03jepler 07TRUNK * 10emc2/src/rtapi/rtapi_bitops.h: include bitops implementation for x86_64, needed on edgy where the /usr/includ/asm/bitops.h header is no longer available
I ran my Mesa 5120 under EMC2 and it didn't complain
I'm guessing that it shows up as a PLX Technology device under lspci
my board had a rohs sticker over that chip, and I didn't peel it off
it's a plx9030
it shares an interrupt, is that ok, or am I going to have to go beat on my bios?
EMC doesn't use interrupts from the Mesa card
so it shouldn't be a problem
that would be a significant factor, wouldn't it?
what, using interrupts or sharing them?
either should be fine, but it's unimportant since they aren't used
jepler: yep, i understand. i cannot write to it, i can read from it.
03jepler 07TRUNK * 10emc2/src/rtapi/rtapi_bitops.h: remove stuff that is not used by rtapi
I would like to add a calibration page to "my" DRO (you fellas really created it).
The only thing I cannot see how to do with the new pyvcp and HAL is store the calibration values for recall when I start up the DRO program again.
there may be ini file functions in the python library, but you can't do it without some programming
Maybe this can wait. I just thought with the multiple page capability, that it would be great to have calibration capability.
toastyde1th is now known as toastydeath
I'll do a little studying with the thought in mind, though.
lerman: around? I have a little more information about that program
the newest (2007 03 07) EMC2_User_Manual says "The EMC uses a tool ﬁle that is read in when a machine control is started. In a standard release this ﬁle is named emc.var, generic.var, or sim.var and is used by the similarly named run ﬁle." i think that oughtta be emc.tbl, generic.tbl, or sim.tbl
and there doesnt seem to be any .run files
use the source, luke!
will try, last time i tried lyx.... argh!
heh - I know the feeling
what section of the manual, by the way?
closed it... well, must be the section on tool tables
oh, that stands to reason
sectionc 20.1 emc2_users_manual (paeg160)
page ( really want me to edit it??) cant type fer sh@@ ireally do know how to spell tho
yeah - I found it
it's all wrong, not just the names
there are no longer a bunch of xxx.run files for the different configs, they're in separate dirs ...
i cant find *.run in ~/emc2-head
right, they're no longer used
there's now one run script (scripts/run), and you supply a config dir for it to use
do you also have the problem with the surrounding text overlapping the sample tool table?
yep, renders badly
03jepler 07TRUNK * 10emc2/docs/src/mkpre.py: a script to help with converting ASCII art diagrams into groff input
you in chris
just ask your question :-)
does emc have servo spindle control
can you say more specifically what you mean? like analog output for a speed control for milling?
does that mean position control (as opposed to simple speed control)?
i am using messa 5i20 to messa 7i30 quad h bridge and would like to use 1 for spindle control
must be a little machine
yes using pittman dc and 500 count encoder just on test bench
if all you want is for EMC to be able to control the spindle speed, then yes
if you are talking about full servo control of the spindle, as for rigid tapping, or boring with tool orient before retract, no
closed loop what? speed, yes, position, no
speed for now full servo if possible
run the motion.spindle-whatever output through a PID loop with feedback from the velocity output of an encoder (or a ddt if neeed)
that gives you PID velocity control
closed loop speed would require some non-trivial design - it can be done with existing hal blocks, but you gotta know something about motor control
what SWPadnos said
so what you are saying is it can be done
you gotta deal with things like velocity feedback quantization noise and such, not outrageously difficult, but not a job for a control theory novice either
yes for speed, no (not yet) for orientation
jmkasunich, it should be pretty easy to make a separate vel (ddt) component that has output like the new encoder velocity output (taking time since last change into account)
because the time granularity is 1 period of the thread which calls ddt
hmmm - why not? (if you use the count output instead of the position output)
the encoder trick works because it knows the time of the last count to base_period resolution, not servo_period resolution
right - thinking about that now ;)
I guess it would help for the case where you are getting 0.6 steps per servo-period
but not the case where you are getting 1.6
yeah, though a spindle is very slow, so you could just about run the PID loop at 10 Hz
yeah, that would work - need another thread
or apply a lowpass to the ddt output
or a skip counter :)
hmm. there's no return from HAL functs, is there?
skip counters aren't a HAL suitable way of doing things
no, hal functs don't return values
they process signals
I understand the idea behind them - that was an implementation question
ie, in the thread function, does that get a return back
I don't think so - I would have to check the code
hal.h has the prototype for hal_new_funct, which in turn has a prototype for a hal funct
ok - no problem, jus twondering
I think I'm gonna do the "ctrl_type=" thing for stepgen
I don't want to document what I have right now, knowing full well I'm gonna change it some more
saved a copy of it as it is now tho, in case we decide to tweak it and backport to 2.1
hmm, nasty decision:
most hal comps have one funct per entity
got 3 mux2's, then you have mux2.0, mux2.1, mux2.2
but stepgen has "stepgen.make_pulses" and "stepgen.capture_pos" and "stepgen.update_freq" that cover all stepgens at once
more convenient for the hal file writer
not as flexible
hmm, I thought this was the dev channel
another truck load of 'stuff' to the new house..
now is when you discover the things you didn't unpack after you moved last time
right. some boxes have been moved without really knowing what is in them :)
I have not tried the tweeks on the gantry yet - maybe I will get some time tomorrow.
see you jepler
sieg has a small grinder http://www.syilamerica.com/product_U2.asp
, looks like pretty small wheel, needing pretty hi rpm spindle
i would not really want to run a small wheel on a surface grinder
03jmkasunich 07TRUNK * 10emc2/src/hal/components/stepgen.c: fix typo, clarify a couple comments, insure against unlikely race condition, tweak parameter init
03jmkasunich 07TRUNK * 10emc2/src/hal/components/stepgen.c: add a comment, fix silly whitespace typo
this is a simple question, but it's oddly hard to google
how are t-slots measured?
with a dial caliper ;-)
do they size the narrow width of the T
or the wide width of the T?
it's usually the narrow
you are trying to buy some, and you don't know how to spec them?
you can't really tell though
as i recall, strap clamps are sized for the narrow
so that's what they give you on the specs
yeah - unless its a "standard" like bridgeport, can be tricky though
if you're trying to mill a t-slot
jmkasunish: worse, I now have a step block set. a very very big step block set
orpheus: strap clamps
much bigger than your machine can use?
I DEMAND INDUSTRY TERMS BE USED
no just kidding.
well, the narrow part of the t-nuts are as wide as the wide part needs to be
got a reciept?
or is it ebay or somethign like that?
it was ebay
I could re-list them, but they cost more to ship than to buy
i mean, they're really heavy
hindsight and all - they often spec that stuff by the diameter of the studs
bport clamp sets are 1/2: studs
the slot is wider than that, no idea what the actual width is
not much wider tho
so if something only has one number on it, you have a 1-in-3 chance of guessing which dimension it is
aren't there like 3/4" studs and 1/2" studs
i just feel really dumb for getting the wrong ones...
so how big is the set you know have?
i guess there are smaller ones, i wouldn't know
toastyde1th: 3/4 would never fit on a bport - bet you'd find them around bigger mills though
* orpheus finds the exact size
I have 3/8 studs for my shoptask
our cincinnati and K&T horizontals at work use big old clamps
made the T-nuts here though
not sure the exact size though
same with the van norman (I use the same 3/8 studs, just differnet t-nuts)
orpheus: thats pretty much industry standard then - save them for when you have a bridgeport ;-)
or mill your table
that was one thought
there'd be no table left
actually i wouldn't do that
what size is your machine? sherline-ish?
unless you were going to regrind it flat
just 4 thin walls
it's a standard harbor freight micromill
probably wants 1/4" studs then
studs are only like 30 bucks, right?
or are they 50
HF probably has clamping kits taht would fit
toastyde1th: a complete set (t-nuts, studs, hex-nuts, strap clamps, and step blocks, is usually not cheap
ebay's pretty cheap
yeah i keep seeing strap clamp kits
the whole deal
for like 30 bucks
chinese ones I guess
just not in small sizes
strap clamps don't need to be american
but so what if you're cheaply made.... they're block o' metal
*i'm going to stop typing so fast now
or correct all your errors on one line, one or the other
orpheus: what size are your tee-slots (the narrow part ;-)
toastyde1th: I'm looking at enco (figured they'd have cheap, and maybe small)
1/2" sets are $76
well, i don't know off the top of my head, but the're almost certianly 1/4" studs
i'm stumbling through msc
mabye they are hella expensive
ok, enco only goes down to 5/16" studs, 3/8" x 5/8" slots
harbor freight has 1/2" sets for $39
yes! gg harbor freight
3/8" for $35
thats the complete 58 piece set
dirt cheap IMO
msc has a 253 dollar clamping kit
te-co I bet
for $250, what all can you clamp down?
orpheus: is this what you have? http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=47158
at work we have a 30 hp horizontal that runs a 10" diameter, 5" wide slab mill
that's what it /used/ to be
i bought it used
you don't want something to come loose when you run a job on that thing
Table slot: 1/2'' x 3/8''
the previous owner... upgraded
I was hoping the webpage would have "accessories for this machine"
heh, shipping weight 103 lbs
it was over 150 when I bought it
you do not need to spend big bucks on hold down stuff
that mill will not generate enough force to rip anything up
the new motor might
a 2hp bridgeport will stall before you run into problems
or snap the cutter
mostly the heavier gauges are for rigidity
heh, the webpage said slots 3/8 x 1/2, the manual says 5/16 x 5/8
rather than fail strength
no, typical HF accuracy
t-nut size versus actual t-slot size?
they're not a tight fit
I trust the manual numbers vs. the webpage numbers
except the actual numbers are probably metric
well, before I buy anything, i'll measure. thrice.
oh, you already have some t-nuts?
err.. no.. 3.
then the real thing to measure is what the thread is
get your self some threaded rod to make studs
get some nuts
oh my god, msc has a 497 dollar strap clamp kit
get a small mill vise (you're gonna want that anyway)
and use the vise to make tee nuts to fit the table slots
already have a vice, that's how i've gotten this far
use the two tnuts you hve to bolt the vise down
and go to town
get a piece of square bar as wide as the nuts need to be, and 6" long
mill the steps in it
but by the time I do it... won't i have spent $30 on materials?
drill and tap every 3/4" or so
and saw apart
probably like 8
6" of 5/8 square bar isn't much
and you could make 8 nuts out of it
that's another question... where do you people get metal?
I use metal express mostly
they are in the midwest/east, where are you?
i don't have home tools, only at work and school
I have a 1"x4"x18" bar i got from ebay that i've been hacking at for over a year
and they buy from bulk suppliers
i'm in atlanta
metal express... that's what i needed someone to tell me
8.99 for 6" of 5/8 bar
oh, someone was so close
and it was me
* toastyde1th dances
crap, $5 for handling, that sucks
small orders are no fun
I think the closest one to atlanta is greenville SC
they have value packs! sweet!
I'm in cleveland oh, so I can usually go and pick the stuff up myself
another question, is there any reason you can't tap something with a mill?
let me explain more
with that mill - torque (or lack of)
and low speed
if you took a wierd but, not square, but more of... diamond shaped
you do not want to tap on a mill without a tapping head
and then ran it around a hole or bar, in a long spiral
does that work?
i mean, for something with a really wierd threading?
sounds like a helicoil
it would not be strong at all, unless the inside of the hole was already threaded, and in that case, why bother?
i'm not sure i understand what he wants
i actually found my own answer
"Threads can be machined with a single point tool on a milling machine using helical interpolation."
toastyde1th: I tap on the drill press all the time (without a tapping head)
what's your process
that works for outside threads, and larger inside threads
are you holding the tap in the chuck
my drill press goes very slow
yeah, i would not really do that
yes - I know, hardened tap shank isn't nice for the chuck...
the worst i do is align the tap handle using the spindle
my drill press also reverses nicely
but it can be done, which is really handy
for those really really wierd threads
orpheus: cnc mills can do many different things
i'm learning this
I've done 6-32 in 1/8 alum, and I've done 3/6-16 in 1/2" thick cast iron
float tapping, rigid tapping
i need sleep
and things in between
10 AM class
night then, folks
toastyde1th: I also have a tapping head, and I use that when I have more than one or two holes
i never really want to risk a tap in the drill press
I would not do it on the average drill press
mine isn't average - a former owner put a countershaft where the motor goes, and stepped down the motor speed to drive it
do you know if yours is marketed as a drilling/tapping machine
I can run it a t30 rpm
so it kind of is
at 30 rpm that is
kind of - but its just a Clausing 15" drill press - not made as a tapper
yeah, but there are normal drill presses that have tapping facilities
that are indistinguishable from other drill presses in the same lineup
between the vari-speed pulleys and the vfd it will do 30-1000 rpm
those have integral tapping heads, right?
it just stops quickly and has a reverse
very similar to what you are describing
must be three phase then
not that i'm aware of?
single phase motors don't reverse for crap
i've never used one, so i can't be all that specific
i think they have brakes
you gotta let them coast down until the starting switch engages
if they are industrial class machines, they'd be three phase anyway
only home crap and small machines are single phase
nah, i've seen some decent drill presses that are single
my drill press is only 1/2HP and it is three phase (hence the vfd)
but yes, i will agree that almost all the good stuff is three phase
I think its 1/2
[05:17:42] <jmkasunich> http://jmkasunich.dyndns.org/shoptask/bsmt-less-3.jpg
i rarely use drill presses, though
i admittedly need to practice more
that's a neat drill you have yourself
used machinery rocks
i'd have some if i didn't live in an apartment =(
if I recall correctly, that was $250
i tried to get my buddy to buy a jig borer
instead of a drill press
500 for the jig bore, 500 for the drill press
he didn't go for it
[05:19:44] <jmkasunich> http://home.att.net/~JEKasunich/vannorman/VN_Home.htm
my mill was $500
that is a cool mill
the one in that pic
that's darn good
took me about 40 hours and $100 in parts, plus a surplus VFD from work, to get it running
how do you like it
a lot of the 40 hours was cleaning
it was covered with caked on cutting oil and sharp chips
03jmkasunich 07v2_1_branch * 10emc2/src/hal/components/stepgen.c: revised stepgen, backport from trunk
it works great
how's the vibration
i can't stand bridgeports
the VN is at least as rigid as a bport, if not more
its 1800 lbs
but it doesn't have that long arm, knuckle, head, etc
that's what I was looking at
it kinda looks like you could mout the vise sideways and run it horizontal
my two standard "show off" pics
[05:25:10] <jmkasunich> http://www.metalworking.com/dropbox/_2001_retired_files/Facemil3.jpg
[05:25:19] <jmkasunich> http://www.metalworking.com/dropbox/_2001_retired_files/Facemil4.jpg
first one is 4.5" wide x 1/8 to 1/4 deep
second one is 1" wide x 5/8" deep
how many hp is that
both single pass, not a bit of shuddering or shaking
gear drive spindle
70 RPM in low
I was doing that cutting at 140 tho
how wide is the cutter
I won't brv
I won't brb
1:30 in the morning
Jymmm has changed the topic to: Welcome! EMC (Enhanced Machine Controller) is a linux-based CNC control. | Latest release: EMC 2.1.2 | http://www.linuxcnc.org
how goes it alex_joni
* alex_joni goes away for the day
[14:24:24] <anonimasu> http://www.metalworking.com/dropbox/_2001_retired_files/Facemil4.jpg
anonimasu, how's the day going?
cut 40 parts earlier
[14:44:47] <skunkworks> http://www.electronicsam.com/images/house/snow.JPG
that should be called "snow+Ice.jpg" :)
unreal - couldn't do that if you tried
03jepler 07TRUNK * 10emc2/docs/man/man1/hal_input.1: markup fix
ok, now this is funny
I wanted to search for X and/or Y motor mounts
so I use the search terms (X, Y, x&y) (bracket*, mount*)
which should match X or Y or x&y plus mount* or bracket*
and the first item that comes up is coffee in K-cups
from 30 miles down the road
"Green Mountain French Roast Decaf Coffee K Cup x 50 NR!"
Green Mount* .... x 50" :)
I find myself doing a lot of subtracting in searches now
dell laptop -battery -adaptor
03lerman 07TRUNK * 10emc2/src/emc/rs274ngc/interp_convert.cc: Fix bug in handling of display of numeric parameters in comments (PRINT, and DEBUG,)
03lerman 07TRUNK * 10emc2/src/emc/rs274ngc/interp_o_word.cc: Fix the previous fix to this file. Was checking the wrong return value.
cradek / jepler:
wanted to thank you guys for all your eagle script help
its working perfectly now, my boards are better than i expected :-)
crepincdotcom: that's great
crepincdotcom: got any photos?
no, i don't have a camera :-( hope to get some soon
crepincdotcom: did you buy that mill CNC-ready, or did you do the addition of CNC yourself (motor mounts and the like)
i just milled the transmitter for my freinds rocket (his senior thesis, if you will)
nah I mad all the mounts and such
in hindsight there are some things I would change
but i am really amazed how well it works
i mean the motors are 7.8 deg/step, and crappy at that. my linkages are rubber tubing
but somehow its ok
crepincdotcom: 1.8 degree (200 steps/rev)? Or 7.5 degree (48 steps/rev)?
I'm half stepping them though
my drive board doesn't chop them though so they get uber hot when sitting still, i need to make a new driver pronto
crepincdotcom: is the limited size of the table (in "Y" particularly) a problem for the boards you're doing
i have to design them with the limitation in mind
* jepler nods
but its what i have until i finish my new mill
I got a nice position "table" I guess you'd call it
cant tell what its from, but its about 3' x and 2' y
ball screws, large steppers, etc
wow that's huge
i have a 54k spindle too for it :-)
there's room for that in your dorm?
ha no, I have an apartment lined up for next year, and until then I have my own space in one of the machine shops
crepincdotcom: :) http://www.electronicsam.com/video/MVC-341W.MPG
crepincdotcom: are you still going to try to use servos?
skunkworks: on the new mill, likely
i need to play with them a little more before i trust them fully
hey i'm new to them :-p
damn skunkworks those are some HUGE motors compared to the table
nice work though
crepincdotcom: (the video was to just sell them on ebay ;))
hehe, compared to any steppers tehy will probably rock..
anonimasu: the issue i think i may have
is that the comparators in the encoders only go so fast
so I can't put the motors at full speed
crepincdotcom: I take it you bought servos without encoders..
and I can't use divider logic, because the loss is at the comparator level not the computer leverl
no, the encoders came with the motors. I didnt buy them though, I liberated them from the robotics lab :-)
(legally, of course)
crepincdotcom: that seems odd..
crepincdotcom: are they dual shaft motors? (encoder on one end)?
crepincdotcom: buy some cheap ones from usdigital.
do they have faster comparators?
they have ones that stick on with adhesive tape ;)
crepincdotcom: like this http://www.usdigital.com/products/e4p/
* crepincdotcom reads
how fast do the motors spin?
haven't clocked them yet. At no load, "quite fast" lol
you'd need a very high speed motor and/or a very high resolution encoder to have encoder speed limit you
perhaps 600 rpm?
the encoders are like 3600 steps/rev though
is that 3600 after the quadrature "multiplication"?
no, its 3600 high pulses per line
* crepincdotcom calculates frequency
well, encoders are rated in cycles, so it's 36 KHz
you get 4x that when you look at the signal in quadrants
but the comparator should be able to do 36khz no problem
36 Khz is not a fast frequency for an encoder
perhaps the motor is faster than i though
ill scope it again and see what i find
you've got to remember the difference between RPM and Hz - there's an extra factor of 60 in there :)
yeah, i calculated it
it's easy to forget about
so if a servo is holding up the Z axis, is would have to constantly be pulsing the motor to keep it there, as oposed to steppers which have wicked holding torque, correct?
do you think it would be bad if i made a pully system with a weight the same mass as the Z axis, and each was cabled over a pully above the Z-axis
so the servo only had to move the axis, not hold the weight?
yes, a servo needs to exert force to keep the axis stationary, but it will only be the amount needed
with a stepper, you can't tell what's needed, so you pump full (or reduced) current into it
but a 2 tpi well lubed ballscrew will take a good bit of torque to keep in one place
a counterweight can be good, but consider that weight is a very good anti-backlash device
I think there are better ways to couterbalance than using weight - but don't ask be what.
some people use air springs
but then doesnt the pressure vary with displacement, as opposed to a constant force of a weight?
(I would not want to add mass to the system
yeah, it would slow rapids
but my Z rapids are never very far
yes, so you use a large-ish reservoir of air, and relatively thin springs
ahh SWPadnos ok
you get some deviation, but not too much
i have a whole pile of pnumatic cylanders, that actually sounds like a great idea
weight has inertia, which is bad
4" PVC pipe 8' long == big reservoir =)
may need better seals, but sure :)
Just NEVER exceeed 120 PSI
or come close to that
lol ill keep that in mind
my potato cannon can attest to that ;-)
we made a 16' long cannon
I was looking for some good springs for my BP, but I never could fine any that had sufficient force and displacement (and insufficient cost)
I actually am going to try a 2' one for vacuum reservoir
crepincdotcom 16' out of pvc?!
was it ever straight?
6' expansion chamber of 4" PVC
then 10' barrel
it was, eh, "fairly" straight
we could put a golf ball through 4 peice of 3/4" plywood
crepincdotcom: do you have pictures of it on your site?
let me see if i have the big one on there....
nope :-( this is just the 8 footer: http://www.crepinc.com/projects/potato-launcher-a-success/
that was probably 2 years ago, I wasnt as good at documenting things
I made a 'normal' one.. what I found worked the best was having multible spark gaps within the chamber. It was scary
I've heard that
LOOK! Up in the sky! Is it a bird? Is it a plane? No, it's GhettoCraft!!!
really a noticable response?
how did you get all the gaps to spark at the same time?
crepincdotcom - the where all in series.
I hear a stun gun works the best ;)
i bet i could rig a stepper motor to do it
dont grab stepper leads when its running!
i found that out the hard way
giant inductor, when it looses its charge it pulses its energy as an "infinite" voltage
lim (dt-->0) V=L(di/dt) == inf
found that the hard way... god i thought i was dying
arced across the chest
I did the same thing with a ceiling fan ;)
don't the stepper drivers have circuitry to decay the inductor current? no, i'm not disagreeing that there's still a lot of voltage
edp: in reallity its not "infinite" voltage, but most steppers (at least the ones I've seen) dont have internal components to deal with it
most drivers tie all stepper lines to the rails with fast diodes to dissipate the pulses
most controller/drivers circuits though have diodes/etc. to allow the current a path to travel so it doesn't increase too much
;-) read up
i have been :)
I think he meant "read one line up" not "read some books, noob" :-P
a-l-p-h-a, how about an Accord?
wrong chan... oops. :)
I had platinum plugs in my toyota for about 100k miles, then sold it
they just don't seem to wear out
do they work well?
they were factory
her car is really sluggish, but still goes...
so I'm figuring just a tune up, with new plugs may help
doubt it :-/
hopefully the spark plug wires will come out.
she did a fuel filter change a little while ago.
110K on it right now I think
it's OBD-II then. check engine light on?
I have a ODB-II reader... :)
it's not misfiring then
is it worse on rainy days?
cradek, I don't know... it's been snowing, and cold most days
I just started dating her 2 months ago today.
ah, already you're fixing her car :-)
actually nothing wrong with it, other than when I hit the throttle, it's slow to respond.
happy two month anniversary - I got you some platinum ... spark plugs
well, when you want to end it, finish fixing the car ;)
dinner and mini putt tonight too!
nice day for it (here anyway)
[20:07:50] <a-l-p-h-a> http://www.theweathernetwork.com/weather/cities/can/Pages/CAON0696.htm
holy crap that's hot
it's uber outside.
I was outside in a t-shirt and light jacket...
we;re up there also - I should break up the remaining ice in the driveway
getting the gear ready when she comes, I can do it quickly, and leave for dinner.
50 deg now
our mud is finally drying out
cradek, where abouts are you again?
very center of the US
Nebraska... what does it look like again?
equally far from, well, everything
is that beside Ohio?
thought Ohio was corn country
corn country is big I guess
three states north of of TX.
what's in NE? besides corn, and flat stuff.
what's considered biblebelt?
/me has goats
* crepincdotcom has goats
where are you?
crepincdotcom, as long as the goats consents, no one is the wiser.
a-l-p-h-a: deep south
atm im at school in mass, but i live in NJ
Is the stereo type of NJ people really bad at driving? Which group of people do NJ make fun of that can't drive... ?
a-l-p-h-a: lincoln is a college town, about 250k people, low crime, a nice place
a-l-p-h-a: second largest nebraska city
there are people in nebraska?
crepincdotcom: yes, but we scurry under the corn when anyone looks
cradek, sounds nice... good parks?
cradek, good night life?
Center for Nontraditional Manufacturing Research, University of Nebraska-Lincoln, a couple of edm publications http://www.unl.edu/nmrc/Researchpublication.htm
Rajurkar is prolific
a-l-p-h-a: i dont think NJ drivers are half as bad as MA drivers
its quite scary to cross a street here
nj drivers are fast
or so i have noticed
we do have a tendancy to move, yes
question: I've my emc2 modified by me in a directory, configured as run in place. if now I want to take this directory, copy it in another PC and recompile all, what should I do?
just copy it.
I've tried, the process went half good...
on the remote machine, run "configure" again
I had the need to run make with sudo
is it a simulator build?
do you have the same user ID on both machines?
UID, not username
is the target the same OS?
uhm...I don't know, but they are completely different machine, one is mine the other is at university...
./configure ....; make clean; make; sudo make setuid
I've done: ./configure --enable-run-in-place
yep - they're probably compatible at the source level, not binary (if that's what you were trying to copy)
at this point it shows a lot of "permission denied"
when trying the depend
so I tried sudo make
is the owner of the files on your machine "root"
you should never "sudo make"
after I type sudo make setuid
that's the one user that has the same UID everywhere
except for "sudo make setuid" or "sudo make install" you should never use "sudo" anywhere in building or running emc
and now I need to run sudo /scripts/emc
what should I do?
Of course, i'm about to contradict myself, because the way to fix this is to go to the top emc directory and "sudo chown <myusername> ."
I suspect you need to chown all the files to your name/group, and then re-tar them
how did you "copy" the files in the first place?
epic root maneuver
well...I clicked on the directory icon, chose copy...
and pasted it in my usb pen...
so I picked them from my usb pen
and pasted the directory in the other machine
curios thing...I've done it one time
and everything worked
yesterday I've done it again and had this problem
I have probably used removable media once in the last year
and I've probably used the gnome graphical "copy by dragging" about as many times
it depends on how your system is mounting the filesystem
lewin1 is now known as lewing
well...what this command "sudo chown <myusername> ." does?
er, it should have been "sudo chown -R <myusername> ."
switches to the root user and then changes all file ownership to <myusername>
it changes the ownership of every file and directory under the current directory "." to be owned by the user <myusername>
for the current directory
when you used "sudo make", it created files and directories that were owned by root
ah...so it should work
you may also need to: chmod -R u+w .
to ensure that all files and directories are writable by you
uhm...I will try
jepler: is there a simply way to read a HAL pin value using python?
simply = that I can understand
from halmodule import * / <something goes here> :)
<something goes here> what goes there?
I don't know :)
'pydoc hal' has one of the simplest example I could think of at the top
but I'm sure it's quite simple
i once shot a man for reading a HAL pin value using python
jepler: can you remember me where I can read something about that...I'm pretty sure I've read it but don't remember
xemet: open up a terminal. source the emc-environment script if you are using run-in-place. then type the command 'pydoc hal'
03petev 07TRUNK * 10emc2/src/hal/drivers/hal_m5i20.c:
-Changed encoder interface to conform to EMC canonical interface. Count
latching, when an index pulse is seen, was replaced with count clearing.
This still needs further testing. Roltek plans to test in the next few
uhm...seems that I can create components and pins
but is there a way only to read an exixting pin while emc2 is running directly
or should I create a component with a IN pin for the pin I want to read...
xemet: you can parse halcmd show pin name
or use an input pin, and connect it
I mean read a pin whit a Python application
xemet: it is wrong thinking to say "I want to read the pin with name X created by some other component"
xemet: because the flexibility of HAL is to hook it up in different ways
alex_joni: if you must suggest invoking halcmd to do this, please suggest "halcmd getp" instead of "halcmd show pin/param" -- "halcmd getp" prints just the value of that pin or parameter with no heading or columns of data, so it mostly eliminates parsing
(there's "gets" for signal values as well)
xemet: as jepler said.. you can do it with getp, but the proper way to do it is have you rown pin, and connect it to that
jepler: getp reads params too?
getp...where? at halcmd prompt?
halcmd getp pinname
target: I've I python gui application, and want to show the value of a pin while emc is running
so I run emc
xemet: then make it export a pin, and link it
i run my application, and I would like that my application shows some pins
ok...so I let my application create a input pin and link it to the pin I want to read
xemet: yes, the proper way to do that is to have your application create a component and associated (input) pins, and use "link" or "net" to link them to the other (output) pins
ok, understood, thanks
hmmm. that may cause issues if you want the other program to run continuously, but have EMC start and stop
yes...I imagine...but I need it only when emc is running
03jepler 07TRUNK * 10emc2/docs/man/man1/halcmd.1: document getp, gets
03jepler 07TRUNK * 10emc2/src/hal/utils/halcmd_commands.c: better describe 'getp'
03petev 07TRUNK * 10emc2/src/hal/drivers/m5i20/bit/ (4 files): -Moved all bit files to the same dir.
i just learned a valuable lesson
when I mill boards, I home the Z axis using the multimeter method
in order to get a good connection to the board, I (used to) press the probe down fairly hard
I was confused that my cuts were suddenly way wider: turns out I was compressing the carpet tape and getting an inaccurate Z reading
just thought people might like to know.