good night all
alex did you read the last paragraph of that page/
jlmjvm: which one?
ill find some info tonite to better explain what im talking about,thanks for the help
jlmjvm: I'm kinda sure I understood what you mean
you're saying that G53 x0y0z0 points to some point in space, regardless of the loaded tool
and loading a tool shouldn't cause any difference with respect to g53
jlmjvm: the company I work for has a laser that has a fanuc om(I think) control. Pretty simple setup though
thanks for the help earlier alex
skunkworks:do they have any cnc mills of any type?
nope. And the machine shop my familly has have pretty old equipment. We are retrofitting an 60's vintage kerney and trecker nc mill.
So - I have not had the experience of tool lenght and diameter offsets. I can't wait
our nc controllers are too old to have that
maybe you will be smart and use a cam package
actually - that is how I had to do it.
every tool lenght/diamter change required a new cad drawing and gcode file.
new cad drawing?
bleh i'll return to fight dpkg-buildpackage in the morning
duh. -nc no clean
offset the drawing - re-output a gcode file
that's not cam
fretless is making those giant parts without cam
"too slow" he says
we can bash out certain parts here without cnc very fast
I can't bash out anything because I haven't made my machine work
eric_My Machine works, as long as I don't want to fabricate anything.
my machine moves, I moved across the room
I gotta take the belts off the servomotors and crank up emc
mine makes gears Im happy
hey guys... greetings
the difference is that I haven't been living right
Jymmmm: did you see the nc file I beamed your way yesterday?
LawrenceG: DCC is disabled here, so I don't think so.
I sent a link on here... I'll have to check the logs.... give me a minute
Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-02-09.txt
if you send me your long/lat I can generate one for you
a great circle map
can you open it in emc?
Jymmmm: anybody home?
as far as your bldc drive.. I still have not finised my project.
yea... too much fun stuff to play with.... I just got a pile of samples from onsemi... now there are even more projects!
I got distracted from real servos when I got some oem750 drives for my stepper shoptask
seems like steppers (with good drives) should be plenty for a shoptask
no problems so far... I would like the servo encoder feedback for manual operation.....
I'd like encoders (with index) for more precise home, and no lost steps concerns
like I say - the oem650s have been great
but I don't use the machine manually at all, in fact I removed both cranks
I have a second machine I am building for pcb use.... xy are servos, z will probably be a stepper
my next major step on the shoptask will be ballscrews
I love the fact that emc can use mixed axis drivers
backlash sucks, even though emc can compensate
you already have z ballscrew though - right?
jmkasunich: yes... I bought the shoptask ballscrew option and it is not that great... still 0..5 or so backlash.. too much flex in some of the machine brackets
flex in X I assume?
thats the worst....
wrapping the nut mount down around the chip guard is very flexy
there are some cnczone people that where very very upset with thier shoptask. Like they didn't have a clue what they where getting.
the Z is original... big rack and pinion although I can still cut pcb's
well, if you are new to machining, you might actually believe the bullshit that shoptask says on their website
I really don't think much at all of shoptask the company, or of JT
any cheap machine is a basket of parts... up to integrator to figure out how to make it work
the machines they sell now aren't even cheap
for some jobs 0.202" is good enough
I paid $1500 for mine in 1998, but now they're asking $5000
(and thats for the manual machine)
pic of one?
I like the shoptask... I have learned a lot.. I think I paid around $4000 with ballscrews and cnc
their 3-axis stepper CNC kit is another $2500
3-in-1 = lathe-mill-drill combo
not bad lathe, mediocre mill, and really bad drill press
wow.. thats a weird looking abortion
* renesis drills with his mill
I drill pcb's
this picture is pretty cool.. http://willepadnos.net/jmkasunich/pc-working-1761.jpg
[02:10:24] <renesis> http://www.darkertechnologies.com/image/audiosam-10.jpg
horrible example =(
i drilled after etch, like some kinda CNC retard
* archivist notes no accuracy spec on shoptask specification page
because accuracy is a function of operator ninja skills
accuracy specs can be misleading anyway
people have been told by 'jt' that the spindle has .0005 runout. That has not been what is actual.
how do you spec things like rigidity
JT is a salesman thru and thru
random rant http://www.cnczone.com/forums/showthread.php?t=51857
(that means lying is congenital)
.0005 runout, cold, on a lucky machine
cleaned an brand new
measured up inside the taper
[02:14:57] <LawrenceG> http://imagebin.ca/view/8fgrLms.html
man, they even lease the tnings
when somebody starts leasing machines, you KNOW they're salesmen first and machinists later
LawrenceG: looks kinda familiar
yup only minor mods so far
is that emc1?
mini display... I now use axis
I've completely rebuilt the Z axis on mine
ah - nice
looks like a deep large milling job will chatter like hell on one of those
they aren't very rigid
and with only 3.3" of Z travel, you either have to use very long tools as in LawrenceG's pic, or raise the work as I do here: http://jmkasunich.com/cgi-bin/blosxom/shoptask/spindle-encoder-bracket-01-07-08.html
I have used up to 1/2 endmill for surfacing, but 1/8 to 1/4 is more comforatble in Al
lathe facing dig in as well?
not really, I've done non-trivial lathe work on it
they have 3.5" Z?
renesis: yes, the quill travel is very short
newer ones can raise and lower the head as well
but mine and LawrenceG's don't do that
even my taig can do 6"
but it looks scary - 4 smallish rods.
now for pcb,s I bolt a 4" thick table extention on top of the y table
the lathe isn't bad
as for the mill - I like it better with CNC than I did before
when manual, I tried to use big cutters at low speed, and make one or two passes
done any deep boring lathe work
it simply isn't rigid enough for that
some - 3.5" deep bearing bores in a gearbox
pic as soon as I find it
I can imagine the tool post leaning in towards the head on boring, Ive had fun on very deep boring
[02:22:59] <jmkasunich> http://www.metalworking.com/dropbox/_2001_retired_files/Worm01.jpg
I too like producing blue chips
I bored the worm bearing seats with both parts of the housing bolted together and clamped to the faceplate
the whole gearbox was made on the lathe, the worm gear was turned on the lathe but the teeth done on my van norman mill
better pic: http://www.metalworking.com/dropbox/_2001_retired_files/Worm04.jpg
can I ask what that was for?
[02:25:38] <jmkasunich> http://www.metalworking.com/dropbox/_2001_retired_files/Worm.txt
^^ wirds that go with the pic
that was my one and only experience with making gears so far
I've made about 10 of those over the last 6-7 years
yes, it is
jmkasunich: have any come back?
very nice :)
looks solid enough to last a lifetime
those things have paid for most of my shop ;-)
I tend to overbuild things
No - I don't get that at all from you.. ;)
the application is simulating the movement of a concrete slab under freeze-thaw cycles, so while it is very slow, there is a lot of force involved
I got involved when the original design bent in the middle - I think it was designed by a chemical engineer
cradek: I got an email thru youtube asking about your maxnc. I sent him to your site.
archivist: any new progress?
just been making gears
noting thing that ought to be changed
and what to finish next
used it for cnc marking out today
I think marking out is a britishism
I is british
you guys are all on drugs about g53
has nothing to do with tool tip or tool offset
so - could you make a hex if you could run the spindle slow enough that your axis could keep up.. using spindle synced motion?
I may have been involved... :)
skunkworks: I was tempted to try that - but if you think about how the tool would be touching the work, it won't be able to cut
if you had a live tool I think it would work fine
like imagine an end mill on its side
the top rake will vary from +30 to -30 degrees as you cut each flat
[02:41:43] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems
this page is good
if something is missing, let's fix it
then people can point users at it instead of guessing
well I didn't mean that to sound insulting
That is ok.
I think "mess with your var file" is bad advice because it teaches superstition ("I don't know how stuff gets in there") instead of gcode ("I should learn how the offsets work")
I missed that part
you could make an oval or similar shape to show off..
ooh ooh - you could make an oval male and female so they would fit togather. ;)
* skunkworks might think spindle synced motion is neat
to see a square turned on a lathe is fun
archivist: did you see the video on the emc-user list?
it would sure be nice to be able to get easy concentric hexes
I saw it at an exhibition
[02:56:04] <skunkworks> http://www.youtube.com:80/watch?v=oGq-9NNmr3o&feature=related
heh it went quiet in here while we watched
funny having the guy speaking bad english in a perfect accent
ssssssh weah hunting wabbits
I'm back to working with the mesa 5i20 again, and have run into a HAL error that I'm not sure how to interpret.
pussy stole one http://icanhascheezburger.files.wordpress.com/2008/02/funny-pictures-cat-bunny-toy-stolen.jpg
this is in attempts to enable homing with index, and, from the looks of it, this gets brought up a fair amount.
tell me all about it, that's what I want to do
I uncommented the index enable HAL code in the motion.hal file
[03:27:59] <acemi> http://www.linuxcnc.org/docs/devel/html/man/man9/sampler.9.html
In this page, is function name true? or it's sampler.N
and am getting an error that the Xindex signal already has one output pin
however, the only location that I find Xindex used is in that index-enable code block.
GNieport1, are you using the dac enable output pin?
[03:30:41] <cradek> http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/emc2/configs/m5i20/m5i20_motion.hal?rev=1.10;only_with_tag=RELEASE_2_2_3
the sample m5i20 configuration already has the index-enables hooked up properly
I also see this file has not been changed since before the 2.2.0 release
[03:32:43] <GNieport1> http://imagebin.ca/view/4lkK8K.html
cradek, I did not alter anything except remove the comment marks
m5i20.0.enc-00-index is the wrong signal name
your file is very outdated then
must be from before 2.2
there was a change
see the one I pasted for the correct hookups
thank you, I dunno how I managed this problem :)
what version are you running?
you just have old stuff, if you have other problems be sure to check the updated sample config files
but you must have kept your old file, it changed from 2.1 to 2.2
since your config is from 2.1 or earlier you should have already read http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?UPDATING
I installed form the CD, however, I had some old files from 2.1.7 that I thought I used as REFERENCE ONLY :)
the change you missed is item 1.1 :-)
YOU WERE WRONG! thank you
LOL indeed, apparently my gedit skills are not strong
oh, we're suppose do read something when we upgrade? I just click the "ok" button
I actually did a 2.2.3 clean install
only for moving a config up a major version (like 2.1 to 2.2)
within a major release set (2.2.x) config file changes are not needed
now it's getting too confusing, I'm going back to surfing porn
hmm, EMC likes the new version better
ah, success! I still need to tie my spindle VFD to the 0-10V analog output. I understand there are some tasty bits in the Mazak Demo config?
I bet so
I don't know if it takes -10 to +10 or 0 to +10 with a separate direction signal, but it does reverse
you need a nice reverse for rigid tapping
my hardware is a bit older, it has 0-10 wiht digital forward,reverse pins
I have a gear change, so the mux block used in Mazak will be informational
at slow speed (300rpm) can you bang the signal from forward to reverse and get a nice controlled reversal?
I'm not sure, it is a manual lever at this point; I plan to add a cylinder to the selector as one of the first machining tasks
lever for reversal? can't the vfd do it?
Oh, I misunderstood, the reverse would be done in the inverter drive
oh you mean the gear change
the gear change is not yet actuated
I was asking if your vfd could handle tapping
to do that, you command a forward speed, emc switches to reverse when it needs to back out, then back to forward
The ramps are fully configurable, I believe it handles a direction change gracefully. At least it does well over RS232 command link
tapping will be great to have on my bridgeport
Would an "S" decel/accel ramp be indicated for a tapping-style reversal?
I don't think it matters, it just has to reverse "pretty fast" and "pretty smoothly"
let me fire it up and try it out
might need a braking resistor if you don't have it
by a fast reversal, are we talking <0.25s?
depends on the rpm - I wouldn't want it to take more than a turn or two to reverse when tapping
with the spindle encoder, I would think that the rate of change would not matter
[03:59:34] <cradek> http://youtube.com/watch?v=JCEwlfJj__A
the encoder lets the axis follow the reversal nicely, but remember during the reversal it continues to tap deeper
hmm, so I need to drill the hole deeper if my hardware is the sucketh
or command a less deep tap cycle
or tap slower
like always, there's just tradeoffs the machinist has to understand
[04:06:09] <cradek> http://www.linuxcnc.org/docs/html/gcode_main.html#G33.1
hmm, I set the accel time to 0.2s and managed to go from ~1200 RPM forward to reverse in 0.5s without a fault. Will taht do? :)
if I could figure out how many turns that is... :-)
sounds great but I bet you will want to tap slower than that
at 300-500rpm maybe you can get down to reversal in one turn
or for thru holes, who cares
it's 4 turns
you have an extra encoder channel on the mesa right?
you've got it made then
yes, lots of projects to cut on the 2-axis.
on my bp the spindle sticks out the top where the drawbar would have been. it will be trivial to put an encoder on it
cradek: does the spindle move up and down when the quill does?
nope there's an internal spline
pretty sure the thing on top doesn't move
so its not really the spindle that sticking out - its the thing that drives the spindle
I agree, it's splined
something that has to do with the top pulley
I don't know what part to call the spindle really
anybody noticed the internets being a little slow?
I thought it was just me, but I ran a speed test while waiting for a page that normally loads instantly, the test was fine
about half the u-tube things I look at load VERY slow, others are normal
cradek, I have a 5HP milacron servo motor that I hope to eventually retrofit to the mill spindle drive
meh, $20 ebay
1.25" shaft :)
that is the step before the pick-and-place robot changes tools
since I don't have a lot of faith that I could fab a real tool changer :)
I suppose I'll get around to all of that about 2015
off to sleep, gnite all
so if I have a servomotor that's big enough to drive my spindle, how hard is it to incorporate that into emc?
you just need a drive to match
have a drive
most drives accept velocity commands - for a spindle you don't need a PID position loop
you can pretty much treat it like a regular motor+vfd combo
do you remember when the digikey catalog was less than 1/2 inch thick?
fenn wasn't born back then :)
now it is like 3 inches thick.
when i started nerding around in the mid 90's it was about 1.5"
heh - I guess that would be mid 80s then.. when I was around 10 years old.
alex_joni: I got a safety relay... let me grab the link...
alex_joni: this thing is pretty fscking cool
alex_joni: For my use, I'll have to add an additional "reset" button, but other than that, it'll work pretty slick for a START/ESTOP
alex_joni: Yeah, it took me a bit to figure out what they were doing, but when you hit BRB, it opens one half, then to reset to the point of hittting start again, you have to reset the other half. So basically once estop is triggered, it's a two step process to restart the system. And if power failed and then was restored, it still won't restart the machine.
yup.. sounds about right :)
Yeah, as long as it's in manual reset mode, and not in auto reset that is =)
that's a dip switch.. right?
It's a single SPDT slide switch inside the LED cover
so unlikely it'll change by itself
Exactly, So unless you RTFM, you wouldn't even know that the cover is removable.
alex_joni: $31 shipped =)
I'd say it's worth it
Oh yeah, thanks for mentioning them to me... besides the safety factor, it'll save a LOT of wiring for the logic aspects too
cool, glad I could be of any help
alex_joni: I think I got lucky in finding one. Since I've never played with one before $200 is a lil ouch on it's own.
fenn: who made this? http://fennetic.net/gingery_machines/index.php?shaper
alex_joni: The only thing I found a bit screwy about it is if you hit the estop REALLY fast, it doens't always disengage the system, but with a twist to reset BRB, it should be fine.
whee this is cool: http://www.prsco.com/media/rotopod_1.mpg
in weighted sum document, it says that wsum.N.sum is unsigned but it generates s32. the same is true for wsum.N.bit.M.weight and wsum.N.offset too
acemi: in a fight between the documentation and the code, the code wins
lol, what a copout!
jepler: unless the code is wrong - don't forget that part
"Defective 30 Day Return Policy" <- ha
EMC Comes with a 100$ Money back guarantee. Just return EMC to your place of purchase for a full refund of your purchase price, no questions asked!
how about shipping costs?
Shipping, insurance, blank cd's not included
100$ money back will be nice
alex_joni: saw that.. Very cool. neat idea
you could mount the computer/amps in the center that spins.. then if you get into it with wireless.. all you would have to couple in would be power. :) then you could rotate the table round and round.
skunkworks: Bruce Metzger, apparently
it was the prettiest example i had found so far (i got permission for the pic)
and now he's cursed by google's indexing of this chat log
heh. Very nice.
the gingery wiki never really took off, and then i took it on a tangent
Oh man... I got some 6ft power strips, and the table I'd like to mount it to is 5ft - DOH!
I'm getting a kernel oops that keeps me from starting emc at boot http://pastebin.ca/897508
new rtai/kernel build?
I think the problem might be with emc, because everything worked fine back in emc2-pre-2.1.5
try running 2.1 again?
wasn't 2.6.17 the "bad" kernel?
some keyboad issues or something
this is 22.214.171.124/rtai 3.5, btw
yes, there have been some experiments with later kernels (than the one on the liveCD), and I thought 2.6.17 had some problems, but 2.6.19/20 didn't
I don't remember the specifics though
also, is this the same compiler you used for pre-2.1.5?
yes, gcc-4.1 for everything
p2.1.5, 2.2.2, rtai, kernel
I note that it's "version 4.1.2 ... (prerelease)"
if you think something changed between emc2.1 and emc2.2, you can verify that by trying to run emc2.1
unless you do that the most likely explanation is that there's something wrong with your kernel or rtai
if the problem's with the kernel/rtai, what should I do?
maddash: the only successfull advice is redo it until it works right
it sucks, but try another version/combination
rtai-3.6 is out..
ideally a combination someone says is working right for them
maddash: how much ram do you have on that machine?
ok, no issues there
what kernel is working for you guys?
dapper has 2.6.15
the best one is 2.6.15 with rtai >3.5 iirc
maybe "best" is not the best name
I'm using 126.96.36.199 on another system
I'll try rtai 3.5/3.6 first, since it takes less time to compile
you are not going to try emc2.1 and troubleshoot systematically?
no, I did that already, and I'm have the same problem
oh.. ok, so it's more likely it's a kernel issue
what's the point where it oopses?
maybe there's similars on google
I'll need to recompile emc if I use a different rtai, right?
since rtapi deps rtai
maddash: hmm.. if you only recompile rtai
without repatching the kernel, it will work without recompiling emc2
emc2 modules (kernel modules) need to have the same magic as the running kernel (same compiler, and version)
funky; rtai just starts and stops now
wow - I have worn the plastic writting of the bottom of the laptop.
eric_U: emc2 on a 486? are you crazy?
you could do it
big problem is the UI
shoulda mentioned that would be for servos
atleast.. :) and a degree in linux hackin
no, people use RTAI on brain-dead hardware
that is what turbocnc is for..
I have real-time stuff running on old hardware because of the ISA bus
and it works just as well as dos
dos has all those crazy memory limits and the networking stinks
dont ever tell me that turbocnc works well.
I thought it was ok?
well, my circles became ( ) with it..
on the same hardware that liked 3d contouring with emc..
skunky, you like my edit?
I ran it for a while. I only really had issues with g2/3 feeds. (well and CV is non existant) reason why I went to emc
eric_U: nice :)
I must admit.. I ran emc1 for a good week before finding emc2 ( at the time you had to build it on the bdi install) It was pretty new at that time. Issues with accelleration and such - but they where fixed very fast.
actually - about a year before that I tried the bdi.. I just could not get it going. So I gave up. 2nd time was the charm.
I remember compiling & fighting about 2-3 weeks before I ran emc for the first time
* skunkworks doesn't feel so bad now.
none of the BDIs worked on my hardware
either graphics was not working, or the CD wouldn't boot, etc
I remember the second time - going thru 4 diffferent video cards before it would boot
I ended up doing my own kernel+rtai.. that took a while
(this was a 300MHz-ish machine..)
omg, what if 188.8.131.52/3.5 don't work either?!
so a kernel compiled in roughly 20-24 hours
way way back - dad printed out a pdf 94' maybe explaining emc. At that time - I really didn't 'have a clue'
it doesnt take that long, more like 8 hrs
unless you have everything turned on
fenn: this was a geode SBC, and back then I think I compiled all modules & co
wait pdf's in 94? whatever.
I left it working at work, came back the second day.. it finished
i dont think emc was on linux until like '96
* skunkworks was just guessing.. I really don't remember. Seemed like a long time ago.
it was printed directly from nist though. It talked about NT hardware.
alex_joni: have not heard back from nist about the video. Might try to email a few other people there.
i think my problem is that i used distcc to compile 184.108.40.206/3.4
which probably was a big mistake, since the helper box was using gcc 4.2
alex_joni: wtf.. your system is super slow for a 300mhz:er
hm, wait - distcc incompatibilities manifest as compile-time errors, right?
anonimasu: it's a single board computer
about 50% of the stuff in there is in the processor, and the rest is in the northbridge :)
lalalalala la la la la diriguim diribom diriguim diribom diriguim bom diriguim bom diriguim rai rai rai bim bombom diriguim bom diriguim bom diriguim rai rai din bom digdigdig digdigdig dig din rai bin bombom digdigdig digdigdig rai bim rai bim bombom!
[18:54:53] <skunkworks> http://youtube.com/watch?v=yzyTtts2S_U
alex_joni: have you gone insane?
[18:56:44] <skunkworks> http://youtube.com/watch?v=oVDFbaVyYSQ
* fenn recognized the mind-annihilation virus before it could do any damage
yahoo mail just jumped the shark, you can't scroll through email messages
wow, I can read the header
jlmjvm: did you ever find me a 13 tooth timing pulley?
the 1 the guy emailed me was a 13 tooth pulley,however when i looked at the pic it has a different tooth pattern than i have ever seen
lemmee find that email
disgusting but funny
its been deleted,i can get him to resend monday.It had grooves running in both directions,kinda looked like a 13 tooth roughing end mill
probably wouldn't work for me then
maybe I should try to find a reasonably priced 26 tooth pulley
i didnt think it would work,looked like it would eat a belt up
guess it depends on the shape of the teeth
alex_joni:all is well now,after i hooked up the x and y motors and switches i can move in mdi
thanks for trying though
no prob,wish it had been what was needed,i am needing 1 also
alex_joni: funny :)
it appears that if all axis arent homed out you can move in mdi
jlmjvm: you can do most anything if the axis aren't homed
but that's a "feature"
I know some controls won't allow you anything else but jogging while not homed
i couldnt move in mdi before with just the z axis hooked up,but after i hooked up x and y everything started working,figured it was a safety feature
just turned machine off and on again,homed only the z axis,issued a mdi move and get error,home all axis and it will move in mdi
jlmjvm: you're running special kins?
dunno,just did a new stepconf yesterday,where can i check that at?
if you are running special kins you would know it
everything appears fully functional at this point
just letting ya know what was stopping my mdi movement
these pacsi steppers are much smoother than my original steppers
freakin' kernel compile takees forever
i know ;)
for what do you compile an kernel?
anyone know how much a dvi to hdmi converter will cost me?
why does iocontrol.cc allocate emcErrorbuffer even though it never uses it?
secondly, is it possible to detect a call to emcoperatorerror() made by emctaskmain from iocontrol's emcerrorbuffer?
maddash: regarding the first question - all NML interfaces have the same buffers - CMD, STAT, and ERROR I believe
regarding the second question, I have no clue
maddash: check the buffer it opens
if it's the same one, then yes
maddash, for what do you compile an kernel?
(same one meaning the same *.nml reference)
dumb question of the day What is NML?
"neutral messaging language" - it's the glue that connects the GUI and the rest of emc
and there's some internal communication
google for nml gets you northwestern mutual LOL
try site:linuxcnc.org nml
fretless85: for AIUR!
fretless85: actually, for EMC2
maddash, what the heck is AIUR?
fretless85: you'd have to play starcraft to know
maddash, actually ive played diablo1+2 and WOW :P
just curios...ive tryed to compile an kernel for my 64bit ubuntu 7.10 and failed ;)
how'd you fail?
cant get emc to run...
maddash, thats the error log http://mitglied.lycos.de/fretless85/emc2errorlog
ouch, a segfault? I can't help you there, because I have roughly the same problem myself
nice to hear ;)
if you solve it, let me know
I'm having some strange datum issues with my new homing system, anyone here know anything about homing and datums?
after you home - it isn't where you think it should be?
I can home, machine coords are fine, the red bounding box is fine, it limits jogs just as it should
but if I touch-off and set a coordinate to 0 the result is as though I had entered a nonzero value, something like 17.3mm (I can't remember the value). the values are different for different axes
ie. I touch off and set z to 0, axis displays the current coord as z=-30, the live plot shows the tool at -30
that happened to me hmmm let me remember
I think the stepper.var had a value in it from running one of the sample programs
It was an offset I think
blew my mind
I can imagine
do # 3 http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems
then see if everything acts normally
good one for the day, just changed each opto and buffer on my breakout board one by one trying to solve a problem with some outputs
skunkworks711; hmm ok, thanks
slap on forhead, the e-stop is in nothing will work
now which opto is bad...
skunkworks that is what got me once after running one of the sample programs that ship with EMC
16:33:20 <lerneaen_hydra> I can imagine
16:34:09 <skunkworks711> do # 3 http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems
16:34:17 <skunkworks711> then see if everything acts normally
16:34:38 <BigJohnT> good one for the day, just changed each opto and buffer on my breakout board one by one trying to solve a problem with some outputs
16:34:46 <BigJohnT> nothing worked
16:34:52 <lerneaen_hydra> skunkworks711; hmm ok, thanks
16:35:05 <BigJohnT> slap on forhead, the e-stop is in nothing will work
16:35:28 <BigJohnT> now which opto is bad...
16:36:52 <BigJohnT> skunkworks that is what got me once after running one of the sample programs that ship with EMC
maddash; having fun?
we can scroll up no need to repaste
crap - middle clicked
I'm so ashamed
now you embarrased him :)
heh poor maddash
oh, he'll be back
yea! all my outputs work
time for some pics & vids
time to machine the rest of a stepper adapter tube
time to start drinking
regarding that touch off thing somebody asked about above?
lathe or mill?
it behaves differently based on whether or not you have a program loaded, and axis loads a program (the EMC2-AXIS logo) by default at startup, lathe doesn't
this is for a mill
all I can suggest is make sure there aren't any offsets in effect (other than the one you just set with touch-off)
hmm ok I'll take a look at it in detail
[21:53:39] <jmkasunich> http://www.youtube.com/watch?v=hjVXbEUZJJs
freaky - at abou 0:40, its threading and drilling at the same time
the work is moving in Z for threading, and the drill is chasing it
skunkworks711 is now known as skunkworks
* skunkworks starts to run
* skunkworks keeps looking over his shoulder and doesn't see anything..
and i thought TCPM on my heidenhain is kind of spooky o_O
hmm nice toy, sliding head with live tooling
tpcm? tpm for mills?
tool center point management
thats the name for it on the heidenhain
what's tool center point managment?
keeping the tool rectangular to the contour on head movements
lets say you move to x100 y100 z100
and you turn the head
about 90deg in A
with M128 (TCPM)
and the head kind of stays on the point
X and Z follows
and its B cuz you spin Y
kind of confusing ;)
i dont use it that often
* archivist wonders if the boss will donate the sliding head auto in the basement for a cnc sliding head project
nice one, coming to emc one day!
tool poin management is important
for 3d milling? yea
as i said i dont use this often
because i dont need it
fretless85; you mean "standard" 4-5 axis kinematics?
we only can do that on one machine so...
the rest got heads with hirth teeths
oh, is that an "advanced feature" on commercial cnc's?
what do you mean with advanced feature?
something out of the ordinary
a feature that costs a lot maybe
it cant be done with orthogonal heads
oh, no, of course not, the hardware bit is quite interesting
you cant move other axis or mill while the head is spinning
because its unclamped and very unstable at this moment so no need for tcpm
we got an zayer with free positionable head
that one can do tcpm
[22:25:45] <fretless85> http://www.iberimex.de/main.html
[22:25:54] <fretless85> http://www.iberimex.de/german/produkte/fraeskoepfe/universal/positionen.jpg
its not that bad
but it moves when you push it real hard
because its hydraulic clamped
no hirth teeth
but orthogonal heads got some contras to ;)
[22:29:34] <fretless85> http://www.iberimex.de/german/produkte/fraeskoepfe/pinole/pinole_xxl.jpg
thats the best one for power :P
jmkasunich, has anyone put emc on a sliding head?
care to join #cam and talk gcode generators?