thx jepler unpythonic :)
tomp: sounds like an interesting project..
im going to get some sleep...good night @ all!
i use the L298 sets from solarbotics, and wondered where the analog velocity went.... there aint one, you just toggle the enable with pwm to generate different speeds ( duty cycle is proportional to VelcityCommand on old amp )
hmmm, i dont need the special 7i33 pwm to analog feature
and seems like solarbotics is outta buisness
no, google couldnt find em http://www.solarbotics.com/products/l298/
[00:10:58] <tomp> http://www.solarbotics.com/products/k_cmd/
19$ cheap, use for 2 teeny motors ( rc car size) or 1 tiny motor ( blender size ) by dbling the outputs
tomp: 2 axis>
i bet pwm at low speed is 'chuggy' ( on wait on wait on )
I mean - that board..
(3 dbld drivers )
board is 2 tiny or dbld for 1 larger
ok - what is dbld?
tomp: cradek ran his lathe with only 5 levels of pwm.
was it something I said?
the solarbotics l298 driver http://imagebin.org/13644
( dang cell phone camera! and no ubuntu usb driver to connect with, had to email to myself )
$16 for some screw terminals
accent on the 'some' ( those are 'some' terminals )
on the solarbotics thing?
they look pretty normal to me
the 16$ ones you mentioned
the way i see it, the solarbotics thing is a l298 with screw terminals attached
l298 - $3
screw terminals ($19-$3)
yeh, i dont wire as well as the pcb does. i assemble the pcb a lot faster
Tomp - what did you change?
the pic shows the outputs paralleled, the dirs and enables are also paralleled ( pic doesnt show ), i used bigger diodes and 7905
how is feedrate handled in a G2/G3 interpolated move?
same as linear - just feed along the path
its porportional to each axis so velocity is constant
GNieport: a better question is 'how is accel handled in a g2 move?'
did I answer the right question?
fenn, not when the post wizard asks me "true radial feedrate calculation?"
hm, what does that mean?
uh what does that mean?
fenn: that's a good question
Maybe its asking if feed is based on spindle centerline or cutter comp location.
so Tomp - in paralleling the H-bridges did you use 8 diodes or 4 larger ones?
[02:04:36] <GNieport> http://pastebin.ca/889665
larger, looks like 1n4002's (orig were signal diode size )
never mind - I'm sick and I didn't look carefully at you pic
jepler has a pcb and schema on unpythonic, i could just buy these ready made
oh and 8 pcs, and all larger
GNieport: emc only calculates the tangential velocity if you are using tool diameter comp
otherwise it will just move at the requested feed rate
I built one of these with the 7805 and 50V? diodes... but its a bad idea to run it at higher voltages that 7805 GETS HOT!
yeh it gotta drop the excess
it really needs a 2 step on the 5v source
I used mother board chipset heatsinks on the L298n
dropping 36V across a 7805 wastes 86% of the input power..
26V is max reccommended. still very hi
for that board? i thought l298 was ~36V
the chip vs the mfct r of board
but remember the source
oh wow l298 is 46V even better
these guys are making items for the toy robot market - units likely powered by 2 motorcycle batts in series so there would only be ~26V on a full charge.
GNieport: update from the back room: emc actually doesnt change the feed, it only works properly in "inverse time mode"
fenn: "works properly" is some serious editorializing :-)
well, depending on your definition of 'works properly'
i suppose i could try it and see
fenn: that post setting makes me think controls tend to work like emc
i plan on running my 1/2 ton pick up on HHO this summer...
re: the solarbotics, thier pdf manual shows they tested at 9,12&18.95V, and have specific recommended device changes to use 26V,30V,35V and 46V
dmess: i'd like to see it, been looking at that stuff
i'm planning on running my dirigible on the stratospheric temperature differential this summer
i have a propane machine that should be easy to retro .. so im in.
fenn: layer inversion will let you change altitude at least 2x a year
dmess: it was a link you posted earlier, they called it 'gravity powered' for some odd reason
ya.. i read. it and forgot it.. seems out there tooooo far.. for we mere mortals.
seriously though, flying windmills are cool
yes.. and would/could work
ive been an alternative energy bug for 30 yrs..
lol how do you power something with gravity
thinkinin about over-unity for 25 or so...
toastydeath: you set up a series of space tethers such that you can lift bags of dirt off the moon and deposit them on earth
i could... you have funding..
60MJ/kg of moon dust
fenn: that would work
or you can stop halfway and turn them into space stations in low earth orbit
i drop from yhe suface of the earth.. cheaper...
I need to build up a small propane sized waste motoroil furnace to cast up some replacement end caps for my servo motors.
just need to find a bottomless hole
3 km is deep enuff... and a return system... secret for now...;)
SkullWorks_PGAB: if you arent doing much casting, you can just use charcoal or even hardwood
I been sick lately - this chest cold allows me no rest so I don't get anything done.
or get some za12 zinc alloy..
i think that melts on a stove top
yeah I'm going to try a disposable Charcoal unit - but I'm afraid of sparks - everything is a tinder box around here
HHO gas apperantly is hot enuf to cast stuff
well duh, "HHO" runs the frickin space shuttle
but its easy enuf to make.... at home by morons like you and i
sure, in small quantities
it doesnt take alot to cook shit..
well, actually a hydrogen flame is hotter, but there's less energy in the reaction so you do need a lot more than an equivalent volume of (propane, natural gas, butane)
33% oxygen, 66% hyrogen
you mean 4H + O2 ?
4 H2 + 2 O2?
as a gas
and its safer than H
Chem, ouch, last studied circa 1979.
with the right generator some vehichle are exclusive to HHO
more physics tough..
dmess: what's going on is the hydrogen flame burns faster, pre-vaporizing the gasoline droplets
chem pcs have been easier to remember...
but i dont plan on running any gasoline/propane in it at ALL
well, that wont work
what to they use as a catylist a super hot magnesium wire screen?
why wont it run on hydrogen... and oxygen.. if tuned correctly??
dmess: where you gonna get all that hydrogen?
from the hho cel onboarb..
and the electricity to run the cell, where does that come from?
conservation of energy - amount of energy to separate
from the front of the engine.. there's an alternator there..
h2o into gases is same as there re-combining
dmess: here's a little experiment in perpetual motion: take a dump in a mason jar, then mix it up with some water and drink it. repeat until you get a satisfactory experimental result
dmess: it takes at least as much energy into the cell as the hydrogen and oxygen out of the call can produce
only FREE engery input would be solar - but then the time factor makes it impractical
well, the coal's just sittin there
there is no free energy
doesnt that mean it's "FREE"
solar comes from the sun, coal comes from long dead trees, oil comes from long dead animals, etc
ihave a dream that YOU are ALL wrong..... sorry.. this is the 2nd way i show a plus in enegy and the only comeback from peeps is dont youknow this.... yes.. i know.. and dont agree..
I call it free cause its allready being wasted
hho energy comes from the minds of true believers
dmess: take an electric motor, connect one terminal to the other
since the motor acts as a generator, it should run itself right?
no.. but it will run another one..
i have nothing to lose JMK..
just your time and energy, and the respect of those who understand conservation of energy
Fenn if you spin it...
dmess: try it and see what happens
or do the equivalent process with an added complication of electrolysis/combustion
been there fenn... peddled a bike...
the HHO thing only works (if it does work) because it catalyzes the gasoline combustion
not sure catalyze is the right word
well iff i still need some fuel i dont care.. but we have to get BETTER
well, using a smaller motor is a start
usually they compromise on efficiency for power by making the stroke too short
in order to fit the engine under a car hood
better way is to have a diesel generator in the back and electric motors to handle peak power
the problem with car engines (that outweighs everything else) is the range of power needed
i was thinking the h have a boost to the combustion and the O vapourized and did the same as it did in the big old make &break engines... HP
cruising even at highway speeds is only about 10-20HP
but for accel and such, the engine needs to put out much more
so its inefficient at cruise
dmess: a standard car alternator is rated at about 500W, and your car uses 15-30kW at highway speeds
thats why hybrids do so well - the engine can be sized for average instead of peak load
so there's no way the alternator can generate enough H2 to run an engine
er, sorry, the engine GENERATES 15kW, it uses about 100kW of input energy
so you'd have to run 200 car alternators to run an engine
ballpark figures.. a car engine
my 4.3l GM on propane wont make anywhere near that.. and runs like a watch
saw a 1.5 liter engine without a cam and valves that had a 70:1 compression ratio - would run on almost anything combustable that could be pumped thru the injector - including wheel bearing grease and wood pulp - made like 400HP - but it tore it self apart - crank and rod bearings just were not up to the kinds of loads it put out
heh wood pulp
send me dwgs... lets build her..
SkullWorks_PGAB: sounds like some flavor of diesel
yeah - for the test they used sawdust mixed with white wine
sweet... eats anything..
trailer of wood has the same energy content as a tank of gas..
it is very much like a diesel - but no normal valves
Infact they used a VW Diesel block for the testbed
2 or 4 stroke...??
I built the cylinder head for it - machined from solid 7075
hemi head.. no holes??
operates as a 4 stroke with something called 180 degree timing
combustion chamber was 8mm x 22mm by 2mm deep
the way i see it, you want to make the engine out of ceramic and get combustion chamber temps as high as possible
every bit of heat lost in cooling the engine is waste
i see.. keep is as low as possible and burn ALL the gas
valveless diesels have been around for a long time: http://en.wikipedia.org/wiki/Opposed_piston_engine
WW2 sub engines for example
but they have emissions issues
Took me a week to understand how it worked - they had to show me a "normal" buick indy block with the 180 degree heads that had only 6:1 compression to understand the intake/exhaust method
or graphite maybe instead of ceramic, backed up with steel
dunno how the expansion issues work out
SkullWorks_PGAB: how does it work?
the opposed piston engines have the two crankshafts out of phase, so the exhaust ports open and close before the intake
think of a cylinder going accrossed the top of the head
still spinin in me head Skull.. but ill figure it out too
dunno how you'd do that in a conventional engine
drill a hole down thru the cylinder
a cylinder at 90 degrees to the normal cylinder?
boxter style.. twin opposed...
another idea is to filter out the nitrogen in air with zeolite, so you have a pure oxygen/gasoline mixture
when the cylinder is rotated 180 degree the hole will line up again
though gasoline would be too volatile with pure oxygen
SkullWorks_PGAB: you are describing a rotary valve I think
so there _are_ moving parts other than crankshaft, conrods, and pistons?
when you said valveless I thought you meant truly valveless, with no moving parts except pistons, conrods and crank
whats the devil of the whole thing was the inventor got the math/timing methods patented
not the actual implementation
so if we figure it out on our own.. we cn use them...
as long as its different
and you dont trip on any of the other billion patents out there
it will be.. at -40 c
well the single cylinder handles both intake and exhaust - the super high compression burns everything utterly
i dont think its possible to make anything without accidentally violating some patent you've never heard of
ok so catch me... ;)
part of the critcal detail was getting the head and piston tops coated with the same ceramic used on the exposed joints on the space shuttle
7075 isnt very heat resistant
oh yeah - no cooling to the head (water or oil )
air cooled only??
increasing maximum temperature improves the carnot efficiency
this coating was great
sulfamate nickel plates on 7075
i forget the mil spec
you could hold the piston in your hand and apply an oxy-acetylene torch to it for 5 minutes - the face was glowing but the heat didn't reach your hand
how thick was the piston??
VW OEM diesel
didn't cut one up
VW bus or car..
like a 79 rabbit 1.5 liter block
they were sweet machines..
was a bear to start - if you didn't get it going in the first 40 sec your batt was dead
I think he went way too high on the compression ratio
but he said it had to be pretty high to avoid NoX
my friends had a $3.00 a day ether habit...LOL
costs more to smoke nowdays
that doesnt make sense - NOx is formed at high temps and pressures
heh just dont get hooked on both :)
fenn - supposedly there is a point above which it isn't an issue - I never saw any test results - I only built the head and other support equip for the project
I think it could have been refined - but the cost was impractical
well it sounds interesting
i will make HHO work in my truck... im funny that way..
a complete rollerbearing crank and rods
SkullWorks_PGAB: google sez that if you let the gas cool slowly the NOx will dissipate
roller bearing crank is something we should have had 50 years ago
a bearing is what - $20?
the tech is there - no one want to pay the price upfront
why would you want to replace a hydrostatic bearing with a roller bearing?
it's more efficient
wnat about wear?
what about it?
what's going to prevent wear in roller bearings?
um, the rolling action
hydrostatic bearings last longer than roller bearings do.
not under these shock loads
it's a hydrodynamic bearing anyway
hydrostatic - the oil is injected under pressure.
not at the con-rod bearing
oh, connecting rods
i was talking about the crank
are you sure it's hydrostatic? i thought injection was just for better lubrication
oriface hydrostatic bearing.
either way - it pounded the bearings to foil in less than 600 miles
an experimental 70:1 compression engine
if you don't size the bearing properly, of course it will fail
there just isn't enough load bearing surface in a 1.5 liter 4 cyl block to handle 450 HP @ 1800 rpm
there is if you make it from scratch
that is true - but the guy didn't have that kind of money to play with
prototype was too big then
hire a model engineer :P
well engine #2 was a 205 CID Buick V8 - it drove normally
then they went smaller to the VW diesel block for #3
what was the first one? a lawnmower engine?
seals kept failing on it
i think i'd rather experiment with something i didnt need a crane to move
seal material in #3 was like a glass phenolic that was 65% Tungsten and graphite from EDM electrodes
it worked for the life of the block - no idea on long term
I think they should have made custom pistons and had the block sleeved down to under 600CC
as it was the flywheel was almost 3" thick solid
maybe that's why the shock load was so high
force didnt have anywhere to go but down
this was a slow motor - supposed to redline about 2800
more like a steam engine
but when each cyl fired it was violent
I think today with better electronic timed injection that could have been managed
long slow boat cylinders...
slow down the injection stream - less fuel initially with a longer duration
but he only had some crazy uber high pressure mechainical unit at the time
i think it would be neat to generate electricity directly from the piston, no crankshaft (dual opposed pistons)
i have a pic of the largest diesel in the world... guys are standing i the case installing the bearings with a crane
"inertial confinement combustion"
piston made of ferrite in a silicon electroplated AL bore?
dmess: it's also the most efficient internal combustion engine
skull we should talk...
SkullWorks_PGAB: i was thinking more just a magnet on the shaft
fenn.. we ned to chat..
we are chatting :)
and very off topic
hey, you could use HAL to do engine timing :P
im on moving magnets... plus some water...
the l298's been running a few hrs now, 24V= supply, the 7805 never warmed up, but i put a fan on the l298, cold now, still running 100% duty cycle ( no ammeter handy )
hal... is the plc for all im desighning
what is it driving?
actually HAL/PC would be a little slow for engine timing
thru coils... generating power...
not much of a flywheel ( disk 1/4 thx, 1.5" dia )
or you could do some kind of switched reluctance with lots of little fins to get more force
electricity from gravity,,,
the hoover dam was only 700 million dollars
dmess mount a windmill on your vehicle
i can do it cheaper... and DIFFERENT,,,, better..
i have 1
(48 million at the time)
Fenn - those were what 1930's dollars on the gold standard?
too many short sighted people these days...
too many no-sighted people
yes them too
all i want to do is save the world... and i cant get a hand...
what do you need?
other than from family...
a mason jar?
if we had another 29 - like depression - no one would work for a pitance on public works projects like they did back then - atleast not until there was no other way to get food.
have them.. the HHO is just wating for more temprate wheather,,,
the public works projects were mostly just busywork.. poor way to build an economy back up
i hope we have a nice good depression so something gets done around here for once
true but we reaped the benefits for decades afterwards
i'm just saying busywork doesn't create wealth
the gravity generator... was put on hold when GOLD went thru the roof... they pumped it and are mining SOME of my cache.
well now you get to fill it back up again
problem with electricity generating schemes is that electricity is so damned cheap
they paid 1 cent per litre... to pump it.... who would have bout in??? right no??
I'll bet the old timers are down in the holes again up in Victor & Cripple Creek
i ran some calcs on brushless motor battery driven bike. it looks charging batteries and accounting for 1 cycle of battery's life (2400cycles) is near 31 cents for near 35 miles. My Honda gets 30 mpg and gas is near 3$, so 30 miles costs 3$. thats 31cents versus 3 dollars! sound reasonable?
what's the range on a full charge?
they say 40, i say 30
max sustained speed?
13 to 18 mph
bikes usually get at least 60mpg on gasoline
they pumped with 4 - 48 inch pumps on discharge.. for 4 yrs.....
aw 18mph that's pathetic
supercapacitors + small turbo diesel is the way to go
i meant the cost, is that proportion seem reasonable ( i can get more watts and more speed )
well, my office chair costs nothing and goes 0 mph
the guy who grabbed it is building a 13 Million dollar hole in Hailibury....
the honda is not free either
imo, solar thermal. fresnel lense traking the sun
ew no, use a parabolic trough mirror heating a collector tube
twice2... you have MY tracking system>>
i want one in my back yard
fenn... you b@#$%
lenses are inefficient, heavy, and expensive
anyone doing solar or telescope work with emc?
parabolic trough made from strips of mirror??
no, there's dedicated software that is better
EMC would do it SWEET
solar tracking is so simple though, you can do it with analog electronics
two led's and two transistors
thats how i did it in 1986..
hmm something like that at least
not leds but photo trancieversers
the focal point of the mirror must stay on target while tracking the sun
dmess: did you generate steam?
it held the TUBE.. where water recirculated and exited.... was a preheater for to hot water tank.
this is what i'm talking about http://www.redrok.com/electron.htm#led3x
too much stuff on one page
on a small CLOSED scale i did generate steam but NOT in december/january. at 49 deg lat.
i wonder about using propane as the working fluid
These guys know the score: http://www.pge.com/news/news_releases/q4_2007/071105.html
whenever you see the phrase "create jobs" you know the project is based on political pork
177 mega watts from the sun
twice2... i started this progect 24 yrs ago...
...most of the time
one one square mile of ground
they stoll it from me in 1998
i had 1 set of MY hand drawn drg's go missing 1 time..... 1 yr later its all over pop science
i was in high school at the time
twice2.. thanks for twistin the knife...
i wasnt bleedin' enuf as it was....
dmess: quit sobbing abut it, you'll get your oats soggy
been eatin them that way for yrs now.... my mom still has my model in the garage...
and a solar powred watercraft i built..... and some early snow boards.... and..
ok have fun
thats why i got into cnc machine tools.... make em do what we need to.......
hmm, ive got an exciting melody down, gonna have to get masternode back up finally I guess
any one awake to test something for me?
is workspaces in emc broken?
like, i home axis, right, so my g53 is fixed
i set a g55 X offset
g92 x0 with spindle over g55 origin
so i move over about 6"
switch to g56
switch to g55 again
it reals -.9" or something for X
so its broken?
workspaces are supposed to ref off g53, why would changing my g56 origin affect my g55 origin?
fuck does the output not reflect current workspace or something?
you don't understand how g92 works - it moves all the coordinate systems
[05:28:55] <cradek> http://linuxcnc.org/docs/html/gcode_main.html#sub:Coordinate-Systems
you set G5x offsets with G10
or much better, use touch-off in AXIS
You can set the offsets of the nine program coordinate systems using G10 L2 Pn (n is the number of the coordinate system) with values for the axes in terms of the absolute coordinate system. See Section [.].
really, use touch-off
nm ill look it up
lol the hard way
no i wont touch those button until
until you put a warning on home button
load up systems.ngc, it's a nice demo to play with
lol what? why do you need a warning on home
becase its right next to touch off
touch off is also "End" on the keyboard
you hit home instead of touch off with soft limits and your setup is fucked
the overide limits apperently doesnt mean the soft limits
yeah end is right next to home =)
override lets you move off limit switches
so what do you do if you accidentally home in the middle of your workspace?
you mean then soft limits don't let you jog back to the right home location?
there's no "unhome" so you just have to restart emc
and then order some home switches :-)
so a 'are you sure you really want to home the axis? [yes] [no]' dialog would be nice
other controls handle home like a different run mode
so that there's no mistake
hell it would have saved me 20min of setup, on multiple occassions
all coordinate systems are saved across multiple runs of emc
cradek: are you huh'ing to me, or the renesis
he's saying the home knocks the setup off the table
i dont have switches so homing isnt exact
so when i lose home, all coord is trash
oh, yuck, so you set up a lot of points that are lost next time?
if you want to keep those setups you should come up with a reproducible home location
needs some switches.
even a wiggler and some feature on the table
but yeah, switches are better
and turn the machine off
sure, if you can assume it doesn't move
anyway, i dont mind the soft limits
id never have problems if there was a nag dialog on home
and you can't assume that with modern 8 microstep drives
im fine with close enough, step off wont kill me
could be 4
its setup to 8
renesis: on some of my axes I have a rough axis position I can see, along with a mark (like handwheel) on the motor shaft
hehe, its alot quiter with microstepping, no more tron sound effects =(
this is fairly reproducible home by eye then
I should have said 4 either way, the drivers init to step 0 at powerup, and you could have been at step 4 when powered down.
not an issue
homing with no nag directly adjacent to touch off on the kbd and gui using soft limits is an issue
I tried it, and gradually lost the position, but TBT, probably less than my ability to pioperly locate the next part on the table without a pallet holder of some kind.
like, damn at least give me quick-keys to set workspace offsets no dialog if youre gonna give me the option of screwing myself no dialog
renesis: there is a key for touch-off-to-zero
control-end maybe? I'd have to look (check the key ref in the Help menu)
i am kind of intrested in why "home" is a button
for each workspace?
or to popup the dialog or to touch off current
yeah i dont see the use of home button
you use it once and then its just in the way
in my case its like the PROJECT DEATH HAHA button
well i use it all the time, but i don't want it to be a button
I don't follow - if not a button what would it be?
in the menus
Yeah, but you ALWAYS use it once. Per axis, at powerup
all the machine controls i've used, the control has a few run modes
yuck, menus are for things you almost never use :-)
edit, auto/mem, mdi, jog, rapid, mpg, home
like on Haas there's a row of softkeys that put the machine into diffent modes
on fanuc it's a knob
I hate too many modes
heheh <3 haas controllers
but the advantage is that you can have different contexts for the commands
okay well suggestion:
on my bp I jog, then I try to turn the wheel, nothing happens. I have to stab at the crappy membrane keyboard until it says KNOB on the crappy little display
and also you can't move an axis unless you want to
nag on the HOME button, because youre fucking soft limit users
and THEN, finally, the knob works
cradek: that's a problem with your hardware, not the UI
well, I disagree
if my keyboard breaks and the "a" key no longer works, i can't complain about how "a" is useless
no I have to look at the screen and take my eyes away from what I want to actually do
the membrane key, which can't be found by touch, has to be pressed several times, while watching the display
when the display is just right, you can go back to what you're doing
it's terrible UI, not a hardware problem
that's horrible logic
you mean to tell me you hit keys on your control without looking?
i dunno the hass did sim and tool/workspace management and jog and etc
I can sure as hell turn the jog wheel without looking at it
a jog wheel is good UI
it would be fucked up to have all on one screen
and a keyboard i'd see how well in bounces off the bottom of the traash barrel
cradek: i think you're not understanding what i am saying
that makes two of us - let's just stop
maybe an option just to remove home button al together, if no nag
there are as many ways as there are people
well, no, there's two ways - what the machine controls do across the board, and what emc does
Chris, can you put a glove on a std keyboard and use that?
gene_h: it's not using emc yet, it's still the way it came
I know I've got to get some of those for mine to keep the swarf out.
totally customn keypad then? Bummer...
yeah I'll replace the control one of these days
emc's jogging is so much better
a key for each direction, and a wheel that works whenever you aren't using the keys
The comment about the shoemakerrs kids comes to mind long about here... *-)
and configurable jog increments!
the bp has one incremental jog: .0001"
I shit you not
gene_h: no kidding :-)
yup, I have three search wheels from a tape editor I'm going to see if I can make work when it warms up out there.
gene_h: that sounds like another good UI (another thing where you don't have to look at it to run it)
And if possible, I'd like to make them active while running so they can put an offset into some var in the program, like the diameter of a hole being circle milled.
you could have a wheel control an analog value in hal (displayed on a vcp) that you later read with whatever that gcode is
The are dual dial units, center dial is a finger socket regular jog, out dial is a zero center varriable speed either way driver when its running a vcr. 8 speeds IIRC.
gene_h: what kind of signal do they produce?
I believe a byte wide hex, they are mechanical so I don't think they are any kind of a grey code.
oh just some switches? sounds easy (if you have enough inputs on your system)
surely the finger wheel part is quadrature?
That's why I bought another dual parport card :)
Maybe, I'd have to look at the schematics again now, its been about 3 months & at my age I have CRS.
Question though, is this something I should opto-isolate from shop floor static?
in your place, I'd figure that since optos would cost you much more than a replacement parport card, and add a lot of complexity, I'd try it without first
That was my thinking too, the dual parport cards are about a $20 bill.
tap 5v from a usb port, find some pullup resistors, wire up the switches, call it done
Isn't there a pullup on the parport? There used to be year ago.
chris what control is on your BP now?
SkullWorks_PGAB: boss 8
gene_h: not sure, maybe they even differ, there are so many parports
SkullWorks_PGAB: UI = 2x20 lcd screen, membrane buttons, estop button, jog knob
I'd have to look that up, these cards come with zip docs of course.
I used to run a Series II NMTB 30 with a Bandit control
SkullWorks_PGAB: anyone who says commercial controls are all better than EMC has not used this control :-)
I here that
SkullWorks_PGAB: that's the slightly older stepper models right?
not sure - know the rapid was 100ipm
yeah the steppers were that
mine does 250 rapids which is very nice
no cutter radius comp
(optional at the time for a huge price)
the jog knob is fixed .002/click which is too coarse to touch off, but way too fine to get anywhere in a reasonable time
that combined with the fixed .0001 incremental jog is all you get
X/Y DRO with a vernier scale in the Quill ( it did not move the knee)
Yes, acc to my amiga 2000 manual, there is 5 volts with a 47 ohm limiter available on pin 14 of the parport. And its labeled 'centronics'
gene_h: that's news to me!
You might measure some of the ones we're using now.
Yeah, about sleepy time here in WV too, goodnight.
renesis: you could just bind the home key to something else
- root_window.bind("<Home>", commands.home_axis)
+ root_window.bind("<Home>", None)
so touch off button is broken?
an control-home too
okay, i go to a position, machine is in g54
i touch off, tell it to set G55
it possibly zeros G54
i switch to G55, and its like at 7" or something
what does this have to do with homing?
nothing im over that
now im using touch off like cradek said
and this is what its doing
which is what i remember it doing last time
so i think something is messed up, i dunno if its touch off, coordinate system, axis coord display, or what
touch off sets g54, which wouldn't affect the g55 location
then why does it have a dropdown
perhaps touch off should change the g92 coordinates instead? i dont know
with values labeled G54 thru G59.3
and also Pn
so dropdown is broken?
it appears to do something to g55 when i select g55
so i have to set workspace offsets in absolute values from home origin with G10 L2 Pn if i wanna use anything other than g54?
well its not zeroing it
because i G55 in mdi, the display switches, but its not zero
work offsets never made any sense to me honestly
in general or in emc?
i've only used emc
because in general i can prob come up with 5 or 10 reasons
i dont have a mill to test touch off with either
like, i dunno but this is basic CNC controller stuff, even the fucked up 80s digital UI bridgeport controller at school does offsets better than emc
touch-off only changes the system you tell it to change
well then its broken
no, I'm pretty sure it's not
because it didnt zero the system i told it to
im in G54
i use touch off to set G55 to 0
i switch to G55
its not zero
what am i doing wrong
you probably also have a g92 offset
[06:25:32] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems
yes ty i read
btw whats L word in G10
clear g92 with MDI G92.1
it has to be 2, I think it was for compatibility with ... something
the P is what specifies which system
ok it was the g92 thing
G92 moves all the systems by the same amount
it's neat if you have several parts at the different systems - with one G92 you can shift all of their origins
(but usually I don't use G92)
have you seen the systems.ngc program? it's nice for experimenting
why would one input for limit switches handicap your machine? sez so in the wiki
load it and move the systems around with touch off, and you can immediately see what happens
truth is, I don't care which limit I'm on
but when g92 is active and you do a touch off, you dont get what you asked for
eric_U: if you have separate inputs, emc can keep you from moving the wrong way if you're one a limit
renesis: one of the distributed gcode programs, or at least I think it's there
k i check
i want to make pcb and play gtr2 and sleep
that is neat
[06:32:12] <renesis> http://www.darkertechnologies.com/image/audiosam-07.png
you want to make love with my layout, yes?
why is usb connector in the middle of the board?
its a dev kit
the connector is actually on the module that plugs into the board
[06:34:03] <renesis> http://www.olimex.com/dev/sam7-h256.html
G10 L2 Px is for defining a work offset. L1 would be for changing a tool table value (not supported by EMC2 IIRC)
i might use it for power from usb, but im prob gonna wire it up to the stk500 for power
SkullWorks_PGAB: I think some controls have even more coordinate systems at L20
so its directly editing the offset table
the haas had a ton
maybe 36 or 99, get confused, had alot of tooling offset values too
well there is also a G54.1 Pxx on some controls - allows 99 additional work offsets
used mainly on machines with pallet changers
was maybe it
i have the manual in my closet somewhere
heh, prob easier to just find online tho
but Haas has a bunch too - and different ways to access them depending on control build date
yay for software
one of ours uses G140
it had pages of tools and offsets
one of them had 99 values
maybe more =\
also erotic: http://www.darkertechnologies.com/image/audiosam-06.png
too bad eeepc doesnt have a parport
would make a neat controller
it probably has terrible realtime performance
they underclock to 650MHz tho
i think for power saving
emc is allergic to intel chipsets and procs or something?
realtime latency can easily be screwed up by power management features that 'borrow' the cpu
turn them off?
if you can
you can disable in bios or kernel
[06:44:18] <fenn> http://www.rtai.dk/cgi-bin/gratiswiki.pl?Latency_Killer
shared mem integrated video is a real killer
CPUFreq/PowerMgmt - ACPI
you can turn acpi off in the bios or kernel
sure but sometimes there are other things that you cant turn off
SkullWorks_PGAB: i have some boards with integrated video. if i dont run an X server they are fine
if i do, latencies of ~1ms
so i will probably just run them remotely
yeah thats about all you can do
so if i go g92 x2.
itll move all origins up 2 inches?
or no maybe it will move the origin down to inches because it specifies current location
this could make paneling pcb easier
i believe 2. moves it up two inches
you should check just to be sure.
yeah ima test
is the time estimate in properties usually accurate?
* renesis writes down current time
it should be
the control knows how long each line will take to run
on mills things always seem to take longer than they should
it ignores accelerations and feed override
and of course wait for input lines
that shouldn't be an issue for a feed of like 10 ipm or less
depends on how many changes of trajectory
BACKLASH=.003 seems to be helping
* fenn reads a b&s screw mchine manual
wonder how long wobble broaching has been around
i dont see any mention of it in this operators manual (yet)
the floating holder looks just like what they use to hold wobble broaches though
cept it moves around and also is not on an angle
no it only moves around when you loosen the clamp screws
which kind of floating holder are you talking about
looks kinda like a stubby pitchfork
the two side tines are clamp screws
yeah, and the center is toolholder
i thought you meant the floating holders that actually stay floating
this says they're used for reamers
yeah, but you line those up and lock them
there's another style that remains floating when it's driving the reamer
is the book cool?
its your typical late 1930's technical book
paid by the word eh
i like how they write in plain english
oh, not what i'm thinking then
no, old stuff usually is much more straightforward
lots of pictures and drawings
they dont quite look like photos, you know, like photos on lathes.co.uk
or whatever they're called
how do they do that? airbrush or darkroom techniques?
if i had any artistic talent i'd be smoking pot in an art class somewhere talking about postmodernism
nah if you're any good you dont need to go to art school
well yeah but all the art kids i know went to art school to smoke pot and talk about postmodernism
art school is if you are highly motivated but no good
not necessarily learn anything
or an excuse to smoke pot and get loans/parental approval
are there many cnc lathes with 'balance turning'
2 parallel axes going in opposite directions, to reduce deflection
any 4 axis lathe will do it
and some single axis machines for high production use balance turning tools
on a single turret
er 2 axis
i wouldn't exactly call it "commonplace" though.
shouldn't you be in bed?
you can stick a balance or quad cutting toolholder in any cnc machine if you want to do it
* fenn shrugs
i've wasted the last week because of my messed up circadian rhythm
i was actually drooling over Okuma twin turret machines on ebay earlier
haha, i don't have a real healthy sleep schedule
ty irc for making us some kind of wingless bats
could you lap an index table with a dished face on one of the gears?
and then use that as a worm gear (with some kind of cycloidal profile)
i don't see what you want?
a really accurate worm gear
so the worm gear is 'automatically generated'
i'm not seeing it, i suppose
one lapped surface is like a bevel gear
the other is the negative surface
then the worm drives the bevel gear in the finished product
i'm not sure that's an inherently self-generating process but i guess it would work
at least to an accuracy anyone reasonable cared about
or you could drive it with a bevel gear i guess
when you lap worms/gears together you still have to map the gear out
and change the pressure based on the error of individual teeth
but you can do that, it just doesn't work the same as a true autogeneration
well, that's not what i'm interested in really
what's the end goal, just an accurate rotab?
yes, something with a continuous movement
that would do it to a pretty good level
better than just lapping a worm to a worm wheel?
if you can lap two components together, to be used together
that's always going to be the best option.
how do you lap two components together without getting abrasive grit stuck in everything?
lappable stuff is usually hardened
so the abrasive does
and then you go back and lap it with pure vehicle with no abrasive, to try and wear the last of the abrasive out of the part
if there is any in the first place.
that's the theory anyway, i haven't done it.
folks like moore made laps in sets, and lapped them to each other
then used them on the hard parts
then finish lapped the "rough" lapped parts.
with each other.
the 1440 index, for example.
so maybe it just doesnt matter if there's a little abrasive left
it will matter if the part is soft, not if it's hard
but i DO have to go to bed now =(
so goodnight sir
enjoy your five winks
04:02 <+renesis> http://img114.imageshack.us/my.php?image=index1kf0.jpg
04:02 <+renesis> http://img512.imageshack.us/my.php?image=index5ai0.jpg
04:02 <+renesis> http://img253.imageshack.us/my.php?image=index6nm8.jpg
long sleeves + lathe = dead
nsfw or children
unless they are like, children who machine
or your work is a shop
* skunkworks is just glad he has all of his fingers..
I know when not to look...
your better than me.. I have to click on links.
I couldn't get #2 or #3 to load
a warning would have been nice
Warning... bah. Consider it your annual safety lesson
heh that was the skinned finger from a week ago
archivist: Ok, consider it your no-interactive annual safety lesson.
We all start off doing things safely at first, then "get comfortable" over the years, I think it's a good thing to have a "reality check" periodically.
I periodically shoot things across the room - that's about it
so far, they've always missed me
a few broken nails and minor cuts keeps you aware
cradek: Have you seen those kevlar shop bibs?
I have done the most damage to myself with manual tools
beard wrapped up in dremel is entertaining
archivist: Do I even want to know how you managed that?
Jymmmm: no, I have a denim apron, but I only wear it to keep oil off my clothes - it's probably no good for safety (might even make it worse)
I guess I did manage to take the tip of my finger off (just a bit of skin - very lucky) in a jointer planer. I was probably 14.
often safety stuff introduces issues
cradek: Yeah, there suppose to prevent flying metal from penetrating vital organs.
but adds loose clothing
I need to get some real safety glasses made - that's the biggest problem I have
I have goggles but they're uncomfortable so I only wear them when grinding etc
I think out of all the tools I have used.. table saws are still the scariest.
I bought a full face shield, the safety googles interfiers with my glasses
Jymmmm: I have a shield too, but I'm even less likely to wear it
I did find goggles that fit well over my (relatively small) glasses
cradek: They kept bugging me right behind the ears.
glasses are a pain
I wears ski goggles over my glasses.. They are comfortable..
skunkworks: the holes are too small so my glasses fog up =)
This works and is really oversized too http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=46526
I would think I would fog that up..
I guess I could just wear my motorcycle helmet
hmm - might have to get one of those.. :)
I want a couple of these http://www.thinkgeek.com/gadgets/tools/6d98/
ooh, blue laser pointer, only ... $900
[17:25:56] <cradek> http://www.thinkgeek.com/gadgets/lights/896a/
I thought green laser pointers were illegal
only if you point them to aircraft
[17:37:32] <Jymmmm> http://www.sti.com/oshaguide/index.htm
jepler:has anyone had a problem with tool offsets after the latest update?mine dont seem to be working anymore
what version? 2.2.3?
exactly what isn't working?
alex_joni, 15 minutes to go on that robot arm and getting to the amount of money I have loose in my pocket :((
* Jymmmm grabs the sledge hammer and breaks open archivist's piggy bank!
I have no bank
not any more ;)
* archivist is a poor clockmaker
archivist Well... make more than a clock a year
hmm we need to kick the boss here to make more money
its pathetic here
archivist: 15 mins have passed.... what happened?
it went for £155
to someone other than you I guess?
only 5 miles from me
archivist: guess it's too late?
* alex_joni just came online
ah ok yup beyond pocket money
too bad :)
I need a better paying job for toys I want
that's only $300
or 4 tanks of gas in my truck
all be damned, im learning cpp accually, and i can read emc source much better now :)
i need to buy a tv.
im debating on a big lcd tv or a realy nice projector.
dont get mad at me once I put my finger into emc
whoot, sorta exciting accually
I'm getting my list of requests ready for you
conversational is going in
if I can do it legaly
for what machine?
lathe and mill, basic functions
from what machine
like RAPID X0 Y0
LINE X2 Y2
no, like conversational move, would open a box up, you tell it where to move to, how to move, then it inserts a line of code into the program
something like g9001(x,y,p,g,f,,,,,,,,,,) and then that pulls the macro that knows what to do withthat
conversational is not a user readable gcode line readable format, however it is a graphical representation of the move
thought you meant like conversational like Anilam and others use on some controllers
I think that aspects of axis would make for a super fantastic conversational method too, cause it can draw the path on the fly, so as the user inputs data, the "picture" changes, versus having a fixed picture of what can be done
does that make sense?
jlmjvm: you didn't say what version you were having trouble in, but I reviewed the differences between 2.2.2 and 2.2.3 and there's only one tool-length related fix: on metric systems the tool offset was treated as wrong by a factor of 25.4 in the preview in axis (the actual program ran correctly).
Imean if your looking at a conversational move path for example, you get a picture of a line with an end point, an option to make it a rapid or a feed, and how to do the move, straight line, move x first, move y first, ..... but if the picture adapts to input that would be easier to see the representation of the move
Hey people, when I try to run something out of a .hal file with the command "loadusr -W pyvcp -c iotest iotest.xml" I get "Waiting for component 'pyvcp' to become ready...."
and, what does -c do? I was able to see any documentation using "halcmd show loadusr"
the -c is an argument to pyvcp; see man pyvcp for that
also see 'man halcmd' for a description of the flags for loadusr
hint: there's a relationship between the -c flag to pyvcp and the flag you need to give to loadusr
think that's why I'm getting the "waiting for component" comment?
since you gave the component the name 'iotest' (pyvcp -c iotest), you need to tell halcmd to wait for the component named iotest: 'loadusr -Wn iotest -c iotest iotest.xml' (whew)
jlmjvm: did you say what the problem is? if so I missed it
doesnt appear to be reading the offset
but arent I making a call to pyvcp? 'loadusr -Wn iotest pyvcp something er other'?
jlmjvm: you have to give me a lot more to go on if I'm to guess what you're doing wrong :-)
say what you do, say what you see, say how that differs from what you expect
"loadusr -Wn iotest pyvcp -c iotest iotest.xml" is working. as everyone else on the channel probably knew already.
love being the newbie. :-)
owhite: not me - I would have had to look at the manpages too
:-) good times.
when you enter a move in mdi that has a tool offset its not reading the offset value from the tool.tbl
jlmjvm: what is the gcode and the resulting move that you think is wrong
g0g43z0.h1,with a tool.tbl value of -1.00 should move the z axis to -1.00
the absolute position of the machine should be -1, but the axis will show 0, since you commanded 0
gracias people. its working.
when I issue that move, the cone jumps up an inch to represent the tool getting an inch shorter, then the cone moves back down to zero. if I had a machine hooked to it, the axis would move down an inch. This is correct behavior - is it different from what you see?
ahh,my machine isnt hooked up right now,maybe thats it,however in 2.2.2 it wouls show the tool move down 1.00 without a machine hooked up
I was using 2.2.2 when I just tried it
will just get my mill back online and see if its moving to position
anyone interested in SHCS Counterbore G-Code generator for EMC2?
sure.. ship it to firstname.lastname@example.org
[21:51:26] <BigJohnT> http://951753.pastebin.ca/890596
all ready an update... http://951753.pastebin.ca/890610
jepler: are there any pluto-step sample hal files?
in the unlikely event that anyone cares, I built emc2 trunk on my ancient 2.4.20 + rtlinux 3.2 machine and it runs fine
in the unlikely event that anyone wants to preview my g code software the latest one is http://951753.pastebin.ca/890654
BigJohnT: I think your shcs program looks neat and I'll probably want to use it when I convert my big mill
cradek: I don't even know what you said
thanks for making it public
I'll put it up on wiki once it's polished up a bit and some feed back comes in
BigJohnT: nothing important - a troll on the emc-users list accuses us of having dropped support for some older machines, but he's wrong, we didn't
the latest version on pastebin addressed the tab/enter key to move to the next field
have you tried it yet with emc?
it is pretty slick you can have a bunch of counterbores into emc in a few seconds
cradek: whee :)
are you going to put the updates on the wiki page too?
when I'm finished I will
but I'd like some feedback first
alex_joni: not a big surprise really - we didn't take it out
cradek: yeah, I know.. the only part I was worried about is forgotten things in the Makefile
like something got added for 2.6, but someone forgot to add it to the 2.4 part of the build process too
you know what looks good to me looks good to me but I don't know others until I get feedback
alex_joni: because of old versions of python/bwidget I just built enough to run xemc. if I cared enough I could update python etc and get a full build - but I just wanted to test the rtlinux/kernel stuff
BigJohnT: I'll try it
BigJohnT: I set it up as an input filter. it comes up, but it doesn't seem to work
I select #10
put .375 for tool diameter
x,y locations 1,1
do I have to fill out everything? I don't know some of them
like the depths?
no just the tool diameter and pick a screw size
did you press the generate button
then what do I do?
generate doesn't seem to do anything
(I bet .375 is too big for #10)
actually they're equal
pick a smaller tool
I should bitch at the user when they do that LOL
well when they're equal you can still do it - it should maybe just use a G1
I see what I was doing wrong: I didn't hit enter in the X/Y field so it didn't fill out that list
now I see how it works, but it wasn't very obvious to me
maybe it could also use a button to clear the gcode window
we should ask jepler to look at this - he's a great gui guy
Ok, I can add a button to clear the g code window
one thing that helps a screen be self-documenting is to have linear flow, either top to bottom, or left to right
actually it's in the menu but you don't have that one
you _almost_ have that
where did i miss the flow
you fill out things on the left, then move right to enter coordinates
then you have to go back to the left and hit generate
that puts stuff on the far right
then you go backward again and hit To AXIS (this button should be named "OK" I think)
if your not in AXIS is says "To Clipboard"
Quit should be Cancel
if you hit Quit/Cancel it should return an error code (exit 1) so AXIS knows not to try to load the gcode
BigJohnT: if running standalone, maybe it should let you save in a file instead?
It does that but you don't have that one yet
quit/cancel just quits only to axis sends the file to axis
"OK", "Cancel", and to a lesser extent "Apply" have magic meanings that all gui users know - using them helps someone see how to work the screen
my intent was to put this in a tabbed notebook some time in the future
should it say "Send to Axis"?
I hope this was the kind of input you wanted
no I'm pretty sure the button should be OK
btw you can double click on an entry in the xylist and edit it
or delete it
I would not have guessed that
LOL that took some time to figure out
can I just remove one?
just hit the delete key after you highlight it
I mean click on it
does the "Insert EOF" button make any sense
I'm not sure how you could make those editing features more "discoverable"
the help file I think
no, I had no idea what that is, so I didn't push it :-)
ok it adds an M2 to the end of the code
oh I see
man, saw the new ti nspire calculator at the store earlier, its freaken nice, but I already forked out the cash for a ti89
if you have several different sizes you only want the eof on the end
of the last one
for that, you should use a checkbutton widget, since it's a toggle
should it say "Insert M2"
it is a check button LOL
it's a "gui blooper" to use a button as a checkbutton
the checkbutton is the little square box you can turn on and off
since it has a verb on it, I expected it to do something (like Generate does) when I pushed it - I didn't expect it to be a toggle controlling later behavior
I think indicatoron=1 will fix it??
ok it's back to indicator on
I just removed indicatoron=0
ok I'll do that here
and the green thing
when I have a setting missing and I click Generate, it doesn't do anything - I have to guess what it wants me to fix - maybe this could be better somehow
Please enter the Tool Diameter. [OK]
yea, I need to add some error checking
I like the generated gcode
helical entry for stuff like this is very nice
BigJohnT: its a really nice start to a conversational method of making gcode too :)
doing some error checking code now...
this feedback is great!
glad to hear it
have to run, bbl
ok see you later
btw there are two algrothims use one if the tool diameter is greater then the hole radius and one where the tooldiameter is <= the hole radius
good night all
good night alex_joni
ok anyone want to check out the MarkII SHCS g code generator it is here > http://951753.pastebin.ca/890706
cradek: where can i find docs on vismach?
Roguish: there aren't any except what you find as comments in the existing machine descriptions
nothing on the formats for the different commands?