is axis supposed to look at the value of [DISPLAY]LATHE, or does its mere presence, even if set to zero, result in lathe mode?
I thought LATHE=0 would get me mill mode but it doesn't look that way
whee -- now on the arduino you get 6 ADC, 6 PWM DAC, and 6 digital pins that can be used as ins or outs
jmkasunich: if you write LATHE=0 then the result of inifile.find is the string '0', which is not the empty string, so it is treated as a 'true' value
defensible behavior? maybe not ..
documented behavior? (I dunno, I just guessed at what it might do and tried it)
it just suggests including LATHE = 1
* jmkasunich removes and replaces the Z axis motor for the 4th time today
(awkwardly located screws and all)
it doesn't say what LATHE = 0 means but I sure can see your expectations were violated
would "MODE = LATHE" and "MODE = MILL" be more of a pain to implement?
oh not particularly
page up and down are the Z jog controls aren't they?
my mini keyboard doesn't have those keys :-(
hm that could be a problem for you then
of course in lathe mode, left/right is Z and up/down is X
you could put some lines in .axisrc to change the bindings. For example, bind_axis("bracketleft", "bracketright", 2) would make '[' and ']' move Z.
which is fine right up until you get a rotary axis
I was gonna ask about rebinding things
I'll probably have a jogwheel long before I have a rotary, so thats OK
now I'm finding another annoyance (probably operator error)
this sort of thing happens when you actually spend time being an operator
I have home_offset for Z = 0, I home it, it finds the switch and the home icon turns on, but the display shows 1.3something
I tried g92.1 and homed again, no change
try using the menu to clear G54 offset?
that did it
thats stored in the var file, right?
(I copied my lathe var file)
now to put limits in for Z (this is the first axis with a switch, so limits are new....)
jmkasunich: did you figure out the 1.3 offset?
I think it was in the var file
have you seen http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems#So_if_you_re_lost_what_should_you_do
good link, thanks
now I have another weird thing going on
my first homing worked (at least the moves were all in the right direction)
now its going the wrong way (and into a hard stop, ouch)
I didn't change search vel, I wonder if it thinks its on the switch
* jmkasunich fires up halscope
did you see the 5-axis rigid tapping in that last vid?
I don't see a link
oh I see it
just one at the very end
duh, figured out my homing issue
max limit needs to be greater than min limit ;-)
ze quill, she works!
SWPadnos_ is now known as SWPLaptop
woo! you have a mill?
SWPLaptop: you missed me saying "ze quill, she works!"
homes and limits and all that neat stuff
and it's semi-portable ;)
soft limits that is - I'm going to do without limit switches
(raltive to the Van Norman or a BP)
700 lbs in the basement - NOT portable
gah - it's hard to type on a laptop
semi, semi :)
on my lathe I use a home switch for one X limit
my quill home switch trips 0.044 before physical end of travel
I've set the offset to -0.040, so I can use that extra travel
wow that seems pretty close
(the total travel is only 3.3", so every little bit helps)
heh, < 1/20" seems like "a little bit"
but I guess if it lets you machine something that's 1" thick with the right tool, instead of something that's 0.95" thick, then it really does helop
you think the switch will repeat within .004 for a long time?
since you don't have index...
I just stuck a piece of 0.005" feeler gage stock between the banjo and the bottom of the casting - it never got pinched during a home and a rapid to zero from 3" away
(the 0.044 measurement is a bit conservative, might be more like 0.048 or so
interesting point though, I can do a repeatability test
well, I should sign off - I just wanted to see if IRC worked through this connection (I hate the Nomadix portal system)
see you tomorrow maybe
I don't see how you can ever use both an end mill and drill bits on the same job with 3.3" of Z and no knee
I crank the knee constantly and I have 5"
I guess you could spot drill in the cnc, and use a drill press later, and do fairly well
now you know why I have a knee mill
and a floorstanding drillpress
I did 6 trials - home ; G0Z-0.35 ; G1F1Z-.40
indicator readings were 0.0000, 0.0002, 0.0005, 0.0007, 0.0000, 0.0001
I wish there was an easy way to get just index (homing for steppers)
you mean a one channel encoder?
yeah, but cheaper than an encoder
opto-interrupter and a disk with a slit in it?
sure, that might do it
my machine is 20 turns of the stepper per inch, a 1" dia wheel has 3.14" of perimeter, so you have a 62.8:1 multiplication factor
any uncertainty in the detector would be divided by that ratui
(and I could easily do a 2" disk)
I have an idea I'd like to explore that would play nicely with that concept
for lathe Z, you can't really do a home switch because the chuck or tailstock is likely to block the ends of travel
a N bit linear feedback shift register generates a stream of (2^N)-1 bits with no N bit sequence repeated
my screw is 10 TPI
make a linear part with the output of an 8 bit LFSR milled in it, 0.100" per bit
when you home, start moving, and sample the opto-sensor on every index
after 8 revs, you have a bit pattern that tells you where you are to within one turn
the next index gets you your location
one opto-interrupter looking at the linear "scale", and one looking at the disk on the screw
and a custom HAL component to match up the bit pattern with the LFSR code
wow, so this is just for homing?
its a very cheap "absolute" encoder
I still want 'typewriter margins'
well, that is SMOP, once you have your absolute location
heck, I want keepout areas
sure, another SMOP
so I can have it avoid the chuck, but still pull to the front and then move left so I can get the tailstock in close for drilling
I want joint constraints
it would have to change with every tool
and every workpiece?
(it's maybe even impossible)
I'm assuming that you want to cut the workpiece
on my list of wants, thats actually pretty far down
well that's true. but the jaws or dog could be anywhere (for instance)
I was just being a smartass
I wish the splash screen g-code wasn't so big
I suppose I could truncate it and just do the EMC2
edit the scale
hmm, where is it?
first couple lines in the gcode
I mean, where is axis.ngc - its not in ~/emc2/nc_files or the examples dir under that
file/save gcode as...
then you have your copy to edit
that's handy with the various input filters too
man, that keyboard is not the nicest thing to type on
I'm doing all my editing and such here
see how it's multiplied by everything later
I guess they were _really_ big
they were in truetype units (postscript points maybe?)
SH is what? safe height?
that program doesn't turn the spindle on!
did you happen to notice at the beginning where it says it's not for milling? :-)
I also noticed the lack of M3 before I ran the program
[04:54:13] <jmkasunich> http://jmkasunich.com/pics/first-milling-1782.jpg
jmkasunich: youtube next :)
I don't have a video capable camera
(I don't think - my point-n-shoot might be able to do some very limited video)
have to RTFM
merry christmas everyone... (sorry chris)
I can't reproduce andy's problem but his screenshot is unmistakable
I'd like to let his message through but it's pretty big. I wish we could remove the image and let the rest through.
ah.. I entered admin interface.. but didn't find anything
(maybe send it yourself? and put the image on pastebin?
I didn't do anything wiht it
hmm.. I also got the full message, I can do the above if you want
I bounced his message and asked him to put it on pastebin
ok, I guess alex did the work for him then
didn't realise it was an image, or that you were discussing it in here
it's a good bug report now, but unfortunately I still can't make it happen
(actually I think I bounced it before the discussion started)
jmkasunich: I think you did.. when I looked in the admin interface it wasn't there anymore..
that was before the conv.