#emc | Logs for 2007-11-27

[01:12:02] <cradek> mdynac: how is your emc edm machine?
[01:12:33] <mdynac> it is fine, just sitting there doing nothing.......
[01:12:56] <cradek> that was a nice leaf you cut out, way back when
[01:13:17] <cradek> I hope that wasn't the last thing it made!
[01:13:18] <mdynac> i am going to build my own personal multi purpose table top machine
[01:13:27] <cradek> that sounds fun
[01:13:33] <mdynac> the leaf was nice....
[01:13:33] <cradek> edm or milling?
[01:13:39] <mdynac> both
[01:13:50] <cradek> when you say general purpose, you are serious then
[01:14:00] <mdynac> yes sir
[01:14:19] <mdynac> i live in a small apt, need space
[01:14:53] <mdynac> i see no reason why one machine bed cannot do more than one function
[01:14:54] <cradek> I have been playing with 5-axis stuff lately. I am trying to figure out how to build a small 5-axis mill or retrofit my tabletop 3-axis machine
[01:15:25] <cradek> that's very ambitious for me...
[01:16:03] <mdynac> 5 axis wire edm is quite common
[01:16:28] <cradek> can you see youtube?
[01:16:30] <DanielFalck> cradek: great job on the 5-axis visualization vid
[01:16:44] <mdynac> if i go there....
[01:17:03] <cradek> http://www.youtube.com/watch?v=fWKYQUj5AOs
[01:17:22] <cradek> DanielFalck: I did the code in emc to run it -- jepler is the one smart enough to make a video and get it on youtube!
[01:17:46] <DanielFalck> that just blows everyone away!
[01:17:52] <cradek> but yes the 5 axis work is very promising!
[01:18:32] <mdynac> quite nice
[01:18:43] <cradek> 5 axis tool length compensation is done and works great (but not well integrated into emc yet)
[01:18:55] <cradek> we need to work on a few things before they can be merged
[01:19:53] <cradek> you can run that sample configuration with visualization if you want. just check out and build the w_tool_length cvs branch
[01:24:19] <jepler> (actually I think I was running trunk...)
[01:25:25] <dmess> nice work CRADEK
[01:25:38] <mdynac> any pluto p people out there?
[01:26:02] <dmess> there is ALOT of potential and real work in that demo
[01:26:28] <DanielFalck> jepler: what's that bit of code on your web site that is just underneath the video?
[01:26:38] <jepler> DanielFalck: probably a mistake
[01:26:41] <dmess> im just a stranger her on this place called earth...
[01:26:59] <DanielFalck> jepler: the 'void URLDecoder' stuff
[01:27:12] <jepler> what URL?
[01:27:19] <DanielFalck> http://emergent.unpy.net/
[01:27:24] <DanielFalck> front page
[01:27:35] <jepler> oh -- that's a piece of terrible code I found on the internet
[01:27:40] <DanielFalck> "As seen on the internet"
[01:27:48] <jepler> its purpose is apparently to convert from hexadecimal to decimal
[01:27:59] <DanielFalck> ok
[01:28:49] <jepler> but it works in a crazy way (by turning the input into a string which is a decimal number) and includes the letter 'g' which is never used as a numeral in hexadecimal (just as the decimal system goes from 0-9, hexadecimal has the 16 numerals 0..9a..f)
[01:29:30] <jepler> here's where there's more information about the 5-axis functionality that chris is working on: http://axis.unpy.net/01196094602
[01:30:49] <jepler> watching that video (yet again) makes it clear that the machine visualization should use a contrasting color for the plot...
[01:31:03] <DanielFalck> ok, it took me a while to get the joke... I see the 'g' at the bottom now
[01:31:09] <cradek> we have colors now - we could "pretty up" the machine
[01:31:23] <jepler> cradek: yeah and you wanted to show some more "stuff" as well
[01:31:25] <jepler> like a "tool change"
[01:31:36] <cradek> yeah that would be nice
[01:31:43] <jepler> I am sure happy to record another video
[01:31:50] <cradek> if you do that, use the branch - now it can change TLO at any orientation
[01:32:08] <cradek> on trunk the tool has to be upright to not jump joint position (which is not enforced)
[01:32:09] <jepler> OK I'll be sure to do that
[01:32:41] <cradek> do you still have the better cone program?
[01:32:46] <jepler> yeah I think so
[01:33:30] <cradek> we should work together on a good demo program before you bother to make another video
[01:33:41] <cradek> this one is not very cool - it was only for testing various things
[01:33:44] <jepler> http://emergent.unpy.net/files/sandbox/cone.py
[01:34:12] <dmess> use APT360 for the code source too
[01:34:21] <cradek> heyyyyy we could fix up 3D_chips
[01:34:30] <jepler> that would be interesting
[01:34:32] <cradek> we could approximate a surface normal at every point
[01:34:45] <skunkworks> that would be cool :)
[01:34:46] <jepler> I fear it's too long for a video ...
[01:34:53] <cradek> yeah it couldn't show all of it
[01:35:09] <skunkworks> 10 min max. I ran the it on the hermes 3 axis in less than that..
[01:35:13] <cradek> or, maybe we could cut out some middle stuff
[01:35:16] <skunkworks> youtube limit
[01:35:17] <dmess> hyper speed the video
[01:35:20] <jepler> I have a feeling that an interesting video will not be one that would give a finished part
[01:35:31] <cradek> that's probably very true
[01:35:38] <cradek> let's think about it.
[01:35:46] <dmess> dream
[01:35:52] <cradek> (but I'd like to see that with 3Dchips anyway)
[01:36:21] <skunkworks> http://www.youtube.com/watch?v=Hdhn_j6PrCw
[01:36:28] <cradek> maybe someone could generate something with APT and see if it's at all compatible
[01:36:30] <skunkworks> less than 4
[01:36:36] <dmess> i could run APT vectors over it easily enuf
[01:36:51] <dmess> i have..
[01:37:06] <dmess> it is
[01:37:13] <jepler> it's easy enough to speed up the video by whatever factor is desired, too
[01:37:49] <jepler> not so easy to make different parts run at different rates, though I am sure I could put it into a format that is suitable for video editing software instead of a format for youtube
[01:38:30] <dmess> there is a 4 axis post available in 1 day no less... and AXIS showed it incorrectly.. although the code was correct
[01:38:45] <jepler> yeah axis won't show those very well, either in preview or backplot
[01:39:00] <jepler> somebody should work on that *cough*
[01:39:33] <dmess> *cough*.. hack.. *cough*.. i have the flu
[01:39:59] <dmess> it looks like someone has been
[01:41:04] <jepler> the backplot you see in that video is unfortunately separate from axis
[01:43:13] <dmess> really// how so??? or from where??
[01:44:21] <jepler> a Python program has a description of the machine's geometry, including how each joint moves (rotates or translates) parts of the machine.
[01:44:36] <jepler> it is hooked up to the HAL joint commands
[01:44:54] <jepler> it was created to visualize various kinds of nontrivial machines.
[01:45:27] <jepler> other pictures of machines created through the same method: http://axis.unpy.net/01170621073 http://axis.unpy.net/01170693566
[01:46:30] <jepler> there's also a hexapod visualization
[01:47:05] <jepler> two examples of what the machine descriptions can look like (they're just python programs that use a poorly documented module): http://cvs.linuxcnc.org/cvs/emc2/src/emc/usr_intf/axis/scripts/5axisgui.py?rev=1.6 http://cvs.linuxcnc.org/cvs/emc2/src/emc/usr_intf/axis/scripts/pumagui.py?rev=1.3
[01:47:16] <dmess> nice stuff
[01:47:56] <jepler> by using some other source of data besides hal, you could use it to visualize trajectories from any other source you like
[01:50:03] <jepler> somebody should document the python 'vismach' module, though I notice that pumagui.py has a nice explanation of what most of the steps are doing as they assemble the machine
[01:57:04] <DanielFalck> jepler: can you tell dmess about that opengl g-code material removal program that you came up with? I recall using it last year.
[01:57:13] <jmkasunich> hi guys
[01:57:18] <DanielFalck> hi john
[01:57:47] <jmkasunich> jepler: cool movie
[01:59:00] <cradek> hi jmk
[01:59:06] <jepler> DanielFalck: well, I can point at the code and images: http://axis.unpy.net/01169521961 http://axis.unpy.net/01169689661
[01:59:55] <jepler> it works for 3-axis milling .. for a grid of predefined size, it tracks the highest bit of material that hasn't been milled away by the tool movements
[02:00:25] <DanielFalck> thanks
[02:02:12] <cradek> that's still very cool
[02:03:28] <jepler> for 5-axis it's a much tougher problem .. you have to keep track of whether each "particle" of a grid of predefined size has been milled away.
[02:03:43] <jepler> or use some more sophisticated method altogether
[02:04:39] <jmkasunich> one of these days I want to mill this:
[02:04:40] <jmkasunich> http://members.home.nl/peregrine/The%20Labyrinth_bestanden/labyrinth.gif
[02:05:03] <cradek> jmkasunich: just use your rotary table...
[02:05:07] <jepler> this library could be a good choice to start: http://www.opencsg.org/ though I don't know if its performance continues to be good after your CSG operation is "remove 10000 cylinders and 10000 spheres from a volume)...
[02:06:04] <jmkasunich> cradek: sure, piece of cake
[02:06:13] <jepler> (and 10000 spheres/cylinders--one for each distinct position that you place the tool on, along the total path) is still a very small amount of milling
[02:06:33] <cradek> jmkasunich: how's your motor mounting progress?
[02:06:49] <jmkasunich> what progress? all I did was bore the pulleys
[02:07:03] <jepler> (hm, that makes me wonder what the circular version of a peano curve is..)
[02:07:04] <cradek> what's next?
[02:07:16] <jmkasunich> the Z axis can be mounted at any time, the Y axis I just need to measure for and order the right belt
[02:07:21] <jmkasunich> X is another story completely
[02:07:31] <jmkasunich> the as-designed location of the motor won't work
[02:07:47] <jmkasunich> it was designed with old round motors in mind, and interferes with a square frame motor
[02:07:48] <cradek> X is the lathe tool post/milling table?
[02:08:02] <jmkasunich> X is Z (when you think lathe)
[02:08:10] <cradek> ok the long one
[02:08:22] <jmkasunich> right, and Y is X
[02:08:29] <jepler> (nice -- opencsg is gpl)
[02:08:31] <cradek> right
[02:08:36] <cradek> jepler: cool
[02:08:45] <cradek> jmkasunich: can you just get a motor that fits?
[02:09:17] <jmkasunich> 1) I already have these motors 2) they have significantly better torque-speed performance compared to round frame motors
[02:09:25] <cradek> hm
[02:09:34] <cradek> that long one is the one you especially want to be fast too
[02:09:41] <jmkasunich> yep
[02:10:54] <cradek> http://www.ktmarketing.com/sttips.html
[02:11:08] <cradek> "if your gib strips are not flat, take a belt sander to them"
[02:12:20] <jmkasunich> heh
[02:13:32] <jmkasunich> hmm, its not quite as bad as I thought
[02:13:42] <jmkasunich> (still some interference, but not as much as I thought)
[02:15:25] <jmkasunich> looks like it can be made to work, but I'll need to re-route the lathe spindle motor cable
[02:16:39] <cradek> that doesn't sound bad...
[02:16:52] <jmkasunich> that is a huge relief - I was envisioning needing to redo that whole mount
[02:19:16] <jmkasunich> now I just have to get the X axis drive pulley to fit over the end of the leadscrew
[02:19:28] <jmkasunich> must be a burr or something
[02:19:52] <cradek> sounds like you're pretty close to making it move
[02:20:38] <jmkasunich> well, if I cobble things together temporarily maybe
[02:21:02] <jmkasunich> proper cables, connectors, nice sealed box, all that takes lots of time
[02:21:19] <cradek> yep
[02:21:21] <jmkasunich> but maybe cobbling things is a good way to get started
[02:21:48] <cradek> it's rewarding to see some life from it
[02:22:15] <jmkasunich> yep
[02:22:26] <jmkasunich> and some of those "proper" things will be much easier with CNC
[02:22:50] <cradek> you can never go back...
[02:23:06] <jmkasunich> like plates to go over a cutout in the main box and hold connectors
[02:23:30] <cradek> yeah that stuff is easy
[02:23:53] <cradek> like cutting out a hole for a DB25. simple to get the right shape with cnc.
[02:25:34] <jmkasunich> heh, no wonder this pulley won't go on the shaft
[02:25:36] <jepler> http://lib2geom.sourceforge.net/ -- * Exact elliptical arcs * Computations such as offset curves can be written with their mathematical definition and still get a bounded error, efficient curve.
[02:25:52] <jepler> it would be nice if the license is OK -- sf says LGPL / MPL 1.1
[02:27:14] <jmkasunich> as close as I can measure, the pulley bore is 11.99mm and the shaft is 12.01mm
[02:27:24] <cradek> http://lib2geom.svn.sourceforge.net/viewvc/lib2geom/lib2geom/trunk/COPYING?revision=615&view=markup
[02:28:25] <jmkasunich> isn't the LGPL even more liberal than the GPL?
[02:28:28] <cradek> yes
[02:28:57] <cradek> jmkasunich: pretty sure the hole has to be bigger than the shaft for it to work...
[02:29:10] <jmkasunich> somebody forgot to tell the chinese that
[02:29:18] <cradek> so, all you need is a 12.02mm reamer
[02:29:28] <jmkasunich> (the pulley is shipped with the machine, so you can do the CNC conversion, but it is not shipped installed)
[02:31:05] <jmkasunich> I don't have a 12.anything reamer
[02:31:16] <cradek> yeah I was being a bit silly there
[02:31:25] <jmkasunich> (well, I guess I have as 12.7mm one, but it is labeled with proper units)
[02:32:40] <jmkasunich> do I dare try boring it?
[02:32:54] <cradek> sure, but only if you can center it
[02:33:07] <jmkasunich> got a 4-jaw
[02:33:17] <jmkasunich> and unlike the motor pulleys, this one has a long hub
[02:33:28] <cradek> I think I'd wrap fine sandpaper around a piece of rod and gently attack it
[02:33:49] <cradek> it'll stay concentric that way, even if it's not centered in the lathe
[02:33:50] <jmkasunich> yeah, I've been pondering that
[02:34:06] <jepler> trying to compile this C++ software, I am suddenly feeling that 256MB is inadequate..
[02:34:20] <jmkasunich> I bet it will take quite a while to remove 0.0015 or so inches though
[02:34:34] <cradek> jepler: when I was your age ... well, it was probably last year sometime, forget it
[02:36:07] <cradek> jmkasunich: I bet if you blue it and sand off the blue 3-4 times it will fit
[02:36:34] <jepler> cradek: when you were your age, you stayed away from template-heavy C++ code, and rightly so
[02:36:38] <jmkasunich> sanding the inside of a 0.472" dia x 1.something hole isn't easy
[02:36:49] <jmkasunich> but I'm gonna give it a shot
[02:44:30] <jmkasunich> heh, where's a sunnen hone when you need one?
[02:52:18] <jmkasunich> seems to be working, opened it up by half-a-thou
[02:52:32] <jmkasunich> more at the end, gotta be carefull I don't bell-mouth it
[02:52:44] <cradek> that's why the blue
[02:53:46] <cradek> it's amazing they sent you these two parts that don't fit together. or maybe the leadscrew has "swelled".
[03:04:01] <jmkasunich> the blue wouldn't have helped - the bore is at least twice as deep as it is round, and hard to see into
[03:04:17] <jmkasunich> but I have a small hole gage, and can judge taper pretty well by feel
[03:04:20] <jmkasunich> and.....
[03:04:23] <jmkasunich> it is done!
[03:04:42] <jmkasunich> light press fit - I can put it on and remove it by hand, barely
[03:05:51] <cradek> nifty
[03:12:15] <jmkasunich> now I have to figure out what length the tooth-belts need to be
[03:12:37] <cradek> 3 insulated walls isn't nearly as good as 4 insulated walls
[03:12:55] <cradek> certainly not 3/4 as good
[03:13:26] <jmkasunich> ?
[03:13:33] <jmkasunich> cold in the shop?
[03:14:04] <cradek> yep
[03:14:11] <cradek> no insulation in the new wall yet
[03:14:15] <jmkasunich> ah
[03:14:20] <cradek> maybe tomorrow.
[03:14:24] <jmkasunich> but the new wall abuts the garage, right?
[03:14:42] <jmkasunich> (the new garage)
[03:14:44] <cradek> yes but the garage is at outside temperature
[03:15:00] <cradek> it's not quite sealed yet
[03:15:02] <jmkasunich> the other three walls abut the house? or the outdoors?
[03:15:12] <cradek> one house, two outdoors, one new garage
[03:15:33] <jmkasunich> unless you have a heat duct into the shop, its always gonna be cold
[03:15:42] <jmkasunich> heat from house to shop, two walls
[03:15:49] <jmkasunich> heat from shop to outside, two walls
[03:15:50] <cradek> sure but it was much better...
[03:16:14] <jmkasunich> better when there was a garage door there?
[03:16:46] <cradek> yes, it's a very good door
[03:16:46] <Jymmm> jacuzzi in the shop after a hard day milling... priceless
[03:17:10] <cradek> heh, 45 degrees now
[03:17:19] <cradek> that's warm compared to outside though.
[03:17:47] <Jymmm> Geeese where do you ppl live in an igloo?!
[03:17:58] <jmkasunich> we live where there are seasons
[03:18:07] <cradek> ha
[03:18:11] <toastydeath> fffff
[03:18:24] <Jymmm> When 45 is warm, I think I can live w/o the seasons
[03:19:05] <jmkasunich> 45 does seem cold for an enclosed space, it's 39 outside here
[03:19:17] <jmkasunich> what did you sheath that wall with, windowscreen?
[03:19:24] <Jymmm> lol
[03:19:34] <cradek> currently it's one layer of drywall only
[03:19:46] <Jymmm> how thick the insulation going to be?
[03:19:47] <cradek> (32 outside here)
[03:19:54] <jmkasunich> well, 3.5" of fiberglass will make it nice and toasty
[03:19:56] <cradek> normal 3.5"
[03:20:16] <Jymmm> Ah, I'd like a home with 6" exterior walls
[03:20:20] <cradek> and then another layer of drywall because you can't leave the insulation uncovered
[03:20:25] <jmkasunich> what about the ceiling? is there heated space up there, or unheated attic?
[03:20:33] <cradek> unheated but insulated
[03:20:43] <jmkasunich> thats good
[03:20:55] <jmkasunich> I've been thinking about adding insulation here
[03:20:57] <cradek> talk about a lot of work
[03:21:06] <cradek> putting 12' drywall on a ceiling...
[03:21:07] <jmkasunich> 1928 house, with haphazard insulation attempts over the years
[03:21:18] <cradek> yep, had one of those
[03:21:35] <cradek> never really warm everywhere inside, but sometimes warm in a few places
[03:21:44] <Jymmm> In a place I was renting, after 5 years I insulated the attic (blown in), my only regret, I didn't do it 5 years earlier. cooler in the summer, warmer in the winter.
[03:21:46] <cradek> (same for cool in the summer)
[03:22:32] <Jymmm> only cost like $225 for like 12"
[03:23:50] <cradek> that does sounds like a good deal
[03:25:14] <Jymmm> free machien rental after buying 10 bags at home depot
[03:25:49] <jmkasunich> 20 bags here, but still, its only $6 a bag
[03:26:42] <Jymmm> And pay the neighbor kid $50 to crawl around for ya
[03:26:58] <Jymmm> It'll be the EBST $50 you've ever spent
[03:27:01] <Jymmm> BEST
[03:27:30] <Jymmm> the shit goes everywhere... ears, nose, butt cracks, everywhere
[03:28:38] <jmkasunich> I wouldn't trust the kid to spray the stuff in the right place
[03:28:48] <jmkasunich> I have a very cramped and odd-shaped attic
[03:28:56] <jmkasunich> (the back roof is almost flat)
[03:29:16] <Jymmm> jmkasunich: you have a good back, right?
[03:29:32] <jmkasunich> I just bought a tyvek jumpsuit with a hood, for crawling around up there without getting filthy
[03:29:45] <jmkasunich> reasonably good back, yes
[03:29:58] <Jymmm> jmkasunich: cause you wont have a back left after your done for a week or so
[03:30:10] <jmkasunich> I can believe that
[03:30:21] <jmkasunich> I'm not really looking forward to the job at all
[03:30:53] <Jymmm> Also, get a light but strong 10ft pole to attache the end of the hose to.... helps BIG TIME for getting into tight spots
[03:31:11] <jmkasunich> already thought of that - I'll need a 15 foot pole
[03:31:14] <Jymmm> I thick I used a 10' 2x2
[03:31:21] <Jymmm> think
[03:31:21] <cradek> good idea - if you think there isn't much room to crawl somewhere, there isn't
[03:31:29] <jmkasunich> that back part, you simply cannot crawl into the space above the back bedroom
[03:32:20] <Jymmm> and never stop feeding the machine.... cut in half dump, break apart bug chunks
[03:32:29] <Jymmm> big
[03:32:36] <jmkasunich> what happens if you stop? it clogs up?
[03:33:02] <Jymmm> jmkasunich: no, you jsut into a rythem, and then it mucks you up a bit
[03:38:02] <cradek> jmkasunich: did you figure out what belts to order?
[03:39:15] <jmkasunich> not yet, got distracted
[03:44:36] <Jymmm> * Jymmm hangs shiny thing on a sitck and hovers just in front of jmkasunich
[03:44:47] <jmkasunich> ohhh, shiny
[03:45:00] <Jymmm> =)
[03:46:23] <Jymmm> I just bought a serial port controlled o/off switch 120VAC@7A for $4
[03:46:54] <Jymmm> requirements... DOS 3.31 of higher
[03:50:17] <cradek> I often see stuff with requirements like "Windows 2000 or better". These are very puzzling because I don't think they work for me.
[03:59:19] <jepler> the lib2geom path offsetting stuff is far from finished
[04:00:16] <jepler> so far they have "find a curve that is tangent to the given curve", but it doesn't give good results for milling applications. green is input path portion, purple is offset path. http://emergent.unpy.net/index.cgi-files/sandbox/offset-test.png
[04:00:47] <Jymmm> alex_joni: wheres that video of your tri motor wire thingy?
[04:00:49] <jepler> the person I chatted with thinks that this code wouldn't work on input paths with square corners
[04:03:43] <cradek> darn
[04:03:57] <Jymmm> anyone remember the video of alex moving the washer around?
[04:05:16] <cradek> I remember seeing it
[04:06:10] <Jymmm> darn, can't rememebr the url
[04:06:14] <jepler> http://dsplabs.cs.utt.ro/~juve/emc/mytoy/ from the wiki
[04:06:22] <Jymmm> ty
[04:19:38] <jepler> 'night all
[11:40:32] <alex_joni> JymmmEMC: videos around emc2 are at: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Videos
[16:26:54] <alex_joni> http://www.proview.se/ <- this is REALLY interesting
[16:46:54] <cradek> no, it can't be a serious project, their logo is done in comic sans
[16:50:04] <cradek> Microsoft admits that the launch has not gone as well as the company would have liked. "Frankly, the world wasn't 100 percent ready for Windows Vista," corporate vice president Mike Sievert said in a recent interview
[16:50:58] <alex_joni> wanna see something funny?
[16:51:34] <alex_joni> open: http://www.microsoft.com/windows/
[16:51:56] <alex_joni> then look at the information of the main picture file (the one talking about windows Vista + windows Live..)
[16:52:10] <alex_joni> "Orientation of image: 1
[16:52:10] <alex_joni> File change date and time: 2007:11:01 14:05:34
[16:52:11] <alex_joni> Software used: Adobe Photoshop CS3 Macintosh"
[16:53:27] <cradek> all I see is a big Flash turd, no images
[16:54:36] <alex_joni> http://www.microsoft.com/windows/images/hero_xp.jpg
[16:56:30] <cradek> not even microsoft could find an ad agency that uses windows...
[16:57:26] <alex_joni> http://dsplabs.cs.utt.ro/~juve/blog/index.cgi-files/sandbox/win-loss.PNG
[16:57:36] <skunkworks941> heh - on windows xp - I cannot get any info on it ;)
[16:57:48] <alex_joni> skunkworks941: sure you can :D
[16:57:53] <alex_joni> but you need a smarter browser
[16:58:01] <alex_joni> IE and Firefox don't give you much
[16:58:13] <skunkworks941> heh
[17:06:50] <alex_joni> bbl
[17:09:02] <skunkworks941> cradek: here you go - for your bridgeport http://www.5-axis.com/
[17:29:57] <cradek> I so want that
[17:30:16] <cradek> but I'd need much more travel especially in Y
[17:33:50] <skunkworks941> yes - you would be a bit limited. It would be cool though ;)
[17:34:27] <skunkworks941> they have videos
[17:37:02] <cradek> huh, 20krpm with a face mill
[17:37:14] <cradek> looks like about 3"
[17:39:30] <skunkworks941> where do you see that?
[17:39:48] <cradek> http://www.hsk.com/vids.htm
[17:40:10] <cradek> and a comic sans slogan
[17:40:21] <cradek> with drop shadow
[17:41:51] <skunkworks941> heh
[17:43:12] <cradek> I think that's 15k sfm
[17:43:39] <cradek> I wonder where you would even find a face mill that's rated for 20krpm
[18:12:22] <dmess> you make them yourself...
[18:49:03] <jlmjvm> can axis display different tool types also,like a ball mill,or drill point?
[18:50:25] <jepler> no
[18:50:39] <cradek> the tool table doesn't have that information
[18:51:02] <jlmjvm> k,was looking at the gdepth screen shots,thought maybe it did
[18:51:56] <cradek> gdepth is actually unrelated to AXIS
[18:52:34] <jlmjvm> its a standalone,correct?
[18:54:24] <jepler> yes
[18:54:39] <jepler> instead of using a tool table, it gets tool information (diameter and tip shape) from special comments in the gcode file.
[18:54:54] <jepler> or maybe it was from the commandline; my memory is going
[18:54:59] <jlmjvm> what is the FMS for in the tool.tbl
[18:55:13] <jepler> it's a number that's not used for anything in particular
[18:55:20] <jepler> somebody somewhere knows what FMS stands for, but he's not telling
[18:55:49] <jlmjvm> k,was wondering what it was for
[18:55:49] <cradek> it's one of the small mysteries that keeps us going
[18:57:32] <cradek> we're not putting you on - we really don't know
[18:58:27] <jlmjvm> cool,my nasa buddie asked me last nite and i didnt know what it was
[18:58:57] <jlmjvm> however that gives me an idea
[18:59:12] <skunkworks941> cradek: the 5 axis machine is fun to jog around :)
[19:00:10] <cradek> skunkworks941: I think teleop mode isn't working very well - I'm surprised you can jog it
[19:00:22] <cradek> often the rotary axes run away for me
[19:00:41] <jlmjvm> could that possibly be a place for axis to get its tool display dia,and the control get the cutter comp number from the Dia slot in the the tool.tbl
[19:00:48] <skunkworks941> it did lock up on me :)
[19:01:08] <cradek> jlmjvm: they both use the same number (diameter)
[19:01:50] <jlmjvm> but does it have to
[19:02:05] <jlmjvm> is really what im wanting to know
[19:02:17] <cradek> I don't understand why you would want them to use different numbers
[19:02:54] <jlmjvm> then the cool display wouldnt change if you used a small comp number
[19:03:15] <dmess> Ive been there.. need 2 offsets
[19:03:39] <cradek> oh you want to use a small number to offset an already-offset tool path?
[19:03:48] <jepler> AXIS gets its information about the tool table from the the "stat buffer", but emc doesn't put the FMS number there
[19:04:07] <jepler> so, axis cannot use the FMS number
[19:04:14] <dmess> maybe it should???
[19:05:16] <jlmjvm> was mainly just wondering if it was possible,next project after the tool button
[19:05:37] <jepler> in ioControl.cc the fms number is treated as an integer, so it would break the existing tool parsing code to write a fractional number such as .0625 there.
[19:07:56] <jlmjvm> well sounds like its a dead end run
[19:10:15] <jlmjvm> once we get the button working correctly will you be able to show me where to read the z axis position
[19:15:18] <jepler> I can try
[19:18:48] <jlmjvm> cool,that will be stage 2,were gonna get it writing to the table first,then gonna add in the offset button
[19:19:39] <jlmjvm> thats when we will prolly need to ask ya a question or 2
[19:20:23] <jepler> here is an example: http://pastebin.ca/799580
[19:20:42] <jepler> I commanded 'G0 X1 Y2 Z3', then ran python in the terminal and typed in those commands
[19:21:38] <jepler> s.position is the position in machine coordinates, s.offset gives the net offset in effect from the active coordinate system (g54 in this case), tool length compensation (if in effect, it wasn't for me), and g92 (if in effect, it wasn't for me)
[19:22:22] <jlmjvm> thank you very much,gonna save and forward this
[19:22:25] <jepler> so by subtraction, you can see that the Z position is 3.0 above the offset origin.
[19:22:26] <jepler> >>> s.position[2] - s.origin[2]
[19:22:26] <jepler> 3.0
[19:23:08] <jepler> also, you have to issue a command to emc after writing the new tool table to cause emc to read it. You can do that with these lines:
[19:23:11] <jepler> >>> c = emc.command()
[19:23:14] <jepler> >>> c.load_tool_table()
[19:23:29] <jlmjvm> excellent info
[19:24:12] <jlmjvm> gonna save all this conversation in a tex file and the pastebin also
[19:24:37] <jlmjvm> had no idea how to get to this stuff
[19:26:50] <jlmjvm> so if the last command was issued when the offset button was pressed the tool.tbl would be reloaded automatically?
[19:27:37] <jepler> yes that should be the case
[19:27:48] <jlmjvm> icing on the cake
[19:32:15] <jlmjvm> really appreciate that info
[19:36:32] <jepler> use it well
[19:39:01] <dmess> its like duct tape.... it has light side and a dark side...
[19:45:00] <jlmjvm> im getting an error when i type the commands in terminal,its on th >>> s.poll line
[19:46:10] <jlmjvm> http://pastebin.ca/799617
[19:46:16] <jepler> type "python" in the terminal
[19:46:19] <jepler> its prompt is ">>>"
[19:46:25] <jlmjvm> i did
[19:46:28] <jepler> if you are writing the lines in a .py file you would strip off the ">>>"
[19:47:04] <jlmjvm> its in the paste
[19:47:09] <jepler> ok looking
[19:47:11] <jlmjvm> k
[19:48:08] <jepler> looks like I cut too many corners in showing you the example
[19:48:30] <jlmjvm> ouch
[19:48:38] <jepler> after 'import emc' you have to set the emc.nmlfile to the one specified in your inifile
[19:50:58] <jlmjvm> >>> emc.nmlfile = emc.nml
[19:51:35] <jlmjvm> is that what you meant?
[19:51:41] <jepler> no that won't do it
[19:52:01] <jepler> it would be more like: emc.nmlfile = "/home/jlmjvm/emc2/something/emc.nml"
[19:52:15] <jlmjvm> ahhh,gotcha
[19:52:34] <jlmjvm> will give it a whirl
[19:54:27] <jepler> you'd do that before the line 's = emc.stat()'
[20:00:04] <JymmmEMC> yo
[20:03:15] <jlmjvm> k,its giving me an invalid syntax
[20:05:10] <jepler> >>> import emc
[20:05:11] <jepler> >>> emc.nmlfile = "/home/jlmjvm/emc2/something/emc.nml"
[20:05:20] <jepler> this is accepted by my python
[20:05:46] <jepler> you want to show me exactly what you typed?
[20:07:16] <jlmjvm> never mind,i had taken the quotation marks off like a dummy,lol
[20:09:26] <JymmmEMC> you dummy! ;)
[20:09:31] <jlmjvm> lol
[20:10:19] <jlmjvm> after s.poll() it says this :emc.error: emcStatusBuffer invalid
[20:10:29] <jepler> did you use the correct path instead of the one I gave you?
[20:10:39] <jlmjvm> but it did take the nml line
[20:10:45] <jlmjvm> yep
[20:10:49] <jepler> emc is running?
[20:10:53] <jlmjvm> yep
[20:11:46] <jepler> you did it all (emc.nmlfile = "...", s = emc.stat(), s.poll()) in the same python interpreter?
[20:13:40] <jepler> if axis works then this ought to work .. it's doing the same steps, but axis does them in a more complicated way
[20:14:24] <JymmmEMC> online coupon.... front brakes labor only $89, called and asked how much pads are $89 WTH
[20:14:46] <jlmjvm> got it,it was me again,was giving it too much of a path
[20:15:44] <jlmjvm> emc.nmlfile = "/home/jlmjvm/emc2/configs/joe/joe.ini/emc.nml" replaced with "/home/jlmjvm/emc2/configs/joe/emc.nml"
[20:16:00] <jlmjvm> showing numbers now
[20:16:35] <jlmjvm> yall are gonna have to nickname me fumble fingers
[20:16:56] <jlmjvm> or fat fingers mcghee
[20:17:09] <jlmjvm> but its working now
[20:18:24] <JymmmEMC> * JymmmEMC does a /ForceNickChange jlmjvm FumbleFunglyFingers
[20:19:48] <jlmjvm> maybe i shoulda took that mavis beacon class
[20:21:06] <jlmjvm> jepler:thanks again,works fine
[20:21:45] <jepler> good
[20:22:07] <jlmjvm> can see the position in python,too cool
[20:23:16] <jepler> as you are working on this, remember that s.origin includes the offset in effect due to tool length compensation if it is active, so you will want to make sure that tool length compensation is not active when using s.position[2] - s.origin[2] to find the length of a tool.
[20:23:42] <JymmmEMC> jepler: You had a good eye last time, see anything amiss with this by chance? http://www.cnc4pc.com/Store/osc/product_info.php?cPath=33&products_id=45
[20:24:57] <jepler> JymmmEMC: I don't have time to look at that right now :(
[20:25:09] <jepler> in fact I should have been doing my job instead of helping jlmjvm
[20:25:18] <cradek> All inputs are outputs are tied to pull-down resistors.
[20:25:19] <JymmmEMC> jepler: within the next 48hours by chance?
[20:25:33] <JymmmEMC> cradek: is that bad?
[20:25:34] <cradek> specs need proofreading.
[20:25:44] <jepler> I think cradek just offered to do it for you
[20:25:47] <cradek> I don't know what it means
[20:26:07] <JymmmEMC> cradek: I suspect so the signlas aren't floating
[20:26:21] <jepler> look at the words again: "all inputs are outputs"
[20:27:20] <JymmmEMC> s/are/and/ ?
[20:28:34] <JymmmEMC> ok, assuming that's a typo, anything else?
[20:29:27] <dmess> i think they can be either inputs OR outputs.. like that
[20:29:40] <JymmmEMC> My parker drives already are opto-isolated, and I plan on using SSR's
[20:29:42] <jlmjvm> i have 2 of those boards
[20:30:15] <JymmmEMC> jlmjvm: any issues?
[20:30:42] <JymmmEMC> or complaints?
[20:32:09] <jlmjvm> had a problem with g203v geckos using it,but when i was using it as a 2nd parport to read encoders i had no problems
[20:33:20] <jlmjvm> the campbell board i just used on a boss bridgeport had no issues with the 203v drives
[20:34:55] <jlmjvm> the cnc4pc board can be used for inputs or outputs,and has a jumper now to change some of the pins from 5v to grd
[20:36:22] <jlmjvm> but the campbell boards are way better for running a machine
[20:36:39] <jlmjvm> its heavy duty
[20:37:34] <jlmjvm> reminds me i was supposed to email him and tell him how well it worked with emc and the 203v drives
[20:39:25] <jlmjvm> JymmmEMC:what are you gonna turn on with the ssrs
[20:40:51] <jlmjvm> the cnc4pc will send all outputs high when the computer turns on,so beware if you have a spindle or anything like that on the outputs
[20:42:01] <jlmjvm> thats when i put in a board with a charge pump,nothing goes high when emc isnt outputting its signal,works great
[21:06:57] <JymmmEMC> jlmjvm: thanks for the info, that's exactly what the SSR's are for... spindle and dust collection.
[21:08:47] <JymmmEMC> Anyone have an idea what front brake job should go for these days?
[21:09:06] <cradek> in california? no clue
[21:09:27] <JymmmEMC> cradek: ok , where your at
[21:09:30] <skunkworks941> Do it yourself.
[21:09:39] <JymmmEMC> need the rotors turned
[21:09:41] <cradek> with rotors turned, a few hundred
[21:09:59] <cradek> replace the pads in the garage and hope for the best: $50
[21:10:15] <JymmmEMC> the fucking stock pads never squeeled, now down to metal.
[21:10:25] <cradek> ouch
[21:10:48] <SkinnypuppY34> Some don't have the squeel wear indicator on pads
[21:10:52] <JymmmEMC> only 38K miles too
[21:11:10] <cradek> if you need to do it on the cheap, see if you can take the rotors off and take them somewhere to be turned
[21:11:14] <JymmmEMC> really pisses me off, pads are easy to replace, WHEN YOU KNOW THEY NEED TO BE!
[21:11:18] <SkinnypuppY34> That blows
[21:11:58] <JymmmEMC> Nah, I can take it in, going to have the bearings repacked too, an extra $15 labor
[21:12:44] <JymmmEMC> So far i got a quote for ~ $160 including turning
[21:12:48] <JymmmEMC> front only
[21:12:55] <JymmmEMC> disc all around though.
[21:13:05] <JymmmEMC> I miss drums on rear
[21:13:06] <cradek> the backs probably need pads
[21:13:10] <JymmmEMC> yeah
[21:13:25] <cradek> they probably wear 60-80% as fast as the front
[21:13:55] <JymmmEMC> Well, if no wear indicator, I'll get them replaced too, before having to replace the rotor on them too
[21:15:08] <SkinnypuppY34> Have any of you used screw micrometers with the replaceable anvils for different pitch? Do these measure the pitch dia directly on the scale? Or measure to the thread root?
[21:15:10] <SkinnypuppY34> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=140182159595&ssPageName=ADME:X:RTQ:US:1123
[21:15:25] <cradek> SkinnypuppY34: pitch diameter
[21:15:29] <cradek> thread root means nothing
[21:16:21] <cradek> but they cost a fortune. get wires :-)
[21:18:31] <SkinnypuppY34> Yeah I know ya could go broke buying starrets size and pitch range .
[21:18:52] <cradek> wires are not too hard to use once you practice a bit
[21:20:39] <SkinnypuppY34> Ok, I was going to order some enco stuff, I'll grab a set of those wires also.
[21:21:00] <cradek> yeah they are cheap and work just as well (assuming you have a good micrometer already)
[21:21:36] <SkinnypuppY34> Yeah got a new set of starretts , who said making straight A's don't pay off.
[21:22:41] <SkinnypuppY34> Our local school had a program called "TOOLS" where the highest gpa's were given tools of their trade. I probably got 1/2 of my tuition and gas money back
[21:25:37] <cradek> cool, only the UPS guy gives me tools (when I use my credit card)
[21:44:51] <dmess> ive used wires to make GUAGES... and was always closer than the mic's...
[21:45:39] <SkinnypuppY34> dmess, cradek thats all I needed to hear. Thank you both
[21:45:55] <dmess> GOOD FOR YOu.. you should talk to my son's... maybe learn em something... CUDO's
[21:46:28] <SkinnypuppY34> ;o)
[21:47:05] <dmess> MASKING tape 2 together... the a tag/tail on the other... they are a biatch to find in the chip tray
[21:47:26] <SkinnypuppY34> I bet !
[21:47:47] <SkinnypuppY34> I was thinking of those tiny rubber bands used on dental braces
[21:48:27] <dmess> experience comes from bad judgement... good judgement comes from experience...
[21:49:06] <dmess> TOO small... TAPE.. with sizes on em.. once used good for life..
[21:49:58] <dmess> just put them back in the case with the chart.. ALL together.. they like to sleep that way..
[21:50:37] <SkinnypuppY34> All my tools like to sleep at their home, no camping round the shop
[21:51:00] <dmess> thaT'S ABOY.. ;)
[21:52:25] <dmess> i usedta could LOOK at my t/b drawers and inventory missing stuff..then tell ya where i left this stuff.. where's the rest???
[21:57:49] <dmess> SkinnypuppY34.... where are you located??
[21:58:31] <SkinnypuppY34> Near Macon Georgia, right in the middle of the state. Yourself?
[22:00:03] <dmess> Oshawa Canada... Toronto area..
[22:00:53] <dmess> sounds like you have a very good tech program at your school... I had one too.. yrs ago...
[22:03:28] <SkinnypuppY34> It's pretty good. They could use a bigger budget, but a lot of that was going into a new building when I was there
[22:10:18] <dmess> what grade??
[22:11:58] <SkinnypuppY34> Ah . Technical school , not High School
[22:12:02] <dmess> i managed to graduate college with an over-all 4.0.... but was a solid c+ average in h/s
[22:12:47] <dmess> they neglected my red seal on the diploma too...
[22:13:03] <SkinnypuppY34> Know the feeling. I just didn't much care about h/s other than math and science.
[22:14:11] <SkinnypuppY34> Not that other things are not important it was just a hard sell in other subjects
[22:14:47] <dmess> likewise... we still had grade 13.. i hit 1 english.. 3 maths and chem and physics .. all but english at an enriched level.... worked full time in the mine too..
[22:15:44] <dmess> mad some $$$ for a few comfy tyrs at college
[22:17:01] <dmess> tutored there for $5/ahead/hour... paid by the college... they supplied the room i retaught the classes... ; )
[22:21:10] <SkinnypuppY34> I can't tell ya how many people I taught trig , new guys. If they didn't get past that it just wasn't going to fly. The guys I laughed at were the ones going to electronics b/c "the math is too hard" ... just you wait.. RLC circuits coming to mind
[22:24:57] <skunkworks941> going to electronics because the math was too hard? odd
[22:25:05] <skunkworks941> did they not have a clue?
[22:25:42] <dmess> we started with 150 in sept...40 came back after xmas......18 made it into the 2nd yr.....9 came back....4 were left at xmas - 2nd yr....2 graduated... me and a very good friend..
[22:26:18] <skunkworks941> heh - our electronic class started with 40 - I think 3 graduated
[22:26:42] <skunkworks941> associate degree
[22:28:18] <dmess> these #'s were going to shut down the program... so we merged with another gov't sponsored course.... THEY got paid to be there....
[22:29:23] <SkinnypuppY34> skunkworks no they didn't have a clue
[22:30:52] <dmess> not skunkworks project..... i wish..
[22:32:14] <SkinnypuppY34> Oh refering to skunkworks941 "did they not have a clue"
[22:37:33] <dmess> gotya
[23:09:19] <skunkworks> the math was the fun part for me at the time. We lost a bunch though
[23:22:59] <SkinnypuppY34> http://video.google.com/videoplay?docid=5533833308228879027
[23:23:49] <dmess> math is the ONLY absolute... its correct... or NOT
[23:29:34] <LawrenceG> JymmmEMC: http://imagebin.ca/view/ph6a2l.html ... found a chunk of heatsink for the drives
[23:31:39] <dmess> ive made H/S from billot.. what do you need... a anice parametric 360APT program might prove some worth to all this..
[23:32:40] <dmess> e-mail dims etc and get g-gode...
[23:35:01] <LawrenceG> hi dmess ... I did use emc to face off the heatsink... it was warped about 1/16"... have room for 4th drive when I find one the right price
[23:36:25] <LawrenceG> I really need a fly cutter... facing with a 1/2" endmill takes a while
[23:36:40] <dmess> once you figure out HOW warped they will get its not too bad...
[23:38:10] <dmess> we had a single pt tool doing the tops of it all... 1 sweep... low and slow.... no vibrationst
[23:38:28] <dmess> sp
[23:39:05] <dmess> do you hav abbar??
[23:39:26] <dmess> sp again... im FLUed
[23:39:57] <LawrenceG> gotta cut down on those meds!!!! :}
[23:42:26] <dmess> chicken soup.. an jonny de'kyper gin..... its really SHITINABOTTLE.... it should be a pharmacumitiacl... and sold it the drug store
[23:44:00] <dmess> but my great grampa kept the plague out with it so since then... YUOU just DO it.... in our family...
[23:44:22] <LawrenceG> dmess: heard from the outlaws yet?
[23:45:07] <dmess> nope.. maybe it WASNT their ship.... can t figure it
[23:45:37] <LawrenceG> strange...
[23:45:45] <dmess> pops sister called us.. first
[23:46:26] <LawrenceG> we had first snow of the year last night.... just enough to turn the grass white... still icy today