the traverse cannot spend too much time on either end of it's travel
[19:20:18] <wd5jwy> otherwise too much wire gets place on the ends and you don't get that pattern
[19:20:43] <wd5jwy> that's the reason for the heart-shaped CAM they used in the older mechanical winders
[19:21:02] <wd5jwy> dimple at the top and small peak at the bottom forced quick reversal of the traverse
[19:21:12] <wd5jwy> if that makes sense
[19:21:42] <jthornton> ok, yes that does make sense and that can be done in the component I think with little effort
[19:22:14] <wd5jwy> but, someone was commenting on here earlier today - we could play with the acceleration settings dynamically to get the same movement
[19:22:28] <wd5jwy> so I think a lead-screw would work instead of a cam
[19:22:46] <jthornton> say you define the start and end of the coil and the amount of accelerated move and the scale of the acceleration
[19:22:56] <wd5jwy> YES
[19:23:01] <wd5jwy> I think that would work
[19:23:21] <wd5jwy> you would need to play with the numbers, but it should work
[19:24:01] <wd5jwy> so, here's my first problem
[19:24:04] <jthornton> say for example the coil is 1000 steps wide and you want the steps 0-25 and 975-1000 to accelerate to a faster speed to the end and decelerate back to normal speed.
[19:24:16] <jthornton> for the middle
[19:24:22] <wd5jwy> yes, exactly
[19:24:59] <wd5jwy> If I was good at trig, I'm sure the numbers could be generated that way too
[19:25:04] <r00t4rd3d> a single speed and tension the magnet wire sounds easier
[19:25:10] <jthornton> so you tell the component how many steps to accelerate in and how fast to add on to the velocity
[19:26:23] <jthornton> you could even tell it how many layers or revs of the mandrel you want and stop when that amount of layers is wound
[19:26:53] <wd5jwy> correct - a lot of it could be automated - and it is in a commercial winder
[19:27:02] <wd5jwy> but I don't have 50K for one of those
[19:27:21] <jthornton> the control side is free!
[19:27:26] <r00t4rd3d> so are trying to mass produce rf chokes or something?
[19:27:38] <jthornton> you just have to build the mandrel and guide
[19:27:58] <wd5jwy> well, maybe - there's a few types of these that no one makes any more
[19:28:09] <wd5jwy> yes, I already have the wire guide
[19:28:22] <wd5jwy> and collecting parts for the mandrel and tail-stock
[19:28:55] <wd5jwy> so, here's my first problem
[19:29:14] <wd5jwy> I haven't yet succeeded in getting my motors to run with any of the code so far
[19:29:38] <wd5jwy> I think I'm doing something wrong in the .hal file, but can't figure out what it is
[19:30:04] <jthornton> ok, install that component and connect the step and direction pins to your parallel port in start.hal
[19:31:01] <jthornton> you should be able to run the guide stepper back and forth at will with what I uploaded to the forum
[19:31:34] <wd5jwy> well, I notice in one of the standard configs that they add the parport driver to the thread - is that required before you can get pulses out of the parallel port?
[19:31:49] <jthornton> yea
[19:32:21] <wd5jwy> ok, so that's what I'm missing I think - and that's not in your .hal file either - right?
[19:32:30] <jthornton> loadrt hal_parport cfg="0"
[19:32:37] <jthornton> should load the driver
[19:32:45] <wd5jwy> that loads the driver
[19:32:56] <wd5jwy> but it doesn't load it to the thread "addf" - right?
[19:33:00] <jthornton> show pin in the terminal will show you if it is loaded
[19:33:02] <jthornton> yes
[19:33:09] <jthornton> http://linuxcnc.org/docs/html/hal/parallel_port.html#_installing
[19:33:21] <wd5jwy> addf parport.0.write base-thread
[19:33:37] <wd5jwy> that's what's missing - wrong thread name in that example tho
[19:33:50] <jthornton> in my example it would be fast not base-thread
[19:33:55] <wd5jwy> yes
[19:34:05] <jthornton> your catching on :)
[19:34:10] <wd5jwy> ok, so let me get the .comp file installed
[19:34:13] <jthornton> will pass me soon I bet
[19:34:14] <wd5jwy> SLOWLY
[19:34:18] <wd5jwy> never
[19:34:40] <r00t4rd3d> wd5jwy,
http://www.bignick.net/images/RadioPix/Gingery.jpg
[19:34:44] <r00t4rd3d> u see that?
[19:34:52] <wd5jwy> yes, I built one
[19:34:56] <wd5jwy> that's not mine tho
[19:35:12] <wd5jwy> the brass wire guide was a bitch to build
[19:35:18] <wd5jwy> had to cut out all the parts on my cnc
[19:35:23] <wd5jwy> but it works - kinda
[19:36:01] <wd5jwy> if you look on the right-bottom corner of the machine - that's the CAM he's using - it's oval
[19:36:05] <r00t4rd3d> http://www.ebay.com/itm/New-Old-Stock-in-box-Morris-Coilmaster-Coil-Winder-for-making-radio-coils-/221085300312?pt=LH_DefaultDomain_0&hash=item3379b5f658
[19:36:08] <r00t4rd3d> Coilmaster
[19:36:14] <wd5jwy> it can't be oval if it's going to work correctly
[19:36:24] <wd5jwy> but, for messing around, it's ok
[19:36:58] <wd5jwy> yes, those were manufactured back in the 1950's - same type of mochine
[19:37:29] <r00t4rd3d> that means it will last forever
[19:37:47] <wd5jwy> yeah, more or less - still a bunch of them around
[19:38:05] <r00t4rd3d> 17 bids
[19:38:46] <r00t4rd3d> who are you trying to ham with japan?
[19:39:03] <wd5jwy> me?
[19:39:21] <r00t4rd3d> isnt that what rf chokes are for like that?
[19:39:40] <r00t4rd3d> some kinda transmitter
[19:39:52] <wd5jwy> well, yeah they are used in all types of radio equipment - not just transmitters
[19:40:00] <wd5jwy> receivers too
[19:40:16] <r00t4rd3d> what do you do with them?
[19:40:45] <wd5jwy> I would use them mostly tube-type equipment - transmitters and receivers
[19:41:05] <wd5jwy> the problem is - no one makes them anymore because there's not a big market for them
[19:45:22] <jthornton> wd5jwy, try this
http://pastebin.com/ZFY88Nd9
[19:45:46] <jthornton> after you get the component installed
[19:46:39] <jthornton> wd5jwy, is that your ham call sign?
[19:49:02] <jdh> its his name. his parents hated him.
[20:08:07] <wd5jwy> Yes, that's my ham call sign
[20:09:00] <jthornton> I thought so
[20:09:30] <Jymmm> ReadError: dis it discharge the uneven cells at the begin of the charge cycle, during, or at the end typically?
[20:10:26] <Jymmm> No, I'm with jdh, wd5jwy parents hated him.
[20:11:00] <Jymmm> Were probably drunk and playing scrabble at the time and that's what tiles they had.
[20:11:10] <ReadError> during
[20:11:15] <Jymmm> ReadError: k
[20:11:22] <ReadError> you want a balanced pack
[20:11:27] <ReadError> out of balance, you can ruin the entire pack
[20:11:37] <ReadError> whatcha makin?
[20:12:48] <Jymmm> ReadError: A sorta universal battery that can provide any common voltage and be charged by mains, car, or solar/human
[20:13:55] <Jymmm> 3.7 to 16v
[20:13:59] <r00t4rd3d> i got a 18650 battery pack that can be recharged with solar panels for my arduino stuff
[20:16:01] <r00t4rd3d> i dont ever use it but i got it.
[20:18:13] <r00t4rd3d> i have a shit load of arduino stuff i dont touch actually
[20:18:15] <jthornton> wd5jwy, let me know how you make out
[20:18:24] <r00t4rd3d> with his tongue
[20:23:28] <Jymmm> ReadError: I have a folding solar panel
http://www.solarwares.com/unisolarpac.html
[20:24:38] <Jymmm> ReadError: I'd liek to compliment it with a nice battery pack I can use for whatever
[20:28:53] <r00t4rd3d> http://www.batteryspace.com/Li-Ion-18650-cell-series-Battery-Holders-with-PCB-installed.aspx
[20:29:00] <r00t4rd3d> run the 18650's in series
[20:31:39] <ReadError> oh
[20:31:50] <ReadError> the stuff i run is like 35C discharge
[20:32:08] <Jymmm> ReadError: What does that mean? 35c 40c etc ?
[20:32:10] <ReadError> i peak out at 120A discharge on one of them
[20:32:15] <ReadError> C = capacity
[20:32:27] <ReadError> so like 1C on a 1000mAh pack would be 1A
[20:33:00] <Jymmm> a multiplier?
[20:33:45] <Jymmm> 12AH @ 3A draw per hour == 4 hours
[20:33:51] <Jymmm> etc
[20:34:00] <Jymmm> bbiab
