#emc | Logs for 2008-02-11

[00:02:22] <eric_k> actually having a monitor that you don't have to adjust every time the screensaver turns on: priceless
[00:06:36] <eric_U> getting cold down there
[00:18:17] <BigJohnT> anyone know where I can get the tooth profile of an "L" timing belt?
[00:18:39] <jmkasunich> google?
[00:19:06] <BigJohnT> didn't find anything
[00:21:30] <eric_U> gates?
[00:21:34] <eric_U> browning?
[00:22:13] <jmkasunich> http://www.sdp-si.com/D260/HTML/D260T015.html
[00:22:29] <jmkasunich> 12 different profiles, including L
[00:22:43] <BigJohnT> been nosing around on gates
[00:22:48] <SWPadnos> or here: http://www.sdp-si.com/D265/PDF/D265T015.pdf
[00:22:51] <jmkasunich> * jmkasunich = googlemaster
[00:23:12] <jmkasunich> SWPadnos: same page, one is HTML, one is PDF
[00:23:19] <SWPadnos> ah, I see
[00:23:25] <jmkasunich> the PDF is nice, since you can enlarge it
[00:23:29] <SWPadnos> I was tricked for a moment by the .zip downloads at the top ;)
[00:23:32] <BigJohnT> * BigJohnT just demoted to googlemouse
[00:23:42] <SWPadnos> googleweenie
[00:24:07] <eric_U> too bad they don't have H
[00:24:13] <BigJohnT> yep looks like an acorn at the moment
[00:24:43] <eric_U> * eric_U google is not my friend
[00:26:48] <Jymmmm> eric_U: you're just googless
[00:27:32] <eric_U> apparently, if google really doesn't like you they'll accuse you of being a search spammer and refuse the search
[00:32:56] <Roguish> hello again, what program can i use to record the animated motion of my 5axis vismach?
[00:33:05] <Roguish> from the screen?
[00:38:04] <SWPadnos> Roguish, http://demorecorder.com/ (from google)
[00:38:34] <SWPadnos> I'd search in synaptic for screen recorder
[00:39:28] <jmkasunich> jepler posted the process on his blog
[00:39:39] <jmkasunich> unpythonic.com, or .net, or something like that
[00:39:45] <SWPadnos> net
[00:46:05] <skunkworks> * skunkworks waits patently for Roguishs' video
[01:06:09] <eric_U> * eric_U still waiting
[01:16:36] <Jymmmm> SWPadnos: get my msg?
[01:48:57] <Jymmmm> jmkasunich: In respect to threads, what is NS vs UN ? "15/16-16 UN-2A"
[01:49:23] <jmkasunich> NS = national standard, UN = unified
[01:49:39] <jmkasunich> don't know what either actually means, grab machinerys handbook
[01:49:39] <cradek> 2A is fit class
[01:49:47] <cradek> yep I was just going to say that
[01:50:18] <Jymmmm> jmkasunich: So, if I bought a tap that said "15/16-16 NS NOS Tin Coated 3 Flute Tap" would it be the right one for "15/16-16 UN-2A" ?
[01:50:59] <jmkasunich> no clue
[01:51:07] <Jymmmm> okey dokey, thanks =)
[01:51:25] <cradek> do you really need to hit 2A, or do you just need to make a hole it screws into?
[01:51:54] <SWPadnos> it's for the ballnut on the X axis of the mill, so - uh, I don't know the answer to that question :)
[01:51:57] <Jymmmm> cradek: Sorry, I really dont know what any of the suffixes mean, those are just what they give in the specs
[01:52:36] <Jymmmm> cradek: Is 2A the tightness of the fit of the threads?
[01:52:48] <tomp> a link to jepler's blog info on dtop vid creation http://emergent.unpy.net/01196105360
[01:52:50] <cradek> yes it's a tolerance of pitch diameter
[01:53:08] <Jymmmm> cradek: Is 2A considered tight, loose, avg?
[01:53:19] <cradek> you should get a machinery's handbook, it doesn't matter the year, so you can get an old cheap one
[01:53:24] <cradek> mine's out in the shop :-)
[01:53:31] <cradek> maybe google knows?
[01:53:40] <jmkasunich> I think 2A is pretty middle-of-the-road
[01:53:40] <SWPadnos> brrrr. the shop
[01:54:11] <Jymmmm> http://www.efunda.com/DesignStandards/screws/unified.cfm
[01:55:26] <Jymmmm> ah, here it is... http://www.efunda.com/DesignStandards/screws/unified.cfm?start=228&finish=301
[01:56:21] <SWPadnos> http://www.sizes.com/tools/bolts_inch_threadFit.htm
[01:57:05] <Jymmmm> ah, so 2A it most common
[01:57:32] <Jymmmm> Now to figure out the tap suffix of "NS"
[01:58:09] <Jymmmm> Series NS = National Special
[01:58:14] <Jymmmm> wth does that mean =)
[01:58:34] <Jymmmm> http://academic.evergreen.edu/projects/biophysics/technotes/fabric/fraction.htm
[02:00:02] <SWPadnos> now why would that be on a biophysics website?
[02:00:18] <Jymmmm> DIY equipment repair
[02:00:23] <eric_U> best machine shop on Penn State campus is at physics
[02:00:33] <eric_U> they make all the cool experiments
[02:00:38] <SWPadnos> I was just noticing the "bio" prefix there ...
[02:01:24] <eric_U> true, they do nothing
[02:01:35] <eric_U> :)
[02:03:49] <SWPadnos> damn. it's nearly impossible to stop eating licorice
[02:03:59] <Jymmmm> red or black?
[02:04:02] <SWPadnos> black
[02:04:28] <Jymmmm> dip it in garlic power, guarnteed to make you stop
[02:04:40] <SWPadnos> hmmm..... garlic licorice ...
[02:05:45] <Jymmmm> SWPadnos: or better yet, dip it in ascorbic acid
[02:05:51] <tomp> licorice the sweetest sugar known
[02:05:57] <SWPadnos> hmmmmm. sour patch licorice ...
[02:06:06] <Jymmmm> SWPadnos: Yep =)
[02:06:24] <SWPadnos> (you foget - I used to work in movie theaters :) )
[02:06:27] <SWPadnos> forget
[02:06:45] <Jymmmm> SWPadnos: Better than "Would you like fries what that?"
[02:06:53] <Jymmmm> s/what/with/
[02:06:59] <SWPadnos> "would you like butter on that?"
[02:07:06] <SWPadnos> about the same I think
[02:07:07] <eric_U> you people are sick
[02:07:12] <Jymmmm> butter licorice?
[02:07:18] <SWPadnos> hmmmmm ...
[02:07:19] <SWPadnos> ewww
[02:07:26] <Jymmmm> yeah ewwwwwwwww
[02:07:41] <kimron> pastis maybe better
[02:07:55] <Jymmmm> eric_U: Hey, ascorbic acid licorice doens't sound half bad at all
[02:08:12] <eric_U> I found a place in california that sells 10 pair of full extension 20" slides for 42
[02:08:16] <eric_U> $
[02:08:24] <eric_U> shipping is then $32
[02:08:37] <Jymmmm> eric_U: ballbearing slide?
[02:08:42] <eric_U> yes
[02:08:44] <Jymmmm> for like rack mount gear?
[02:08:52] <tomp> drawer slides?
[02:08:53] <eric_U> no, heavy duty
[02:08:54] <SWPadnos> drawer slides most likelt
[02:08:57] <eric_U> drawer slides
[02:08:59] <Jymmmm> eric_U: how long do you need and how many?
[02:09:00] <SWPadnos> heavy duty drawer slides
[02:09:26] <eric_U> I probably need about 6 pair, 21" -- 20" would work
[02:09:36] <Jymmmm> 6 pair?! wth
[02:10:04] <eric_U> tool/hardware cabinet
[02:10:14] <Jymmmm> Oh, heh.
[02:10:28] <Jymmmm> I thought you were making a foam cutter or something.
[02:10:40] <eric_U> could do that
[02:10:57] <Jymmmm> eric_U: Why do you think I have the ones I do =)
[02:11:10] <eric_U> they kinda accumulate around here
[02:11:15] <Jymmmm> I'd give up some, but not 6 pairs worth =)
[02:11:31] <eric_U> did you go ebay?
[02:11:44] <tomp> lista brand cabinets had great slides. made to hold tools in the drawers. was a good ladder.
[02:11:49] <Jymmmm> eric_U: Nope, I went in the trash can at work =)
[02:11:56] <eric_U> best place for them
[02:12:20] <Jymmmm> eric_U: Especially in a data center where most ppl throw beand new ones away still sealed in the box
[02:12:38] <eric_U> my experience is those are extra
[02:12:48] <eric_U> $150 a pair extra
[02:13:14] <Jymmmm> eric_U: Nah, ppl just dont have/order back rails so they cant use them, they order a shelf instead.
[02:14:07] <Jymmmm> Or if they're in a shared cabinet there is no backrails, so they CAN'T use the,.
[02:15:21] <eric_U> I have some full extension rack slides, but they aren't ball bearing
[02:16:17] <Jymmmm> Ah, these are, but not for drawers.
[02:19:15] <eric_U> some guy in Georgia is paying half the shipping I am, origin is in california
[02:19:22] <jmkasunich> axis preview just saved another endmill
[02:19:34] <jmkasunich> typo: G0Z-1.25, instead of G0Z-0.125
[02:20:15] <eric_U> someone posted a pic of a bridgeport here that I swear had a half-moon 2" diameter divot gone from the table
[02:22:07] <cradek> they all have that
[02:22:27] <eric_U> mine has some nicely filled 3/4" divots
[02:22:40] <cradek> I mean it's supposed to be there
[02:22:47] <eric_U> don't believe it :)
[02:25:19] <eric_U> when I mentioned it on here, someone thought I was talking about the slot that goes the wrong way in the y direction in the center of the table
[02:25:41] <eric_U> which I never really understood, but it's on all series 1 cnc bp that I've seen
[02:27:17] <tomp> Stuart Brorson ever show up here? just read some interesting gEDA/EMC stuff from the user list
[02:28:02] <eric_U> that was interesting, I didn't realize that anyone could compare geda to eagle
[02:28:43] <tomp> the discussion was about used gEDA to visually describe Hal configs
[02:29:07] <tomp> and to make the Hal file from the visual
[02:36:39] <cradek> eric_U: trust me, it's supposed to be there, the manual even talks about it
[02:36:57] <eric_U> whazzit for?
[02:37:20] <cradek> it's at the machine origin - you can bore a hole there for some reason
[02:39:15] <cradek> http://www.umaine.edu/amc/images/machines/Series1_07.jpg
[02:39:20] <cradek> hard to find a good picture of it
[02:40:02] <eric_U> machine origin is way off to the left?
[02:40:16] <cradek> yeah left front like normal
[02:40:17] <jmkasunich> that was a pleasant surprise - the Shoptask is capable of more than I've given it credit for
[02:40:32] <jmkasunich> 1/4" end mill, 1/8" deep, 4 ipm, in aluminum
[02:40:47] <jmkasunich> (I've been chickenshit up till now, doing 0.050 or so stepdown)
[02:41:09] <eric_U> what kind of cut?
[02:41:16] <jmkasunich> 0.2" stepover
[02:41:20] <cradek> I bet you really need more spindle speed for that
[02:41:29] <jmkasunich> 2400 is as fast as it goes
[02:42:07] <jmkasunich> thats only 0.00083 per flute
[02:43:27] <cradek> wonder if you can get roughing mills in smaller sizes
[02:43:40] <cradek> I bet that would help too
[02:44:39] <jmkasunich> I don't think a rougher helps until the depth of cut is more than a couple flutes
[02:44:53] <jmkasunich> and most roughers have flutes about 1/8" or so, IIRC
[02:45:01] <jmkasunich> flutes is the wrong word
[02:45:10] <jmkasunich> the ridges or whatever they are called on a rougher
[02:45:11] <cradek> jaggies
[02:52:47] <tomp> teeth ( on flute )
[02:54:07] <toastydeath> a good rougher will help at all depths of cut, but crappy ones do not
[02:54:12] <cradek> yay, max is in one piece and on one parport again
[02:54:27] <toastydeath> the jaggies (i like that) are ground differently on expensive vs. inexpensive cutters.
[02:54:51] <cradek> they help because they chew little horizontal chunks out
[02:54:59] <toastydeath> uh, yes
[02:55:07] <cradek> instead of trying to make a cut that's the entire depth
[02:55:16] <toastydeath> right, that's not at all what i am talking about.
[02:55:20] <cradek> heh
[02:55:27] <toastydeath> i'm talking about the form of the individual teeth on the cutter
[02:55:45] <cradek> I'm still back thinking about what jmk said about them not helping if there's not much depth
[02:55:57] <toastydeath> that's what i'm addressing
[02:56:23] <toastydeath> a crappy rougher has those teeth cut likea thread, and that changes the engagement of the teeth
[02:56:27] <toastydeath> *like a
[02:56:43] <toastydeath> a good rougher has every one of those teeth ground straight, without any pitch to them, and those work great at small depths
[02:56:57] <cradek> I see
[02:57:21] <cradek> but if the work is only as thick as one jaggie, won't one flute/jaggie do all the work since the other flutes won't even touch it?
[02:57:34] <cradek> or do they overlap quite a bit
[02:57:44] <toastydeath> there's lots of overlap on good cutters.
[02:57:53] <toastydeath> one tooth will do more work though.
[02:59:15] <toastydeath> for really, really tiny depths of cut i could see only one doing any work
[02:59:28] <toastydeath> if you cut shallower than the cutter had overlap
[02:59:57] <cradek> I bet jmk could cut several times as deep with one though
[03:00:13] <toastydeath> also tru
[03:01:18] <toastydeath> you have to be careful at that point, because if your cutter has helical teeth it's going to pull out of the collet
[03:01:27] <toastydeath> endmill holders are a better choice.
[03:01:38] <toastydeath> and even with straight teeth it sometimes does that
[03:01:46] <cradek> I think he has Tormach stuff
[03:01:53] <toastydeath> not familiar =(
[03:02:08] <cradek> holders that go in a collet
[03:02:23] <toastydeath> cool
[03:02:29] <toastydeath> i will google it
[03:03:12] <toastydeath> as long as there is a positive locking (not just friction) holding that toolholder in, he will be fine
[03:03:22] <toastydeath> but if it's a straight collet w/ no lip, there can be issues
[03:03:52] <cradek> tormach is for cnc retrofit of R8 machines, there's nothing to keep them from pulling out
[03:04:15] <eric_U> I didn't realize they went in a collet
[03:05:01] <eric_U> http://cgi.ebay.com/Tormach-Tooling-System-for-R8-Quick-Change_W0QQitemZ230221114288QQihZ013QQcategoryZ25295QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[03:07:06] <toastydeath> yar, that will pull out
[03:07:12] <toastydeath> but how big is his machien
[03:07:15] <toastydeath> machine, even
[03:09:42] <jmkasunich> my machine is 3/4 HP
[03:09:53] <jmkasunich> and I'm using a 1/4" cutter
[03:09:59] <jmkasunich> I don't think pullout is gonna be an issue
[03:10:26] <toastydeath> nah, not on 1/4"
[03:10:37] <jmkasunich> heh, something else subroutines are good for
[03:10:44] <jmkasunich> commenting out a block of code you want to skip
[03:10:47] <eric_U> I could manage it
[03:10:59] <jmkasunich> just turn it into a sub with O400 sub / O400 endsub
[03:11:01] <eric_U> that and playing 4 part harmony
[03:11:39] <cradek> O777 ignore / O777 endignore
[03:11:47] <jmkasunich> heh
[03:12:30] <jmkasunich> well, my part cut MUCH faster with these feeds and speeds
[03:12:39] <jmkasunich> need to do another pass tho
[03:12:50] <jmkasunich> I expected (and got) significant tool deflection
[03:13:05] <jmkasunich> I told the program I was running an oversize tool, so it left some metal
[03:13:11] <jmkasunich> now I need to go back and clean it up
[03:14:10] <cradek> nice to see someone using radius comp a lot
[03:14:39] <cradek> historically I think that's not a well tested area of emc
[03:14:45] <eric_U> I need a shop(s) like this: http://www.garagejournal.com/forum/showthread.php?t=2122
[03:18:06] <cradek> I've never seen that much empty space at once
[03:18:22] <eric_U> he fills it up later in the thread
[03:19:39] <SWPadnos> we looked at a place that had a 3-car garage, very deep bays, and a full-size upstairs which was a workshop
[03:19:46] <SWPadnos> oh, there was also a house on the property
[03:20:18] <eric_U> upstairs workshop seems limited without a crane
[03:20:38] <SWPadnos> on 10 acres, and even affordable. too bad it was (a) outside he high speed internet access shpere and (b) in the middle of a 5-acre "lawn" which would have been better named a swamp
[03:21:00] <eric_U> ain't that the way it always is
[03:21:07] <SWPadnos> would have been fine for almost anything not requiring large machiners, which would have fit just fine downstairs
[03:21:26] <SWPadnos> the garage was something like 45 feet deep - just about 3 car lengths
[03:21:35] <eric_U> we looked at a house that would have been very nice, but it was sooooo ugly
[03:21:44] <SWPadnos> and it was "over-wide" too - something like 1200-1500 square feet total
[03:22:02] <SWPadnos> well, this one was a raised ranch, which at the time was our least favorite style
[03:22:11] <eric_U> I like ranches if done right
[03:22:14] <SWPadnos> and of course what we eventually ended up buying ;)
[03:22:15] <eric_U> prefer
[03:22:27] <SWPadnos> ranch isn't bad, as long as there's a basement
[03:22:32] <eric_U> they don't build them much here in PA
[03:22:47] <SWPadnos> raised ranch = ranch with basement that counts as living space - ie no storage space
[03:22:51] <eric_U> walk out basement a must
[03:23:19] <SWPadnos> also, the garage is generally half od the lower level, so you have no ceiling height in the garage
[03:23:24] <SWPadnos> s/od/of/
[03:23:29] <eric_U> you can keep that
[03:23:35] <SWPadnos> yeah, it's annoying
[03:23:56] <SWPadnos> the BP fits in there, but if I have to replace the drawbar, I'll have to re-tram the head
[03:24:08] <eric_U> lots of places charge extra tax if ceiling in basement is higher than 8'
[03:24:42] <SWPadnos> it's 8 feet, but the lower level of a raised ranch isn't considered a basement
[03:25:01] <eric_U> basically a split level?
[03:25:43] <SWPadnos> sort of - most raised ranches have a half-down level and a half-up level, with the front door in the middle
[03:26:11] <SWPadnos> the basement isn't quite deep enough to be a basement, and they finish half of it off and call it living space
[03:26:31] <eric_U> know the type now
[03:26:48] <SWPadnos> so you build a 1500 SF foundation, put a "second floor" on it, finish off 750 SF in the basement (the other 750 is the garage), and call it a 2300 SF house
[03:27:04] <SWPadnos> (that's basically our house - a little larger than a "normal" RR style)
[03:27:48] <eric_U> once my Parkinson's sets in, I'll need a real ranch
[03:28:19] <SWPadnos> or levitation technology
[03:28:33] <eric_U> reminds me, I need to build an exoskeleton
[03:29:04] <SWPadnos> http://www.dekaresearch.com/ibot.html
[03:29:07] <eric_U> every male in my father's family has suffered from parkinson's by the time they reached 80
[03:29:18] <SWPadnos> well, at least 80 is a reasonable age
[03:29:33] <eric_U> true, but why give up then?
[03:29:35] <SWPadnos> if you haven't done anything by then, you probably don't deserve to last much longer ;)
[03:29:36] <SWPadnos> heh
[03:29:59] <SWPadnos> of course I'll cahnge my tune when I get to 60 or 70 :)
[03:30:01] <SWPadnos> change
[03:30:26] <jmkasunich> I'll have to remember that conventional milling deflects the tool toward the work
[03:30:40] <jmkasunich> I've taken another 0.030 off the profile and it hasn't all cleaned up yet
[03:30:43] <SWPadnos> hmmm. I sense an undersize workpiece
[03:30:51] <eric_U> I'd like to make wheelchairs, my BIL had a bad experience with his
[03:31:05] <jmkasunich> I allowed 0.020, so I'm not too far under
[03:31:23] <cradek> 1/4 long is a pretty floppy endmill
[03:33:40] <jmkasunich> yep
[03:33:47] <jmkasunich> about 1-1/4" flute length
[03:34:02] <jmkasunich> the parts where it was cutting full width were the worst
[03:34:22] <cradek> other than pride will the undersize hurt anything?
[03:34:41] <eric_U> that's what _she_ said
[03:34:56] <SWPadnos> hmmm. she never said that to me
[03:35:18] <eric_U> didn't want to hurt your feelings
[03:35:40] <jmkasunich> no, it won't hurt anything
[03:35:51] <jmkasunich> I'm stopping at 0.020 under though
[03:36:05] <jmkasunich> about 80% of the perimeter is cleaned up
[03:36:22] <jmkasunich> that means there was 0.040+ deflection in some places
[03:36:31] <cradek> wow
[03:36:32] <eric_U> scary
[03:36:54] <jmkasunich> funny thing is the mill sounded happy during the cut
[03:37:10] <jmkasunich> made nice chips, no nasty noises
[03:37:18] <cradek> does the endmill still have both points?
[03:37:40] <jmkasunich> I believe so - now that I'm done milling I can take it out and look
[03:38:37] <jmkasunich> nice and sharo
[03:38:40] <jmkasunich> nice and sharp
[03:38:46] <toastydeath> deflection doesn't really hurt cutters
[03:42:16] <gezar> how deep was the cut?
[03:42:41] <gezar> interesting that a 1/4" end mill would deflect that much .040?
[03:43:06] <toastydeath> .040? or .020 a side?
[03:43:28] <gezar> even if it was .020 worth of deflection, thats still ALOT
[03:57:07] <jmkasunich> 0.040
[03:57:19] <jmkasunich> it is a long flute mill, about 1-1/4 flute length
[03:57:28] <jmkasunich> and some of the deflection was probably machine, not cutter
[03:57:42] <[1]a-l-p-h-a> do any of you use remote desktops and multimonitor? (for windows)
[03:59:15] <jmkasunich> if I want to helical mill a deep hole (many revs), I need one G2 or 3 per rev, right?
[04:06:32] <jmkasunich> shit
[04:06:42] <jmkasunich> G0 straight down
[04:09:27] <cradek> oops, hurt anything?
[04:09:53] <cradek> and yes (arcs)
[04:10:15] <cradek> see the hole milling in useful-subroutines.ngc, I bet it's what you want
[04:11:51] <jmkasunich> I had one written already, just had to find it
[04:11:58] <jmkasunich> I apparently botched the touch-off
[04:12:13] <jmkasunich> that _was_ a nice 1/4" long flute center cutting endmill
[04:12:22] <cradek> dang
[04:12:31] <jmkasunich> and that _was_ a flawless vise
[04:13:09] <jmkasunich> fortunately I don't think I ruined the part
[04:13:18] <cradek> good
[04:13:31] <cradek> I bet cnc vises are never flawless for long
[04:23:50] <jmkasunich> heh, I was able to resharpen the end mill well enought that it can helical mill a hole
[04:24:08] <jmkasunich> as long as I don't try to ramp too much
[05:21:47] <eric_Unter> that didn't work so well :)
[05:27:01] <eric_Unter> murphy's law of screwdrivers :(
[05:28:34] <eric_Unter> whoever the jerk was at giddings that sourced this screw, I'm glad he's looking for work
[06:56:18] <toastydeath> http://icanhascheezburger.files.wordpress.com/2008/02/funny-pictures-agnostic-cat-shadows-window.jpg
[08:39:27] <Vq^_> Vq^_ is now known as Vq^
[12:52:13] <BigJohnT> anyone know of a good flowchart for linux?
[12:53:45] <archivist> inside rear of skull
[12:56:41] <BigJohnT> that one is forgetful
[12:57:12] <archivist> then dia iirc
[13:01:46] <BigJohnT> just installed it last night will give it a go this evening
[13:04:02] <archivist> Ive been writing a web based database diagramming tool
[13:04:16] <archivist> similar application almost
[13:04:53] <BigJohnT> I've tried OO draw but it is clunky to try and get things aligned
[13:05:19] <BigJohnT> played with kivo some yesterday but the symbols are lacking
[13:06:24] <archivist> http://www.archivist.info/erd/erd.php?drawing=1
[13:07:24] <archivist> symbols being drawn from the database field names etc
[13:07:35] <BigJohnT> cool
[13:09:00] <BigJohnT> well it's time for breakfast here then off to my shop ttul
[13:14:50] <alex_joni> heh, lexmark is censuring cradek's site :)
[13:15:42] <skunkworks> ?
[13:17:42] <skunkworks> logger_emc: bookmark
[13:17:42] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-02-11.txt
[13:18:08] <jepler> skunkworks: someone on the emc mailing list reports that he can't retrieve a file from chris's website
[13:18:30] <skunkworks> odd
[13:22:58] <[1]a-l-p-h-a> [1]a-l-p-h-a is now known as a-l-p-h-a
[13:24:15] <archivist> must be a problem his end as he says, works for me
[13:45:21] <skunkworks> works here also
[13:48:10] <archivist> some corporates have evil internal firewalls
[13:49:11] <archivist> it could see guy in the url and assume a pornsite and block
[13:50:51] <archivist> AOL had a similar one which affected all emails to a town in England "Scunthorpe"
[13:51:13] <alex_joni> haha
[13:54:53] <archivist> we as clockmakers expect cock in emails so blocking software gets in the way
[14:01:39] <skunkworks> how about cnc'ing this http://www.unimat-1.com/unimat_classic.htm
[14:03:08] <eric_U> don't think it would be worth cnc'ing a unimat. anyone else here surprised that clockmakers are into porn?
[14:04:12] <archivist> cnc a real lathe
[14:04:17] <maddash> porting sai to win32...
[14:09:21] <skunkworks> holy crap - they sell a stepper kit for it..
[14:09:42] <skunkworks> http://www.unimat-1.com/images/unimat_acc_cat.pdf
[14:09:45] <skunkworks> page 3
[14:10:02] <skunkworks> (I don't know german) assuming it is german
[14:11:57] <eric_U> there are U.S. importers
[14:14:41] <skunkworks> a guy here was wanting it for model work
[14:16:17] <archivist> keep you kids off your real cnc, givem a plastic cnc toy
[14:23:03] <maddash> arghhhhhhh
[14:24:47] <skunkworks> http://www.thecooltool.com/produktgruppe.php?language=e&status=20&sh_id=75&ptitel=MetalLine+CNC
[14:25:21] <skunkworks> http://www.thecooltool.com/produktgruppe.php?language=e&status=20&sh_id=73&ptitel=CoolCNC+-+Linux
[14:25:25] <alex_joni> http://www.thecooltool.com/produktgruppe.php?language=e&status=20&sh_id=73&ptitel=CoolCNC+-+Linux
[14:25:33] <alex_joni> darn.. you were faster :)
[14:25:51] <alex_joni> those are really nice guys :)
[14:27:06] <alex_joni> skunkworks: http://dsplabs.utt.ro/~juve/blog/index.cgi/projects/01157974398
[14:32:47] <skunkworks> neat
[14:32:51] <skunkworks> Never heard of them
[14:33:10] <alex_joni> they're from austria
[14:33:20] <alex_joni> maybe you remember a guy called cncuser
[14:33:25] <alex_joni> who did the puppy CD
[14:34:19] <maddash> alex_joni: you work with DSPs?
[14:35:10] <skunkworks> I do remember that.. I was wondering if it was the same guy
[14:35:41] <alex_joni> skunkworks: he used to work with them
[14:35:48] <alex_joni> maddash: I used to at some point
[14:36:04] <fenn> alex_joni: got a spindle on that mill yet?
[14:36:23] <alex_joni> fenn: yeah, a small one
[14:36:45] <maddash> alex_joni: how different is a DSP from a uc?
[14:37:20] <alex_joni> fenn: http://dsplabs.utt.ro/~juve/emc/mycnc/10070328/PICT0001.JPG
[14:37:27] <alex_joni> maddash: the instruction set is different
[14:37:44] <maddash> alex_joni: I mean, performance-wise
[14:37:47] <alex_joni> it's optimized for large amounts of data
[14:38:00] <alex_joni> maddash: depends what you want to compare
[14:38:15] <alex_joni> if you take a DSP which has been optimized for voice codecs
[14:38:24] <maddash> alex_joni: I read that a DSP is just a uc with hardware FP
[14:38:44] <alex_joni> yes .. ish
[14:38:56] <alex_joni> it's absicly very similar to a uC
[14:39:04] <skunkworks> alex_joni: you bought the unumax spindle also?
[14:39:10] <alex_joni> skunkworks: I got it :P
[14:39:54] <alex_joni> here's a slightly better pic: http://dsplabs.utt.ro/~juve/emc/mycnc/10170329/PICT0004.JPG
[14:42:50] <skunkworks> alex_joni: what is your work envelope?
[14:43:29] <alex_joni> tiny
[14:43:34] <skunkworks> aprox?
[14:44:50] <alex_joni> XY : 2.75" x 13" (70 mm x 330 mm)
[14:45:07] <alex_joni> and Z is also roughly 200-300mm
[14:46:05] <alex_joni> skunkworks: sorry.. I lied
[14:46:19] <alex_joni> Table travel: X-axis = 8.38" (213 mm), Y-axis = 4" (102 mm)
[14:48:56] <skunkworks> Thanks
[14:49:51] <archivist> more xy than my build
[14:56:53] <maddash> wooooooooot kernel oopsies are gone (
[14:58:31] <maddash> is it possible to redirect all output from scripts/emc and the emc2 subprocesses (milltask,io,motmod, etc) to a file?
[15:05:14] <alex_joni> maddash: yes
[15:05:40] <alex_joni> (don't ask me how, but it sure is possible :)
[15:05:52] <alex_joni> emc > file.foo 2>file.foo &
[15:05:55] <alex_joni> or something like that
[15:06:14] <jepler> you may need to disable the "normal redirections" in scripts/emc
[15:06:45] <alex_joni> emc 2>&1 > file.foo &
[15:06:48] <jepler> if ! tty -s; then
[15:06:49] <jepler> exec 2>> $DEBUG_FILE
[15:06:49] <jepler> exec >> $PRINT_FILE
[15:06:49] <jepler> fi
[15:07:01] <maddash> jepler: doh!
[15:07:14] <alex_joni> or simply redefine $DEBUG_FILE and $PRINT_FILE appropriately
[15:07:20] <jepler> ^^ the emc script does this by default, redirecting stdout into the 'print file' and stderr into the 'debug file', which is used to display the error dialog
[15:07:49] <jepler> let me also point out that motmod (and other realtime modules) output cannot go to stdout or stderr, it goes to the kernel message log
[15:08:23] <maddash> noooo
[15:08:27] <alex_joni> unless compiled for sim
[15:08:49] <alex_joni> maddash: you can dmesg -c, then dmesg > file.foo
[15:08:52] <maddash> then can I get milltask and other userspace friends to print to the same place?
[15:09:13] <alex_joni> maddash: I'm not sure it's very nice to print to the kernel log
[15:09:57] <alex_joni> maddash: timings won't be consistent anyways.. (order of the stuff printed might differ from actual happenings..)
[15:11:33] <maddash> "2>&1 > foo" is easier, and scripts/emc deletes DEBUG/PRINTFILES after it terminates
[15:16:26] <maddash> brb
[15:23:28] <jepler> huh, you learn something every day. You can use bash redirections to open tcp connections using the format /dev/tcp/host/port
[15:23:47] <jepler> oh darn, the next paragraph says that it's disabled on debian
[16:13:40] <bill20r3> http://es.youtube.com/user/PraeMortem someone's DIY X-Y table, made from old dvd drives.
[16:57:54] <qw> ls
[16:59:57] <alex_joni> this is wild: http://www.youtube.com/watch?v=f2UtfHRdi68&feature=related
[17:00:34] <lerman__> lerman__ is now known as lerman
[17:37:25] <alex_joni> lerman: around?
[17:37:42] <lerman> Yes.
[17:38:56] <lerman> Barely... I just returned from a week in the hospital involving major surgery to repair a problem from a previous surgery. :-(
[17:39:31] <lerman> Sitting on the couch with my laptop sitting on my incision with 30 staples in it. -- But, I'm getting better.
[17:40:14] <alex_joni> ouch.. sorry to hear that
[17:40:24] <alex_joni> hope you get well soon
[17:40:31] <lerman> So, what can I do for you?
[17:40:42] <alex_joni> uhmm.. not much :)
[17:40:58] <alex_joni> there was a german guy around, asking for someone who knows his way around interpreters
[17:41:07] <alex_joni> so I sugested he should bug you when you're around
[17:41:21] <alex_joni> basicly he wants a new interp for heidenhain code :)
[17:41:35] <lerman> And what does that look like?
[17:42:00] <alex_joni> * alex_joni knows nothing about it
[17:42:06] <lerman> And.... aside from him, how many people would like it?
[17:42:23] <alex_joni> oh, I think there's plenty heidenhain users
[17:42:42] <alex_joni> Vq^ here might know more about heidenhain then me
[17:43:01] <lerman> I'd be happy to advise though. If you here from him again, he can email me at lerman dash emc at se dash ltd dot com.
[17:43:17] <Vq^> * Vq^ wakes up
[17:43:19] <lerman> here -> hear (still under anesthesia)
[17:43:54] <quiteBIGeye> * quiteBIGeye kicks Vq^ in the nads
[17:43:58] <quiteBIGeye> 'lu
[17:44:06] <Vq^> hiya :)
[17:44:35] <alex_joni> Vq^: you had some converter and sample .h code
[17:44:39] <quiteBIGeye> Vq^: did you scare that you with crap h2iso yesterday?
[17:46:07] <Vq^> alex_joni: yep, thought quiteBIGeye is the one who wrote the sample code
[17:46:44] <alex_joni> lerman: http://arda.no-ip.org/h2iso/test.hei
[17:48:31] <lerman> Well, that's a long way from a spec, but my guess is that it is doable; but probably not an easy job.
[17:49:13] <lerman> If someone could produce a spec showing the equivalent gcode for each consruct in heidenhain, it might be a reasonable thing to do.
[17:50:13] <lerman> The big issue is whether the canonical Hcode machine is the same as the model that we use. If it is, than it would be a reasonable thing to do (in the sense that I could do it).
[17:50:16] <jepler> if it is a radically different language (and it looks like it is) then it makes sense to point out the "canon" interface, and program a different interpreter to talk to the canon interface
[17:50:17] <alex_joni> lerman: I think the trickiest thing will be converting programmative stuff (procedures & such) to g-code
[17:50:33] <lerman> If not, then it would not be something I would consider doing.
[17:50:35] <alex_joni> jepler: that's what I was talking about..
[17:51:15] <cradek> LOOKS LIKE EVERY
[17:51:17] <cradek> LINE STARTS WITH
[17:51:18] <cradek> L
[17:51:32] <lerman> You can't use the canon interface if the "model" is not the same.
[17:52:03] <cradek> I think canon is basic enough that it represents the basic machine capabilities
[17:52:20] <cradek> that's assuming linear and circular motion are the primitives I guess
[17:52:24] <jepler> that was certainly the intent of canon
[17:54:30] <quiteBIGeye> cradek: sorry but only linear lines begin with L
[17:54:40] <alex_joni> http://filebase.heidenhain.de/doku/tnc_guide/pdf_files/TNC124/bhb/284_679-24.pdf <- contains some examples
[17:54:48] <lerman> But then throw in things about constant time and feed speed. And about commands with mixed rotary and linear axes.
[17:55:41] <lerman> Some of those things are (1) not obvious and (2) not the only way things might reasonably be defined.
[17:55:42] <jepler> if you need new kinds of movement that can't be represented in canon, then you'll also find that they can't be executed by motion
[19:08:58] <skunkworks> * skunkworks wonders how gamma-x is doing
[19:10:14] <alex_joni> catching smoke?
[19:11:38] <skunkworks> hospital..
[19:11:41] <skunkworks> :)
[19:11:48] <alex_joni> hopefully not
[19:11:50] <skunkworks> that might have been harsh
[19:12:02] <skunkworks> how are you doing today?
[19:15:42] <alex_joni> pretty ok
[19:16:25] <skunkworks> supposed to get another 6 inches of snow tonight
[19:16:51] <alex_joni> hmm.. it was close to spring over here
[19:17:01] <alex_joni> we had ~10C last week
[19:17:06] <alex_joni> now it's down to 4-6C
[19:17:51] <lewin1> lewin1 is now known as lewing
[19:18:30] <jepler> skunkworks: yuck, I hope we escape that down here in nebraska
[19:20:43] <skunkworks> it has been quite a winter here.
[19:21:02] <skunkworks> been a while since we this much snow has fallen
[19:23:31] <alex_joni> http://www.youtube.com/watch?v=69FqfVpZVvA&feature=related
[19:23:36] <alex_joni> looks very much like vismach :D
[19:25:52] <skunkworks> saw that - there is one where they are machining a porche also
[19:26:31] <gezar> dangit, soo much work to do tonight :(
[19:26:35] <gezar> how are you guys doing?
[19:28:05] <gezar> I got to go to comp sci class now..ill be watching but I dont know if ill have much time to talk tonight
[19:49:48] <skunkworks> dave~!
[19:50:41] <dave_1> hi there
[19:51:49] <dave_1> hey skunkworks ... whatcha up to?
[19:52:54] <skunkworks> work
[19:52:58] <skunkworks> nothing exciting.
[19:53:04] <skunkworks> well - kitchen
[19:53:12] <dave_1> oh... sometime necessary
[19:53:22] <skunkworks> http://www.electronicsam.com/images/house/kitmud.JPG
[19:53:35] <skunkworks> That hole in the wall is new
[19:54:26] <dave_1> pass thru
[19:54:52] <dave_1> that real tile on the floor?
[19:56:26] <dave_1> when we built 20 yrs ago we made the kitchen/dining room/living room all one ...
[19:56:37] <dave_1> makes it easier to visit while cooking
[19:58:37] <skunkworks> yes - well - it is ceramic tile
[19:59:28] <dave_1> good stuff, cleans up easily ... and always looks nice
[19:59:40] <skunkworks> that is why we did this. when friends are over - they are always bringing chairs into the kitchen to talk. this way the breakfast bar on the other side of the hole will be nice.
[20:00:11] <dave_1> yep! that is why we did ours the way we did.
[20:00:33] <dave_1> otherwise the cook is just a slave with no fun
[20:00:46] <skunkworks> http://www.electronicsam.com/images/house/prefloor.JPG
[20:00:58] <skunkworks> http://www.electronicsam.com/images/house/postfloor.JPG
[20:01:49] <dave_1> that does look a bit familiar
[20:02:15] <skunkworks> really happy on how the tile looks
[20:02:26] <dave_1> indeed
[20:02:37] <skunkworks> it is big 19" square. I think
[20:02:39] <dave_1> but the best way to start a remodel is with a match
[20:02:59] <skunkworks> yes
[20:03:02] <skunkworks> I agree
[20:03:09] <dave_1> then you can put things where you want them
[20:03:10] <skunkworks> this house is so out of wack
[20:03:44] <skunkworks> but that is what you get with a turn-of-the-century house that has been moved atleast once.
[20:03:47] <dave_1> it is difficult to hold tolerances even on a new house but old ones are always out of square
[20:04:27] <skunkworks> did you get your new drives tuned?
[20:04:40] <dave_1> well kinda
[20:05:12] <dave_1> I'm working on homing right now ... on the X once it all runs then I will do the Y
[20:05:45] <dave_1> I'm also going to upgrade from an 80 V supply to 125 so that will change the tuning.
[20:06:25] <dave_1> I just got new brushes for the Mazak
[20:06:34] <skunkworks> I saw that on the list.
[20:06:59] <dave_1> It is a real pain to get the old ones out and cleanup the commutator and then install new brushes.
[20:07:11] <skunkworks> we are going to need brushes for these monsters.. http://www.electronicsam.com/images/KandT/DSC_0242.JPG
[20:07:14] <dave_1> access to two of the brushes is almost impossible
[20:07:18] <skunkworks> 8 brushes per servo
[20:07:30] <skunkworks> easier to take the servo off?
[20:08:36] <dave_1> not in my case ... apparently the gear case is filled with oil and not sealed.
[20:08:43] <skunkworks> yeck
[20:08:48] <dave_1> indeed
[20:09:34] <skunkworks> going to need a 150v supply for ours. Think we have a transformer though (3 phase)
[20:10:10] <skunkworks> (could almost rectify 120) :)
[20:11:47] <dave_1> skunkworks I used these people http://www.schunkgraphite.com/en/sgt/home
[20:12:26] <dave_1> in my case there were slow because of the holidays but the prices are certainly better than going to ... in my case Mazak
[20:12:49] <dave_1> Stede George or Melinda Schall are the contact people
[20:12:57] <skunkworks> Tjamls
[20:13:03] <skunkworks> thanks
[20:13:04] <skunkworks> heh
[20:13:36] <skunkworks> I think they are standard brushes.. 1/4 by 9/16 or something like that
[20:13:42] <skunkworks> hopefully
[20:14:09] <dave_1> ... then you may not need custom .... however the grade of graphite is critical for good results
[20:14:38] <skunkworks> it is inland motor http://www.electronicsam.com/images/KandT/servotag.JPG
[20:16:01] <dave_1> buying from inland may be a good deal but I wouldn't bet on it.
[20:16:08] <skunkworks> heh
[20:16:09] <skunkworks> right
[20:16:19] <BigJohnT_> when an ADC is described as 8-Bit does the output on the 8 pins come out like BCD? http://www.chipcatalog.com/Datasheet/6F3EE1B54C1A704E43F54325D6A5B1C8.htm
[20:16:56] <skunkworks> binary most likely
[20:17:02] <skunkworks> bcd is for us humans
[20:17:08] <dave_1> BJ it should be straight binary unless they say bcd
[20:17:13] <BigJohnT_> from the lsb to the msb by 2's?
[20:17:59] <dave_1> usually sign ..msb .... lsb
[20:19:09] <dave_1> or unsigned ... may depend on how you configure
[20:21:01] <BigJohnT_> I've printed out the data sheets and will try to understand how it works from that
[20:21:16] <dave_1> yep ... you can select 0-10v or +/- 5 ...
[20:21:24] <dave_1> by jumper
[20:22:11] <dave_1> pretty straightforward
[20:22:38] <dave_1> unless your signal is very stable ... you need a S/H in front
[20:22:55] <dave_1> adc's don't do well when the signal changes during conversion
[20:23:24] <dave_1> unless the chip is one of that that has track and hold
[20:25:38] <dave_1> did some of this stuff 30+ years ago. .. converted a P-E Atomic Absorbtion to digital
[20:25:58] <dave_1> 4 KHz a/d ... into a PDP-11
[20:26:21] <jepler> in fact figure 12 on page 6 proposes a S&H circuit using AD582 together with AD570..
[20:27:17] <dave_1> it may be easier to find a chip that is track and hold instead ... but S/H works well and has more flexibility
[20:28:25] <dave_1> 8 bits isn't nearly as touchy as 12 and up
[20:33:19] <dave_1> going to go back to shop and try to get something done. .... see ya later
[20:33:33] <dave_1> good luck on the a/d
[20:38:24] <BigJohnT_> sorry, I had to step away from the computer
[20:40:24] <BigJohnT_> jepler: I see the figure 12 and it looks kinda straightforward
[20:42:29] <fenn> seems like one could just get over-size brushes and chop them down to size
[20:44:19] <skunkworks> as long as the spring and such is the same size
[20:44:29] <skunkworks> /lenght
[20:44:37] <fenn> large brushes have copper pigtails
[20:45:08] <joern23> hi all
[20:45:27] <fenn> hello
[20:45:32] <joern23> hi, is here a lucky emc profi?
[20:45:43] <joern23> i have a big problem with it...
[20:45:44] <fenn> skunkworks: do you have the pulley that goes on the huge motor? how does the triangular keyway work?
[20:46:34] <fenn> joern23: what's a profi?
[20:46:36] <skunkworks> it was direct coupled to the leadscrew
[20:46:55] <skunkworks> huge clamped coupler
[20:47:00] <fenn> i'm just wondering how the keyway works
[20:47:23] <skunkworks> we haven't really decided how we are going to couple it into our machine.
[20:47:33] <fenn> like, are there setscrews pushing on the key? and if so, which direction
[20:47:40] <joern23> I've installed the ubuntu cd. I've configured it with the step tool.... the axes test there works fine. But when I start Axis, there is no movement.... can you help ?
[20:48:02] <skunkworks> I really don't remember. I would have to take a look at the coupler
[20:48:10] <fenn> joern23: can you jog the axes?
[20:48:33] <joern23> @fenn no
[20:49:09] <joern23> it works only with the config tool... with the axes test option
[20:49:13] <fenn> does emc show movement on the back-plot?
[20:49:41] <joern23> yes on the pc the axes move... but not on the machine
[20:50:00] <joern23> but with the config tool the axes move on the machine fine.
[20:50:57] <joern23> any idea where the problem can be?
[20:51:02] <fenn> joern23: could you post your .ini and .hal files on pastebin.ca for us to see?
[20:52:00] <fenn> i've never used stepconf so i'm not going to be much help
[20:52:56] <joern23> i'm can currently not get the data because the machine is not here.
[20:53:19] <joern23> it is a standart stepconf generatet config
[20:53:27] <fenn> ok
[20:53:54] <joern23> do you think that stepconf is missing some parameters?
[20:55:18] <joern23> in Axis there is a button, "override limits" I can't set this button, so i can not override the limits... can that be a reason for the problem?
[20:56:11] <fenn> no, if a limit switch were tripped emc would give you an error
[20:58:15] <joern23> but it can't until now there are no limit switches installed on the machine
[21:00:11] <joern23> is there any chance of help without posting the ini file? If not, I will post the ini files tomorrow
[21:00:48] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[21:02:31] <joern23> In my opinion there is something missing in the config...
[21:09:53] <jymm> tap tap tap... is this thing on?
[21:10:08] <SWPadnos> sqweeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee!
[21:10:11] <BigJohnT_> I ran the stepconf and the only problem I had was I left the e-stop in and it wouldn't run
[21:10:33] <jymm> SWPadnos: I'll take that as a yes =) New irc client, never used it before.
[21:10:50] <SWPadnos> joern23, stepconf seems to work in many cases, so it's hard to tell what the problem is withoug seeing the files
[21:11:13] <SWPadnos> one thing to be sure of - make sure you're loading the correct config when you run EMC
[21:13:47] <joern23> i think that it will load the right files.... but at the moment i'm not able to get at the files....
[21:15:29] <BigJohnT_> I don't think the stepconf creates a desktop start icon does it? You might be using an old config there...
[21:15:45] <SWPadnos> I don't know about desktop icons
[21:18:25] <joern23> everytime i start emc ... it asks for a config... then i choose meine-confi
[21:18:57] <SWPadnos> ok, that should be the right one
[21:20:11] <joern23> i don't understand that... with step conf the machine works fine... but under emc the machine won't move... only on the pc
[21:20:47] <SWPadnos> that is pretty strange, but you should realize that stepconf is a separate program that loads a test environment only
[21:21:06] <SWPadnos> when EMC loads, it uses a different set of parameters to start (ones written by stepconf)
[21:23:47] <joern23> hmm
[21:25:10] <joern23> but it seems that EMC can communicate with the lpt port. Because the emergency stop button on the machine works fine
[21:25:22] <SWPadnos> ok, that's a good thing :)
[21:25:36] <joern23> there must only be a little problem ;-)
[21:25:41] <BigJohnT_> did you have a pin to enable your drive or anything like that?
[21:25:47] <SWPadnos> so what does EMC do (or not do) exactly?
[21:26:31] <joern23> i have a drive with step and direction signal on lpt
[21:26:58] <joern23> on the screen in Axis everything moves .... but only there :-(
[21:27:58] <SWPadnos> well, I think a look at the inni and hal files will be a lot more informative than speculating about what's going on
[21:28:19] <joern23> no i ve only connectet the drives with step and direction ... all other singals are set to not used...
[21:28:28] <joern23> yes i think too...
[21:28:31] <joern23> ;-)
[21:28:55] <joern23> can i install emc without using a real time kernel?
[21:29:39] <BigJohnT_> did you disconnect and reconnect the drives inbetween running stepconf and emc?
[21:29:57] <joern23> I would than install it on my kubuntu system and i would reproduce the config files
[21:30:17] <joern23> ... by the way... sorry for my bad englisch..... ;-)
[21:31:18] <BigJohnT_> your english is better than mine, I'm not from England either
[21:32:05] <joern23> I'm not form England too... I'm from germany ;-)
[21:33:07] <joern23> But I think writing here in English is a good training.... but the problem is that I will never now if there are mistakes ...
[21:33:23] <BigJohnT_> so far so good
[21:34:03] <BigJohnT_> in that case we will let you know if you make a mistake LOL
[21:37:24] <joern23> :-)
[22:05:01] <jlmjvm> alex_joni:are you running a robot with emc
[22:16:38] <alex_joni> good night all
[22:27:13] <jlmjvm> jepler:has emc ever been used on a robot yet?
[22:29:27] <skunkworks> I think you need a space between the : and what you type..
[22:29:42] <skunkworks> like jlmjvm: Hello
[22:30:40] <skunkworks> it has run quite a few hexapods.. I think alex knows a guy that is running a puma robot..
[22:32:54] <jlmjvm> thanks skunkworks,never knew i needed a space
[22:33:09] <skunkworks> here is a test
[22:33:16] <skunkworks> logger_emc: bookmark
[22:33:16] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-02-11.txt
[22:33:22] <skunkworks> logger_emc:bookmark
[22:33:22] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-02-11.txt
[22:33:27] <skunkworks> oh - maybe not.. :)
[22:34:21] <jlmjvm> skunkworks:hello
[22:34:42] <jlmjvm> did that work?
[22:35:20] <skunkworks> heh - I am running the linuxcnc.org irc client.. It doesn't seem to have highlighting of names
[22:35:27] <jlmjvm> k
[22:37:00] <jlmjvm> so a 6 axis fanuc robot arm hasnt been done yet?
[22:38:47] <skunkworks> Not that I am aware of..
[22:39:03] <skunkworks> dave is back
[22:39:26] <dave_1> yep ... with questions ... or is it puzzlement ;-)
[22:39:43] <skunkworks> uh oh ;)
[22:39:48] <dave_1> indeed
[22:40:06] <dave_1> not making sense so I have to presume I have something set wrong
[22:40:28] <dave_1> HOME_OFFSET 0.0
[22:40:42] <dave_1> HOME_SEARCH 0.2
[22:40:51] <dave_1> HOME_LATCH 0.01
[22:41:09] <dave_1> HOME_USE_INDEX YES
[22:41:28] <dave_1> HOME_IGNORE_LIMITS YES
[22:41:53] <dave_1> goes neg and finds home switch
[22:42:14] <dave_1> reverses and stops with following error when it comes off switch
[22:42:37] <dave_1> if I use HOME_USE_INDEX NO
[22:42:49] <dave_1> then it homes coming off the switch
[22:42:59] <dave_1> what am I missing
[22:44:00] <jlmjvm> do your encoders have an index?
[22:44:05] <dave_1> yes
[22:45:31] <dave_1> I can prove with a scope that the index pulse is getting to the interface board ...
[22:45:43] <skunkworks> I am not going to be much help
[22:46:11] <jlmjvm> did you try a minus sign on your home latch yet ex: HOME_LATCH -.01
[22:46:26] <dave_1> yep ... it goes the other way
[22:46:54] <dave_1> and I think still does wierd things or never finds the index
[22:47:32] <dave_1> I assume that the index pulse is too fast to follow with halmeter and Xindex (that doesn't set and stay set does it)?
[22:48:07] <jlmjvm> i could never make my index work properly either
[22:49:02] <dave_1> I suppose I should try to catch it with halscope ... not necessarily easy
[22:49:33] <SWPadnos> dave_1, what hardware are you using for encoder counting?
[22:49:45] <dave_1> jlmjvm what motion board?
[22:49:45] <dave_1> motenc100
[22:49:52] <jlmjvm> usc
[22:50:25] <SWPadnos> ok, then it's unlikely that you will be able to see the index pulse - it goes to the motenc and resets that counter (or latches the count or something) - it may never make it into HAL
[22:50:26] <skunkworks> dave_1: where did you get your servo dynamic drives?
[22:50:54] <dave_1> direct from Servo Dynamics .... talk to Lillian
[22:51:08] <SWPadnos> bbiab
[22:52:24] <dave_1> there is a pin for index ...at the motenc level called out in hal ...
[22:54:40] <jlmjvm> dave_1:did you try to slow your latch speed ,HOME_LATCH 0.001
[22:55:13] <dave_1> yes I did ... I think I got impatient waiting. 0.01 is pretty slow
[22:55:57] <dave_1> it seems a pity to have an index pulse and not use it
[22:56:23] <jlmjvm> yes,would like to get mine working also
[22:57:21] <skunkworks> jlmjvm: I would talk to jonE about it.. seems to me they had the index working correctly\
[22:57:38] <dave_1> I guess I should try to verify that I can see it at the board level
[22:58:24] <dave_1> my board is pretty much the same one they used at Galesburg ... and I think they homed on index.
[22:58:32] <dave_1> have to check the ini.
[22:58:36] <dave_1> brb
[23:00:26] <gezar> yar, made it home, how yall doing?
[23:00:31] <skunkworks> HOME_USE_INDEX = YES
[23:00:54] <skunkworks> what home sequence are you using?
[23:01:02] <skunkworks> HOME_SEQUENCE = 1
[23:01:58] <skunkworks> ^ MAZAK ini
[23:04:43] <dave_1> ah ...
[23:04:48] <dave_1> duh
[23:05:20] <dave_1> I think I need to go look at the chart ...
[23:05:44] <dave_1> I just checked the Mazak demo ini and they don't use the index ...
[23:06:07] <dave_1> maybe that is just stale but it is in the current distro that way
[23:06:20] <jlmjvm> says it does on my mazak demo
[23:06:29] <dave_1> OK
[23:06:32] <skunkworks> I just checked it in the cvs - that is where I got it from
[23:06:47] <skunkworks> http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/emc2/configs/demo_mazak/demo_mazak.ini
[23:06:59] <dave_1> hmmm wonder where I downloaded mine from ?????
[23:07:30] <skunkworks> dave_1: http://www.linuxcnc.org/docs/2.2/html/config_ini_homing.html
[23:08:27] <jlmjvm> skunkworks:doesnt the home sequece number determine the order of homing if you use home all
[23:09:20] <dave_1> I think so ...
[23:09:21] <gezar> question about the steper torque emails recently issued, doesnt voltage have a direct impact on the output torque of a stepper?
[23:09:43] <skunkworks> yes
[23:09:55] <jlmjvm> k
[23:09:57] <skunkworks> * skunkworks is reading the homing part of the manual
[23:13:44] <skunkworks> for the most part.. Voltage gives you higher stop speed - current gives you torque.
[23:14:32] <skunkworks> * Top speed
[23:21:04] <gezar> ah, thanks
[23:27:05] <gezar> oh sweet, ive done rotary broaching in the past