stustev: You around?
zephyr is now known as john_f
i just bought earbuds
Tomorrow I figure out why I am turning tapers...
toastydeath: Yes, but just where I am not sure - suspect tailstock, but will do some reasearch tomorrow - easy enough to compensate by turning it end for end and running another pass.
Time to fall over - all in all a good day even if it was manual work :-).
Night for real!
uh wont turning your piece make it taper the other edge too?=
well, make it bow (
anonimasu: that's correct.
also you aren't supposed to test taper with the tailstock.
as you probably know that taper is almost always caused by the tailstock
it's hard to set up right
(dial indicator and test cuts
dial indicator and a straight piece of something..
toastydeath: how's things?
also yeah using the right test proccedure for what you want to measure is rough
if you do cut first.. then you end up with the taper of your ways :p
or the misalignement of the headstock.
I adjust crossslide during travel to offsset bed wear and tapaer error (old worn out machines)
hmm.. yeah that works
though for most stuff I dont use the crossslide as it dosent have powerfeed
compound or cross
I think compound is what I mean
the little slide
compound also known as top slide
archivist you're just talking about issuing little x moves to cancel out the majority of the error, right?
yes (hand control)
I'm talking crooked ways /
(I'm kind of beat how to make thoose machines work :p
error correction table
manual machines do not have correction tables
scrape! file first if its bad
I'd buy a new lathe :P
(new used lathe)
if it was bad enough
it's been done
me southbend is very worn
my grinder shipping is fixed
^_^ the spindle bearings are in good condition so are the ways
hi all :_)
I am back for my nightly stupid question...
Not stupid... http://www.cnczone.com/forums/showpost.php?p=473026&postcount=354
Wow, I am impressed !
that was fast
any idea's ?
although - off the top of my head - I don't know.
I wonder if it is a soft limit kind of thing
if it where me - I would hook z up to one of the other axis that works.\
just to see if it was emc or hardware.
when I press the + or - the DRO does not change at all
but in stepconf I can jog away merrily
keyboard +/- or screen?
Z and Y are fine
X and Y
could you post your ini and hal files?
[12:22:05] <skunkworks> http://www.pastebin.ca/
also - what version of emc are you running?
not right now, router PC is powered off, I was closing up for the night, late here and I have to start work early...
erm, latest I think
I am not too worried, just spent five minutes jogging around to see if everything works
I am pretty sure it will be something simple, I will go through tommorrow and see what the travel is set for on Z
may be something like that, seems strange that stepconf can jog without prob's
ok - if there is a bug in step_config.. we would like to know about ti.
no prob, I will help if I can
let me check tommorrow just in case I have done something stupid
before we blame stepconf :)
doesn't help doing this late at night
most of the time got used up by wiring a data point into my garage :)
heh - nice to have internet on the machine. ;)
will u be around at this time tommorrow /
heh sure is
gotta resist the urge to install frozen bubble tho
router will never get finished !
n e way, thanks and I will get back to you tommorrow
you know what many people are liking, in modern office buildings with windows that don't open, on a hot day?
ding ding ding
is the heat on also? ;)
no, it's actually not
but I wouldn't have been surprised.
we have about 30 space heaters that take breakers out thruout the day.
I just love it when I notice that my credit card number has been stolen 2 days before leaving for Europe
I wonder if this is a problem:
not too many charges - maybe $300 total
and they didn't share the porn. can you believe it?
at least in europe it's mostly cash you need
heh - that is an "advantage"
although you can't rent a car without a CC
but other than that I can't think of anything you can't do
we're using rail, so that won't be an issue
I can't get miles, damnit!
actually, I noticed a while ago, but they haven't sent out a replacement card yet
you can put them on later
so I need to call again
see you later
oh man. do *not* leave Firefox 3 running on an eeepc overnight
hi to all
is there anyone ?
hi, my name is Alessandro from Italy
i have a question
[15:43:06] <cradek> http://vts.bc.ca/pics4/drillbent.jpg
i would put two button on axis who open two Python file,is it possible ?
sorry I would want to put two push-buttons on axis that they open two file of Python is it possible ?
I don't think cradek is running mach..
yes, two new buttons
it takes alot of skill to bend a drill :p
or bad luck
[17:09:53] <BigJohnT> http://www.youtube.com/watch?v=ZCXkEgMo1hE
somebody watched it :)
that was me
unelss there are two views now :)
the first view was me
btw, that is me in the video
oooohhhh - I'll have to look again ;)
thanks it tested 5 things in < 5 seconds
* BigJohnT goes to lunch
* BigJohnT is FD&H
Fat Dumb & Happy :)
anonimasu: did you get your grinder?
well, I paid for it today
the shipping guys will pick it up tomorrow
I talked a bit to the guy
the spindle sounded good and the ways looked to be in good condition
cool, also all the surface grinders I've use the nut for the spindle is left handed...
good to know
a pin spanner and a plastic hammer to remove the nut
one or two taps on the spanner and it should come off
* anonimasu nods
im happy about the ways
the spindle isnt as important.. that can always be fixed..
ways are more problematic
everything else can be fixed without much problem..
as I see it
* BigJohnT gets back to running little engines
how many do you have to test?
I'm just testing my PLC code
then we deliver the machines to the plant and they do the testing
that's what I were going to say
my machines are so expensive there's no other way to test them out then deliver them and then do the final trials with customers
(I've got one customer that does help out with testing of them, so I can try out new features before the rest gets them)
we usually do a run off here in the shop then assist the customer in final installation
* anonimasu nods same with us
* BigJohnT gets back to work
anyone mess with icsp on pic chips?
i have an icsp header to 40 pin socket adapter board designed and i'd like to have someone double check it
*before i cut
hm.. is it hard to get mastercam to post circuit boards?
not so much... if you ahve a dxf it's cake
though I need to buy a dremel or something
I've got some boards I want doublesided and I havent figured out how to do that with uv yet :) (no doublesided lightbox yet)
i'd say, after using the dremel for my spindle... buy or make your own precision high speed spindle... unless you jsut want to cut something
oh.. I'm going to use the big mill for it.. as that's +/- but the spindle is slow..
err +/- less then I can measure in position
it's a bit slow to mill pcb's..
got any idea of runout of your dremel?
more then 0.02mm ?
the runout of the dremel is about .001 - .002 (if you push it) if it's a light cut it's not bad
like on circuit boards it's not running out because of the light cut
oh.. that's good enough
that's the only thing I'd cut with a dremel...
i'm not really happy with the knock-off dremel spindle deal, but it works
* anonimasu nods
im hoping on making some kind of spindle once I get my tpg going
i'd like to make a spindle with skate-wheel bearings that would accept tiny solid holders
[18:42:22] <chr0n1c> http://imagebin.org/22084
<- screenshot of the dxf for my icsp to 40 pin pic18f452 programmer board... if anyone wants to take a peek (i know i need to re-cad it and make the traces neat)
i don't know if i got the pinouts right., so don't use that layout until i comfirm it...
(i think i got the pinout right)
if you rotate the header 180, you'll get one roundabout trace instead of three
yes, i wanted to make sure the traces are right before i messed with the layout
that was a quick, place this here and here by hand layout...
dont you use like eagle or something for making boards?
draw up a proper schematic first :)
that makes you very much less prone to errors
well, i haven't learned enough of eagle to make my own boards as quick as i can in mastercam ;)
well, it takes a few hours :)
less if all components are there already
i found all the parts in eagle and can do traces between parts.. i'm still trying to teach myself the auto layout bits
i think icsp is a black art or something... nobody is fessing up to knowing what i speak of.. lol :D
chr0n1c: autorouting basicly sucks
unless you pay a sh*tload of money for it :)
so far i can't get it to do anything but scatter the parts all over the screen
and it lines all the wires up neatly
but not attached to anything...
chr0n1c: don't attempt to do auto-placing too
oh, that was PCB and not eagle...
eagle i can't even get that far
you need to place components by hand, and then make it autoroute
although you usually get better results with manual tracing
ahh, good tip joni
I used Orcad and Protel a while ago
but as anonimasu said, you need to have a proper schematic
after that doing the PCB is slow but straightforward
(and when doing the schematic try not to cheat, and identify the proper packages & all..)
probably about the same as doing a board in cad?
chr0n1c: you use the move tool then move them until you like the placement..
then use the "route" tool
and drag pins to where they should..
it helps you buy showing you where you want stuff to go
from me messing around with eagle/pcb it seems the only difference with plain cad is that eagle keeps track of the traces' connections
oh, and it lets you make a proper schematic before you start making boards..
create a schematic when you start not a board
then you have a go to board button on top
where it populates the board with your components
and gives you connection lines to help when you do the routing
it's not dead simple, but it helps alot to reduce errors
I hate scrapping boards
perfect etch/stuff and then there's this stupid thing you havent bridged because you didnt make a proper schematic..
like putting a 305 H.O. engine in your 84 monte ss and forgetting to put the rear main oil seal in?
... "a friend" did that once... lol
car nerds! :P
well, a single bridge isn't hard to fix... :PO
no, but when you dont have one..
like, your part pin is soldered without a connection
so drill a hole right next to it in the pcb
and the pin we are talking about is beneath a lcc chip..
put in wire
a bit harder, yes
[19:00:32] <anonimasu> http://pt.farnell.com/productimages/farnell/standard/136560007-40.jpg
like that ;)
that's a sexy chip!
anonimasu: at least it's no BGA
Got a problem cutting tapers on the lathe (7*12 mini). Got an idea to fix it, would appreciate feedback on it.
Hi chr0n1c !
I have an old dot matrix printer (letter size), in that there is what is supposed to be a precision machined rod for the printhead to run on.
What I am planning to do is to loosen the tailstock with that bar chucked and with either a collet or the drill chuck clamp it in the tailstock...
K`zan_emc: does your lathe tapre without tailstock?
tighten it up minimally (tail stock) and use a dial test indicator to check it and when it is straight tighten it down an tap until it is straight.
anonimasu: dunno never turned anything really large on it.
everything I do is between centers since it is 3/8" max to date.
That tailstock is sloppier (adjustment wise) than a bb in a boxcar...
Like maybe put in a piece of 1" stock, leave it out around 6" take a thin cut and measure?
well, for all I know the taper might be ways/spindle/something
Over about 4" I go from 0.250 to 0.260 +/- .002
is the headstick on crooked?
Well, I also noticed that that carrage is a bit loose (can rock it back and forth some small amount.
like to the y axis?
in this direction ^
Don't think so, I checked that a long time ago.
er x axis i mean
if sofix that first..
I'll grab a piece of 1" and see what I get...
lol, i typed headstick instead of headstock
* anonimasu has been thinking about the lathe build
tubing hot rolled 500x200x100mm 10mm
:-) /me very confused but the taper was obvious...
and isel slides
Got a piece of ~7/8" stainless round. That seems like a good choice?!
and something like http://www.solectro.se/Archive/Images/Products/244/lfs127.png
Considering that God knows where the tailstock is now that I tried to adjust it :-(.
i think you are onto something anonimasu
this box is slow, working....
or two http://www.solectro.se/Archive/Images/Products/244/lfs89.png
3 or 4 jaw chuck ? .01 over 4" length sounds like the headstock not in line with ways
put a dowel pin in your tailstock and try to indicate around it with an indicator in your chuck
That would be nice, but how to incorporate it into/onto the mini mill seems problematical
im thinking if I can get the grinder working I can grind the tubing flat
Or a parallel if you have a 4 jaw...
Lemme chuck this up and see what happens...
ream in two locator pins and bolt the spindle block there
then grind it all flat
it's just a idea..
seems like square tubing is the easiest way to make it stiff
isn't round tubing more structurally strong?
how do you make a lathe base out of round tubing?
yes, for a given cross-sectional area
lol, anonimasu, same way as with square... i agree it would be a pita to mount things to
im going to throw it at the fea program at work and see what it says
it's probably good for a few 10kN's..
but it'll flex a bit :P
Heh, just had a set screw drop out of somewhere non-obvious...
K`zan_emc: I'd check the headstock first, then go on with checking the tailstock
So we stop while I find that.
IIRC I tested the headstock a good while back - TRO ~0.0025
But that doesn't say anthing about it being straight to the ways does it.
Well if the TRO is .0025 find if it is in line or looking up/down
This is a cheap chinee 7x12...
Don't mean you can't tweek it into working better for you
that's some horrid throw
shim stock washers!
Looks like I am going to have to disassemble the carrage (saddle?) as I don't see any missing from the cross slide or compound...
think shim and geometry
I am so lost...
hello, anyone here with a m5i20
K`zan_emc: one thing at a time
I've got one on this box?
mikeb, I may be able to help with setup questions, though I'm not running a machine with one
chr0n1c, are you programming dspics?... pinout for programming is slightly different
Kzan you need a good precision round like a piston wristpin, check your runout at both ends of the round stock you use
thx sw, has the card low or high active i/o?
depends on what you attach to it
chr0n1c, its a start on some eagle files if you want to play.... I have used it and it does work.... also, the pickit2 plugs into the adapter and works
if you use a 7i37, the outputs are isolated, and I believe ther'e active low (from the card)
Then taking the runoout into account run the indicator the length of the roundstock with the pointer at 9 oclock and 12 oclock to get an idea outta how the spindle is pointing down the ways
7i37 works like a opto isolated inverter?
SWPadnos: what was the name of your fancy PCB software?
it's opto-isolated, and its inputs are active low
altium .. right..
Altera is an FPGA manufacturer :)
chr0n1c: try looking at Altium :D
SWPadnos: I think I remembered that Altium does FPGA stuff too.. so somehow altera popped up
the outptus on the 7i37 are contact closures though, so they can sink or source current
yes, it does
you can embed a CPU into an FPGA, mark functions or blocks of code to put into the FPGA (instead of executing them), swap pins around while routing the PCB, and propagate all the changed back and forth between FPGA tools, schematic tools, PCL tools, and the copilers
man, that was full of typos
looks like i had pins 10-11 labeled as pins 11-12, someone over in #electronis on efnet pointed that out... TIP: don't deisgn circuits when you are half asleep ;)
but they say my pinouts were right expect for that mistake
What have I gotten myself into this time :-)?!?
LawrenceG, you have some other nice projects on your page, thanks for sharing!
To get at the carrge retainers I got to get it out in the middle of the floor and disassemble a chunk of it. No biggie, just what it takes and it *NEEDS* to be done. No time like the present as they say...
stone everything and deburr the castings while you have it apart ;)
Ain't moving much, but you can hear it thunk when you crank tthe cross slide
\I'd like to stone it and be able to affor a Harding or something :-).
Thank the gods for a workmate :-) :-)!
chr0n1c, thanks.... I like doing a quick web page to document projects.... helps me find the stuff when I need it later!
Still haven't found where that setscrew came from...
i found a strange nut on my mill table yesterday, i know i didn't use any nuts like it when building my machine so it must have fell though the floor from upstairs or something? or my roomate found it on the floor and set it up on the mill??
LOL, magic :-). I still see no place where one is missing, digging. Got it out in the middle of the floor now on the workmate (shudeer(.
No earthly, nothing mussing that I can see.
I am lewss than impressed with the workmanship on the slide....
good night all
Nighto, rest well
Just courious, should the part of the carrage that rides on the bed be smooth or should it look like someone hit it with about 80 grid grinding wheel? and mayne then broke the sharp edges?
heh, I read that as 'carnage'
are you seeing the hand scraping marks?
Heh, dunno if it has some putpose like to float mostly on oil or not. Took pix of it will post mater.
some ways look like ( ( ( (
to keep oil there
K`zan_emc: have a photo to share?
If there are scrape marks, the go across and up and down the V way.
Yes, but my hands are greasy and oily atm...
lemme go wash up and I'll post them
OK got them and the first part off the CNC mill (plus drilling mounting holes):
[20:50:12] <K`zan> http://wrlabs.shacknet.nu/~vw/tmp/ShopMisc070708/
hmm, I can't quite tell what's supposed to be the bearing surface there... it's pretty awful.
what kind of machine is this?
angle grinder anyone?
it is not even remotely flat.
[20:52:48] <anonimasu> http://wrlabs.shacknet.nu/~vw/tmp/ShopMisc070708/IMG_0007.JPG
hey, cool motor mounting plate.
it's worse then it'd get with a angle grinder
it looks like the end is somewhat scraped..
I'd at least strap that to the mill and machine it flat...
I think it is bearing on the very right end
For Gwiz, I'd like to be able to read the .ini file. What's the easy way to do that?
I see that axis does inifile.find(...), but I think axis has some stuff compiled in (whatever that means).
inifile = emc.ini(...)
right then a = inifile.find("...","...")
Where does emc.ini(...) come from?
(are you sure you want to read the emc ini file?)
I think so. The ini file needs to tell the interp where to find the gcode for the wizard subroutines.
It would be nice to coordinate that with where GWiz thinks they are.
I'm at the point where I generated some code using gwiz and then tried to run the program. Of course, the interp couldn't find the subroutines for the wizards.
yeah just copy inifile related stuff out of axis.py. you might have to mess with sys.path so import emc works (also see that stuff in axis.py)
Starting to think I should have had this thing delivered to a machine shop and said "here are the parts, build me a mini-lathe"...
K`zan_emc: what brand / where did you get it?
Cummins. From them...
huh, the engine manufacturer?
* anonimasu nods
ah, looks like a similar thing to what they have at harbor freight
Dunno, my ignorance is legion, I just wanted to get into this. Paying for my education, I guess :)
In theory... All these 7x12s are pretty much the same...
this is why im going to build a lathe myself :/
I'm just a bit discouraged...
some of that stuff definitely comes as nice castings that are ready to be used to make a machine tool out of
* anonimasu nods
But is it better than no lathe at all, got to keep that in mind...
it'd be better if they just sold you castings sometimes
The adjustment for this does not seem robust....
K`zan_emc: does it work badly? it might work fine if it is bearing at both ends. maybe they ground this center part away because it was rocking.
it's hard to guess.
Again, cradek my ignorance of this is legion, sigh, so far I have been able to do pretty much what I needed, but it ain't prefect, mostly just functional.
The turning of tapers has been the reallly big issue.'
functional is good.
what do you mean issue? what kind of tapers?
"Snug SHCSso the carrage can move but without play. Then snug setscrews gently and tighen nuts being careful not to break the retainers."
hey, all you programmers, if i modify pyvcp_widgets.py do i have to recompile (anything) to get what i changed??
Wish I had taken pix of the dop sticks I am making for Jean, it was obvious. Got it out by turning the work end for end and setting up on the narrow diameter and turning it again. Pretty close to straight at that point.
so you are getting an unwanted taper when you move just the carriage?
Yes about ~0.010 over 5"
Narrower toward the tailstock on the first pass
between centers? or do you have 5" hanging out of the 3 jaw?
Between centers, pretty much always (small stock usually 3/8"
Got one I buggered badly, lemme measure that one.
then the tailstock is offset, or the bed is twisted...
Would bet on tailstock, spent about 3 days screwing with it a while back and improved it slightly.
Tailstock end: 0.2495
Chuck end: 054
Chuck end: 0.255
maybe you just have to get the tailstock aligned with the headstock then. mount a longish bar between centers, take a light cut on each end. measure with a micrometer to get the difference. adjust the tailstock and repeat until it's close
you don't have to tediously cut the whole thing, just a bit at each end.
thanks for the suggestions on the adapter board cradek, LawrenceG, anonimasu i'm working on a revised version now...
it's best to bolt the lathe to something sturdy before you bother getting it close, since twist in the bed will mess it all up.
cradek: Next step after I fix the carrage. Got an old dot matrix and the print head rod is supposed to be quite precide, figured that might tell me something?
if it has centers, that could make a nice standard for running a dial indicator along.
Will tear that up later :-). Damn screws are over the leadscrew - FUN to get at :-)
it's nice to have a standard rod with centers on the ends. but I don't know about you but I don't have one.
so you can use the cutting method instead.
google probably knows a lot about how to set up a new lathe... I'm no expert.
and I have to run! bbl.
Old printers are supposed to be very good for those...
Probably enve inkjets (<= $3 at thrift stores
rr scul es thanks
chr0n1c: np :)
K`zan_emc: this is a pretty good guide to fixing the sort of problems with alignment you're having: http://www.gadgetbuilder.com/Lathe_Align.html
the key is to be systematic about it
what's that taste like in a pie??
dmess: like honey, but bitter
and a little sour
naw i probably wouln't like it... ; )
fenn: Thanks, checking.
aligning a lathe is pretty straight forward.... level it... dial up the spindle nose as best hou can and snug up the headstock... bring in the tailstock and find IT off the spindle nose... snug it up and start turning longer bars tan you plan to cut on this machine.... eliminate the taper and lock her down.. check level again...
carrage doesn't rattle anymore when I turn the cross slide handle - tight but not horribly so.
bes is also not even - gets tighter toward the end.
what did you do??
dmess: Adjusted the carrage so it didn't slop.
sound like an alignment of the screw to me in the up/down orientation
Back side was real llose and *everything* else was loose too.
Feels better now.
be carefull of over tightening tapered gibbs... they lock up at the least oportune time..
dmess: you're assuming the lathe is properly made and has actually cut a straight bar before
Box of parts loosely and more or less assembled to look like a lathe :-).
no assuming anything... if you do that you should be in the ball park ... and able to tweak it in... the carriage... is another thing...
Have seen *nowhere* for that set screw to have come from.
K`zan_emc: i've read a lot of reports how 'lapping' the gibs with sandpaper or a stone will improve things a lot
sounds like my new/old hardinge
hard stones ONLY
right, we dont want lots of grit hanging out in the crevices
I'll just sprinkle some of Jean's rock tunmbling grit on the ways and let it work while I do :).
you scared me for a sec K`zan
mine is apparently 9"x16" but with a cross slide only no tailstock... i cant see being 16" out on it
LOL, slow, crazy and ignorant but NOT stupid :-)
probably that's to make room for drill chucks on a turret
And working hard at dispelling the massive amount of ignorance.
Coffee break, bbiaf
yes... there is a cut little 6 stn turret on a nice cross slide option i dont have..
a tiny tool turret?
hmmm.. "RDM does assume that the prism which guides the carriage along the ways is absolutely straight..."
6" across the flats of the hex i figure
i cant really think of a good way to measure straightness of a bed in the front-back direction since you can't use a level, and turning it sideways would cause distortion due to gravity
without a straightedge*
straightedge is about the only good way
on a small lathe.
fenn: then the question is how precise do you need it to be.. any measurement is better then none
toastydeath: other ways involve interferometers etc right?
you need a gound piece of flatbar to put a level on
fenn: indeed, on longer machines differential levels are used
and interferometers when things start to get pedantic.
toastydeath: have you seen the silicon sphere kilogram artifact?
er, read about it
not personally, but i've seen pictures
what is it??
it's "open source" metrology :P
they're trying to lap a single crystal silicon sphere to get a better kilogram artifact
[22:21:29] <anonimasu> http://en.wikipedia.org/wiki/Image:Silicon_sphere_for_Avogadro_project.jpg
oh theres a job to send to china...
hard to believe they use hand implements to make it
not to me..
nice garden ornament
the lapping fixture is hand operated
anonimasu: lapping one of those things is not terribly complicated, machinery wise.
the measuring, etc, is.
do the mention the sphereicicty of it any where??
i think it's 0.3nm
if it was as big as earth the highest spot would be 8ft over sea level
use that in place of the 8 ball for billiards
[22:24:20] <anonimasu> http://www.popsci.com/files/imagecache/article_image_large/files/articles/vwFigure2.jpg
to check that they have to bounce it....
the gyroscopes for gravity probe B were lapped using sewing machine motors
SWEET lookin rig
and they're 80 nm sphericity
* anonimasu nods
oh, 0.3nm roughness, <75nm sphericity
still pretty ROUND
you sure about the 75 vs .3?
[22:25:54] <fenn> http://www.acpo.csiro.au/spheres.htm
i thought it was into angstroms
[22:26:12] <anonimasu> http://einstein.stanford.edu/TECH/technology1.html
toastydeath: no i'm not sure, there are two 'very special' ones that are maybe a lot better?
i talked to the guy who was in charge of lapping the gyros
he seemed to be adamant they were in the angstroms
and the same guy was equally adamant that the most circular spheres were lapped on that tetrahedron lapping rig
so i'm confused as to why they're using a traditional spherical lapping setup on the silicon kilo
they certainly ARE measuring to Angstroms...
* anonimasu nods
they need to take it slow... to stop at EXACTLY 1 kilo
i dont think it has to be exactly 1 kilo
er, but it is.. so, eh
the tet would want to finish a cycle in order to maintain its sphere...
spin-down time constant of approximately 15,000 years for the four GP-B gyros.
amazing what frictionless bearings will do for gyro longevity
and a turbo to
but it requires helium
i lied regarding the straightness thing
there's one more way, which we are exploring at work, but it's almost never used in anything
peaucelier cell? :P
if you get two spherical bearings, and attach a fairly accurate rod between them
and spin it
stick two measuring probes on the rod, and subtract the values
to get a decent estimate of the center of rotation of the rod
you get an instant straightness reference
limited by your ability to discern the center of rotation, and the accuracy of the two bearings
it's used in some very accurate CMMs
toastydeath: btw.. I saw porous bearing materials for sale
fortunately, lathes already have a bearing that can grip relatively straight rods
it's gotta be off the equipment you're trying to test.
this is pretty much the same as the "rollie's dad's method" isnt it?
spin a rod in the chuck, measuring runout with DTI
no, it isn't
toastydeath: with homogenous porosity
anonimasu: the homogenity of the material doesn't matter
toastydeath: that could be ground to mirror finish for bearings and stuff..
and maintain the porosity..
tuning the bearing is where the trade secrets are, and that process can shut down parts of the bearing so there's no flow at all
if you had a good tuning process, you could turn a sponge into an air bearing if you wanted
the material almost doesn't matter
which, i imagine, would be very popular with diy guys
"Turn your sponge into a bearing with 30 nm repeatability!"
well, if you can grind something and have homogenous airflow through it, I dont see why that wouldnt work as a bearing :)
i can't tell you why it doesn't work, because i signed stuff, but try it and see how it fares
you are one boring guy.
if the material is cheap, and you do a really good bearing design, an untuned bearing works.
not as good as a tuned one, but it works, and better than most stuff available to diy guys.
well, I didnt research it alot..
like i said man, if you want to experiment and the material is cheap
i suggest you go for it, because they're not that difficult to at least get flying
I doubt the material is cheap
I saw a ad in a paper
it's just graphite.
I dont know about that
there was something new about that material though
I didnt put it to memory.. i've got too much other stuff to think about
unless i can make something that's stiffer and throws less then a pair of ok spindle bearings it's not really any point to it
air bearings can be really, really stiff
and are very high speed.
_if_ you have a process for making them right..
well, sort of
like, you can get around having a bad process with good design
you just couldn't make them on a production basis.
just like traditional oriface or slotted air bearings, you can make em and get them running pretty darn good without much equipment
you just can't mass produce them without knowing the tricks, i guess
there's literature out there on air bearings that will tell you how to do everything, just with older sealers and stuff
if you actually look, nobody has had any patents on air bearings in a long time because it's all just been process improvements, it's all 1890's technology
lots of things are like that
the process to do stuff is too complicated for anyone to copy it
but like, alexander slocum's precision machine design
has all the formulas for flow requirements and clearances for both aerostatic and hydrostatic bearings
if you look up how to lap a bore, and practice
there's zero reason why you couldn't have an air bearing lathe or mill
or a hydrostatic lathe or mill, if you wanted a heavier duty but slower bearing
ive had an air bearing mill
hydrostatic has bigger problems I guess..
cavitation and stuff
but air is good - we have a 20 hp lathe at work with an air spindle, they're not wimpy
hydrostatic bearings are problematic mostly because of the filters
cavitation is an issue, but moving to lighter and ligher weight fluids helps
had hydro static ways on the BP series machines...
you could run one on acetone if you wanted, and get a darn fast bearing
just an intake...
hydros are stiffer than aero, which are already stiffer than rolling element bearings
and they have a lot more capacity before they ground out
toastydeath: the problem with lapping is probabl measuring..
but the filters have to be able to filter everything bigger than the gap size
of the bearing
its all about oversizing the pump in the 1st place
and they have to filter it at a rate faster than the bearings are using the clean fluid
nah man, lapping is easy
my y key sticks..
for flat bearings, a minute or two on any old lapping plate will do the trick
i lap shit ti .0001" daily these days...
hm maybe it's worth to call them and ask for a qoute on a bar..
and hear how small qty's they sell
for spindles and stuff the process, done by hand, requires practice
bar for what??
we hone ours, saves time
you can try without buying any porous media, man
we used to hone and lap ALL our ejector pins
or air bearings..
porous media actually just complicates things
all the really, really high accuracy bearings are oriface bearings and such
hm.. dosent the porosity give you a much stiffer bearing?
media is cheap... if you buy 3 cans ov various grit you'd be ok for LIFE
because it's easier to grind and lap tool steel than it is porous stuff
or heck, even aluminum
we had ceramics in ours....
if you find a copy of precision machine design, it shows you all kinds of layouts for grooves and stuff for slot bearings
all you need is a lathe or a mill
hmm.. I could buy the book if it's still in print
you couldnt spin the collet without FULL air... or the spindle was pouched..
it had its OWN compressor attatched to the machine
minimum rpm.... 8000....max 7200
that is a good point though, never let an air bearing starve for air
toastydeath: im going to think about it, it'd be a cool thing
at best they'll ground out and get scratched up
at worst the bearing will be designed with preload and weld to the other surface
scratched up bearings no workey so well
consider it more! read up on it.
just need to finish some more things first..
and bounce forever more.... my partner took out a spindle on THAT machine... after it was replaced in the showroom Japan said i was the ony one to play with her... ; )
like the grinder(when it arrives)
i babied that thing for 3 yrs... they never did sell it before i left
zephyr is now known as john_f
toastydeath: diy 30 000rpm spindles would be cool :p
yes they would
break a bit and die.
seen the datron machines?
i could cut Rc 60-62 on that thing like butter... for a short period of time...
hah, i think i have heard of those
look at the videos :)
very high rpm and low power spindles
I think they are made by "jeager"
flooded table with water... and about 10 % isoproply alchohl
keep the collet out of the water or ther IS a WATERSPOUT up thru the machine
positive presure thru the spinde saved her
dmess: why only for a short period of time?
TOOLs didnt last...
well, you were ahead of your time :p
didnt they make lots of improvements with tooling lately
i had 9 layers of TiAlCN on a .125" cutter and get 30 minutes
END MILLS.... i seen a ceramic ballnose from Hanita... EVERYONE i speak to about that tool remembers NOTHING about any tool like that...
cermet has a chance but it needs a SWEET spindle to mill with it
there are ceramic ballnose cutters..
and you need to FIND the sweet SPOT
I've seen ad's about them..
this was in 1988
Nicel little rod, ~12mm dia and 14" long an inch or so on each end eccentric for print head adjustment, but this should work.
i was pretty close with both hanita and iscar... (same thing i beleive now)
Wonder if it should wobble at the opposite end from the chuck, lets see...
there should be 2 centers in it then...
1 dia should be true...
downloadin' and viewin'
guess who i work for.
Only one center...
toastydeath: I think this is a way to do it.. once i get the grinder running grind a flat piece of something then machine a alu slab with orfices and test.. :p
1st one was a small center......then it got blown away by the 2nd
and read.. :)
i approve of your plan
you are tru to the smaller dia ... correct.
(While I dont need it it'd certainly be a cool thing)
easy enough to test too
you cal buy a 1x3x18 of precision ground stock for 30 bucks..
but im in sweden
you can't go wrong with trying it on aluminum first.
it costs twice as much and they wont sell me unless I buy twice as much
i could lend you mine if you'll cove shipping back and forth
she has a place in my heart... she's 22 yrs this yr
I've got ground alignment blocks setting up stuff on the mill
(I'm making up uses for the surface grinder)
Under 0.0005 right at the chuck!
i could pack a temporary tool box
dmess: it's ok :)
open offer my friend
toastydeath: that's a good pdf :)
the bar alone is 3 kg
lol i had nothing to do with it
you can thank mike, our marketing guy
toastydeath: it contains lots of info..
he puts together all the cool pamphlets and videos
toastydeath: my next problem is how to measure :P
"you need a 1µm indicator"
here's a philosophical question for you
Now I guess I get out the tailstock, stick a live center in it and get it down to the same thing at the other end and lock it down.
if the error is so low that you can't measure it
do you care about the error?
less then I can measure is no error.
if you lap it and, heaven forbid, it comes out to ~50 millionths flatness
which i would find way, way bizarre
would you care?
im reading about preload for flat bearings
oh.. it was just about what's avaiable..
50µinches :p is the minimum recomended
no, for setting up preload
for flat bearings
toastydeath: if a spindle throws less then I can measure it does not throw.
toastydeath: as far as im concerned..
if the air bearing flies, chances are it's more accurate than any spindle you've seen
whether it can handle cutting loads is a different story
ie.. if my needle on the indicator dosent move it's good..
but they tend to rotate accurately no matter what you screw up on them
there's always more pressure :P
then you have to vent it
or the bearing will hammer
though there's a limit to when things will explode.
also cfm requirements
non-porous bearings are air hogs
relatively, that is
* anonimasu nods
also i guess this is a good time to learn scraping
because if you start making your own bearings, you're gonna need a guide surface
though you are getting that grinder, right?
and I have a surface plate at work
hot stuff man!
really cheap too
get a scrap bar of something and scrape it up for a straightedge
* anonimasu nods
* anonimasu has no remote idea how to scrape
i have got the basic technique down, on some aluminum
but lack a surface plate, or the paitence to autogenerate some flats from these bits i have
I wonder how close a grinder will be to flat..
we'll see about that..
pretty close, i'd bet.
I hope so..
but for 250eur..
well consider that grinding and scraping
are really for two different types of material
you don't want to grind iron if you're going to be using it for ways
you grind and then hand lap hard ways, versus planing and scraping soft ways
i got my hardinge for 500 canadian... the cheapest ive found on e-bay is 2900 ther is 1 for 5995
hm, why not grind iron ways?
or you know, just grind since you'll probably be able to get a couple tenths in your machine's envelope
grinding softer stuff tends to embed grit
and assist wear, rather than impede it
you can grind it if you just need it flat
but if it's going to be doing any heavy bearing, it's not a good plan
grind to where you can and lap or scrape the rest...
how small is a tenth of a inch?
uh, .002 i think?
hold on lemme convert
scraping ISN't a lost art... its learnable..
isn't that 0.254?
.0001 inches is .00254 mm
eh? tenth of a inch == 0.1 inch right?
google says you guys are on crack
its a tenth of a thousands
Well, that bar isn't straight from the looks of it :-(.
oh I see..
tenth is slang for ten thousandth
it facked me up too as a kid in shop... we were the !st METRIC class
right, but I never heard it called out as a tenth of an inch to mean 0.0001; just a "tenth"
that's good enough
i just said "tenths," not tenths of an inch
i am so confused!
i.e. a "tenth reading indicator" not a "tenth of an inch indicator"
* anonimasu rolls over dead
canada used to be british imperial...
I know how to calculate from " but not the slang involved :)
that IT.. you just got it all
Out of ideas :-/, break time.
how much is a tenth of a tenth?
water level help any??? might get ya close
or is it 100
K`zan_emc: roll your bar on a piece of plate glass to see if it's straight
100 mil is a tenth
shit i hate this stuff some times
(but it doesnt need to be straight with enough math)
toastydeath: who in the right mind would define that?!
toastydeath: 1 mil = 0.001 inch
ds2: define what?
fenn: "mil" in the sense of microinches
toastydeath: "100 mil is a tenth"
toastydeath: "inch" in the sense of furlongs
no fightig boyz
the nonmachinist folks define a mil as a 0.001 inch
haha this is scary
* fenn thinks slang has no place in technical subjects
slang is inherent in technical subjects
check out the labeling on the plastic sheets home depot/lowes/etc sells.. they go by 6mil, 8mil, 3mil, etc
crap... what did we start..... i reworked THE inch about 8 yrs ago...
* anonimasu punches toastydeath
* toastydeath dies
* fenn flings poo at all
not uncalled for ^_^
mess throws up a fannn
most people don't call stuff out in tenths of tenths
what a scary world it'd be
after tenths you go straight to microinches
so who flung poo wears it
so .00001 is 10 millionths
i do beleive
yeah that makes sense
then you have to go into special rooms ;)
temperature/air controlled; not rubber ;)
special rooms where everything is brightly colored and everyone is so nice
no just sit in a special chair and face the corners...
and there are no sharp corners so i don't hurt my dainty head
toastydeath: I thought about padded soundproof rooms
and men in white coats
the specieal hade of Colgate Green
you try measuring to the millionth w/o a temperature controlled room ;)
it's kind of funny, people think we can measure .00001 on a surface plate that's .0002" in the area we're using
but the men NEVER come inside to chat.... its always thru the Bars
sometimes the people i work with are brilliant
"yes, this will work"
sure... but NOT out THERE
"yes, got it dead nuts!"
congradulations, sir, you've very accurately machined a .0002" dish into your part
YUP and its SCRAp
no, it's good, because the indicator reads zero!
the machine says it's .0002 concave!
the machine must be wrong.
never 2nd guess your equipment
just the calibration ;)
toastydeath: what machine says that?
anonimasu: the milling machine
that we just made the part on
how did he make a bow?
the machine bowed during the cut
or when we clamped it, rather
you set it up bass ass wards...LOL
you see, if you vary the clamp pressure very carefully, you can impart the exact error the surface plate has
into the part
toastydeath: now if you can just control that dish on say... glass...
are you mapping the part... every time??
i'd use the phrase "mapping" very loosely.
surface plate and clamping? eh?
the milling machine clamps.
warp the part.
thought a surface plate is suppose to be in the 'free condition"
the surface plate is .0002 out.
the milling machine is dead nuts.
why do you use the plate on the machine then?
NO machine is dead nuts... trust ME
we don't use the plate on the machine
the plate is supposed to be the inspection process.
My 18 x 24 surface plate has lapped granite plates.. IT is the NUTS
whatever man, if they want a .0002" bend, i'll give em a .0002" bend
just push down in the center while you clamp it.
plate says it's okay!
if the surface plate is certified for 0.0002 and they are using it for more, someone needs a remedial class in sig digits
inspection devices are a key to manufacturing
toastydeath: good idea I cut a groove \ because the customer's axle was crooked so the measuring didnt work out at friday..
we also have a cmm that's posessed
ds2: the plate is certified to .0006
but they found a spot that was only .0002 in the envelope they care about.
heh I see
like g1 x100 z0.02 f100
thought sweet spot measurement is not permitted for proper metrology?
ds2: a lot of things aren't permitted if you're doing things properly
you first have to decide you're gonna do it properly
toastydeath: I had to machine keyway on a shaft..
i farcing HATE cmm's we have a large hunka junk and a small cadilac...
oh, you were bumping the Z up to keep it in tol?
toastydeath ramping the Z 0.02 over 100mm :p
toastydeath: and the guy wanted it precise.. and the shaft was crooked..
toastydeath: well, if they want ~ they can have it.
you're more honest than i am
i'd have just said "yes sir"
and done it
"is it accurate?"
it was for a plastic gear in some machine..
heres an invoice
i am picking up bad habits like that, especially when management says "well the engineering drawings are more of a suggestion"
that would have worked with +/- 0.1mm
NOT in aerospace..
it's like this damn press fit people talk about..
commercial OR military
dmess: i have no doubt that in places that have decided to do everything right, then yes
but again, it comes down to "do we want to fire everyone in engineering, or just make the damn parts"
the machinists there have been making good parts for so long they make them right, no matter WHAT engineering puts on the drawings
"this is +/- .03
do you want to have planes droping out of the sky with friends in them.... NO
What, that needs to be +.002, -.000
dmess: we don't make planes, though.
the housing of an air bearing is not exactly an aerospace part.
toastydeath: you guys work in wood or styrofoam? ;)
ds2: feels like play-dough and an enchanted castle sometimes
hey some one needs a Kaizan on design for manufacturability
toastydeath: oh yeah.. you should keep a normal caliper(non digital) close to your machines for work like that..
i once asked them if "design for manufacturability" was one of the goals
not a ISO9000 facility?
the response was "that's not our job"
and i quote.
ds2: wil be soon
the problem is that we've got it documented.
that's all they want, really.
"this process is broken, but at least we have a flowchart"
heh so any idiot can come in and make mediocore parts? :)
well.. IF someone ABOVE their boss lights a spark... it'll burn..
it's not so much a matter of mediocre parts, because we have really good machinists