woo-hoo - no more blue chips for me, I'm making yellow ones now
400 SFM, 0.005 DOC, 0.004 feed per rev, hardened steel
the tool is the insert boring bar?
no, the boring bar was to make the steady
I'm now running one of the spindles in the steady
just taking some experimental facing cuts
boring comes after I get comfortable with facing
unfortunately this part has lots of little holes in it - so just about everything I do will be an interrupted cut
the spindle is a lathe spindle?
one of those HSK spindles that I want to re-work
yes - but you are using it as a lathe or are you milling?
anyone ever made a smoking pipe?
a-l-p-h-a: smoking what?
as in tobacco?
Jymm, you know I'm Canadian.
I'm modifying the spindle, in the lathe
a-l-p-h-a: Oh, pot... long ago
I got a few designs I'm curious in trying... http://www.smokingpipes.com.cn/viewproduct.php?productid=577
change the end to a square or hex head... to it could lay flat.
a-l-p-h-a: make the whole thing hex
pic of setup in a minute
you are facing the end of the spindle - using the steady rest you built
the picture finally came into focus
a-l-p-h-a: spiral the grooves
biab - dog is whining
Jymm, that would be cool... but I don't have a threading attachment to be able to do so.
can I use that excuse? my wife is calling
[01:48:57] <jmkasunich_> http://jmkasunich.com/pics/spindle-facing-2580.jpg
jmkasunich_ is now known as jmkasunich
question for anyone who has done any hard turning (or other turning with carbide for that matter)
right now I've done test passes at 0.005 DOC and 0.004 feed/rev
if I want to get more agressive, which should I increase
I have no experience with hard turning - I would think if you have parameters that work and give you the finish you need you should not try to fix it. If I were going to try to get it faster speed is the parameter I would try first. DOC and feed are the parameters that load the tool. Inserts can take a lot more heat than impact especially if the heat of the cut is in the chips.
interrupted cuts are the killer here
just for grins I tried cutting coming out - that gives a very shallow entry angle, thin wide chip
I thought it would be bad, but it cut nicely
about 300-350, varies with radius
that should help on the interrupted cut
stustev: yeah, it sort of spans the hole
I was afraid the chip would be too thin and it would start rubbing
yay, I got Z feedback
must be a good solid setup or it would rub and not cut
did you see the pic?
I got some metal there
yes - the diaper is to keep chips out?
yeah, protecting the back bearing pari
looks solid - you never can tell though
ah, neat, I didn't read back
there is definitely some flex - I repeated the cut 3 times without moving Z, and it cut everytime
you got Z feedback but no read back - what is read back?
looking back at the conversation here (to see that url)
he didn't read back in the IRC channel to see what we were talking about
cradek: good evening
can I configure with --enable-run-in-place and --enable-simulator at the same time?
I should have asked the question different. Will it corrupt anything when I do that?
upped the speed to about 450 FPM, DOC 0.002, feed 0.0035 per rev, cutting nice (cutting on the out pass only)
RIP shouldn't touch anything else
with the DOC and feed less you will have less pressure on the setup - the chips should turn blue when you cut - that would be what I would shoot for - no coolant either
lol - I'm way past blue
chips are coming off red to yellow
that sounds as if you are fast enough - I cannot be sure as I have no GOOD experience with hard turning - the only hard turning I have done is to try to remove the work/heat hardened areas after a tool has broken
that sounds hard
I changed the emcmodule.cc and the axis.py
after compile the preview changed but not what I wanted - this is only the first try - the B axis command moved the A axis display - hopefully it will get better
[02:16:04] <jmkasunich> http://jmkasunich.com/pics/hot-chips-2582.jpg
HOT - be careful - when those chips land on skin they do not bounce off!
nomex overcoats help
the flying chip will find the ONLY open skin - it's a law
they are very very fine chips - they don't fly far
also is that ceramic/cbn?
or should i scroll up
its one of some inserts that my dad gave me - quite a few years old'
I think your carbide won't last long like that, but I'm not completely sure.
there were some ceramics in there, but I don't think this is one of them
did it last
so far so good
total cutting time is probalby only a minute or two
I'm still in the "make one pass, change something" phase
I noticed some chipping on the top early on (with a loupe)
i don't know much about the older high temp carbides, so i'm not sure what kind of tool life one should expect up there
it hasn't gotten any worse since I stopped doing in-feed cuts, now I cut coming out
that's weird, man
chipping = tiny flakes of carbide missing from the top of the cutter
on the edge or a little in
this is an interrupted cut
edge chipping that you need a loupe to see is probably microwelding. Something in the part is similar to the tool material
look closely at the end of the part in this pic: http://jmkasunich.com/pics/spindle-facing-2580.jpg
see all the little holes
and the notch on the outside? there are two of those, and 6 holes
oh snap these are the spindles you got
finally got the steady made, this is my first experimental cutting
so far I've taken 0.047 off that raised part on the end
up to 0.003 DOC - it takes about 5 spring passes to stop cutting when I stop advancing the tool
hey does anybody have any quick tricks to center a drill chuck on a turret lathe
there's a hardinge hc at work and it's got a drill chuck on it, but nobody bothered to mark where the zero is on it
and i don't want to indicate it because it's not worth the effort
it shouldn't take long to chuck a dowel pin in the drill chuck and dial it in
not fast enough
what i did was just poke the drill and watch the deflection
it worked pretty good, the hole wasn't that oversize
how fast do you want to be? how much are you willing to invest? how close do you want to be - cheaper faster better - chose two
uh, i'm not trying to get philosophical here dude
what is the hardinge hc?
what is it, like, a picture?
a lathe - mill?
[02:32:15] <toastydeath> http://www.machinemanuals.net/web_pages/hardinge_hc.htm
we keep it tooled with some common crap so you can just walk up and do quick modifications to stuff
chuck a dowel pin in the spindle - tighten the drill chuck on the dowel pin - clamp the holder in place
that'd work pretty well, thanks
better than what I did, anyway =/
you could mount a block (with a pin in it) beside the spindle - determine the distance from the spindle center - tighten the chuck on the pin - you would know how far to move the drill chuck to align with the spindle center
you would leave the block mounted at all times - could be useful for other tools also
that's a kind of neat idea but it's not my lathe
anything that was actually important i'd dial in
scratch the lathe modification :)
yeah i don't own any machines, just work in a machine shop
do you use the cross slides too? if you have an extra then mount the block in one of the slides
it's not that important man, i was just looking for a quick and dirty method to use that requires no effort
chucking some rod will work because the parts that fit in the machine almost always fit in a drill chuck, so that works
interrupted cuts suck
they are UGLY
where there are no interruptions, it builds up enough heat to soften the metal, then goes to town
where there are interruptions, it doesn't get hot enough, and it bounces and rubs and chips the tool
very light cuts very fast rpm
my last attempt was 0.003 DOC, 0.002 feed, and 1700 RPM
I doubt I can go any lighter and have it still cut at all
the machine just isn't tight enough
there is little vibration in the steady, but the tool vibrates during the interrupted cut - the ways are just too lose
not enough mass - can you set some more mass on the cross slide?
I could probably come up with 20 lbs or so of metal
200 lbs or more
not gonna happen
you need something to dampen the tool motion when it enters the cut
the table is only about 6 x 16"
* jmkasunich thinks about grinding
all you need is a quart or two of mercury :)
stustev shouldn't I be able to grind it with a regular wheel
Osmium would be better, but there may not be 200 pounds of it on eart
yes - a regular wheel will do it
I have some cermet boring inserts on the way, I'll try boring with them before I give up
[03:41:08] <jmkasunich> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=120267949543
cbn will wear very little - the soft aluminum oxide wheel will wear a lot - you will want a soft wheel to grind the hardened spindle nose
those are supposed to be good for hardened steel
rare to find an ebay vendor who gives feed/speed data in the auction ;-)
will be a good test
Hi all, I am coding my first complicated parts. I am wondering what each of you use to get from the CAD model to cutting a part the fastest and/or easiest.
with my spindle grinding set up the cbn wheel wears not at all if it is sparking during the cut. if it is not sparking the wheel is wearing and almost not cutting anything. I only take about .0004 DOC max with it.
when you do a spindle, do you just regrind the taper, or do you put an insert in?
depends on the spindle
I don't have 50 taper inserts so any 50 tapers I just grind the existing taper
when you do the insert, how is it held in? shrunk?
I use .001/.0015 shrink fit - insert in liquid nitrogen - 2 in dia shrinks about .005 - very easy to put in
assuming I can bore these things, if I take them to MT3, I'll have holes and such inside the bore - coolant, and tooling key from the old taper
I'm thinking of taking them to MT4, and then inserting a section of an MT4-MT3 hardedned/ground sleeve
then just finish grind the MT3 ID for concentricity - very light grind
GNieport1, I suspect the answer depends on whether you're looking for "fastest/easiest for this part" or "fastest/easiest in the future"
that should work just fine
unfortunately, I don't have any specific suggestions for you (I've only used TurboCad, and that only for a review)
nice thing about MT is that I don't have to have the diameter perfect
as long as the taper is right, it'll go
complicated is a big word with many meanings - is this a mill/turn part, lathe part, mill part, multi axis
jmk: that's right
if I chill it first, with dry ice or liquid N2, then gently whack it in there and it should be in for good
SWPadnos: We learned Solidworks in school, and I was able to dig up an 2005 educational version to use. I have tried importing the models into both Surfcam 3 and Mastercam 9, but the tool paths have been incredibly difficult or impossible to add, as if the file conversion was faulty. I'm not sure at this point if there is any automated way to go from the 3D model to GCode. I could always export 2D models and use something like DXF to GCode.
if you have infinite cash, you can experiment a bit ;)
there's a SolidWorks plug-in (CAMWorks maybe)
I should say, automated conversion that doesn't cost $$$$$
if you chill it you shouldn't need to whack it
did you export as STL?
SWP: That, and IGES and ProE
well, there's a pretty basic, $120-ish program called STLWorks
it's reasonably good, though pretty basic
(I guess I've messed with that also)
The problem would appear to be that the Solidworks is a year or two newer than the CAM packages
STLWorks will only import STL, so if SolidWorks isn't broken, it should work
SWP: I'll take a look at STL Works; I hope to not have to redrw each part in GCam
I did a model for Bridgeport X and Y axis end bearing brackets, and STLWorks seemed to generate OK toolpaths
I don't know enough to know if they sucked though (and I didn't know much about generating toolpaths or tool selection at that point either)
yeah, I'm not running a 5-axis, but there a lot of arcs :)
stustev: I didn't mean whack it with a hammer - just put it in quickly, same as putting a chuck into a drill press
I'm uploading that toolpath, so you can see some of what it can do
are your cbn wheels something like this: http://cgi.ebay.com/BORAZON-CBN-MANDRELS-PLATED-500-DIAMETER_W0QQitemZ270162348747QQihZ017QQcategoryZ58198QQrdZ1QQssPageNameZWD1VQQcmdZViewItemQQ_trksidZp1638Q2em118Q2el1247
sure. here it is: http://www.linuxcnc.org/dropbox/bracket1.nc
I don't even know if it loads in EMC, to be honest :)
hm, the velocity loop oscillates
it moves fine but buzzes most of the time
is the scale the same as the other axes?
you tune it by changing resistors and caps
SWPadnos: no scale - all analog
right - got that now :)
that leaves me out ;)
stustev: did you get my PM?
anyone here know much about encoders?
resting with beer ;)
anonimasu btw thanks for your pointing in the right direction re my spindle speed counter
i have now integrated a display into EMC that measures and displays the actual spindle speed
woohoo sorted my encoder:)
now just gotta get the rest of it working lol
colinb_ what encoder are u working on?
on my lathe
nice. i'm just working on making my mini-mill motor controller work from an external pwm input which will come from emc
although the power supply is a bit crappy, all the onboard pwm stuff floats at 1/2 mains
this isnt really a mini lathe tho
had some problems powering the mesa boards enough due to lengh of cables
think im gonna have to install a second mesa io board due to all the IO's
righty F1 race starts shortly
Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-08-24.txt
nap time anonimasu ?
I'm watching the OS end
err olympic games
anonimasu: When you said OS was that suppose to be Olympic Slaughter by chance?
I keep wondering if the fireworks are real too :)
I watched the openin cerimony, but that's it
When they have Olympic Female Jello wrestling... then we can talk
how are things going?
Pretty good all in all. FINALLY solved the network problem for the EMC box. Not quite optimal but it works and at this point I am tired of screwing around with it, I wanna get to throwing chips :-)!
What you been up to, it has been a while.
speaking of throwing chips: http://jmkasunich.com/pics/hot-chips-2582.jpg
jmkasunich: what kind of freaky material is that you are turning?
oh, is that the spindle :)
not going well - it actually cuts nicely where there aren't any interruptions
but much of the cutting is interrupted, and it sucks
I'm waiting for some boring inserts - that will be the real test
Got a 1" round of stainless of some ilk and where I tried to turn that in the beginning it looks more like I was knurling it :).
let's talk about cast iron now :p
Got better after that
* anonimasu chuckles
heh, cast iron = nice
Easy stuff but messy as I found out when I modified the tailstock...
haha.. depends on what kind of cast iron it is :)
Gets even more messy with oil/coolant
should have seen the pile of chips from making the hole in the center of that steady rest
K`zan: always cut cast iron dry
jmkasunich: Yes, have since learned that <blush>.
thermal shock kills tooling pretty nicely too :)
NICE to have the EMC box networked, finally.
network is good
* K`zan does happy dance!
And I get away with NOT having to run 100' of cat5...
Not so much the length here but the routing. 25' by wireless, 100' by ethernet.
Of course by the time I got wireless working I could have done the routing manually.
did you make any chips yet?
gnome IMO just reeks, but that box would be a klunker under KDE.
I've been procrastinating cutting plastic all day :)
I first need to make a design to cut..
brb dinner I think
Only got the NEMA23 mount done. Ended up just using the mill to center drill the mounting holes and then did them on the drill press to the right size.
jmkasunich: amazing photos of yellow chips. EMC hard turning! congrats on the steady
tomp: the faux ivory is cast urethane resin, + i think the links are fixed now
dgarr: thanks, you've done some beautiful work.
oh, and sorry if critical about the php
i'm kinda free with the 'show source' when it's interesting
tomp: do you know anything about mastercam?
no thanks for mentioning it -- i broke a php function the other day and didn't know it
dgarr: ah! np
anonimasu: i dont use it myself, and have problems when handed their files
haha, gcode is more unified than cad formats!
STEP is pretty unified
anonimasu: ordered an Design Express
a bit under 1k
it's the most complete version
not sure how much CAM it contains
oh not very mch
alex_joni: that would be expert
crap, that's what I meant
Xpress is the free one :)
[19:01:44] <tomp> http://www.alibre.com/xpress/
check out their contest: 2nd place rc controlled duck decoy
and 3rd is a cnc router, but with 2 motors moving Z platform ,and Y is a single, all at one side
I like the ATC they have on the "free Design Express" page
(just under the ATV)
has anyone tried Design Express under Wine?
hmm.. not me
I'm assuming it works OK in a VM ...
SWPadnos: url for ATC?
[19:09:26] <SWPadnos> http://www.alibre.com/xpress/software/alibre-design-xpress.asp
it's the 4th photo down - the spider-like thing
I don't recall where they're for sale, but I did see a video or domething a year or two ago
yeh, looks familiar, for mini mills
I think fenn planned to build one like that
I think it all runs on air, even the spinning and lifting
whoops, stumbled into another min ATC http://littlemachineshop.com/products/product_view.php?ProductID=2937
they had one of those at the CNC workshop
no wonder I was having fits last night - there are no less than 24 holes in the face of the spindle
I knew they were there, but untill I sat down and transferred them to my drawing I didn't realize how many
the spider http://home.insightbb.com/~joevicar3/Automatic_Tool_Changer_Plans.htm
jmkasunich: which was shown? horz chain or spider?
will it help to plug holes?
hey, Thomas Kramer (EMC pioneer) wrote about boring and checking bores http://www.nist.gov/msidlibrary/doc/kramer89.ps
regarding holes - I'm doing some exploratory work, first with measuring, later with cutting
the spindle is made of at least two, maybe three pieces shrunk together
I'm not exactly sure where the borders are yet, but if I'm lucky, I might be able to shed a piece or two and have a nicer chunk left over
well, one joint line confirmed
but there still seem to be things that are impossible about this thing
I'm sorry Dave, but I can't do that
HAL9000 is now known as SkinnYPup
jmkasunich: what the heck are you doing?
the spindle project
you're going to have to update your blog.. :)
when I actually have some progress to report I will
you saw the hot chip pic, right?
that and a few chipped insert corners are all I have to show at the moment
Anyone had any luck converting the output from pcb to gcode ?
[20:48:32] <SkinnYPup> http://sourceforge.net/projects/pcb/
quick question, What is a good program to use in linux to convert a dxf file into a file (ngc??) that can be milled with emc?
Sheetcam works well
do you know of any that are open source, or at least not bricked?
[21:44:39] <SkinnYPup> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Cam
REVIEW THE GCODE
on my version it ignored Z feedrate parameters during etching
its an eagle ulp
if youre not using eagle, that sucks
twingy is reworking the gerber to gcode in gcam
i'm not up to trying to mill pcb's yet. :-(
so itll work with all the retarded aperture macros
I've been using eagle with the gcode ulp , my friend has been using PCB and trying for the same outcome since it isn't size limited
its not so hard just need good setup
i made a clamp fixture from aluminum
blitz: i've had success on simple outlines with dxf2gcode on that wiki page
fenn, kk I am trying to see if it is already on the fedora repo quick
no, i really doubt it
renesis: pcb-gcode from http://www.brusselsprout.org/PCB-Routing/
Is that the one you are refering to ?
no, can't find them.
download it from the cadsoft ulp downloads page
thx, I found what I needed, I need to head out to the shop to test this quick. :-)
ok python gurus... what is the syntax if I want to print the hex value of a char? ie char = ' ' print("char = %04X", %char) chokes..
skinnypup: ive had good results with single flute and pyramid tip conical cutters
30deg ones did down to like .008 space/trace ok
but homebrew boards at that resolution are a bitch to assembles
design sucks because drc on the eda apps isnt set up to catch problems that only occur on homebrew boards
like via under chips or no continuity via thru hole plating
LawrenceG: print "char = %04X" % char
renesis: thanks I'll check out that ulp for eagle also. I'd gotten fairly good results out of the gcode-1.zip ulp on their site
fenn, self.gcode.append("(character %04X)" % char)
TypeError: int argument required
maybe in a month or two there will be a better solution
twingy is using libgerbv to generate bitmaps and converting to objects used by his cam app
Hoping to hear something for PCB , belive he used this converter and it made gcode that looked more like silkscreen than trace outline
[22:00:22] <SkinnYPup> http://www.chiark.greenend.org.uk/~alanb/ps2gc.html
cuz funky gerbers were killing the existing converter
skinnypup: this should work with any pcb app
but yeh prob wont be done for weeks or months, iunno
renesis: refering to what twingy is working on ? "work with any pcb app"
LawrenceG: i think you want something like ord(char)
feen.... thanks... I think that works.... amazing how something so simple can escape the brain
skunkworks: ha, gcam
but youre gone
i thought i already said
s = "A B"
for c in s:
ord( c) Given a string of length one, return an integer representing the Unicode code point of the character when the argument is a unicode object, or the value of the byte when the argument is an 8-bit string.
SWPadnos: u are a python interpreter :)
for small values of python ;)
ord, car, started to look like Pascal (gag!)
using CRT, DOS;
well, I think the "spindle project" is dead, or at least seriously wounded
I've finally figured out all the details of how they are made, and I don't think I can avoid an interrupted cut
fill the holes with solder, then melt it out when you're done ;)
solder != hardened steel
I did consider suggesting thatyou put screws in the holes, but there's no easy way to get them out afterwards
some of the holes are threaded, 2mm and 3mm
others are not
and there is a huge 10mm wide slot (one per rev)
those can get pins, but they have the same removal problem (unless they're through)
I don't think any of the pins/screws would need removed
oh - are you trying to make a flat face?
the face is the lesser problem
boring is the real issue
I figured you'd need the holes later, but if you don't - well - stick pins/screws in them ;)
thats actually not a bad idea
hopefully a SHCS is close enough in hardness to the spindle material to keep the cutter happy
rayh: I think I've got a duff amp... I can't get it stabilized.
I checked everything for mechanical slop again. I can't find any problem.
it's the velocity loop oscillating.
the spindle problem in a nutshell: http://jmkasunich.com/pics/HSK_spindle_machining.pdf
green = shrunk on ring, red = the sleeve I'd like to install (and the bored hole it goes in)
black = hardened spindle
cuttable, but not if there is an interrupted cut
too damn many holes in that thing already
jmkasunich: the problem is the holes shown in section B?
those are in a terrible place
primarily, although the ones in A are a problem too (angled ones)
are those coolant nozzles?
I think, except that they are obstructed when the tool is installed
maybe they mated with the holder for through-tool
can you just plug the holes with some CRS?
it might be enough to keep the tool from bouncing around
most of them, probably
although I'm not sure how to keep the plugs from rotating
but if you look at the front view, you'll see that B also includes a huge square slot
hm, the B might want to come out when you cut nearly half the diameter away
one coming out would be a serious problem
if I don't try to use the sleeve, then the inner red lines are the bore lines
in that case, the main problem is the big slot
I don't understand the lower drawing of section B. is the big slot not open to the inside?
looking at the front view
the right hand hole is just a drilled hole, it is on the bottom of the section dwg
the left hand hole goes to and thru the slot
I understand so far
the slot is "buried", they machine it from the outside before they put the green ring on
if you took the green ring off and looked from above the top of the lower section B drawing, would you see through the slot into the bore?
thats how I figured out that the green ring was a shrink-on - otherwise it would be impossible to machine the slot
as is, its still difficult - the corners are square
ok the extra lines there confused me
it is a very confusing part
oh I see, you're showing where the curved and flat parts meet at an edge
took me all day (and cutting the ring off of one) to figure it out
so if you don't do the sleeve, you only have to deal with the slot and the holes in section A
and the section A holes barely (or don't) need to be cut
the ones on the face are a nuisance, the ones in the bore could be left for final grinding
you could just grind some clearance and leave it
oh I forgot about the face
but the two small sections are way too much metal to remove by grinding, and the one closer to the opening has the slot
needs to be flat for tts?
yeah, flat to 1.5"
I could probably use the 4 outermost threaded holes to put a ring on, instead of machining off that step
true you could cover the whole front
the thing that scares me about not sleeving it is the lack of support for the collet where the slot is
I think you have to close that anyway, to bore it
hmm, I wonder if I could kill two birds with one stone - mill a block that fits in the slot - drive a pin thru the upper B hole and thru a hole in the block, to lock it in
it would have to be 10mm x 15mm
the B pin hole is 5mm behind the slot, and 5.25mm in front if it - makes the pin a bit more challenging
a tiny bit of taper would help it lock in
the holes are straight tho
I'm thinking spring pin(s)
mcmaster has 5mm spring pins
because mcmaster has everything