how do machine ways ever come out straight on some of the 4x8 wood routers some of these guys build
strait is relative
and also i see they complain about nonlinear paths a bit
i guess it all makes sense
on a router striaght is kinda important
flat is not so important
you just skim the bed with the router and its "perfectly flat"
or "perfectly wavy"
just like the ways!
it matches the ways, so relatively its perfect
jmkasunich_ is now known as jmkasunich
pew pew pew
In typical PGAB fashion I tore apart one of the Ametek Servo motors...
the mountig plate has positions for 4 brush holders - so they do make a 4 brush version - unfourtunately these are only populated with 2
the lack of cogging is due to a high skew armature design - looks like almost 60 degrees
4 permanent magnets
there is plenty of room to drill & tap the rear cover for any Encoder mounting method
the 3/16" rear shaft that protrudes out is part of the main armature shaft - not a pressed in addition
the rear bearing is like a skate bearing with a 8mm I.D.
Made in USA NSK bearings on both ends
front shaft is 5/8" dia
Wiring diagram labels the brown wire as unused - it is a chassis ground inside the motor.
Servos arrived individually bagged as new old stock.
Not a bad deal at $32 ea.
Camtronics sells these ( or a variant of ) for $165 ea.
[03:42:21] <Skullworks-PGAB> http://www.surpluscenter.com/item.asp?UID=2007062621461319&item=10-2351&catname=
whew. typity type type
"premium gold angus beef"
* fenn weights his mailing list post with baited breath - see if we get any biters
O-word subs - Stack them all up at the top of the file? or at the end?
I'm guessing at the top.
the new frappr interface is atrocious
fenn: I agree
hmm this doesnt work so well; doesnt seem to go to the right line http://cvs.linuxcnc.org/lxr/
for example http://cvs.linuxcnc.org/lxr/ident?i=init_display_window
thats cool how it shows the last author next to each line
yeah, cvsview does that too
[07:03:15] <alex_joni> http://cvs.linuxcnc.org/cvs/emc2/README?annotate=184.108.40.206
oh the one i never click on
yup.. I discovered it by accident too :P
here's some stats on emc2 http://eneas.juve.ro/~juve/stats/
built them a while ago for fun.. no interest in automating it so far
what do lines in bold with ! marks mean in annotated mode?
hah 36% paul_c
sed will do that
fenn: that's on a branch
fenn: and dos2unix ran twice over all files
i think CIA provides similar stats
lemme dig it up
[07:08:08] <alex_joni> http://cia.vc/stats/project/emc
the nicest thing imo is http://eneas.juve.ro/~juve/stats/commitscatterauthors.png
you can really see who's doing the work there..
oh i was wrong, it was something paul_c had on his machine, not cia
yeah, same software
woah i need something like that scatter plot
how did you draw that?
I have nfc ;)
but you can probably look at the sources
hmm all this java chart stuff is too much bother..
will roll my own from cairo
a big turd ball :D
fenn: do you happen to know where I can find a 3D model of a human?
for what purpose?
the blender makehuman script does pretty good models, and its parametric
fenn: I just need a 3D model to put into some simulation software
it doesn't need to move or anything
basicly a human operator standing upright
no textures needed
what formats are available as export from blender?
quite a lot, with varying levels of support
[08:36:16] <fenn> http://www.blender.org/download/python-scripts/import-export/
do you have one running?
looks good to me, want me to pose him or anything? what format?
ACIS sat if you can ;)
how about obj, dxf, 3ds
yeah, its just a mesh
have stl too
ok, that'll work
let me check for imports in my program
but I think dxf I can work with
[08:47:57] <fenn> http://fenn.dyndns.org/pub/irc/tool.png
[08:51:06] <fenn> http://fenn.dyndns.org/pub/irc/tool.dxf
let me look how it looks ;)
hmm.. it loads ok in autocad.. but it's not a solid, so I can export as sat
hmm i dunno if its possible to make a solid model from this guy
[09:01:53] <fenn> http://fenn.dyndns.org/pub/irc/tool.stl
trying to convince my proggies to import STL now..
obviously they all know how to _export_ it.. but importing is too hard :D
alex_joni: heh, me & awallin did stl import it's a few hours to implement
anonimasu: not if it's binary stl
alex_joni: no, but binary stl isnt all that complex either
just can't find a nice solid model of a human
it can be low-poly or whatever
just put some spheres and cylinders together, jeez
managed to stitch him up to solid
alex_joni: let's see some proof
fenn: coming up
we got home from London Monday, and I was damned ill yesterday
I think I'm better now, but I haven't eaten in nearly 48 hours, so that may be contributing some dizziness :)
I think I should try that oo
I don't recommend it :)
saw a nice welding table (turn/tilt) last week in germany
saw that for welding cars?
nope, parts of excavators
ok, even bigger :)
that's 10t parts
it had a vertical axis of 2m
and the turn/tilt mounted on that
a HUGE biest
hi people. i have determined maximum speed of the stepper by using Feedrate override slider. after that the stepper loses steps. is it really the maximum speed or should i conjure with INI file?
were you cutting when you tested this, or just moving the machine in air?
Dent: I wouldn't set it as high.. you need to leave some headroom
yep, i understand that i should, alex_joni.
I did set it like that on my machine, then after some cutting and clogging of the screws I was getting stalls
SWPadnos, no matter
Dent: does it lose steps imediately? or after getting to speed?
maybe accel settings help it too
yes, yes. all i want is more speed. i know e.g. that max is 25 and for process use 24-23. but i need more, does that mean that i need to get a new driver/hardware?
and nothing can be done with available?
probably. EMC can't make the motors output more power than they're capable of outputting, no matter how good it is :)
Dent: what voltage are you driving the steppers with?
and what stepper drives?
yep, but i just wanted to know if i missed maybe some parameters in INI or program
(for instance switching to 50-60V from 20V will make a HUGE difference)
also using microstepped drives with current control will help some too
yes i know that, i only want to know if this is the maximum at the given voltage.
SWPadnos, no, i meant that my driver cannot take maximum of the stepper
SWPadnos: how are the projects coming?
technical hurdles? or people problems..
well, the requirements keep changing on one of them
the A/D converters I like aren't available for a few weeks
I would call that 'cutting edge' if the parts you want are not available ;)
heh - they're not new, just unavailable
they are too good to build up stocks :D
must be. it's a 16-bit, 6-channel, 250 ksamples/sec, simultaneous sampling converter, for $27 or so
with SPI, "quad SPI", or parallel data bus
jepler_ is now known as jepler
Anyone tried ace converter on linux?
I vaugly remember someone using it in wine or such.
it is open source - maybe it could be ported.
there have been several reports that it runs OK in wine
Yeah, it probably could be ported. I just want to test it to see wether its worth the effort
wine complains here... some unimplemented function is called
I looked at the source code -- it is very dense and the user interface seemed to be entwined with the guts of the converter
making it difficult to convert to gtk, qt, or whatever linux toolkit you would prefer
OK, i just glanced through the source, and i got the opposite impression. But, i just glanced through it
maybe I'm mistaken -- that would be nice
if anyone have it running under wine i would be glad to know which version of wine they used
a-l-p-h-a2 is now known as a-l-p-h-a
you don't have to look far to find stepper motors for small mills .. but where do you look online for small servos with encoders, to buy in small quantity?
[14:37:38] <alex_joni> http://ajaxcnc.com/cnc_components.htm
not quite cheap though
normally you luck out and find them on ebay.. or surplus outlets. If you want new - then I don't know.
-r-------- 1 crhilton develop 3377 Aug 14 17:27 include/RCS/shared_warning_list.h,v
but I think JonE has some
[14:38:49] <skunkworks> http://cgi.ebay.com/Dynetic-Servo-Motor-Brushed-48V-DC-3-1A-W-encoder_W0QQitemZ180147279522QQihZ008QQcategoryZ78195QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[14:38:48] <alex_joni> http://pico-systems.com/japservo.html
alex_joni: yeah, jon_e had those for sale at cnc workshop
seems pretty cheap at the price (including encoder)
yeah it seems like a good price
depends what you want to drive I guess..
58 oz-in might be too weak
got ace to work with wine-0.9.29
jepler, how much power would you want from these sevo motors?
SWPadnos: it was more an idle question -- I'm not buying a servo motor today
for a "sherline size" machine -- the ones that people try to run with steppers 200 oz-in and smaller
I have about 15 pancake-style motors, a lot like the ones JonE has. they have 500-cycle encoders with index
I wasn't sure if you wanted any of them :)
sure -- when my birthday comes, and you're also shipping me the new parport mesa board as a gift, put a half dozen of those in too
unfortunately that's not till april :(
I guess you're out of luck until then, then :)
anyone have a CO2 cutting system? what's the best resolution people get from these? would like to cut some plastic film with 50um or 20um resolution
steves_logging is now known as steve_stallings
steve_stallings is now known as steves_logging
cradek_ is now known as cradek
SWPadnos: I know you found some ribbon-cable-to-screw-terminal boards on ebay. I've had less luck. Any advice on search terms?
cradek: how goes the battle with R2?
hmmm. I think I just searched for "50 breakout"
without quotes, of course
today instead of error 4000, I'm getting error 400, which means the tool table checksum is bad
I got the serial console to work, I can painfully enter a program now
and, I think there's no oil pressure
terrible line editor? or just "buffered mdi"?
so, "fine" I guess
terrible line editor
That is a good start.
I am excited for you :)
it does run if I put in a program
so - you took today off also? ;)
XYZ G00 moves Z last if it's going down, or first if it's going up, which is surprising and interesting
That is interesting.
actually I woke up with a killer headache, but didn't think anyone would believe me if I called in sick, so yes I took another vacation day
I intended to go to work today
that's a good way of not breaking tools (unless you're using tools that undercut)
makes sense I guess - less apt to run into a vice or somethinhg
it's a good way of breaking tools if you every try to use another machine though
well, that's true
cradek: we had a user who insisted all CNC controls did (something like) that
I think we dismissed him as a crank
before we laughed and made fun of him..
nah, we just said it's a bad idea, which it still seems like it is
that sounds a lot like the photographer who told us "All professionals use 1/125"
what is the oil pressure for? oiling or other uses?
SWPadnos: if it's the guy I'm thinking of, I'm pretty sure he was giving me practical advice
skunkworks: just the way oil
I thought the pump wasn't getting power, but I finally traced enough wiring to find that it only runs when the spindle is on
so it's running but doesn't seem to pump - I'll just take it apart and look for obvious things
SWPadnos: "don't use ultra-short shutter times with flash" was the advice he was trying to give
hmmm. for some reasoni, I don't think so :)
and he was probably right, because when I changed from 1/250 to a longer shutter time I got more uniform exposure from shot to shot
OK, then I didn't listen to him very well :-P
jepler, very true for focal plane shutters as it needs time to open
there is a minimum time
I was using 1/250 and now that I look at the spec for my camera it says "Flash X-Sync: 1/200 sec"
this was a wedding photographer telling a few techies and a commercial photographer (who was charging ~1500/day, about 10 years ago) that there was no reason for our radios to support 1/250 sync
ok -- I thought you were talking about the guy who stopped by, watched a lathe demo, and gave me photography advice at this year's fest
ah - nope
between the lens shutters were faster than focal plane
I think I'm glad I missed that :)
leaf shutters can have faster X-sync, since they always expost the entire film surface
lens shutters are damn hard to clean
when you say "your radios", you mean RF flash triggers?
or something else?
jepler, yes. my old company makes the PocketWizard
hah -- I clicked on a google ad (something I do rarely) and the linked page is 404'd
I hope they're not paying much per click
on a search for Pocketwizard?
cradek: itching to convert?
SWPadnos: no, something else
gah - I hate the pocketwizard site. stupid inconsistent flash shit
ahem. excuse me
huh -- what's the 2P screw terminal for on here? http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=160142177485
.. bbl again ..
steves_logging is now known as steve_stallings
cradek: the Bridgeport lube pump may have lost its prime if it has been sitting for years
and it only cycles auto every few minutes, so could take a long time to reprime on auto
jepler count the pins on the long screw connectors only 24 ea, the other 2 probably wired power ground
steve_stallings: what is the metal knob thing that looks like a plunger, but doesn't plunge? I can unscrew it ccw, or it turns fairly hard cw
I confirmed the clock motor is running (or at least making what sounds like the right noise)
ah, yours does have the plunger, pull it straight up and it should suck up oil and return slowly under spring pressure
it takes a fair pull
ah ok I didn't try pulling (very hard)
so much I don't know
osmosis is a slow process
there's what appears to be an oil pressure gauge so I assume it should be indicating something if it's all working right
I don't remember an oil pressure guage, but then the one I am familiar with has suffered a few "enhancements"
I do seem to remember an air pressure gauage
yeah this one has both
the air stuff seems to start working around 40psi so i'm running it there
the oil gauge goes up to 120 psi or so! can't imagine it could build up that much pressure
stiff spring and small diameter plunger.... I would expect over 30 psi, but ... whatever
then again, way oil is stiff stuff, so higher pressure is expected
also since the metering orifices are small, don't just fill it with whatever is handy
ok, I'll try to get what it says (there's a label on it)
it's fairly full, who knows if it's the right stuff
OK, RTFM, it says lube pump discharge pressure is 20-60 PSI
there is supposed to be a low oil switch to inhibit operation
cradek: itching to convert?
cycle time on plunger is 27.7 minutes
steve_stallings: ok, I did see some kind of sensor, didn't know if it was level or pressure
you are instructed to pull the plunger "several times" on initial startup or after running out of oil
ok that I can sure try, hope it works
it should take 10-20 seconds to return, if it's like the manual Bijur units
(unless mine's broken :) )
they list several lubes, the two that I still remember seeing in todays catalogs are Mobil Vactra #2 and Shell Tonna 68
yes the plunger return is slow, if it returns in under 10 seconds, suspect a leak somewhere
I'll go try...
swpadnos: Im having a problem with some ground noise making my g203v pulse slightly when i turn on spindle or coolant
jlmjvm, uh - bummer
they are manual switches,not running through relays or breakout board
noise doesn't care if it's controlled manually or electronically :)
will a ground rod help
that depends on how things are grounded now
through fuse box
like, using 3-prong line cords?
are you using a parallel port for step output?
the noise only affects 1 drive,the 203v,the g 201,s are unaffected
I think the 203v has a lower current requirement for the step/dir inputs
like 3 mA instead of 15
but as so often the case im going to have to be able to use the new drives
that may be the case
what kind of breakout board do you have on the aprallel port?
1 from cnc4pc,c10 i think,lemmee check
can a crappy breakout board let this happen?
it's a matter of grounding and shielding, I think
let me rephrase,a faulty breakout board
could be, but I'll bet it's just bad grounding/shielding
you cuold try swapping the 203v with one of the 201's, and see if the problem moves with the drive or stays with the specific parport pin
can do that
The G203V has a much lower current requirement on the step and direction inputs and would likely respond to noise that a G201 would ignore.
sounds like i need to clean up my act,lol
i been trying to get this fixed since last week
I think the general recommendation is to use shielded cable from the PC to the geckos, and only ground the shield at the PC end (not at both ends)
no, step/direction cable from parallel port breakout to gecko
just have wires on that
those should be shielded
i have some shielded i can use
lemmee try that right quick
grounding issues are a pain.
say ground the shield to the breakout board ground
not the drain
actually all the gounds should be at the same point.
but it depends on how anal you are
I think the drain wire and foil shield can both be grounded
I'm not sure where that matters, to be honest
SWPadnos: did you see this? http://www.youtube.com/watch?v=dKz81fY1OjY
skunkworks: sweet, what's the maximum speed?
swpadnos:still doing the same thing
i thought that would do it
gonna put it on some different pins
jlmjvm, you should also look at grounding of the power supply to the breakout board. that shouldn't be connected to the ground of the motor power supply
lerneaen_hydra: 120ipm so far.
3000mm/min, not bad
that is using the existing 6 wire steppers as 4 wire.