Jacky^afk is now known as Jacky^
hehe when come les come the sun ! :P
this has been a busy place the last few days.... great to see the excitement and progress
les_w: looks like you might see some progress on HSM.... you will need bigger bolts to hold down your router :}
Well it's great to talk about it
doubt if anything can happen...too hard to integrate into emc I think
can't just unplug a box and plug another one in
I was looking at the Hurco VM-1 today
it depends on how much code one wants to save..... In some respects, it should not be hard to take a list of instructions (gcode) and turn them into nice sampled motion niblets
Seems a nice small machine
in some respects, the nml libraries have been a hinderance...
got a link?
vm-1? let me check
check out the Haas mini and VF-0 too
the Haas mini is a bargain
robin_sz: question for you
whats the lowest feedrate usable for cnc plasma cutter ?
[00:34:11] <les_w> http://www.hurco.com/Hurco/English/Products/Machining+Centers/VM+Series/
or 'normal speed'
Jacky^: how thick metal?
max 3 cm
so you need 120A plasma for edge start
250A for piercing
at 30mm, with 120A, expect speeds around 1m/minute
uhm.. thats too fast for what I was expected
thats WAY SLOW for plasma
oh, sorry ..
I need to take windfall profits and buy cap equipment I guess or pay big tax.
in 1.6mm sheet, with 30 amps, expect 10m minute
Need to talk to an accountant I guess, as much as i DISLIKE THEM
les_w: that is a nice looking machine
I need to do the maslow " need and wants" thing
les_w: yea... I have a related problem... several stocks I held were bought out so it triggers a bunch of capital gains I had not planned on this year
A bit irrational about it now
so far I am just confused and have done nothing
les_w: new or used machine?
fitted you new spindle yet?
It's on but doing wiring
when is first chips expected?
Week or so...just doing it in spare time really
robin_sz: thanks, I taked note, my cousin will start to build his plasma cutter in january too
The articulated dust shield works good
runs on vertical thomson bearings
Jacky^: plasma is a funny thing, good luck. I think to stand any chance you need arc-voltage height sensing
750ipm rapids should hop it across the floor
20hp spindle... lots of chips to come ... VM-2
robin_sz: I just hope in my cousin experience .. he's the expert in that :/
its not the speeds that makes em hop, its the accels
Saw used vm-2 for about 20k...4 years old or something
yep... mostly the full speed into the stops!
the Haas I play wiht is *very* smooth
also good looking machine
well thought out
and, you can cook breakfast while you work :)
anyon wanna play a round nbetris ? :)
other than protos How could I keep that spindle turning/
anyon wanna play a round netris ? :)
ugh, I like gtkatlantic
well how about building a coke can refinery and an al rolling mill, then using all you output to build hobby router tables?
I only play monopoly. I always win too.
les_w: have you seen the spindle brake dump resistor on thos mills?
Oh I did kinda design up hobby router this weekend just for amusement
sort of metal clad ...
in a flat spiral design ...
like a stove top element?
not *like* ....
it is ...
hey... I was just looking at those.... cheap
lots available at the dump
the Haas mini is an amzing little machine for the money
tools are so cheap to buy now!
well, no netris, no reason to dont stay asleep ;=)
thats because there is no work
itsx 1:50 in austria
still early :)
so...for hobby machine....
ive doen this analysis many times
I always end up with
realizing that little and big linear rails cost about the same (within a range)
[00:50:01] <LawrenceG> http://www.haascnc.com/VMC_MODEL_MINIVMC.asp#VMCTreeModel
they do look nice.... would just about fit in the garage
use master linear rail...and spring loaded bearing (casm follower) on a flange on the other side
of course, moving table fixed gantry
thing is...it needs steppers
les_w: use a rollerskate/skateboard wheel as the tensioner... some give if it rolls over a chip, but wont rattle
one fixed, one spring loaded
cam follower on the other side
the rubber wheel is the spring
I tried to think how such an arrangement would compromise performance
and I couldn't think of anything
it eliminates a lot of expensive overconstrained construction.....semi kinematic....
and halfs the number of rails and blocks
a moving table is a good app, because there is a decent baseline to keep table motion stiff and square... space linear bearings on one side about the width of the table apart
nice triangles are GOOD
im doing some photos at a 'strange' slide ive here ..
yes...seems to have damn few disadvantages really
want to upload it in few min
no pushing on a rope.... I still havent got that figured out yet, but there has to be a use
jacky, I think this outrigger thing would be good for your machine too
funny how one remembers 1st year engineering training!
no compromise, but linear rail cost cut in haf practically
les_w: I want to show this to you ..
les_w: that pretty much the way my big laser is built
les_w: would you use screw drive near the constrained bearing side??
lawrence: haha yeah. Three legged stools don't rock.
les_w: one side has a rack and a 30mm linear rail
les_w: the other side has a rack and a cam-follower bearing
well, 3 actually
les_w: I think the outrigger would be nice
two on top, one underneath I think
if moving table, drive placement depends on friction angle
well, yeah, but build the table long
concept could be extended to any type of ways really
it doesnt save much though does it?
the reason they used it on the laser was not to save money
it was to allow for thermal expansion of the bed
but jus use one master rail to for lateral and vertical stiffness
othe floats laterally
yes...thermally stable too
it makes it easier to build
you can be sloppy with tolerances
especially for hobby kits
but not much cheaper I dont think
grout steel with epoxy. machine one nice flat datum in the epoxy for the master
soon as you said "machine" I saw $$$ signs
I'd probably settle for a piece of rolled bright bar
cheap enough, and flat to a few thou
I can't get out of the industrial machine ilk
routers are less demanding
That's why i tried this ...for entertainment
you dont even need a flat base ...
never tought about hobby maschine design.
you can have the router skim itself flat afterwards :)
did that even on the big machine
the cheapest way is fixed gantry and moving table
bit more space, but ...
welded steel I am thinking
not 80/20 or something
I have laser, press, powder coat ...
Well we were just looking at those hobby routers on ebay for fun
a welded steel one would cost no more
designed by a chimpanzee...
and be so much better
ok, that is absolute not for my machine, but my cousin sayd he can build a similar rails with home and cheap materials
[01:09:22] <Jacky^> http://digilander.libero.it/jackydgl0/guida.jpg
how it seem to you as 'type' ?
it use 6 bearings
yeah .. costs too much
les_w: yes, for a cheap machine
Igus rail is better and cheaper
cost too much ?
same as Jymmm's machine ...
he sayd could be cheap
4 skate bearings on the outrigger side
it costs more than proper linear rail
the guys who err "designed" jymms machine started doing that
then they realised and started shipping linear rail instead
remember...abba rail 20mm $93/ meter unit quantity
les_w: yeah, sure
but just use one per axis!
yeah, but an equivalent linear rail to that sort Jacky^ showed, would only be 15mm rail
its just so bendy
and costs too much to buld the roller system
6 cam followers cost way more than on linear carriage
15mm carriage is less than 30 euros from HiWin
heck a polymer slide would prob be ok on an outrigger
15 mm rail for master.... polymer slide and tensioner for outrigger....easy...thermally stable....semi kinematic
slider or skate bearings could be on bright steel yeah
cold drwan we call it
hey guys, look at that nice machine here: http://www.infodelta.it/meccaniche.htm
i was found this website mounth ago .. and I send some email to ask
they do not sell to private people ..
alex tried to ask for the small cnc too
and they replyed they ship only a container for at least 20k euro
hmm 1700 euro for the router
no one can buy there
cannot see details from the picture
E. 1.300 the small cnc seems nice
[01:22:22] <Jacky^> http://www.infodelta.it/index_en.htm
this is the homepage, that i cant see very good with my browser
and alex ask as society too
hard to see but I think y axis is single rail and outrigger!!
nothing to do ..
not well ..
bottom rail on y seems to be a linear rail like abba
top rail seems different
yes, could be
like I describe before
see no bolt hole on top
outrigger design. Just looking at the same picture more closely.
another good system for the Y axis is V shaped track with err, V groove wheels
used on the tool changer on the Haas
plenty of line contact, so not much scuffing wear
the differential radii of the contact patch tend to wipe muck away from the centre of the rail to the outside
les_w got a few?
that sign you made was quite nice, impressed.
I didnt think that little router would ever actually, work, but the Harley sign was pretty neat
yeah things looking better
robin_sz: seen my jobs ?
[01:31:58] <Jacky^> http://digilander.libero.it/jackydgl0/photos/lab/lab.html
after I burned 15-20 bit
and ruined 50-60 pieces
i got them ! LOL
robin_sz: really* nice for the machine I use !
an you ? are cutting metal in these days ?
we had a good spell a few weeks ago
cut 22 tonnes in 6 weeks
right bedtime ...
yeah for me too
G night robin_sz
Jacky^ is now known as Jacky^afk
03rayhenry * 10emc2/configs/sim/README: Sample of readme function
djb_rh_ is now known as djb_rh
03rayhenry * 10emc2/tcl/bin/setupconfig.tcl: partial directory copy script
rayh is now known as rayh_away
03swpadnos * 10emc2/tcl/bin/setupconfig.tcl: Changed the search to only look for directories. This will also prevent the setup tool from seeing .ini files in the configs/ dir.
03paul_c * 10emc2-auto/wiki/ (15 files in 11 dirs): "Auto update wiki from a cron job. Tue Dec 13 05:30:01 GMT 2005 "
SWPadnos_ is now known as SWP_Away
time for another fight with the server
can some one here maybe help me? i got a transformer with 110v out, and now i wanna put it through a rectifier, and then through a cap. what will my end voltage be? is there a formula or something?
Jacky^afk is now known as Jacky^
hi Yuga , 110 V ac ?
need to convert it to dc
i know there is a voltage change when rectifying... just dont know what it is
I think would be 158.4 DC
hmm... so if i wanted 120v dc... what size transformer would i need :)
that is where my problem is lying
im not 100% sure, but the formula would be AC * 1.44 = DC
well atleast i have something to go by now :)
before i was shooting in the dark
just didnt know it was asmuch as 1.44
seems you where close :) they say 1.4
for the capacitors about 1000 uf/ampere would be ok too
and for the caps voltage you should stay about 1/3 up the nominal DC voltage
i think youll get big size capacitors
well i think i will be needing 150v cap
problem is that they are REALY expencive :)
yeah.. may you can find something as surplus
problem with that is if i ever need more i am not going to be able to find them :)
how much current ?
would perfure to buy something i can probibly find in a years time again
do you have
need 10 amps at 120v dc
220 input voltage
toroidal transformer amy would be a nice choice ..
what is ? hi-fi amplifier ?
yes... i know... but not the easyest thing to find here in sa :(
even nicer would be a switching powersupply
Yuga 110 * sqrt(2) <--- the answer :) what's the sqrt of 2 thou :)
yeah, probably right
lol... and to think the chap i phoned said i will need about 115v to get 120volts :)
glad i thought he was talking shit
how do you get more voltage with less voltage?
converting from ac to dc
you get a gain when you put in the capacitor
hm why not use a rectifier?
I do that with my welder..
well first a rectifier... and then a cap
ah ok :)
you get the gain when recifying it btw, not from the cap :)
is it possible to make 12v dc higher?
something i have always wondered
Yuga: probably possible, as in the hi-fi car amplifiers
I think not good in your case
any one know of a website where i can buy cap's? the only place i was able to find was in london... and london prices suck
if you have a inverter you can..
* anonimasu has a neat inverter in a project at work
it's a magic device..
costs like 200$
but its magic..
anonimasu: from to ?
no I think it's a bit more then 200$..
12 to 24
the input voltage can drop to 9v without affecting the output voltage if you dont load it too much..
the guys that designed it must be magicians..
we had trouble with the other inverter off mascot fscking with us..
did drop in output voltage or vary..
that's death to electronics :/
Yuga: rscomponents could be a nice place to look
but expensive ..
friggin hell.. i have a lag
i just got that all now
anonimasu... i'm sure the website i was just at said u get the voltage gain from the cap... but hey... i could be wrong :)
hm I am running debian off a usb stick now
atleast the install
taking a look
they may work on request at any voltage
making your own transformer
I bought a toroidal here years ago, but the voltage was about 70 V DC (dual output)
I payed a lot of money for that :(
friggin hell.. i dont wanna pay alot... they must give it to me cheap!!!
now 120 V dc out seem to me a lot
it seems to me a bit 'out of standard' too
probably hard to find ready to use
well it needs... 220 in... 85v out 10 amps
thats what i need
but not to find the bugger
whats the final side ?
transistors, Igbt , ?
85v out 10 amps
ive no idea about transistor or mosfet using that voltage
in most cases ive seen to get more power peoples use bridge amplifiers
well that just sux...
I wonder if is possible to use 2 transformer in series
transformers suck... wonder if there is any one out there who makes custome switching power supplies :)
that's what i need
not a clue
jacky: yep you can use em in series to boost output voltage, or parallel to boost va rating
Yuga: you are looking at several 1000$ for stuff like that
large power supplies :)
anonimasu... well the last one i found was 160$
not quite several 1000 $ :)
Yuga: it seem to me a nice price
jacky: 85v 10 amps isnt too large, have 80v 20amp torroid that im using in the lab, think we paid �50 for it
50 pounds isnt that bad
delacroix: nice, first time I eard
delacroix... about how much the bugger weigh?
I got my own transformer just requesting it made by hand !
yuga: is in an old unregulated rackmount psu, bloody heavy
[11:46:45] <Yuga> http://www.trcelectronics.com/Cosel/pdf/pba1500f.pdf
<---- nice... but to little voltage :<
but i guess the 35 amps is abit excessive :)
ugh from 10 to 35 ..
are just 25 amp :)
still to low voltage
half of what i need
Yuga: id look in a surplus shop
but have no idea about websites right now
wish that there just wasnt sooooo much crap on google
one thing you might be able to get is a transformer winding kit, where the primaries done, and you wind your secondary of choice, maplin.co.uk used to sell em, depends how much work you want to make for yourself
easy to find switching powersupplies... just difficult to find the correct voltage
jacky: in case your still looking farnell stocks a 2x45v secondary transformer with 1000va rating, that would give u just over 10 amps per transformer
and its cost new is �48.85
have been shopping for psus myself today, have a robot to build
delacroix: thanks, but Yuga is looking fot that
sorry, not really with it today
delacroix: are you interested in robot ? nice
im building this: http://www.deagostiniedicola.it/frontend/home_prodotto.asp?prodID=90
im studying for a phd in cybernetics, so i get to build loads of cool toys
[13:03:27] <delacroix> http://www.isrg.reading.ac.uk/haptics/index.htm
some of the stuff we do
really nice , congrats
friend of mine has built a cool little robot recently, it works like a segway, it has two wheels and uses accelerometers to balance on them whilst moving, so it can go over rough terrain and stay upright
Id like to learn more about aero-robot
such small helicpoter with autopilot
had some friends doing a small helecopter project two years ago, it wasnt that great, but the idea was sound. It was an electric rc helecopter kit, with an onboard embedded computer, ultrasonic collision detection, and a webcam for vision
could lift off, navigate the corridors of the building, and land on a target on the floor
delacroix: this is weird http://stor.altervista.org/4c/4c.php
i was really impressed by videos on this page http://www.draganfly.com/
I like the Draganflyer X-Pro
the dragonfly is a pain in the ass, thats what they were using, and its load capacity is tiny, plus it has like 10 mins flying time on a good day
also for what it is (4 motors, gyro board and battery) its quite expensive
i reckon you could build a far better one for a fraction of the cost
anyway ive got to go do some work on campus now, so will be afk for a while, later
Jacky^ is now known as Jacky^afk
jmk and I are both working on tcl configuration.
I need to get home to look at it. still at work right now
Let me email you his. Mine is in sf.
does anyone know anything about measuring arms? like this: http://www.sheffieldmeasurement.com/sigmaarm.shtml
rayh: please do ..
fenn: do you know anything about 6-dof measuring arms?
like this: http://www.sheffieldmeasurement.com/sigmaarm.shtml
they are expensive
high res resolvers at each joint, a little math, and presto! - you've got 6D endpoint info
SWP_Away is now known as SWPadnos_
SWPadnos_: that's what I want
which, the math, or the arm? ;)
i'd doubt you could very good precision out of it
I need +/-0.1 mm at the tip
so not very much precision
not sure I can
and get something ok out of it
I love how it doesn't actually tell you the resolution
it needs to be pretty robust
I had a thought ;)
wanna use such a device for teach-in programming for the robots :P
for your robots, you should use the encoders that are already on the joints
put the controller in "DRO" mode, and just hit a button to save the pose
yeah.. but the robot is 250 kg
can't easily move it by hand
it's rather cumbersome
make a manual powered mode
there are probably safety regulations to deal with also
how would one figure out how much amperage is required to drive a amplifier well... any one have any idea's on this?
EMC's probe stuff could capture the touch contact at the end of the arm.
the current depends more on the motor, I think
yuga a bit more than the motor rating
get one of these instead: http://www.ascension-tech.com/products/6dmouse.php
or any of their other measurement devices - they're all cool
though you'd need one of the more accurate ones
cant you do this stuff with LED's and a couple cameras nowadays?
fenn: not industrial grade
that arm is +- 0.024 mm for the 6-foot length, and 0.11 mm for the 17-foot version
The encoders would have to be very high resolution to see small rotational differences.
I'm thinking 10000 counts / rev
or the like
closer to 50k, I think
I may have missed an atan in tehre somewhere, but 0.11mm in a 17-foor radius is pretty darned accurate
i bet you could do better with a couple telescopes and some LED's
even digital cameras and LEDs wouldn't be that good
accuracy issues aside
well - it's all about accuracy
but you'd also have issues with setup
you need the cameras to be very well aligned on perpendicular axes
uh, not really
or apply kinematics to the returned images;
if only I still had the 3D untethered measurement system we made about 15 years ago ;)
talk about a monster.
but you'd still need very precuise knowledge of the actual position of the cameras
you would have to calibrate it with an object with known dimensions
or some very nice length standards to calculate it from
ok - so your accuracy would be no better than the standard used for calibration
fenn: imagine I need to use this inside a workpiece
or some kind of fixture is blocking 50% of the view
not gonna work
the sonar system we made would work, but setup can be a bit of a pain
these days, it could probably have the resolution you're looking for (it was around 2mm over a 200 foot cubic volume, back in the 8051 days)
need to run it like the transmit/receive stuff on an MRI or CAT setup.
no - 200 foot radius hemisphere, not cube
SWPadnos_: think you miscalculated that..
I don't want one big arm
this was untethered
I have small joints, each with it's own precision
you carry a small unit around, and hit a button to get the coordinates
or only XYZ?
no, only 3 for that one
I thought of DGPS
but .. problems inside
hard to do inside though
what's the budget (if there is one)?
so the best way would be an arm
on devel? or product?
offline software is about 15-20k
so less than that
or is this something to be sold to customers for teaching their robots?
ah - OK, then devel ;)
big customers, which have more than one robot
ok - so they want to move the thihng around
check out the flock of birds from Ascension tech
it's not intended to stay on the robot
you move the base, and the sensors go with it
alex_joni: custom stuff?
not fully sure
alex_joni: arent you building something?
jepler: are you there ?
cncuser: what's up?
* alex_joni smells python stuff
I'll answer your Python question, but in return you have to contribute a 40V 10A power supply to the AXIS developers.
keep dreaming :P
the one that runs cradek's machine died
oh - so now you want to test with real hardware - great ;)
jepler: i didnt manage to get your ulp to output gcode. i allways get a empty ngc :(
jepler: I bought that supply from someone on the channel - I just don't remember who it was
jepler: I wish whoever it was had another one to sell me!
cncuser: oh -- I think that version from cvs may just create a file in /tmp that you have to run with the "scr" command
once a sucker, always a sucker
fenn: it was a great supply, but like all switchers, one day it didn't work anymore
cncuser: try entering this command at the eagle prompt: scr /tmp/command.scr
cncuser: after running gcode.ulp
if the screen flashes several times that means it probably worked
cool, it did something :)
did you get an ngc file now?
jepler: yes, you da man :))
i need to reboot to see what emc says :)
see you later
oh well, too late
jepler: what was all the scr stuff about?
fenn: eagle's scripting is a bit .. strange. There's the "user language" which can do things like if, while, and create output to files, and then there's the "script language" which can do things like change the board.
jepler: cool, should i run it against the gcode data ?
fenn: to find the tool path to mill the traces, you create a polygon that eagle intersects with the traces, and then output ngc commands that go along the border of the polygon
fenn: but that involves running an ulp that creates an scr which creates a polygon and then runs the same ulp again but with different parameters
cncuser: running the program from that url on the .ngc file that gcode.ulp created will reduce the length of the G0 moves by reordering them in a safe way.
any one know of a cheap online store where i can buy some capasitors?
Yuga: In the US I buy mostly from www.mouser.com, though www.jameco.com and www.digikey.com both have large catalogs too.
yuga: however, shipping is generally $6 at least, which makes small orders impractical
jepler: i see
yuga you should probably mention that you need 150V 10,000 uf capacitors
jepler: nice :)
[17:20:14] <fenn> http://www.thlaudio.com/bhcapitme.htm
6$ is nothing in shipping really
over here you pay like 14$f
oh yes... i need a 10, 000uf 150v capacitor :)
or well about 10 eur..
hmmm that page used to have links to each product's page
anonimasu: it is if you want 20 LEDs @ $0.17 each or something like that
finaly found what friggin current i need to be able to draw with these amps...i was 1 amp out :)
I've flown stuff that weighs like 20grams, ;)
Jacky^afk is now known as Jacky^
Yuga: 150V 12000uF $37.32USD qty 1, $31.72 qty 25: http://www.mouser.com/index.cfm?handler=displayproduct&lstdispproductid=442484&e_categoryid=387&e_pcodeid=75007
kind of horris
actualy difficult to believe how expencive they are
those are "computer grade", whatever that means
maybe cheaper to buy 2 smaler ones instead :)
anyone remeber a command to search only for file up to 100 mb in a disk ?
"computer grade" means "i will blow up if you look at me wrong"
find -size 100000000c is it correct ?
Jacky^: find ... -size +100000k
the + means greater than
jacky's will only find with exactly the stated size
any file up to 100 mb
then you want -10000k, for less than
with + option seem work ok
/dev/hda4 36384624 29692880 4843468 86% /
found some zombie ripdvd file ..
(200V 10000uF also around $40 qty 1)
hmm, damit, i allready forgot everything about how to calculate input / outputscale :(
cncuser: thats not simple at all
in the ini file?
depend on a lot of things :(
especially using pinion reductors
Jacky^: well, the steps and the length of movement on one full turn
but i dont remember how to calculate
cncuser: steppers or servos?
SWPadnos_: steppers, and i have 3 millimeter movement on a full turn
that's per full turn of the motor shaft?
in german its calles spindel steigung, dont know the english word
ok, and how many microsteps (or halfsteps) per full step?
lets assume 1/4 steps
and this is final movement of 300, after all drive belts / pulleys, etc?
not 300, 3mm
well - that would be good to know ;)
the general formula is:
(steps per motor revolution) * (motor shaft to screw drive ratio) * (number of screw turns per length unit)
in your case, you have 800 turns, I'll assume 1:1 drive, and 3mm traveled, so the scale is 800 * 1 / 3 or 266.6666666666666666666666
(i inverted the 3mm to get 1/3 turn per mm)
steps/rev * rev/rev * rev/mm = steps/mm
yep - always balance units
swpadnos, my calculation ;)
man - 800 steps, that was supposed to be (I need to proofread better)
hmm, well need to gather some infos, i millimter is about 6 millimeter :)
do you have the drive in 1/4 step mode?
SWPadnos_: i assume it, how can i tell ?
you could set emc to 200 steps/mm, then command a 1mm move, and see if the shaft makes 1 turn or 1/4 turn
INPUT_SCALE = 266 0
OUTPUT_SCALE = 1.000 0.000
so i say input_scale = 200
you may want to disconnect the motor
that way, you can do experiments without breaking anything
just a general question.................
I can give you a sergeant answer :)
I'm looking for something to convert small segmetns in g-code to arcs
heh - that's what I was going to say ;)
swp...you know what 2/3rds of a pun is?
i gett that many out of smething ad other errors, is there a way to diable all secutrity ?
just for getting it going i mean
gonna run...go make chips
hm slashdot 503:ed
* chinamill is away: eating
fenn: hehe, i mean the emc security stuff ;)
cncuser: I dont know what you mean
emc has security?
jepler: you around?
03alex_joni * 10emc2/scripts/emc.run: marked emc.run as obsolete, eventually it will get removed
03alex_joni * 10emc2/scripts/emc.in: added sudo to the KillTaskWithTimeout, some processes might have been started as root, only root can kill them now
03alex_joni * 10emc2/docs/INSTALL: added info about the new runscript
03alex_joni * 10emc2/README: added quickstart info to README, this also exists under docs/INSTALL
Jymmm is now known as Red70sShow
Red70sShow is now known as Jymmm
03alex_joni * 10emc2/docs/NEWS: added some information to news, developers should add important changes & bugfixes in here
03alex_joni * 10emc2/scripts/emc.in: added sudo to the gentle kill aswell, the delay killing an app with -9 is too big for normal use
I just got back in..
going to try to get the Z axis assembly finished this week
rotating nut :)
I know you are
alex_joni: I've got a final design done ;)
the machining of the nut/bearing assembly isnt too complicated
but there's a thread there
that needs to be turned..
a bit tricky...
wow found a free video editing software with fantastic effects ! http://digilander.libero.it/jackydgl0/cinelerra.jpg
a bit hard to learn .. :/
could someone look at the circuit and see if there is an issue with it. seem to be getting noise into it. http://www.electronicsam.com/images/KandT/test.jpg
Sorry about the quality.
waht are ? optocouplers ?
skunkworks: I would put an RC on the signals aswell
you got a scope handy?
any ripple on Vcc?
I will look tonight - at work right now. mainly looking at the pulse train so far.
the blocks on the upper right is the drive opto - isolators
step - direction
ok.. if Vcc is ok, I would move the RC to the signals
R in series with the signal, going to the opto, C facing ground
the rc - are in the drive itself
then add yours
what noise do you get?
false / lost steps?
ok - we kinda tried that - but a resister and cap from b+ to the oc output.
seems most of the time we are randomly gaining if it doesn't stall
I see.. well.. an RC surely will make a difference
would be good to scope the signals after the opto in the drive
stepper sounds like crap
sometimes the optos don't work on some voltages
so what rc do you think If I may ask - are you also saying between the b+ and the output of the buffer?
and resistor in series (parport-resistor-drive) and a cap to ground
it is a buffer/inverter chip (don't remember which) between the parallel port and the drive. Open collector. will try putting a resister then cap to ground and see what happens.
I think I mean a cap to ground on the output and then a resister to the stepper drive if I understand it right
what values - wait wait
heh.. a few kOhms
but you need to test, and look with the scope
.01mf or so?
I guess I could calculate the bandpass I want huh
anonimasu: thinking with autocad ? :P
thinking about my machine
spinning the nuts or the screws..
pen and paper ?
in my head right now
oh .. id ask to les_w
could have some interesting suggest
some real high end machines do that since it allows faster speeds..
never spun a nut - sounds like it would hurt. depends on what you are trying to acomplish if the nut is lighter than the lead screw - less mass to spin
yah - what you said. slow typer.
but I dont know
I dont really have a need for that kind of speeds..
there's no control to push my machine along at thoose speeds..
so I dont really know which way that's the best
either one is fairly easy..
the only issue would be grease slinging, but I can mount central lubing to combat that..
what are your thoughts for a spinning nut? timkins? thrust + bearings?
anonimasu: talking about Z axis ?
Jacky^: hm, X and Z
maybe Y too..
not thrust bearings..
just normal ones...
it should be plenty for the kind of load..
if you wait a moment I'll draw you how my Z looks
im not a designer, but im interesting in see it
this is the first machine I designed as engineer: http://digilander.libero.it/jackydgl0/photos/lab/img011.jpeg
bauahahaha lol :)
done in a bit..
I'll draw the bearings red..
jack - looks like wide carrage printer parts ;)
jacky - sorry
I have some linear steppers from line printers if you need them ;)
thanks, im switching to servo systems :P
I found some hp printers using small servos and liner optical encoder
nice toys, really :)
drawing bearings sucks ;)
done in just a bit
drawing the gear + ballnut there..
saving the img
Jacky^: still there?
[22:36:15] <anonimasu> http://io23.net/img/
the ballnut sits in the middle..
should be a |-
or well a flange the nut screws too but I dont have the specs for the nut right now
it looks really nice
not too hard to machine either..
the only hard thing is the thread on the thing in centre..
thats what I was thinking ..
should be turned so it's concentric..
perferably the wrong way..
may using a 'special tool' ?
the part isnt too large :)
not really a problem though..
cool, really nice design
2 bearing surfaces and a mounting flange..
as simple as it gets..
the thread allows for preload ;)
I could have decided on ><
but it's not nescessary
since the Z is loaded down always anyway
I want to show it to les_w if youre not around later
but looks a nice solution
youre lucky if able to machine that
heh, it's just some turning
I suppose it should have little play too
if there's play it's a scrap piece..
that's the issue
I've got a machine that's accurate way beyond 0.01
it seem to me youre taking the advantage of lathe use in this case, right ?
but if my mounts for stuff have play and stuff it'll ruin it..
I'll turn the bearings surfaces at the lathe at work..
I can turn them in a single pass ;)
yes, i know
Jacky^: have you ever been turning anything?
no, I just heard of that by my cousin, he use a 3 lathes
ah ok :)
Ive also seen some nice job hes done
it's very very simple compared to milling
I have a change of the design..
I'll mount the nut on top..
with the flange into the |--
that's better then I drive it in the opposite end of the bearings..
um, sorry if I rant a bit
hm should be a easy part to machine..
what about internal lubrification ??
I think I'll go with sealed bearings..
should be ok
and lube the nut at the lube port on the flange if there is one..
if I could get the flange(nut) on the machine away I'd make one today..
I am starting to itch for a cnc
im fighting with cinelerra, nice software but hard to tame..
a bit wilde :/
but funny :D
hm, I wonder where les is
i think hes busy with his work ..
hm, the ballnuts needs 28mm
I might machine it there directly..
but I need to ask somone about screw dia
16 or 2mm ballscrew is the issue
ah well not too large anyway..
let me get a rulet
how much long the travel ?
for all axis ?
I've got ballscrews for the other axis:es
the issue is only with Z
the ballnut ends up as 38mm <-->
way too large to fit..
with a 25mm screw
besides the pitch of the 25mm screw is larger..
the travel is like 0,5m
hm, if I can mount it in the bottom anyway.
and preload a bit more I can fit 25mm
[23:09:14] <anonimasu> http://www.solectro.se/Archive/Documents/Produktblad/213700%200005.pdf
hm, I think I'll go with 16mm
but when I think about plunging into stuff..
but fitting 25 isnt possible
yeah, 16 mm should be enough
but the load rating for the 25mm one..
I dont think it do a great difference
the screws are from isel, so they have a ok price
like 130eur for the nut..
and like 150 eur for the screw..
or something like that
any one here know the standards for a 19" rack? trying to find out where those holes are placed that hold it in the rack?
they dont have as high specs as the skf screws, but since the z axis has gravity it's not a issue..
im not updated about quotes
the skf screws costs 12 times as much
or something like it..
but like 3x
than these are cheap
skf are very good brand and quality materials
but when it need to work for a life
skf is great..
but the price isnt
I agreed :(
but the good thing is
it seems there are good materials from chine in the latest time ..
hi robin_sz :-)
it seems chine is coming up and were coming down ..
hm, chineese mills suck.
for what I know, yes
but the truth is that they can offer really good materials at lowest cost
aaarrrggg... can find the rack width.. but cant find out the height... this sux... google isnt my friend any more
Yuga: are you sure youre search isnt filtered ?
let me check ..
what youre searching ?
+international +standard +19 +rack aswell as going to some company's websites that actualy make rack mounted stuff
lol... kinda guessed that one :)
i know the size of the box... battling to find the placement for the holes that are used to fix the rack into the case
remember me tomorrow and I'll measure
need to know tonight... need to send pdf's away
cant help you then :/
Yuga: tha only thing it comes to my mind is rscomponents website
try hp or cisco..
wait a bit
* robin_sz hates RS
lol... some launguage i dont understand :)
try Farnell.com instead
it would have eng version too :(
anyway looking for some data ..
there are not pdf
Module rack for 19" mounting system allowing the integration of all modules in EUROPE format,. (100 X 160 mm). Useful width 84 units (1 unit = 5.08 mm) ...
[23:38:13] <Yuga> http://www.hammondmfg.com/pdf/CCR35TZPL.pdf
<--- this look correct?
is that '
no clue :(
something soooooo simple... but yet so complex :)
that doesnt look correct
back to square 1
what are you tryingto do? a eurocard rack is not THAT hard to find surely?
or are you looking for racking strip to fix in a cabionet?
robin_sz... to find the measurements it is
robin_sz: he needs some details about it
making my own rack... need to know the exact measurements
1U = 1.75 inches
3 holes per U
that become to be lithening :)
size of the hole?
depends what size bolts ;)
or captive nut
5mm drill, 6mm tap
so there would be 59.267 mm between the 2 holes on a 2u rack?
there are 44.45 mm per U
thats one hole every 14.81 mm
I make it 59.24mm
I am not sure if being out by .01 will have any major problems ;)
but you never know :)
you can be fairly sure about it