what are you using for a backlight?
I thought it was busted?
i found that if i plugged in CN02 halfway it turns on
i think it's a min. brightness still
that's the one with backlight enable and brightness control
i had the little one working fine and then i foolishly reset the display settings to defaults (and it broke forever)
need to replace the fuses in the other big one now
or figure out what's going wrong with this one
i'm using a 12V 2A wall wart to power it
it draws about 1A
which one is that, the one that was gutted?
SWP_Away is now known as SWPadnos
03alex_joni * 10emc2/src/emc/usr_intf/halui.cc: added some minimal functionality to prove that it works, right now from HAL machine on/off can be toggled by using 2 pins. Names have changed a lot, final naming is not 100% decided
03jmkasunich * 10emc2/bin/ (.cvsignore emc-config.in): removed emc-config.in, it is no longer used
so, it looks like the average temperature there for this time of year is around 70-ish
SWPadnos: But if you're going into the city (SF), you bring a jacket with you all year long.
heh - I know that one ;)
I may do that one day. I'm not sure I can go to that area and *not* go to the House of Prime Rib
SWPadnos: It's been raining on and off, and had snow on the hills at night.
snow - how novel. maybe I'll see some this year
you know - in California instead of Vermont
weird damn weather... sunny this morning, raining right now.
err more like showers
I usually bring a raincoat-like thing
I have a pendelton in the car, but the rest of the time I'm in shorts and a tshirt
all year long.
jmkasunich: its starting to make sense now.. when the brightness is up past a certain point something (current or temp limiter) kicks in and resets the display/turns the backlight off
and apparently the power supplies are all busted
plural? I thought only one was
i replaced the fuses on the other panel with wire and it works at low brightness setting
Jymmm: just lcd panels with vga input
fenn does it hum?
then you turned it up and blew up the power supply? thats what I did
jmkasunich: well this other big one never worked at all until i plugged it into the wall wart
inverter boards go bad alot in LCD displays.
having those little arrows is handy too
or is this JUST the backlight?
its the inverter or maybe the brightness control chip
those power supplies are too big anyway
unless what is turning off when the brightness is too high is the wall wart :P
is anyone using a tool setter by chance?
like a touch probe?
I guess, or non-contact
no but it doesnt sound so hard
Well, I wanted to see others, but the non-contact ones are "break a beam", instead of " measure a distance"
not all that much on ebay
well you run into problems with capacitive sensors because the geometry is different for each tool
I was thinking laser measuring distance, not capacitance
not ultrasonic either.
how does that work?
i dont understand
you use a fast oscillator for long distances, and interference for small changes
how do you reflect a laser off an endmill of arbitrary geometry?
I remember using a special microscope that you would focus on the low part, reset the meter (which was a modified TI calculator) then focus on the new height. and the difference was the thickness.
fenn they use lasers to measure chemical content/type/amounts in smoke stacks and in medical fluids
ands gases in fluids I mean
I have nfc how, but I have seen it when I worked in a BioMed lab.
but those things are transparent and of a known geometry
well how does an optical mouse work? No more metal mouse pads these days.
it has a little camera that picks out specks of dust moving around
fenn a laser tape measure --> http://www.professionalequipment.com/xq/ASP/ProductID.4134/id.15/subID.125/qx/default.htm
here's a shitload of them... http://www.professionalequipment.com/xq/ASP/laser_measuring_tools/id.15/subID.125/qx/default.htm
+/- 3mm is better than i expected
much better than the laser rangefinders we had in the military
(15-20 years ago or so)
well, they were designed in the '60s or early '70s, I think
so it's not too bad
SWP: I would think a laser rangefinder and a laser tapemeasure are completely different
those were either 1M or 10M accuracy back then
these look more like rangefinders - there' sno tape on them, it's a beam only
+/-3mm over say 20ft (tape measure) and +/- 1M over a long distance are two different things
+/- 3mm over 100m is pretty good
but you're right - when the normal range is 1000-3000M, +/- 1 is pretty good
though these days, all you need is a faster oscillator. 150 MHz gives you 1M accuracy
better than carrying a 650 foot tape
that tapemeasure thing is +/- 3mm typ, +/-10mm max
typ appiles to less than 30m total
bah what does "max accuracy mean"
max error would be a better term
typ is specmanship
max is what they are promising to meet
never heard that one :)
* Jymmm smacks fenn for goin off the subject!
* fenn tosses Jymmm on a specmanship and launches it into outer spec
now that fenn knows "how" a laser can measure shit...
nope still clueless
a laser can't measure shit, its not reflective enough
1) turn on the laser beam for a moment
has anyone seen a optical non-contact tool setter that measures distance?
2) start a counter at a known frequency
3) stop the counter when the reflection comes back
see you later Alesx
err - Alex
if they have laser radar guns, they can calc distance too.
not unless you are counting at 50GHz
that's how the old ones worked, which is why they were so inaccurate
so to measure within .001" we only need a 12 teraherz oscillator
150 MHz is all you need for 1M accuracy
i must be doing something wrong
so 1.5 GHz, plus a phase comparator, can get you very good results
I suspect they moduleate the beam at some moderately high frequency, and use phase relationship between outgoing and reflected
Modulated Beam Systems use the time light takes to travel to the target and back, but the time for a single round-trip is not measured directly. Instead, the strength of the laser is rapidly varied to produce a signal that changes over time.
that can only get you the "precise" portion of it, it can't do the gross measurement
good enough for me
The AR600 laser displacement sensors are a series of high accuracy measuring devices for industrial Acuity laser displacement sensorapplication. Models vary in measuring range and can achieve resolutions better than 1.0 micron.
successive approximation - modulate at 10KHz, then 100KHz, then 1MHz, etc
ok, if you use the phase relationship of the modulation signal, not the "carrier" frequency
the carrier is the light itself
thats how you get down into the micron region
with wavelengths in the micrometer range
but that takes more $ than those tape measures want to spend
The AR600 optical measurement sensor projects a beam of visible laser light that creates a spot on the target surface. The reflected light from the surface is viewed from an angle by a CCD line scan camera and the target's distance is computed from the image pixel data. The size of the measurement triangle is a function of the sensor span and case length.
for short range sensors, that's great, but I'm pretty sure the long-range laser tape measures don't use optical phase
ah - the easy way
that's a high-school physics experiment to measure the speed of light
can't remember the name
[02:49:24] <Jymmm> http://www.acuityresearch.com/Resources/time-of-flight.shtml
that doesn't measure the speed of light (IIRC) it measures the absence of aether
it makes relative speed-of-light measurements in two paths at 90 degrees to each other, but neither measurement is absolute
I guess I should read more than the google summaries ;)
(I went to CWRU, where the Michaelson and Morley did the first experiment)
well, I have the excuse that it was over 20 years ago that I did the experiment
SWPadnos your old excuse is getting old now =)
I guess I'll have to learn a new trick some day
you need a new old excuse
or recycle and get a old new excuse
the dog ate my excuse book
ok, so optical tool setters dont exist then, huh?
sure, but they're beam break devices ;)
thats a optical switch to me.
you could use a camera, and write software to determine the tool length
I find that hard to believe... I seen that EONS ago and worked great
ther eare some that do that, and even get the diameter
I was thinking two part...
one to set the tool length (zero out), then another to set to top of material
you never have to specify the material height.
fenn you use a touch probe?
SWPadnos Heh, I'm clueless to figuring out a laser + ccd as a 3d scanner
OK. Foucault and Michelson both did the rorating mirror thing, but not together
not 3D, you backlight the tool, and look at the shape
if the camera and lens are chosen well (and set up correctly), you can get very accurate profiles of the tool
even the shape
Jymmm: no not yet
[03:05:30] <Jymmm> http://www.sjbaker.org/projects/scanner/
i think you'd have to spend a lot of money to get a laser with a tightly collimated beam
fenn ah, ok.
you dont even need the laser actually
[03:06:47] <fenn> http://www.vision.caltech.edu/bouguetj/ICCV98/index.html
its amazing how fast bookmarks go stale these days
there was another one that did it with just the silhouette of the object
[03:08:13] <Jymmm> http://www.vision.caltech.edu/bouguetj/ICCV98/images/demo_scanning.mpg
you could only do convex things of couse
i liked the structured light with a video projector link giacus posted the other day
gets more info than just geometry
I have a digital video camera... might not be a bad thing to try out if all that is needed is some software and a stick
look at debevec.org if you're interested in this sort of stuff
btw jymmm i thought it was funny you were so impressed by that rendering of a warehouse.. check this out http://www.3delight.com/Gallery/Collection/01A/Bailey_HR.htm
that's very realistic
thats the understatement of the week
what the heck? http://flexbar.com/Merchant2/merchant.mvc?Screen=PROD&Product_Code=18507
heh - it's optical - you look at it ;)
that's ok.. nutting wrong there - if it works
that's effectively a touch probe that you go 0.02 past touching
you gotta look at it though
oops - 0.002
right - so it's a non-automatic touch proble
you think a movement differencial of 0.0003" in a switch is good enough?
for wood, very likely
aerospace, maybe not
what kind of switch is good to .0003?
you don't say
i like this guy's renderings too http://www.oyonale.com/ldc/english/accident_gold.htm
he has a network of nutty stories for each illustration and they are all linked together
those are good
you can almost make out the license plate numbers too
this is my favorite.. it's just a tutorial scene http://www.oyonale.com/histoire/images/makecloud2_demo_rain.jpg
hard to believe that's pov-ray
After seeing what can be done with a video camera and a ruler, I'm thinking a CCD tool setter might just work =)
what's wrong with a contact probe?
wonder if those $40 cameras would be good enough
SWPadnos why not?
$0.02 cents worth of wire hooked to a spare parallel port pin
you need a camera that can have auto-focus turned off
SWPadnos those $40 caremas are fixed lenses
ah - *those* cameras
well, you turn the lens and lock it in
I ignore them, sorry
again? what keeps happening?
jmkasunich did you fall in?
messing around with stuff in other windows
jmkasunich ah, ok.
Does emc have any built in 'touch probe' stuff, like tcnc does?
Roguish_ is now known as Roguish
knock, knock, anyone here?
lerman_ is now known as lerman
knock, knock, anyone here?
nobody here but us bots
fenn: you're a very clever bot then
no, i still havent managed to take over the world
that's hard to do using only irc
could be worse, i could be stuck in some government mainframe
or in a toaster.. *shudder*
03jepler * 10emc2/src/hal/ (utils/Submakefile classicladder/Submakefile): don't build some things when gtk is not available (GTK_VERSION is empty)
03jepler * 10emc2/src/depend.sh.in: hopelessly old versions of bash don't have PIPESTATUS. Work around with more temporary files
03jepler * 10emc2/src/rtapi/rtl_rtapi.c: fix function name
03jepler * 10emc2/src/Makefile: Use BASEPWD instead of PWD; don't use CFLAGS for rtdeps, that's for userspace only
I'm looking for a BMP image to gerbercode converter,
Where would I search it?
TorbaX: try this http://timeguy.com/cradek/image-to-gcode
it takes a .gif file
I've lost any bookmarks on Firefox (bookmarks.html was 850KByte in size)
Does It compatible with EMC2?
yes but it requires some code tweaking in normal use
It's a bet
in italian I say "...un terno al lotto"
I have a PCB image (png file) and now I'm trying to convert it in a B/W gif image
I've lost ALL bookmarks on Firefox (bookmarks.html was 850KByte in size)
do you have the original schematic files?
I havent any schematics
only one image of PCB
you can use autotrace to export hpgl (a vector format) and hp2xx to convert to gcode, but i've never done it
autotrace has a online service
ugh, I'd hate to start with just an image of a pcb and try to mill based on it
[16:51:35] <Jymmm> http://www.roitsystems.com/cgi-bin/autotrace/tracer.pl
Sorry, does Gimp make a HPGL format of an imagefile ?
InkScape can trace too (and is very good)
inkscape is very good but doesn't export hpgl
i want to make a .svg to gcode converter anyway
no it exports svg and can be converted to hpgl
how od you do that?
Note: InkScape has Floating Point issue that you have to address in the settings somehow (I forget)
If you draw a 4" square, in the SVG it might be 3.999568345"
when i need a pretty drawing accurate to .0005 i'll worry about it
InkScape could be used for a simple cad program if you leave off the added features.
when you get a rainbow reflection effect on carbide, what does that mean?
there are microscopic grooves that are evenly spaced apart
Ah, I just realisted it's mirror smooth except for the rainbow (grooves)
for some silly reason I thought it might have been a heat treatment thing.
jymmm have you actually done inkscape->gcode conversion?
svg to gcode? Not directly, no.
ok hp2xx chokes on the hpgl file
when i do convert blah.svg blah.hpgl
and then hp2xx -m nc blah.hpgl blah.ngc
I dont know what 'hp2xx' is.
how did you do it then?
fenn: oh my bad... Im only supports READING of hpgl, not writing.
I guess I used Corel and just forgot
because hp2xx converts hpgl to gcode
hpgl is a plotter language, very similar
I know HPGL, I have the 700pg spec manual here.
Dont be hatin!
oh jealous one
and hes read it is the scry part
I have the dead tree version, but if you want to dl it, let me know.
i have the hpgl spec manual and yes i have read it several times.
i dont want to have anything to do with hpgl
i have the APT manuat (Automaticallt programmed tool)
you just said you're converting them
i want to go .svg to .ngc, and i thought i could do it with an intermediate hpgl step, but i guess not
general question, what precision do most of you require? +/- 0.001 or +/- 0.0005 or other
if i have to write code i might as well write what i want in the first place
fenn which is?
.svg to .ngc
more precision is better.. look at a drawing..
whatever precision is required for the application :)
makes for better accuracy and repeatability figures
dmessier, more precision generally costs more up front. precision groung ball screws as compared to precision rolled ball screws.
and you know if you built .0001 int the machine you dont need to check +/-.005 dimensions
welcome to the wonderful world of engineering
fenn hp2xx accepts svg
pay now or struggle forever after ... your pick
it's called hp2xx because it goes *from* hpgl to other formats
fenn: did you bug dmessier ?
about that song?
fenn try it
dmessier: we need your french skills
heh it's your turn no alex
* fenn throws a cheese doodle at dmessire
dmessier: a song, partly in french, but we can't make out the words
maybe you can do better
possibly in french
fenn: please post the link
fenn: it is in french
[17:20:05] <fenn> http://fenn.dyndns.org/pub/irc/Cowboy
Bebop OST 3 - 14 - We Qui Non Coin.mp3
the first half is japanese
i am trying to decide between precision ground ball screws or precision rolled ball screws. they would be used in building an xyz table for waterjet cutting.
waterjets need heavy table load capacity..
if you're going to spend the money for a waterjet you might as well get ground ballscrews
dmessier, not so much a table but more like an xyz gantry with a water tank beneath it.
id say a good set of rolled screws would handle it
thats NOT french
dmessier: only the second half
the first part is japanese
i can get a word here and there ... but they are being VERY creative with the language... : 0
dmessier: right :)
but the first phrase is something like: "je suis sans mon 'sokona'?"
yeah... japanese inter mixed all over
ok thats what i thought
ici et la
dmessier, actually, the xyz gantry is pretty straightforward. the water tank is not that straightforward. one of the biggest problems is sheer weight of the water inself. that may even increased if the water tank is partial filled with ball bearing. than internal to the tank is the supporting grid/grates to hold the material.
anyone ever heard of a Lucida console terminal
sounds like a font
Yes, I also think "Lucida Console" is a font name
i have a postprossing application that pucking
[17:40:33] <jepler> http://www.myfonts.com/fonts/microsoft/lucida-console/
general question, at what point would one switch from using a ball screw in the x-axis to using a helical rack and split pinion?
"number of glyphs: 666" -- microsoft.com page on lucida console
when cutting forces are low and speed is high
jepler, you mean you did not know that bill gates is the beast mentioned in revelations?
3000 lbs or so on big boring mills... and 3Meter travel or more
what is a "split pinion"?
fenn, hang on i will get you a link.
split nuts along the lead screw
n/m found it http://www.atlantadrives.com/zerobacklash.htm
hey les how's your chicken.. died of neglect yet?
have two now
no they were fine
I have been out detailing the car
how can i run a 16 bit app in XP ooop
seems a truck I was behind spilled some paint
the little spatters came off easily with a detailer's clay bar though
oo this is neat http://www.globalspec.com/FeaturedProducts/Detail/NexenGroup/Ball_Screw_Alternative/26600/0
that is neat
dunno what "cumulative errors in ballscrew systems" they are talking about though
i saw a DIY version of this before somewhere
do the rollers float?
[17:52:49] <terrylr> http://www.americanmachinist.com/304/Technologies/Turning/Article/False/8605/
twin opposed master slave drives... i love it
seems silly to me
we used it in a nutating head system to keep the head from tilting all the time
hmm can both motors accelerate in the same direction?
but one is technically turning the other direction
but do they both accelerate the load?
ahh crap-n-marshmellows... any one else wanna try a kick with this post-processor.. or the .cl file to make man-readable??
i've managed to find out that my transistor h-bridge is a disaster
hi Jymmm :)
Bo-Dick let out the magic smoke?
giacus: Somebody said you posted a link to 3D scanning using a video camera. Do dyou remember what that was?
Jymmm: yeah, a moment ..
i haven't discovered until now that transistor bridges aren't very good when the resistance of the load is low.
the interesting part for me is the c file ..
I'm studing around it
and I have to try it yet
giacus: We found that one last night. found another one that uses just a desk lamp and a camera (no laser)
it'll work quite decently on my small 75 ohms motors but it won't perform at all when the coil resistance is in the region of about 1 ohms.
Jymmm: any scanned sample about how it work ?
yeah. let me find the link
[19:35:47] <Jymmm> http://www.vision.caltech.edu/bouguetj/ICCV98/
btw, when i try to connect my pwm current limiter in real-life the comparator blew up. it looked good in the simulation but in real-life there were nothing but smoke.
i think i'm gonna go linear for the current regulator.
Jymmm: nice, I already seen it time ago
what i haven't tried yet of course is if the bridge would perform better in a "tilden" configuration. anybody heard about tilden-bridges?
I bought a small laser level, I'd like to test it in the next days (if I found enough time)
...so my dream about the universal stepper driver is long lost.
i made a simulation with FET transistors and the difference in performance was stounding for low impedance loads.
BJTs simly have to high on-resistance!
Bo-Dick: what about mosfet ?
didn't have any models in pspice for mosfets.
i base my designs on simulations
but i know mosfets are the best
is mosfet the same as hexfet?
Im not familiar too much with electronics, but I remember somewhere mosfets are used for that
that why I asked you
the problem is that mosfets are soo much more expensive than bjt:S
gimme a break they're like $1 each
have you looked into how much endmills cost? hmm?
switching mosfets aren't just 1$ each. bullshit
not at my dealer anyway
Bo-Dick: Try a local crack-whore instead. =)
Jymmm: don't be rude
i'm not smart enough to make the right priorities
fenn: Im back from a very nice day :P http://giacus.altervista.org/photo/hobby/stallman/index.html
* giacus uploading some nice video :D
looks like fun
what did he talk about?
programs was here http://catania.linux.it/
did he absolve you of all your sins?
there's and incredible difference of quality between my sony camera and the Nikon camera of k4ts :(
the mines are dark and poor
k4ats have a fantastic camera, really
I used the mine to take some video..
Bo-Dick: what unit are the prices in on elfa.se?
oops "All prices are in Euro"
i bet shipping from thailand would be less :)
so your conclusion is also that elfa is expensive
[20:04:32] <fenn> http://futurlec.com/TransMosIRF.shtml
well on elfa they are more than 4$ each!
irf530 is $0.24 on digikey
* fenn looks up to-220AB package
i would benefit from ordering them from the U.S
oops out of stock for the $0.24 ones
well i've already spent over 80$ on my non-successful design <:(
it was supposed to be a 4-axis system
after all a four axis gecko system would be slightly under 500$ so quality isn't free.
but 80$ is a lot for a non-successful system
someone should write a script to sort digikey by price automatically
oh that's hilarious.. someone uploaded the binary for a PDP-8 edm machine, and tom decided to take it down because it was proprietary software
now, that is funny :)
wot's cookin' ?
heh, I was gonna say that
* alex_joni had some nice chutney today
alex_joni: I found a neat way to double the production output of the welding plant this week
I wonder what a "management port" is? (on an HP server)
looks like serial
maybe a serial console?
it might be for a serial console (screen & keyb)
jmkasunich: hook a serial cable and start a minicom
robin: how'd you double production
you'll find out ;)
alex_joni: yeah, changed the has to a trimix, with 2.5% oxygen, new liner, 1.2mm wire and most importanly, fitted another torch holder to the workshop
90 8 2 ?
the new torchholder was the key part though
right, that's a good gas for stainless usually
you know .. the 2m tall torchholders, with boots :)
err.. crap, the one for stainless is without co2
robin_sz: lol ;) I use the adjustable type
this is one of the clever self-levelling types
sigh ... clearly my humour is not working today
jmkasunich: yeah that's for console
we're all tired ;)
you know what hard and software are right?
well, most errors can be attributed to faults in the wetware
ahh.. ok got it
so ... in my half crap joke "fitted another torchholder to the workshop" means "employed another welder"
yeah, I got that
first rate guy too, but one of my customers is not very impressed
but when you said levelling type, I though maybe you replaced the welder with a real torchholder
why not impressed... you steal him??
yep :) ... used to be their welding supervisor ;)
i love it when that happens
been there 10 years ... offered to work for me for less than his current rate
get over it folks
well, what does that tell you?
that your customer isnt a skilled manager
* fenn gives some ammo to the terrorists :)
thats one way of putting it ;)
what sotr of welding you doing??
or that he treats employees like shit..
but that equals the same..
good management usually keeps workers atleast content..
the guy did a welding course, in his own time for a year, 2 nights a week
year 2 asked if they would pay the euro 500 fee for the year ....
no way they said ...
it just goes on and on ... ah well, there bad management = my pool of willing labor
yeah as long as you treat them marginally better ;)
I do my best to treat them as well as I can afford to
lear them GOOD welding... and ALL will be yours...
flexible hours, choice of what they do etc
robin_sz: yeah, I were just lidding
cant be beat...
I'm on the cell so I am laging a little bit
I had to do some welding myself today ... first time in 6 months
you up to snuff
sorta ... not as good as the guys though
2m long vertical fillets in 3mm mild steel
they do them in one run, I had to stop a couple of times
i pull a good weld outta my ass when i need it and stop... lol
its a useful skill, but i I wouldnt want to do more than a bit of it from time to time
long and thin was never my strong suit
3mm isnt thin..
0.5 is thin
when youve run 3mm wire... it is
good to be able to pick up the torch and use it, stops staff going "well, lets see you do better then!"
3mm isnt wire
dmessier: if you weld 3mm studd with 3mm wire you are badly equipped for the job
1 mm is about right
cant make steel mill ladles and tundish's without it
coo. big welding
no i understand.... but ive worked on BIG crap.. too
and I guess they like good welds too on that sort of stuff
some guys would burn 300# of wire per shift
pounds = #
errm, only on american kbs
on uk kbs pound = �
goin in was the motto
thats better :)
thats waht, 10 normal realls?
american and english?
well night guys..
our fabrications are much smaller, max 500kg I guess
ive seen BIG fillet fills... get x-rayed
150mm x 45 ged - 300mm x 45 deg
more than one pass huh?
we did a shrink fit on the trunnions of 2mm press fit
* robin_sz wonders haw many amos 3mm wire pulls
nitrogen on the pin..... heat the hole for 5 hrs with TIGER totches
close... wire feed is CRITICAL
I can imagine ...
you want weld, and NO spatter
but when it works RIGHT... you can tell bu the sound..
I was listening to some [ulse synergic the other day
very odd sound
more like AC tig
not odd, that's the real sound
835 A at 1.1 M / min
I'd love a pulse mig ... but not rich enough yet
short arc sounds sick
like it's choking ;)
spray transfer dood ;)
up the power to max and slow down some
I did some 10mm brackets the other day .. only had 0.8mm wire ... so just set 400 amps and turned the wire feed up full speed
had to practice a bit first though ... you have to be gast in spray transfer
400 a little high fo .8
fine for spray mode
but is this a spray transfer machine
or a welder>
if you turn up the big control it is :)
now YOU need a new weldin shield... right>
well, it was alittle hot :)
robin_sz: spray is ok, but pulse is better
I did some tests with Tandem a while ago
2 x 1.4mm wire
went up to 2 x 430A
that was intense
430A median current
10mm plate 1 pass?
pulses were up at 650A
I did a A7 with ~90cm/min
oh, those robots I looked at?
A specifies the height of the triangle
right, probably worth the money, but lots of headaches
the wetware versions are easier to program
btw, I got word from cloos uk a while ago, they had 3 robots for 10k pounds
each or the set?
all of them
with welders & all
spent all my money on a new factory :)
yah, I know what you mean ;)
we're moving too
these days sometimes
a bit smaller, but ours
it'll expand lateron
mine will be 250m^^2, but not mine
we have about 60m^2, and 150 office space
because office is the main activity
rest is only for service & repair & testing
60 is quite small
a big bot would fill that
and bang on the walls ;)
well, no big bots there
it's 15 x 5
75 m^2 ;)
my math is crappy ;)
enough for 1 medium sized bot
and no walls if you press the wrong button
this is mostly for manual welders & such
the walls are 80mm sandwich (metal sheets on the outside ~.8mm, foam between)
almost bot proof
we worry about heat
not about bots :)
keeping the office warm?
a run-away bot can be really bad :D
no, this is for repairs
office is a house
keeping the shop warm?
we have a heat problem too
yup, we do have heating & all (well planned at least, not 100% done yet)
but we solved ours ...
but 80mm insulation all around should be enough to keep the heat inside I guess
we just open all the doors and put some fans on :)
you'd be amazed what a 30kw laser does for your heating :)
30kw laser ????
well, input power
ahh.. ok, got scared
for a moment ;)
30 out would be SCARY
I've seen 5kW out, and it's scary
I think they got co2 upt to about 12
what do you get out for 30kW in?
* robin_sz nods
so efficiency something like 3%
jmkasunich: 3% on that stuff is good :)
yeah, same as about 1500 to 1800 co2
I'm just accustomed to vfds with 97% efficiency
robin_sz: just had a spotwelder at the fair recently that sucks 50kW
and that's a SMALL one
but only low duty cycle
they go up to 630kW
a bit smaller I think
or a bit more
tack .. tack .. tack
just spotting a bit of 10mm together
the drive we were testing a couple weeks ago was running 850kW continuous
tack - sfrrr - sfsshhh
OK, 850 into what? ...
a big motor?
a big wheel
another motor acting as a generator
returning most of the power back to the line
and what was the purpose?
testing the drive
thats one way of keeping the costs down I guess
2x500MCM cables per phase, and they got hot
and this is one of your designs?
next project is a half size one, then we go on to the real fun
paralleling 2 or 3 of the big ones
couple of IR mosfets and a big heat sink?
jmkasunich: how much is your life insurance?
four Eupec 450A x 1200V IGBT modules and a water cooled heatsink
on of my custoemrs uses nic wc heatsinks
like a block of ally
a U shaped groove
and a copper pipe pressed in
our welding sources are usually stainless
well the wc heatsinks I mean
I've seen those sinks
not good enough for our stuff, not enough surface area, and too much thermal resistance between aly and tube
I saw a really crappy welder last week .. cemont? italian thing
open the top, pcb on stick-pad plastic pillars,
IGBT mounted onthe 1.6mm thick case
we take a 1" thick aly plate, mill multiple parallel water grooves in it (about 3mm x 15mm each, 10 or so in parallel) then vacuum braze a 1/2" plate on to cover the grooves
that was about it really
an outside vendor does it
thin foil of braze alloy between the two parts, clamp, put in furnace
comes out hermetically sealed
you can do that with pressure too
the part is about 6" x 1.5" x 24" when finished
seens some machines that do that (instead of heat)
must need a very accurate furnace
robin_sz: btw, with a good power source you can do brazing aswell
you mean temp control?
not only welding ;)
jmkasunich: yeah, temp
alex_joni: the best power source for that has bottles attached ;)
speaking of which ...
no,I mean arc-brazing
just like welding, with wire
prepare for a long but funny story
but bronze wire, and lower amps
I'm not sure how critical temp is, I think the braze alloy melts a couple hundred degrees (F) lower than the aluminum
and the vacuum takes care of oxide issues
we have a long running dispute with our gas supplier ...
the local agent has poor service etc
the rep called in .. I moaned at him about crap handling, lack of bottle trolley etc etc
doesn't ring a bell.. go on ;)
anyway ... dropped a bottle in to swap out ...
the bottle guy was unhlepful .. and the rep turned up bang on cue ...
I asked for the bottle trolley ...
the unhelpful guy just said "oh ... drag it .. move, I'll do it" ...
dragged it 10m down the concrete yard ...
and going backwards, tripped over a palette
and the bottle fell over him?
which had a plastic sheet full of rainwater that he sat in
then the bottle got him :)
how I laughed
that'll teach him
I usually rotate the bottles
very easy to move around
I use a trolley, or I dont bother
need to tilt them 15 degs, then you can rotate with one hand, and steer with the other
yeah, done it .. but i'd rather not
insurance and all that
think these 90,8,2 mixes have anyting going for them?
yes and no
special gases are good, I can't argue that
but they are expensive too
local gas suppliers swear for He mixes
which are darn expensive
well, they would
I usually prefer Corgon-18
82-18 or better (90-10, 95-5 if I can ;)
normally 85-15 for us
right, that's pretty good
80-20 is the limit for pulsed welding
anything with more CO2 doesn't work
good for the 3 to 5mm sort of thing, short arc
short arc should work on pure CO2
but it'll spatter all over the place
you'll get better welding shape though, as CO2 arcs are hotter
and this needs to be watertight when done
so CO2 gives more weld penetration, at the cost of spatter
so for your setup as much CO2 as possible (85-15 or more)
the heavy mix is 25% co2 i think, medium is 15
light is 5 I think
if you add 2-3% O2 it makes it better
thats what they said
yes, they are right
if the cost is ok, go for it
they suggested a 90,8,2 would match up wwell to our 85-15
but I advise against He-gases (where the cost is double)
90,8,2 is probably 10-20% more costly
this comes in a smaller 300bar bottle, same cap as a big bottle
less CO2 (which is cheap), more Argon and O2
with integral reulator
huh, never seen those?
what kind of regulator?
the flow rate is set by an orrifice inthe fitting you connect with
so easy for the welders to swap bottles
yeah, but not accurate
and stops them turning it all up and runing the gas ot in an hour
it costs money to waste gas :)
well, welders always go for more gas than they need .. "to be safe"
ok, so about 12l I'd recomend
15 is ok for 1.2
what gas nozzle size?
you know the wire tip
where the wire goes through
that's current tip or nozzle
outside there's the gas nozzle (we call it), between the two travels the gas
the metallic one
I know ...
which you usually pull-off to put another one
usual sizes are 10-15mm diameter
the one that gets spatter on it ;)
id say, 20mm
its a Binzel MB25 ...
and creates that nice ring.. which falls into the weld when you are not carefull
or MB15 for light stuff
ah .. yes the nice spatter ring
do you use anti-spatter-spray?
sometimes use spray
grease is better
must get some tip grease
welder only dips the whole welding head into it
with wire & the MB25 & the current tip ;)
lots faster, and works better
btw, how do you spray ?
like this ...
that's one of the most common mistakes on MAG welding
<spray> ... <spray> ;)
inside the gas nozzle? or perpendicular?
inside is wrong
but they all do it like that
the idea is to prevent the spatter ring
spatter doesn't get inside the gas nozzle
but if you spray in there, you'll get a sticky surface
where sh*t builds up
we often use a soap-based anti spatter on the workpiece too
which gets you more into trouble then when you don't spray at all
basically a sopay water mix, stops spatter sticking to important bits of steel
but it might cause defects if it gets in the welding area
water & arc-welding don't mix well ;)
03alex_joni * 10emc2/src/emc/usr_intf/halui.cc: halui.motion.is-on now shows the state of emc
as a rule, it evaporates long before
DARN.. I meant machine.is-on
robin_sz: then it's safe
OK, explain to me ...
how did I swap a software contract in geneva for a job as a welder fabricator?
* robin_sz is baffled
you were tired of stamps?
they wanted me in NYC this week too
you know what I remember learnig as a first thing when playing with stamps?
I suspect they may decline to pay me this month in response :)
dont stick them in a book?
no, try not to let your reflexes work
if you drop a stamp, don't let your legs close
just let it fall to the ground :D
and then look BEFORE you move
you can safely pick it up later.. but if you act on reflex and try to catch it with your feet ..
robin_sz: btw, did you know emc2 does threading?
as in on a lathe?
yah, that kind
what is slaved to what?
needs some testing though.. code is still wet
TP to spindle
just basic threading?
basic for now
rest will come later
long awaited !
thanks go to chris ;)
anyone actually tried it yet?
he's our personal TP guru ;)
not yet, only simulating & scoping with HAL
planning on using what for the spindle?
what sort of sensor?
but it's basicly a HAL pin
so anything goes
the code is _very_ wet right now
fitting encoders to spindles can be tricky
it can be a tach, or whatever
I wouldn't go around saying "emc has threading" for a while yet
nah, me neither
don't want to give false impressions
emce has a threading egg laid :)
jmkasunich: but I knew robin_sz wouldn't run off to start his lathe with the latest emc2
but everyone else on the channel heard you say it ;-)
huh? anyone around?
crap - I have the lathe all set up. - no not really.
* alex_joni shuts up
well, I assume they all are clueful enough to know it means "the basic bit of code is there, no gcodes, no GUI support yet"
Mach2, for its sins, uses a clever tach system
about 6 black line son the white spindle
one line is double width for index
and a software PLL
simple reflective IR sensor ...
thats it really
robin_sz: btw, do you know how to do proper grounding?
earthing I mean
on a shop
that spindle sensor sounds interesting
oh, that depends a lot on local codes
maybe at some point I'll make a hal module to do that and generate a more conventional output that can go to emc
local codes only say <4 ohm
alex_joni: really? coo
ground wires should be fat, short, and green ;-)
jmkasunich: I meant the sticks I stick in the ground
and connect the short fat wires to them ;)
alex_joni: star earth point is my personal preference
fat, long, and pointy
alex_joni: foregt the spike it the ground, mostly useless
jmkasunich: they are 1m, 1.5m, 2m
VERY useless in fact
I stuck about 5 of those in the ground
with 5x20mm strip between them
do you have buried pipes? water lines, etc
and salty water around them
but waht did you hope to acheive?
robin_sz: earthing point
well, if it helps you sleep at night
robin_sz: it's mostly for getting approvals
here, we just bond all the water pipes etc to a common point
and connect that to the incoming neutral
oh, that's not allowed here
what about building steel?
we need L1,L2,L3,N,PE
so do we
so N & PE need not to be connected
oh, but here they are connected in the incoming cable
called PME ...
everyone connects neutral to the incoming services and building framework
your system is TTN ...much more difficult to set up
and useless insummer
you need live to earht resistance low enough to pull your biggest fuse in 20ms usually
4 ohms may not be enough
otoh, what I'm setting up is about 16kW
very difficult to achieve i would think
so not crucial
* fenn notes that "ground" is tied to "neutral" in his breaker box..
we call that PME
what does PME stand for?
I know PMS
but PME is new ;)
protective multiple earthing
basicallym the house is a farday cage
its not earth .. but hey, locally, it is
robin_sz: what if it's a wooden house?
oh, this can be a Bad Plan (tm) for farm buildings :)
why is that?
you have this barn ...
and the steelwork is connected to neutral right ...
and one pahse gets broken at the local substation
so neutral becomes a bit "live" ...
no problem in the barn .. and no problem for humans as they walk out ...
but cows are quite long ..
it shorts at wherever neutral is tied to ground right?
the entire building is going up and down at say 200V
robin_sz: at my mountain house, I measured about 30V on neutral
enough to light a lightbulg
its sitting on its own little potential hill OK>
human would get, say max 20V between his feet on a stride of 1m
a cow could get say 100V ...
(mmm .. 10m cow?)
they cover a longer part of the potential gradient
i think my brain just turned off
fenn: what ? you've never seen 10m cows?
robin: its not neccessarily that the cow could get enough to kill it
* fenn goes to raise his blood sugar or something
well, it does happen apparently
I think even a couple volts makes dairy cows stop giving milk
I've read that somewhere
thats even worse than dead
at least dead you can cook em
yeah, and they keep eating
here, we have to incinerate ALL fallen stock
robin_sz: we do that on birds
jmkasunich: dont think we have that any more
hence the incineration to make sure?
jmkasunich: yeah, it was an EU regulation afaik
just general EU stuff I tink
and it made quite a difference
I cut a lot of parts for animal burners
from lots of cases to almost none
although all that might have been propaganda
alex_joni: well, no one owns up to any cases anyway
esp the french
yeah, well the bird stuff worries me lots more than BSE did
I think that was some political scheme to sell more chickens :D
I laser all the parts for these guys http://www.wastespectrum.co.uk/derwent.asp
laser the cows, it's easier :D
I find the whole idea of buring animals a bit repulsive .. bt I just cut steel ...
gorge, n. 3. The throat; the gullet: The gory sight made my gorge rise.
[23:34:47] <jepler> http://www.answers.com/gorge
more like 6 - the contents of the stomach
* alex_joni burns parts of animals
they are called steaks usually
mmmm, cow slices
don't burn them tho, just singe them a little around the edges
you need a very hot fire, to burn the outside, keep the juices in ;)
only 2-3mm on each side
from the 15-20mm slice :D
but not burned, that implies black and crunchy
just seared - brown and tasty
* robin_sz hasnt eaten meat for nearly 25 years now
I thought curries and such had meat in them?
most of india is vegetarian remember
rices and spices eh?
beef curry is a western thing
I'm also vegetarian
I shot way too much wildlife when I was young ... put me right off eating the stuff
cradek: wanna read something funny?
ok, no more discussion of steak
talk all you like ...
robin_sz: most people let the factories do the dirty work for them, and not worry about it
a nice snip: "Change method of passing commands from motion planner to interpolators to allow for control of maximum velocity and acceleration to be set independently for each axis."
cradek: exactly, but I do worry still, as Ive seen it and done it and wont do it muyself
cradek: to "try and not think about it" would be hypocritical for me
robin_sz: I know, I'm not saying that's the way to be
* robin_sz nods
robin_sz: letting someone else do your dirty work doesn't make it ok, it makes it worse.
alex_joni: too many words for me to read!
just what I pasted
he's saying most folkw hide their head inthe sand and pretend meat doesn't come from animals
cradek: I dont get too concerend, as I know I cant change the world, and if other people are happy then I wont try and chagen them, but I have to be happy with my deciisions and conscience
robin_sz: me too.
jmkasunich: yeah, im agreeing with him ;)
oh, right, thats my only other bit of news ... got a new barrel on my rifle ..
now, that is an engineering masterpeice ...
first one on a LONG while
I thought you talked about that a month or so ago?
thats when it went off
or maybe you talked about planning to do it
you know how they are made,
with grooves at just the right angle to give 1 turn in say, 16 inches
not cut rifling?
no, cut is not really that common any more
this is button rifling
how do they deal with residual stress from the displacement of metal?
heat treating after?
well, they do .. or sometimes cryo treat
and the OD near the muzzle is bigger
so more metal
so tighter bore at the muzzle
by a few tenths
but get this ...
this one has progressive twist
I shot a ruger 22 pistol with a 6" bull barrel a couple weeks ago - nice
starts off straighter ...
ends up at 1:16
the bullet accelrates laterally in the first 4"
but the radial accel is spread over the lenght of the barrel
what I wanna know is how the hell they make that!
the change can't be too radical, or the land impressions in the bullet would be all fscked up
anyway .. be intersting to see how it tests on the rig next month
this is what I have been spending too much time on
[23:54:08] <skunkworks> http://www.electronicsam.com/images/TUB.JPG
I could be tinkering ;)
hope that window is well insulated
sitting naked and wet next to a drafty window in winter.....
not a problem ;)
that is snow on the roof outside itsn't it?
the best hottubs are outdoors anyway :)
no - but there was a few days ago
tiling is going to be fun :(
tiles on wood ...
for some definitions of fun
tons of cemet board that has to go up.
probably a ton
grout the tiles likely to get wet with silicone :0
that would be a lot - to the right is going to be a walk in shower
silicone grouting takes a while, but works well on for tiles on a wooden substrate