#emc | Logs for 2008-01-25

[00:01:26] <eric_U> sense should be in
[00:02:14] <Gamma-X> ok
[00:02:17] <Gamma-X> makes sense
[00:02:26] <Gamma-X> how would I write the estop out line so its always open.
[00:02:35] <Gamma-X> unless estop is hit so then it closes.
[00:06:17] <Gamma-X> ? anyone
[00:06:51] <eric_U> is there an estop input into your machine?
[00:07:16] <Gamma-X> yes theres already an existing hardware and setup tied into the amps.
[00:07:28] <Gamma-X> it has 24v going through it.
[00:07:44] <eric_U> so if you interrupt the 24v, it is in estop?
[00:07:51] <Gamma-X> correct
[00:08:53] <eric_U> my understanding of how that works says you should be able to hook estop line through the out high/low of the 7137
[00:09:00] <eric_U> but I haven't tested it
[00:09:05] <eric_U> I would probably use a relay
[00:09:45] <Gamma-X> y
[00:09:58] <eric_U> 'cause I'm special that way
[00:10:15] <Gamma-X> ok
[00:12:32] <eric_U> do you have any idea what the current is? limit through that mosfet is 1A
[00:13:44] <Gamma-X> prolly not 1 amp
[00:13:46] <Gamma-X> thats a lot
[00:13:59] <Gamma-X> its 24v that turns on a relay
[00:14:23] <eric_U> might want to check it with an ammeter
[00:14:40] <Gamma-X> ehhh
[00:14:48] <Gamma-X> whats the worst that could happen i blow that one circui.t
[00:14:53] <GNieport> lol
[00:15:02] <GNieport> that's the spirit :)
[00:15:12] <eric_U> just move it all to the next one
[00:15:20] <Gamma-X> exactly!
[00:15:47] <GNieport> be sure to put a diode across the relay coil
[00:16:07] <eric_U> you would think it has one
[00:16:08] <Gamma-X> ?
[00:16:16] <Gamma-X> it was there existing
[00:18:36] <Gamma-X> im jsut gunna use a piece of wire for my estop
[00:18:44] <Gamma-X> and be ready with a pair of snips.
[00:18:54] <Gamma-X> im soffisticated
[00:20:08] <GNieport> uh, be sure to revisit that FIXME before powering up the servo amps
[00:20:17] <Gamma-X> ?
[00:20:23] <Gamma-X> fix me?
[00:20:42] <GNieport> I crashed my X into the hardstop because my estop button was too far away.
[00:20:55] <Gamma-X> lol
[00:20:58] <GNieport> You might not have time to think about the pliers
[00:21:01] <Gamma-X> limit switches?
[00:21:06] <GNieport> I'm serious
[00:21:31] <GNieport> limit switches can only do so much, when the motor is spun up to max rpm, it coasts a while
[00:21:57] <BigJohnT> then your limit switches are too close to the end of the travel
[00:22:13] <GNieport> lol tell the machine builder :)
[00:22:27] <GNieport> this is factory
[00:22:41] <BigJohnT> If your installing emc you are the builder
[00:23:04] <GNieport> the iron is all original
[00:23:20] <GNieport> the limits are factory installed.
[00:23:36] <BigJohnT> then they put the switches in the wrong spot
[00:23:36] <GNieport> I plan to adjust them and add a home switch...
[00:24:00] <BigJohnT> what kind of machine?
[00:24:12] <GNieport> apparently. or perhaps the steppers it originally had could not spin the screw at this speed
[00:24:20] <GNieport> Shizuoka AN-S
[00:24:48] <BigJohnT> the latter is most likely the case
[00:25:33] <GNieport> Since I calculated the critical speed of the screw and set my motors for that - 5%
[00:26:24] <BigJohnT> bridgeport type of mill... with 120ipm max speed... what speeds are you getting?
[00:26:41] <GNieport> hmm, one sec
[00:27:20] <GNieport> 380 IPM
[00:27:32] <BigJohnT> hmmm that could be it
[00:27:36] <GNieport> theoretical
[00:27:48] <BigJohnT> beefy looking table
[00:27:59] <GNieport> it made the uhaul groan
[00:28:07] <BigJohnT> what are the two handwheels on the knee for
[00:28:08] <Gamma-X> wish me luck biotches! TESTING!
[00:28:16] <GNieport> good luck! :)
[00:28:37] <BigJohnT> Gamma-X you don't need luck you need a fire extingusher
[00:28:41] <GNieport> mine doesn't have those handwheels, it has the same layout as a Bport
[00:28:59] <BigJohnT> ok, I just googled up one to see
[00:29:13] <Gamma-X> * Gamma-X is going to kill himself
[00:29:29] <BigJohnT> take pictures so we can all see
[00:29:34] <Gamma-X> haha ok
[00:29:36] <GNieport> teh knee is on the left, perhaps the handwheels are offset by the motor and belt system
[00:29:48] <Gamma-X> cant phones dead
[00:30:16] <BigJohnT> 51"x11" table is nice
[00:30:29] <BigJohnT> 5 1/2 Z is nice too
[00:30:41] <GNieport> I'm not sure what table mine has.
[00:31:45] <GNieport> Enough to put the Kurt to one side and have room for direct clamping
[00:31:59] <BigJohnT> sweet
[00:39:12] <GNieport> http://home.cinci.rr.com/bog/stuff/100_1808.JPG
[00:39:41] <GNieport> my machine after arriving home and assembling the turret and head
[00:43:50] <jmkasunich> looks heavy
[00:44:21] <BigJohnT> looks clean and in the original paint
[00:44:45] <Jymm> poor pallet jack!
[00:45:17] <Jymm> lol
[00:45:27] <BigJohnT> had to bring his own transformer he stole off the pole LOL
[00:46:12] <Jymm> BigJohnT: Nah, batteries not included... but supplied with 20KVA transformer, cause you gonna need it!
[00:47:16] <Jymm> lots of Y travel in it
[00:47:26] <Jymm> oh, nm.
[00:48:02] <Jymm> I didnt see the head was all the way back
[00:50:53] <GNieport> the turrent, ram and head weigh 750 lbs. The cable hoist was creaking a bit...
[00:51:30] <GNieport> The guy I got it from had a scale on his jib crane
[01:02:55] <BigJohnT> hmm, I don't see a mushroom cloud to the northeast so I'm going upstairs and make a pizza
[01:19:39] <Gamma-X> omg
[01:20:09] <Gamma-X> OH MY GOD!
[01:20:27] <Gamma-X> that was crazy
[01:20:32] <Gamma-X> it liked worked but it didnt work.
[01:20:38] <jmkasunich> all body parts intact?
[01:20:57] <jmkasunich> machine intact?
[01:21:25] <Gamma-X> uhh
[01:21:28] <Gamma-X> i think so lol
[01:21:43] <Gamma-X> i got the machine to disable estop... so thats good.
[01:21:52] <Gamma-X> scales worked
[01:21:56] <Gamma-X> amps turned on.
[01:22:26] <Gamma-X> but the axis were kinda like moving themselves, veery slowly
[01:22:40] <Gamma-X> not noticable without scales
[01:22:54] <jmkasunich> this is with amps turned on but PID turned off?
[01:23:09] <Gamma-X> dont realy know what pid is haha
[01:23:15] <skunkworks> ouch
[01:23:25] <jmkasunich> then you shouldn't be working on that project
[01:23:35] <Gamma-X> hahaha
[01:24:17] <Gamma-X> couldnt realy get teh scales to move manually either.... like from the console they just slowly crept.
[01:24:37] <Gamma-X> servos *
[01:26:23] <Gamma-X> jmkasunich what do i do!! haha
[01:29:32] <Gamma-X> im kinda lost at the moment.
[01:29:32] <GNieport> The axis was probably creeping along do to incidental noise on the amp analog input
[01:29:42] <Gamma-X> GNieport I agree.
[01:29:43] <jmkasunich> or offset voltage
[01:29:44] <Jymm> GNieport: Now you know why I said "poor pallet jack", the'yre only rated to something like 600 lbs or so
[01:29:53] <jmkasunich> Jymm: not so
[01:30:02] <GNieport> Mine is *5500* lbs :)
[01:30:07] <Gamma-X> damn, my pallet jack was rated to 4000 pounds
[01:30:12] <jmkasunich> most of the pallet jacks I've seen/used were between 4000 and 5000
[01:30:32] <jmkasunich> Jymm uses weeny pallet jacks
[01:30:34] <Jymm> I've seen some rated at 1200
[01:30:57] <Jymm> and POS ones at 600
[01:31:54] <GNieport> * GNieport hugs 3.5 ton pallet jack of his dreams.
[01:32:03] <Jymm> some forklifts can't lift 5000 lbs
[01:32:11] <GNieport> agreed.
[01:32:17] <Jymm> w/o tipping over at least
[01:33:11] <Gamma-X> anyone got any ideas on how to get my servos to actualy move when i tell them too instead of creeping cause of a fuzzy signal?
[01:33:30] <Jymm> GNieport: So, you gonna have that all fired up next week ;)
[01:33:31] <GNieport> PID
[01:33:49] <Gamma-X> GNieport p/m
[01:33:55] <GNieport> Jymm: Yes, by this Sunday. My new 7i33 arrives tomorrow.
[01:34:04] <Jymm> GNieport: cool
[01:34:36] <GNieport> I got the machine 6 years ago. it's about time.
[01:34:43] <Jymm> GNieport: oh, LOL
[01:37:30] <Gamma-X> GNieport hey check ur pm please
[01:38:11] <GNieport> G-X it's broken
[01:38:35] <Gamma-X> ahhh
[01:38:47] <Gamma-X> ok well... anyway u can talk me threw the pid?
[01:39:01] <GNieport> aol Geek1229 yahoo geek1229
[01:39:46] <Gamma-X> got it
[01:39:48] <GNieport> I recommend reading the example tuning in the docs
[01:39:53] <Gamma-X> ok
[01:40:03] <GNieport> It's not going to be a two-minute thing :)
[01:41:21] <fenn> Gamma-X: you can't learn how PID works over IRC unfortunately
[01:41:35] <Gamma-X> : (
[01:42:12] <fenn> take one of the motors off the machine and play with the settings
[01:42:26] <Gamma-X> no way...
[01:42:28] <Gamma-X> lol
[01:42:32] <Gamma-X> there on there!
[01:42:33] <fenn> try each of P I and D separately to get a feel for it
[01:42:40] <Gamma-X> ok
[01:42:45] <fenn> gimme a break, it's not like its a honda crank pulley bolt or anything
[01:42:56] <Gamma-X> hahahhaa
[01:43:04] <fenn> guess what i've been working on all week
[01:43:19] <GNieport> "Rock of Gibraltar" Honda bolt?
[01:43:30] <Jymm> fenn: At least it's not the balancer on a 302
[01:44:12] <Jymm> I swear I was cranking the engine by hand, HAD to get a compressor and impact wrench to get that one damn nut off
[01:44:29] <fenn> oo well the impact wrench slunk off with its tail between its legs
[01:44:47] <Jymm> fenn: Sears Rental Dept
[01:45:04] <fenn> the pulley wont turn because i have 2" angle keeping it from moving.. i'm jumping up and down on a breaker bar and watching the frame flex
[01:45:31] <GNieport> I had a monster bolt at work the other day that wouldn't budge with the 3/4" impact... I forgot it was left-hand thread...
[01:45:33] <fenn> i've got a silly 1" drive impact wrench from HF but no adapter
[01:45:40] <fenn> so i'm making one..
[01:45:55] <Jymm> GNieport: DOH!
[01:46:21] <Jymm> fenn: mill, grinder, or welder?
[01:46:44] <fenn> milling attachment on lathe
[01:46:54] <fenn> i thought about welding one out of 5 steel plates
[01:46:59] <fenn> in order to make the square hole..
[01:47:22] <Jymm> fenn: Well, if you have a torch you could just heat the bolt some
[01:47:33] <Jymm> err nut I mean
[01:47:43] <fenn> i have a little more confidence in the 1200 ft*lb wrench
[01:47:58] <fenn> but if it doesn' work i can do both
[01:48:03] <Jymm> lol
[01:48:24] <fenn> (unless i shear off the 1/2" drive adapter)
[01:48:31] <GNieport> likely.
[01:48:41] <fenn> what's likely?
[01:48:47] <GNieport> shear
[01:48:56] <fenn> but.. but.. it's scrapyardonium alloy
[01:49:05] <GNieport> lol, nevermind then
[01:51:12] <Jymm> Bah... I have a 20A toggle C/B, and a 15A push-to-reset panel mount C/B... I hate buying more crap when I have half of it already =(
[01:53:58] <SkinnypuppY34> What brand mill is that squashin the pallet jack?
[01:58:23] <Jymm> SquishyMill
[01:58:38] <SkinnypuppY34> I hadn't scrolled up far enough , Shizuoka AN-S
[01:59:15] <skunkworks> put a rotory table between the head and the ram - one on the table - boom - 5 axis ;)
[01:59:17] <GNieport> It is a lot more naked than the Google images
[02:00:24] <GNieport> I'd say it's solid enough to handle the additional load :)
[02:00:58] <SkinnypuppY34> It looks SOLID
[02:02:03] <GNieport> it's okay, thanks 8-)
[02:02:45] <Jymm> To thy mill be true and square damnit!
[02:03:37] <GNieport> I used the mill to bore its own X screw for a cross pin
[02:03:53] <GNieport> had to fab a motor mound and coupling
[02:04:01] <GNieport> mount, oops :)
[02:07:01] <GNieport> http://home.cinci.rr.com/bog/stuff/100_1943.JPG
[02:07:06] <GNieport> the mount
[02:07:27] <GNieport> Mild steel, weldment
[02:07:34] <Jymm> GNieport: nice workbench ya got there =)
[02:07:43] <GNieport> Brought the faces into parallel on the surface grinder
[02:08:14] <GNieport> I rpotected it with a highly durable coat of spray lithium
[02:08:31] <Gamma-X> what could cause my servos not to move at all? PID is obviously out of "sync" but I loaded the chips example file and it would not move at all....
[02:09:54] <GNieport> Jymm: one more pic, better view of the shaft coupling and mount that I fabbed http://home.cinci.rr.com/bog/stuff/100_1907.JPG
[02:10:35] <fenn> GNieport: if you dont like grease, i find furniture paste wax works great for rust protection on precision surfaces
[02:11:19] <GNieport> G-X check out your DAC values as the part file is executed
[02:11:20] <Jymm> GNieport: Nice, is that bracket hollow (so you can see the dial if you look down)?
[02:11:24] <GNieport> fenn: okay, thanks
[02:11:56] <GNieport> Jymm: yes, two 3/4" by 3" bar stock form the legs
[02:11:58] <Gamma-X> GNieport got it.
[02:12:30] <Jymm> GNieport: Cool, you could toss a nice chunk of acrylic between them too.
[02:12:44] <GNieport> Yes, notice the tapped holes
[02:13:16] <Jymm> GNieport: I do now =) My crt is kinda dark
[02:13:40] <GNieport> I ran a strain analysis on the bracket and finally decided to settle for an 0.0002 mm deflection under full load :)
[02:14:14] <GNieport> Although I'm sure my fillet welds are far from textbook ideal
[02:19:43] <GNieport> why is it preferred to define an axis with zero at one extreme, instead of splitting the travel and defining zero as the center of travel?
[02:29:58] <toastydeath> because it becomes difficult to run multiple parts
[02:30:21] <GNieport> oh, the fixture offsets get messy?
[02:30:38] <toastydeath> if i define an edge or some fixuring thing as an offset, i can run 80 parts and not have to find the origin for each
[02:30:58] <toastydeath> if i define the center of the part, i have to touch off four surfaces
[02:31:00] <toastydeath> to find the center
[02:31:23] <GNieport> well, I did not mean work offsets, I meant defining the machine axis
[02:31:37] <toastydeath> then yes, the offsets do get messy
[02:31:42] <GNieport> but I can see the implications now
[02:31:50] <toastydeath> if you are talking about true machine 0
[02:31:57] <GNieport> yes,
[02:32:01] <toastydeath> because you can suddenly misplace a - and screw everything up
[02:32:36] <toastydeath> that's why machines use the right rear corner as 0,0
[02:32:40] <GNieport> that happens? :-O
[02:32:41] <toastydeath> because that makes everything negative
[02:32:47] <GNieport> ah
[02:32:57] <fenn> that's stupid
[02:33:09] <fenn> why not just make everything positive?
[02:33:16] <fenn> then negatives are out of bounds
[02:33:16] <GNieport> left front corner?
[02:33:31] <toastydeath> because left rear is industry standard
[02:33:53] <GNieport> you said right rear a moment ago
[02:33:58] <toastydeath> er sorry
[02:34:01] <toastydeath> i MEAN right rear
[02:34:16] <fenn> so you have to put a - in front of every number?
[02:34:21] <toastydeath> no
[02:34:28] <toastydeath> it's just assumed for the purposes of work offset
[02:34:40] <toastydeath> on many controls, anyway
[02:34:50] <toastydeath> it's painfully obvious which one wants - and which doesn't care at all
[02:34:58] <GNieport> the the machine axis is always in a negative x,y
[02:35:02] <toastydeath> yep
[02:35:08] <GNieport> but part offsets will be positive in many cases
[02:35:16] <toastydeath> i believe in the ASME machine tool standards
[02:35:34] <toastydeath> there is a place that describes where to locate the machine origin, and it works out to being the rear right corner
[02:35:34] <GNieport> hrm
[02:35:43] <fenn> so.. if you forget a - sign you're screwed?
[02:35:52] <toastydeath> no, it throws an overtravel and nothing goes on
[02:36:01] <GNieport> well, I would read the ASME documents, but since they charge money...
[02:36:02] <fenn> but if you're using a work offset
[02:36:25] <toastydeath> if you enter a positiven number into a machine that wants negative work offsets
[02:36:35] <toastydeath> the machine throws a overtravel as soon as you hit go
[02:37:00] <fenn> that's dumb
[02:37:48] <fenn> at least they all use right hand coordinate systems
[02:38:01] <fenn> (afaik)
[02:38:59] <toastydeath> what's a right hand coordinate system?
[02:39:31] <GNieport> so, standing in front of the vertical mill, looking down onto the table from above, 0,0 would be with the table all the way right, and toward the column
[02:39:38] <toastydeath> yep
[02:39:55] <SWPadnos> it you home to the center, then the maximum error from thermal expansion is halved
[02:39:56] <GNieport> thanks
[02:40:38] <toastydeath> swpadnos: why do you say that
[02:40:40] <cradek> 0,0 is at the left front of the table
[02:40:41] <toastydeath> *say
[02:40:52] <cradek> it's true that means the table is to the right and back
[02:41:17] <cradek> but the working coordinates are positive, not negative
[02:41:18] <SWPadnos> because you reference at the midpoint of the thermal error, so you can only get an excursion to error/2 in each direction :)
[02:41:40] <GNieport> What does Fanuc do? lol
[02:41:50] <toastydeath> goddamnit i am screwing you up
[02:41:51] <SWPadnos> whereas if you reference to the left of the table, then move to the center, you have already moved though half the error range
[02:41:57] <toastydeath> gnieport: listen to cradec
[02:42:03] <toastydeath> fanuc does it how he just said
[02:42:03] <SWPadnos> this will be unimportant if you use work coordinate offsets though :)
[02:42:06] <GNieport> No, I follow you both.
[02:42:06] <toastydeath> i was thinking of where the table is
[02:42:19] <toastydeath> not where the spindle was OVER the table
[02:42:42] <GNieport> yes, I followed you both because 0,0 is defined by the cutter location.
[02:42:46] <SkinnypuppY34> Say, why is the display in EMC always at x.005 and y-.009 when I start up ?
[02:43:06] <SkinnypuppY34> s not really a problem as I g1x1y1,look at the DRO on the mill, and enter its coordinates into touchoff....
[02:43:15] <toastydeath> swpadnos: also the largest error comes from the head of the machine
[02:43:18] <toastydeath> in terms of thermal error
[02:43:23] <toastydeath> not the table.
[02:43:28] <cradek> SkinnypuppY34: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems#So_if_you_re_lost_what_should_you_do
[02:43:29] <fenn> SkinnypuppY34: probably your g54 offset
[02:43:56] <cradek> it could be one of several offsets
[02:44:07] <fenn> but emc starts with g54 active
[02:44:09] <cradek> that page tells you about them all (and how to clear them)
[02:44:20] <cradek> right
[02:44:55] <SkinnypuppY34> Yeah its in g54, I use 55 for the left corner of vice and 56 for the other 57 for a smaller tilt vice , I'll check that wiki page
[02:47:30] <fenn> tiny buglet in AXIS: the bounding box dimensions dont change until the line after you change coordinate systems
[02:47:48] <fenn> or is that the intended behavior?
[02:48:39] <fenn> its strange because the origin moves but the dimensions dont change
[02:49:35] <SWPadnos> hmmm. check the manual - there's a specified order of operations for lines with codes from multiple groups
[02:50:58] <GNieport> Teh manual also says "don't do that"
[02:51:13] <fenn> uh, don't do "g54"?
[02:51:34] <GNieport> no, don't do multiple commands, such that one would need to know the order or op
[02:51:42] <GNieport> order of op
[02:51:53] <fenn> i'm just doing mdi, g54; g55; g54
[02:52:23] <fenn> or even better: g55; g54; g54
[02:52:38] <cradek> I don't follow what you're saying is wrong - what should I do
[02:52:52] <SWPadnos> cradek, take a break :)
[02:53:13] <cradek> oh? hmm I'm not sure how to do that
[02:53:17] <fenn> heh
[02:53:21] <SkinnypuppY34> ha
[02:53:58] <SWPadnos> geez. do I have to do everything? now I'll take a break
[02:54:00] <SWPadnos> ;)
[02:54:36] <cradek> oh, bizarre, I see what you mean
[02:55:20] <cradek> I had never noticed because I only did that in a program
[02:55:32] <fenn> woah its a display thing
[02:55:41] <Gamma-X> SWPadnos before i go to bed i wanted to know if u ever seen a machine that the scales work but the servos dont move at all? GNieport said to get the hal cmd text file wich i will tomorow, but any thoughts for my dreams?
[02:55:46] <fenn> if it's exposed by moving a window over it, the numbers change
[02:55:51] <cradek> oh ok
[02:56:07] <fenn> i was trying to take a snapshot and was like "wtf is going on"
[02:56:19] <SWPadnos> Gamma-X, welllll - should the servos be moving?
[02:56:31] <Gamma-X> yes
[02:56:35] <SWPadnos> if they become very "stiff" when you power up the amps, that's a good thing
[02:56:42] <cradek> surprising that it redraws to move the axes but the numbers aren't corrected yet
[02:56:49] <SWPadnos> if they then don't move no matter what you do, that's not so good
[02:56:57] <Gamma-X> i put in the tux example file and hit run... only creeps from what i think is analog servo feedback.
[02:57:08] <SWPadnos> but the encoders are only used for EMC to know the position of the table, the drives don't use them at all (I think anyway)
[02:57:44] <SWPadnos> there may be some limit set somewhere that prevenrts the Mesa from outputting a reasonable motor velocity commadn
[02:57:56] <Gamma-X> hmmm
[02:58:00] <cradek> you should not ever have amps enabled when emc is not keeping position with pid - creeping when you stop is very bad
[02:58:18] <fenn> if there's no PID then emc isn't controlling the amps
[02:58:21] <cradek> right
[02:58:30] <cradek> if that's the case they must not be enabled
[02:58:37] <Gamma-X> so that could be the reason why i cant control them?
[02:58:44] <cradek> because they'll always drive the motors one way or the other
[03:00:39] <GNieport> EMC is ignoring the following error when Gamma-X runs a part. Where is a likely place to look for an incorrect HAL setting?
[03:01:28] <cradek> I don't understand what that means
[03:02:45] <GNieport> Gamma powers up machine. Loads 3-D tux image to cut. Depresses cycle start. The machine does not move (much). EMC does not fault for following error and contimues to pretend it is cutting out Tux
[03:03:45] <GNieport> The motion controller is not caring about the real world feedback form Gamma's scales.
[03:04:14] <SWPadnos> well, if FERROR is set to 20, that would do it
[03:04:20] <cradek> probably shouldn't be running a program yet if such basic things are not working
[03:04:30] <GNieport> Nope, default 0.001
[03:04:39] <GNieport> I agree :)
[03:04:47] <SWPadnos> set to 1 for X and Y in his ini (unless that was changed this afternoon)
[03:04:49] <cradek> he can troubleshoot with halmeter etc
[03:06:21] <SWPadnos> http://pastebin.ca/871417 <-- gamma X's ini from earlier today
[03:06:29] <Gamma-X> yes
[03:06:32] <SWPadnos> http://pastebin.ca/871416 <-- one HAL file
[03:06:40] <SWPadnos> http://pastebin.ca/871414 <-- another HAL file
[03:06:56] <GNieport> I would like a show all
[03:07:24] <Gamma-X> >
[03:07:26] <Gamma-X> ?
[03:07:41] <GNieport> n/m you were going to get that tomorrow...
[03:08:07] <Gamma-X> yes
[03:08:32] <GNieport> good night all
[03:08:32] <Gamma-X> could my dac be messed up? weong setting?
[03:08:38] <Gamma-X> gnight GNieport thanks!
[03:08:55] <tomp2> alex_joni: Beckhoff sells a control system (aka twincat ) and sells parts for it ( monitors, operator panels, handrads... )
[03:08:59] <SWPadnos> well, if you just unhook the motor from the table and push it, EMC should give you a following error
[03:09:12] <GNieport> #
[03:09:12] <GNieport> linksp Xenable => m5i20.0.dac-00-enable
[03:09:12] <GNieport> #
[03:09:12] <GNieport> #linksp Yenable => m5i20.0.dac-01-enable
[03:09:12] <GNieport> #
[03:09:12] <GNieport> #linksp Zenable => m5i20.0.dac-02-enable
[03:09:14] <GNieport> #
[03:09:17] <SWPadnos> if that doesn't happen, then EMC isn't seeing the encoders correctly, and you need to fix that before continuing
[03:09:18] <GNieport> there is one problem
[03:09:43] <Gamma-X> ?
[03:09:55] <GNieport> Xenable must turn on all DAC
[03:10:03] <eric_U> my cards weren't set up properly from mesa, no power
[03:10:21] <cradek> do you not have all your limit switches hooked up?
[03:10:25] <eric_U> that would also do it
[03:10:26] <Gamma-X> i do
[03:10:31] <GNieport> eric_U No power? please explain...
[03:10:39] <eric_U> there are jumpers
[03:10:56] <GNieport> eric_U the 5V/3.3V jumper?
[03:11:02] <eric_U> ya
[03:11:12] <GNieport> eric_U okay
[03:11:17] <Gamma-X> where it says 3.3v and an led the led is lit up if that helps?
[03:11:24] <Gamma-X> right by the jumper
[03:11:45] <Gamma-X> bottom left corner of card by the 50 pin ribbon cable.
[03:12:47] <SWPadnos> I still think the butterfly for $21.28 is a good deal. there's already a phototransistor on it, and all the I/O is available if you want it
[03:13:17] <SWPadnos> err - ops
[03:13:23] <eric_U> what you talking about bevis?
[03:13:26] <SWPadnos> no, oops!
[03:13:30] <SWPadnos> wrong channel
[03:13:34] <SWPadnos> AVR butterfly board
[03:13:35] <eric_U> avr
[03:13:41] <eric_U> can you still buy one?
[03:13:48] <SWPadnos> sure, at DigiKey
[03:13:51] <SWPadnos> 363 in stock
[03:14:05] <eric_U> I have the isp
[03:14:19] <SWPadnos> ATAVRISP2 - the USB one?
[03:14:35] <eric_U> never used it :)
[03:14:44] <eric_U> parport I think
[03:15:29] <eric_U> that looks like a good deal though
[03:15:31] <SWPadnos> heh
[03:15:34] <Gamma-X> eric_U was that led lit up when u tried powering it on, or was any led powering on for that fact?
[03:15:38] <SWPadnos> I hate those parport programmers
[03:15:54] <SWPadnos> switched to the AVRISP (serial) and never looked back
[03:15:59] <eric_U> Gamma-X: I can't see it
[03:16:23] <eric_U> I'll go look
[03:16:33] <Gamma-X> well im sayin from when u didnt have power.
[03:16:53] <eric_U> I don't know, I tend to take things apart to troubleshoot -- mechanical engineer
[03:17:15] <Gamma-X> ahhh
[03:18:43] <fenn> is there any other way?
[03:18:52] <eric_U> use a hammer
[03:19:00] <fenn> hire a troubleshooter
[03:19:36] <eric_U> usually, I'm fairly confident that if I can't fix it, it can't be fixed
[03:19:58] <fenn> well, anything that can be made can be "fixed"
[03:20:15] <Gamma-X> i think i found my problem
[03:20:27] <Gamma-X> wich both SWPadnos and GNieport said
[03:20:29] <eric_U> curses
[03:20:46] <SWPadnos> heh
[03:20:50] <Gamma-X> gotta have x enable , enable all the dac's
[03:21:00] <SWPadnos> does X move?
[03:21:04] <SWPadnos> but not Y and Z?
[03:21:09] <Gamma-X> haha... x amp is in my hand! hahaha
[03:21:13] <Gamma-X> what a bitch...
[03:21:18] <SWPadnos> well, that isn't helpful ;)
[03:21:29] <SWPadnos> can I smack you?
[03:21:38] <SWPadnos> (because I told you about that this morning)
[03:21:47] <Gamma-X> if u come down here and help me u can do what ever u want to me
[03:21:56] <SWPadnos> um - no, I think I'll stay up here, thanks
[03:21:59] <Gamma-X> well... that sounded pretty bad
[03:22:03] <SWPadnos> yep
[03:22:04] <Gamma-X> hahaha
[03:22:08] <Gamma-X> never mind! haha
[03:23:09] <Gamma-X> SWPadnos i did ask why u wanted me to do that, and wasnt thinkin.
[03:23:11] <Gamma-X> thanks.
[03:23:17] <SWPadnos> no problem
[03:24:23] <SWPadnos> I may have left before answering that part ;)
[03:24:23] <eric_U> why does x enable have to enable all dacs?
[03:24:23] <Gamma-X> cmon! make it a " business trip
[03:24:23] <Gamma-X> "
[03:24:23] <Gamma-X> and u can claim it on ur taxes
[03:24:55] <SWPadnos> well, if you pay me I can do that
[03:25:28] <Gamma-X> how much u want?
[03:25:36] <SWPadnos> I charge $75/hour (plus expenses)
[03:25:39] <SWPadnos> :)
[03:25:44] <Gamma-X> eric_U cause i dont have a y or z enable only 1 enable
[03:25:46] <eric_U> cheap
[03:25:54] <SWPadnos> yeah. I've been thinking of raising my rates
[03:26:09] <SWPadnos> been the same for 7 years (plus a few more at the previous company)
[03:26:24] <SWPadnos> I should probably be in the $125/hour range at this point
[03:26:32] <fenn> too much business: raise your rates; too little business - lower your rates
[03:26:41] <eric_U> I think people don't take you seriously unless you charge that much
[03:26:46] <Gamma-X> SWPadnos i think thats out of my price range haha. thanks though! if ur in the area though..... remember I cook and clean!
[03:26:47] <SWPadnos> damfino amount of business, leave well enough alone :)
[03:26:54] <SWPadnos> excellent!
[03:27:05] <SWPadnos> maybe the next time I visit my father or Nikon, I'll stop in
[03:27:26] <fenn> "will whine for food"
[03:27:34] <Gamma-X> id appreciate that a lot. i got a nice comfey couch, or ill put u up in a hotel for a few nights w/e u preffer
[03:28:08] <SWPadnos> couch! haven't been on one of those in a long time ;)
[03:28:26] <Gamma-X> if u aint married ill buy u a hooker? lol
[03:28:39] <SWPadnos> oh, too bad
[03:28:43] <Gamma-X> hahaha
[03:28:57] <Gamma-X> well i mean if u "tell" me that ur not married
[03:29:17] <SWPadnos> sure, on a google-indexed IRC channel - riiiiiight
[03:29:23] <Gamma-X> hahahahahah
[03:29:53] <Gamma-X> i wonder how much i could sell my machine for with a working emc control...
[03:30:05] <SWPadnos> probably more than you paid for it
[03:30:29] <SWPadnos> but it depends on the mechanical accuracy (ballscrew condition, etc)
[03:30:34] <Gamma-X> hmmm im thinkin 5 grand.
[03:30:39] <SWPadnos> keep thinking
[03:30:48] <Gamma-X> SWPadnos as far as I know its verry precise.
[03:30:57] <Gamma-X> SWPadnos 3500?
[03:31:00] <SWPadnos> hard to tell when you can't make it move :)
[03:31:05] <Gamma-X> heh.
[03:31:12] <Gamma-X> sure go ahead,. Throw it in my face haha
[03:31:13] <SWPadnos> I don't know, but I suspect that $5k is a bit much
[03:31:16] <eric_U> that's a good thing to say in the ad: "as far as I know..."
[03:31:35] <Gamma-X> SWPadnos u know what I have right?
[03:31:39] <SWPadnos> look on ebay to see what machines that work are going for
[03:31:50] <SWPadnos> I think so - a Bridgeport Boss with Anilam controls that don't work
[03:31:55] <Gamma-X> 10k with the original control
[03:32:03] <Gamma-X> bridgeport boss? lol no
[03:32:08] <SWPadnos> sure, but the controls are $5-10k of that ;)
[03:32:16] <SWPadnos> oh right, a supermax
[03:32:19] <Gamma-X> correct
[03:32:22] <Gamma-X> ycm-18
[03:32:29] <Gamma-X> all the fancy stuff!
[03:32:44] <eric_U> what did you buy it for?
[03:32:55] <Gamma-X> 2 grand.
[03:32:59] <Gamma-X> working.
[03:33:11] <eric_U> that's about what it is worth :)
[03:33:27] <Gamma-X> and like 20 tool holders, collets, all that jazz
[03:33:30] <Gamma-X> eric_U u think so?
[03:33:48] <eric_U> part it out, it's worth 5k
[03:33:54] <Gamma-X> yeah right! lol
[03:34:00] <Gamma-X> im gunna part out the older anilam control
[03:35:02] <eric_U> SWP: what's the butterfly part number?
[03:35:16] <SWPadnos> err
[03:35:29] <eric_U> nvm I see it
[03:35:30] <SWPadnos> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=ATAVRBFLY-ND
[03:36:45] <eric_U> thsx
[03:37:25] <fenn> yay pdf catalog pages!
[03:37:35] <SWPadnos> yeah - they have both now :)
[03:37:57] <SWPadnos> pdf link and useless crappy slow catalog-app link - which do you choose?
[03:38:19] <eric_U> they still have an incredibly slow web site
[03:38:34] <SWPadnos> it seems pretty fast here
[03:38:59] <eric_U> I guess it's just the catalog pages
[03:39:01] <Gamma-X> whats the butterfly for exactly?
[03:39:13] <eric_U> smackin' you upside the head
[03:39:18] <SWPadnos> it was a trade show demo board when they gave me one
[03:39:19] <Gamma-X> dont hate
[03:39:26] <Gamma-X> ok
[03:39:30] <SWPadnos> btu it's kind of a cool tinkering board, so they sell it now
[03:39:44] <Gamma-X> eric_U is buyin me some beeeeersss
[03:39:54] <eric_U> maybe
[03:40:04] <eric_U> I don't see chips yet
[03:40:42] <eric_U> I'm sitting here thinking, if I help Gamma-X, I'm buying beer
[03:40:45] <eric_U> hmmmm
[03:40:52] <Gamma-X> lol
[03:41:04] <Gamma-X> and u gotta drive out here
[03:41:48] <fenn> Gamma-X: does your X axis move now?
[03:42:07] <Gamma-X> i aint goin outside and fuckin with it.... i got work tomorow. ill do it when i get home tomorow
[03:42:23] <eric_U> damn, all-nighter
[03:43:16] <Gamma-X> tomorow will be an all nighter saturday and all dayer,. going to hook up my vfd and 220 line. gotta put in new electrical panel
[03:43:54] <Gamma-X> and get my neighbor to cut out the wholes on my computer for the connectors.
[03:44:52] <fenn> Gamma-X: http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=46061
[03:45:18] <eric_U> that's cool, how long does it last?
[03:45:29] <fenn> dunno
[03:45:41] <fenn> it didn't self destruct the first time i used it
[03:46:04] <eric_U> that would be worth it, considering a real one costs $250
[03:46:22] <fenn> its much better than a hand powered nibbler
[03:46:36] <fenn> actually cuts the metal instead of licking at it
[03:47:14] <Gamma-X> wow nice
[03:47:43] <Gamma-X> im just gunna use some whole saws...
[03:47:50] <Gamma-X> only need 2 holes
[03:48:30] <fenn> ah, well if it's a round connector that works
[03:48:44] <Gamma-X> yup
[03:48:47] <eric_U> closest HF to me is right by three mile island
[03:48:53] <eric_U> probably glows in the dark
[03:49:23] <Gamma-X> 4 round connectors, its gunna look good. all military style connectors and such
[03:51:20] <Gamma-X> http://cgi.ebay.com/HARBOR-FREIGHT-ELECTRIC-CUTOUT-TOOL_W0QQitemZ130193053906QQihZ003QQcategoryZ11704QQcmdZViewItem
[03:51:41] <Gamma-X> haha kinda defeets the purpose of owning a cnc machine but w/e
[03:51:52] <cradek> fenn: it took me all this time to find the one-line fix for that display problem
[03:52:17] <fenn> * fenn looks
[03:52:58] <Gamma-X> gnight all!
[03:53:15] <Gamma-X> thanks again. I can honestly say i look forward to comin to the chat room when i get home from work.
[03:53:44] <fenn> that cutout thingy would not be fun to use on sheet metal
[03:54:13] <fenn> or any type of metal
[03:54:20] <tomp2> a step drill is very handy for panel work , many hardware stores carry Irwin Unibit Step drills
[03:55:12] <fenn> yes step drills are good for starting holes to get a cutout punch in there
[03:55:32] <tomp2> greenlee
[03:55:42] <fenn> yeah
[03:55:43] <cradek> I hate drilling sheet metal
[03:55:56] <cradek> it always wants to turn into a whirling blade of death
[03:56:32] <fenn> cool greenlee sells a death-ray gun
[03:57:52] <tomp2> stepdrill = cheap multi purpose, greenlee = expensive and single purpose... me, I use step drill, file & 3 corner knife
[03:58:56] <fenn> cradek: its even more fun when you use a whirling helicopter of death to cut the hole
[03:59:10] <cradek> oh yeah, good point
[03:59:11] <tomp2> trammeling tool :)
[03:59:15] <cradek> I know just what tool you're talking about
[03:59:24] <toastydeath> trepanning
[03:59:39] <fenn> adjustable hole saw
[03:59:43] <fenn> with one tooth
[03:59:54] <toastydeath> trepanning
[04:00:16] <toastydeath> http://www.jtsmach.com/jtswebshop/Pr_CuttingTools/D315.asp
[04:00:16] <cradek> Your search - "whirling helicopter of death" - did not match any documents.
[04:00:28] <cradek> we can't be the first ones to call it that!?
[04:01:06] <fenn> i think a relative of jmk called it that
[04:01:34] <tomp2> trepanning for cutting holes in skulls, trammel for cutting the circumference and center leaving big slug
[04:01:43] <fenn> the amazing drag cut principle - also known as negative rake
[04:01:56] <toastydeath> fenn: not negative rake, positive rake
[04:01:57] <jmkasunich> my stepson, several years ago in a highschool shop class
[04:02:33] <jmkasunich> toastydeath: that trepanning tool uses negative rake, to avoid the otherwise inevitable chatter
[04:02:51] <toastydeath> jmkasunich: the only rake angle that would be negative is radially
[04:02:57] <toastydeath> and there is no cutting tooth radially
[04:03:03] <toastydeath> so there is no negative rake
[04:03:07] <jmkasunich> hmm, might be positive rake, but with the cutting edge well behind center
[04:03:08] <toastydeath> and the axial is a high positive
[04:03:43] <toastydeath> that's all they mean, what you said.
[04:03:47] <fenn> behind center?
[04:04:05] <toastydeath> yeah
[04:04:49] <toastydeath> the axis of the tooth face does not intersect the center point of the cutting tool itself
[04:05:41] <toastydeath> so if the thing binds, it doesn't jam into the workpiece and kill somebody, it just flexes back and shatters
[04:05:47] <fenn> hm. the tooth face is a plane, and the center axis of the tool is a line
[04:05:56] <toastydeath> we're talking 2d
[04:06:32] <toastydeath> if you were to look at the tool over the X-Y plane
[04:06:53] <toastydeath> the tool is angled back
[04:06:56] <SWPadnos> wow. it would be really cool if I could follow what you guys are talking about
[04:07:11] <fenn> yes i see the angle
[04:07:17] <toastydeath> that's all it's referring to.
[04:07:59] <fenn> swp it's like putting a cutter below center on a lathe
[04:08:09] <fenn> so when it flexes it doesn't dig in
[04:09:01] <tomp2> http://www.holesaws.com.cn/products/holesaws/adjustable/adjustable.htm
[04:09:04] <fenn> but that still doesn't solve the "hooking" problem in sheet metal
[04:09:11] <SWPadnos> ah, OK. not scything into the work
[04:09:14] <fenn> whereas a negative rake cutter would
[04:09:20] <toastydeath> what hooking problem?
[04:09:49] <fenn> when it cuts all the way through and pulls up on the sheet, turning it into a whirling blade of death
[04:09:54] <toastydeath> oh, yes.
[04:10:05] <toastydeath> you guys don't use a strap clamp stud to hold it still?
[04:10:10] <toastydeath> so that it doesn't rotate?
[04:10:27] <fenn> yeah of course i do
[04:10:29] <cradek> whirling blades of death don't care about your attempts to clamp them
[04:10:40] <cradek> they laugh in your general direction
[04:10:45] <toastydeath> hahah
[04:10:45] <tomp2> cut into sacrificial plate, allow no gap between, and clamp it down
[04:10:59] <toastydeath> build a punch press!
[04:11:03] <jmkasunich> if you clamp it well enough that you'd be willing to mill it, it will probably hold with a drill bit going thru
[04:11:05] <fenn> use an air nibbler!
[04:11:09] <cradek> sorry it must be getting late
[04:11:17] <jmkasunich> of course sheet is so fragile you can't clamp it that well
[04:11:17] <toastydeath> cnc thermite machine
[04:11:17] <cradek> the air nibblers work great
[04:11:44] <fenn> not as easy to get a perfect circle though
[04:11:48] <cradek> heh true
[04:11:50] <tomp2> draw the circle on the plate. & label 'this hole intentionally left full"
[04:11:53] <cradek> or a perfect anything
[04:11:56] <fenn> but it might end up looking better because it isn't mangled
[04:13:33] <toastydeath> how fast do you guys mill HDPE
[04:13:56] <toastydeath> if uh, anyone does
[04:14:51] <fenn> why do you ask?
[04:15:08] <tomp2> http://www.plasticsmachining.com/magazine/2001-09/PP-HDPE.html
[04:15:33] <toastydeath> a co-worker was doing it at what i thought was ridiculously slow
[04:16:08] <cradek> I think you use a low surface speed and fast feed so it doesn't melt
[04:16:16] <toastydeath> something like 1600 rpm on a 1" cutter and 10 ipm
[04:16:18] <toastydeath> 3 flute
[04:16:24] <toastydeath> and .5" doc
[04:16:45] <cradek> I'd feed much faster than that...
[04:17:17] <SWPadnos> more like 100 IPM
[04:17:18] <toastydeath> i think you could max that machine out in the material
[04:17:21] <fenn> .002"/tooth is probably too little
[04:17:22] <SWPadnos> that's 4800 teeth/minute
[04:17:23] <cradek> and that's 400sfm which seems way too high
[04:17:37] <toastydeath> 4000 rpm spindle and 100 ipm max
[04:17:43] <toastydeath> that's what i would do
[04:18:14] <toastydeath> and just see what happens, it's plastic
[04:18:20] <toastydeath> can't hurt things too much
[04:18:23] <cradek> I'd try 50-100sfm or less, and .020? /tooth
[04:18:29] <toastydeath> why so slow?
[04:19:11] <cradek> to avoid melting
[04:19:21] <fenn> with a properly shaped tool, most of the heat goes into the chip
[04:19:32] <SWPadnos> HDPE machines reasonably well - you don't get too much heating with it
[04:19:39] <SWPadnos> or am I thinking of Delrin (UHMW)?
[04:19:47] <fenn> delrin is very nice
[04:19:55] <fenn> hdpe is a little tougher
[04:20:02] <toastydeath> .020 a tooth?
[04:20:06] <toastydeath> is that mm or inches
[04:20:10] <SWPadnos> inches
[04:20:23] <toastydeath> that's a huge feed per tooth!
[04:20:27] <cradek> for 1" that's 400rpm 30-40ipm... I bet you could double it
[04:20:29] <toastydeath> i guess it is plastic though.
[04:20:42] <toastydeath> cradek: we got just peachy results at 1600 rpm
[04:20:47] <SWPadnos> it's plastic - you can take off .012 of aluminum with no trouble. so double that should be possible in plastic
[04:20:56] <SWPadnos> I'd think
[04:21:03] <toastydeath> my only concern is the chip evacuation
[04:21:12] <toastydeath> at a shallow cut i wouldn't have a problem going .020"
[04:21:20] <toastydeath> but at like, .5"?
[04:21:26] <cradek> slotting or side milling?
[04:21:32] <toastydeath> good point.
[04:21:38] <toastydeath> i was thinking slotting
[04:22:01] <cradek> you definitely don't want to make dust
[04:22:05] <cradek> (beware I'm no expert)
[04:22:24] <toastydeath> .020" isn't anywhere near dust
[04:22:31] <cradek> oh I know
[04:22:47] <SWPadnos> but 10IPM with 4800 teeth.min should be ;)
[04:23:01] <SWPadnos> that's around 0.002
[04:23:02] <cradek> yeah that's .002
[04:23:05] <toastydeath> it came out fine
[04:23:16] <toastydeath> it was just slow
[04:23:59] <toastydeath> i think there's some spare stuff
[04:24:13] <toastydeath> one day after work or something i'll have to cut a block off and turn it into chips
[04:24:29] <cradek> that's the way to find out!
[04:24:54] <SWPadnos> a shop-vac works wonders for chip evacuation, incidentally
[04:25:07] <toastydeath> you'd think we'd have one of those
[04:25:13] <cradek> I do that for wood too
[04:25:23] <toastydeath> i'm just going to edge mill it
[04:25:25] <toastydeath> no slotting
[04:25:26] <fenn> keeps you from getting a wood dust allergy
[04:28:10] <tomp2> if the chips melt into each other... bad for slotting, keep chips cool, slow sounds good, need lotsa chip clearance, like skip tooth saw, fewer teeth & larger diameter
[04:28:31] <fenn> lol: http://www.docsmachine.com/fabshop/anvilmill.jpg
[04:28:56] <fenn> pretty mill
[04:29:07] <cradek> fenn: they're often just steel on top
[04:29:14] <cradek> a plate welded on
[04:29:53] <fenn> he built up the dished top with a mig welder
[04:30:08] <fenn> i guess you'd have to use low carbon steel for that?
[04:30:29] <fenn> otherwise it would harden
[04:31:09] <fenn> carbide can cut through hardened steel though i suppose
[04:31:48] <fenn> oh, n/m he used a TIG and screen door springs as the filler material :0
[04:32:04] <cradek> screen door springs? haha
[05:00:15] <Jymm> http://www.syilamerica.com/product_U2.asp
[05:10:16] <fenn> $4699
[05:10:30] <fenn> looks fairly sturdy for a desktop machine though
[05:11:28] <fenn> mt3 spindle :P
[05:14:57] <fenn> i wonder why they EDM the motor/screw mounts
[05:16:50] <toastydeath> advertising
[05:19:06] <fenn> its expensive though
[05:19:40] <fenn> even if you own the machine
[05:25:46] <toastydeath> tru
[05:35:57] <tomp2> no idea why they'd be edm'd. maybe the mounts are a 'government job' by some edm operator?
[05:42:42] <toastydeath> maybe they only like, touch them with an electrode
[05:42:46] <toastydeath> after milling it out
[05:42:50] <toastydeath> and just say it was edm'ed
[06:17:02] <fenn> what do you call the thingy that cleans crud out of a taper?
[07:07:32] <tomp2> taper cleaner
[07:33:55] <Jymm> fenn: an air gun
[10:02:39] <fenn> 7.3 hours to machine an impact adapter with inadequate equipment.. not bad
[10:03:26] <archivist> 1 thou per cut?
[10:04:37] <fenn> not quite that bad
[10:05:06] <fenn> lathe milling attachment with a built-in vise
[10:05:24] <fenn> the vise kinda sucks and wasn't big enough
[10:06:37] <archivist> I have stretched the machine capabilities now and again, its fun
[10:11:16] <fenn> and here it is in all its glory: http://fennetic.net/pub/camera/DCP_0790.JPG
[10:12:31] <archivist> er wot that adapter? or is that the wrong pic
[10:15:16] <archivist> 2" down to 1/2" ?
[10:15:19] <fenn> its the cone shaped thingie
[10:15:23] <fenn> 1" to 1/2"
[10:16:01] <archivist> I cheatz and buys them
[10:17:22] <fenn> well i'm doing that also
[10:17:38] <fenn> but in the meantime i'm not doing anything
[13:00:45] <fretless85> hey everybody
[13:53:29] <maddash> i'm baaaaack
[14:10:45] <skunkworks_> <maddash> i'm baaaaack
[14:11:00] <skunkworks_> *** maddash has quit
[14:11:08] <archivist> hehe
[14:11:27] <fretless85> hihi
[14:21:17] <skunkworks_> Guest824: HI
[14:21:35] <Guest824> Howdy folks. I'm looking for some advice on how to proceed. I have a cnc router with no software and would like to eventually run it with emc.
[14:22:02] <skunkworks_> you where going to give us links to the hardware that is in the machine
[14:22:11] <skunkworks_> (if your the guest from yesterday)
[14:23:23] <Guest824> I am, and the one board I could track down is DC2-PC100 with the url being:
[14:23:59] <Guest824> sorry, hit rturn www.pmccorp.com/products/dc2pc.pcp
[14:24:40] <Guest824> arrg, http://www.pmccorp.com/products/dc2pc.php
[14:25:10] <skunkworks_> only 2 axis?
[14:25:39] <skunkworks_> ah - 2 axis stepper and 2 axis servo
[14:25:53] <Guest824> I have three axes, but the third is run through the mysterious other board.
[14:27:03] <Guest824> and by three axes I mean stepper motors for xyz
[14:29:18] <Guest824> skunkworks, am i correct in assuming that this board is not meant to control 3 stepper motors?
[14:29:21] <skunkworks_> I can tell you right now that the board is not supported by emc
[14:29:40] <skunkworks_> What I would do is now figure out what is actually driving the steppers.
[14:30:06] <skunkworks_> I would assume the dc2-pc100 is hooked to 3 drives that run the steppers
[14:30:26] <archivist> two stepper on that card plus another card
[14:31:48] <skunkworks_> yes
[14:31:57] <skunkworks_> what archivist said
[14:32:55] <Guest824> it seems that the xy steppers are directly connected to the dc2 card, while the z stepper goes to other card
[14:33:48] <Guest824> if this sounds hopelessly naive, you're not far from the truth, but what exactly do you mean by drivers?
[14:33:59] <archivist> I would get new drives and use emc
[14:34:18] <archivist> what s on the card
[14:34:33] <Guest824> I only have a vague sense that a driver is some device for controllinf the steppers
[14:34:40] <archivist> as the card is unknown by emc
[14:35:41] <jensor> hello - anyone familiar with interfaceing a wireless adapter with linux as installed by the emc installation disk. The wireless adapter is an zyxel G-220 v.2
[14:35:48] <archivist> what country are you in Guest824
[14:36:01] <Guest824> usa
[14:36:36] <archivist> gecko drivers seem to be recomended
[14:41:25] <ALS> Guest824: stepper motor type unipolor?
[14:42:07] <archivist> unipolar are often drivable as bipolar
[14:42:15] <ALS> yes
[14:42:27] <cradek> only if they have 6 or 8 wires
[14:43:36] <Guest824> Hmm, stepper motors are rapidsyn 23d-6108c, looked on product home page but was unable to tell uni-polar, bipolar.
[14:44:40] <jepler> jensor: the internet indicates that the driver for that card, known as zd1211rw, is only available in kernel 2.6.18 and newer. ubuntu dapper, which the emc2 live cd is based on, has kernel 2.6.15. I fear that you won't easily get this adapter working.
[14:44:44] <jepler> http://linuxwireless.org/en/users/Drivers/zd1211rw
[14:44:55] <cradek> I was just going to say that same thing.
[14:44:59] <jensor> My version of Linux is Dapper 6.06 LTS kernal 2.6.15-magma
[14:45:43] <jensor> jepler: Can I interface with an ethernet card?
[14:45:56] <archivist> yes
[14:46:02] <cradek> almost any ethernet card will just work if you plug it in
[14:46:54] <jensor> okay, I'll go with that rather than struggle with the wireless adapter route
[14:47:38] <jensor> Bye
[14:47:40] <cradek> if you need wireless you can get an ethernet<->wireless bridge for cheap
[14:49:05] <Guest824> bell again, thanks folks
[14:49:09] <Guest824> bye
[14:49:12] <archivist> cradek, jepler can you point me at the axis code that rotates the cone about A axis I want to read and hopefully understand
[14:49:24] <jensor> Cradek: I already have a netgear ethernet card. I'lluse the wireless adapter to interface my desktop with the routrer.
[14:49:49] <cradek> sounds perfect
[14:49:59] <jensor> thanks and bye
[14:51:35] <cradek> archivist: line 896 in trunk
[14:53:37] <skunkworks605> skunkworks605 is now known as skunkworks_
[14:54:59] <archivist> cradek, thanks will have a read, lots to learn
[15:16:04] <archivist_emc_> cradek first guess am i close ? http://pastebin.ca/872088
[15:17:15] <archivist> I must write something from scratch to get the feel of python and graphics
[15:18:18] <cradek> axis_mask is a bitmask of axes defined in your config
[15:18:46] <cradek> so, 1<<3 is the bit for A
[15:19:03] <cradek> position[3] is the A axis's value
[15:22:05] <archivist_emc_> hmm ok thanks (suppose its end of dinner time here and back to work)
[15:22:22] <archivist_emc_> I need more play time
[15:49:25] <alex_joni> bbl
[16:12:05] <skunkworks_> off topic - why do all these reviews add up to a C+? http://movies.yahoo.com/movie/1809833630/user
[16:13:29] <skunkworks_> B+,A+,C+,B,B-,A+,A+
[16:19:28] <SWPadnos> because three of the reviewers cheated, so their grade was divided by 3
[16:21:08] <archivist> adding monkey had a headache
[16:30:40] <fretless85> anyone got experience with emc2 and the uhu servo controllers? just curious...because im planing to use emc2 with these controllers
[16:31:35] <skunkworks_> uhu are step and direction input - correct? with internal tunable pid (seraily) or something like that..
[16:31:57] <skunkworks_> serially
[16:32:00] <skunkworks_> whateer
[16:32:05] <fretless85> right
[16:32:22] <seb_kuzminsky> what's the appeal of step/dir servo controllers?
[16:32:30] <skunkworks_> * skunkworks_ has no clue..
[16:32:30] <jepler> seb_kuzminsky: you can use them with mach
[16:32:35] <skunkworks_> heh
[16:32:37] <seb_kuzminsky> hehehe
[16:32:55] <seb_kuzminsky> to me part of the appeal of servos is you get to use the encoders as dros
[16:33:07] <skunkworks_> fretless85: they should work fine.
[16:33:08] <jepler> yeah that's real cool
[16:33:09] <seb_kuzminsky> but that doesnt work with step/dir control
[16:34:49] <fretless85> http://www.uhu-servo.de/servo_en/index.htm
[16:34:59] <seb_kuzminsky> what book should i read to learn about mechanical stuff like leadscrews and ways and bearings and stuff?
[16:35:09] <bill2or3> uhm
[16:35:12] <seb_kuzminsky> those things are not like bits & bytes, right? hehe
[16:35:13] <SWPadnos> machinery's handbook?
[16:35:52] <skunkworks_> interesting IRFP260N max.30A/150V
[16:36:30] <bill2or3> there's a pretty cool book called '507 Mechanical Movements', not really specific to metalworking or anything, but worth picking up.
[16:36:43] <seb_kuzminsky> hm, i thought M's H was more of a reference
[16:36:49] <SWPadnos> it is :)
[16:36:49] <fretless85> the technical data for the UHU's http://www.uhu-servo.de/servo_en/UHU_Servo_300_short_en.pdf
[16:37:01] <seb_kuzminsky> bill2or3: thanks, i'll check that out
[16:37:22] <archivist> seb_kuzminsky, some stuff on this page http://www.thewarfields.com/cnccookbook/MTLatheCNCAccuracy.htm
[16:37:43] <bill2or3> it's from the late 1800's or so, just pages of illustrations and old-timey descriptions
[16:38:24] <fretless85> good ive learned "a few" things in school about mechanical stuff ;)
[16:38:26] <bill2or3> there's also a US Navy manal called something like "Basic Machines and how they work" that I thought was ok.
[16:38:28] <seb_kuzminsky> archivist: that looks good, but a lot of the later sections say "Blah blah blah"
[16:38:56] <archivist> I know
[16:39:14] <archivist> 4/10 must try harder
[16:39:25] <fretless85> get some "old" books about mechanics...they are the best...just my opinion
[16:39:45] <archivist> I like the old books part
[16:39:50] <bill2or3> ahh, I re-found it on amazon, search for 'Basic Machines and How They Work'
[16:40:04] <jepler> fretless85: hm that datasheet says the avr is clocked at 24MHz, which is out of spec for the attiny2313 (spec says 0-20MHz at 5V)
[16:40:50] <seb_kuzminsky> i've seen references to a work called "Precision Machine Design" or something by Alexander Slocum, but haven't been able to find the book itself.
[16:41:43] <bill2or3> it should still run ok at 24mhz, that's not very far over.
[16:42:15] <bill2or3> those avr chips are pretty hearty, if it's warm stick a heatsink on it.
[16:42:51] <fretless85> okay
[16:43:30] <fretless85> ive scored them for 80€...so ill give it a try ;)
[16:45:48] <fretless85> 3 controllers btw
[16:46:33] <bill2or3> nice.
[16:47:19] <bill2or3> there are different versions of most avr's too, maybe you're looking at a datasheet for a Dip, and they used a SMD version, or something.
[16:48:22] <archivist> seb_kuzminsky, expensive http://www.amazon.com/Precision-Machine-Design-Alexander-Slocum/dp/0872634922
[16:48:52] <bill2or3> Ouch!
[16:50:10] <cradek> http://www.abebooks.com/servlet/SearchResults?an=slocum&sts=t&tn=precision+machine+design&x=0&y=0
[16:50:31] <cradek> here's a used one, or you could order it from england for only twice as much
[16:51:01] <cradek> oh, save $5 for used, sorry forget it
[16:51:32] <archivist> * archivist notes total rip off british price
[16:52:09] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[16:52:13] <archivist> us poor clock makers cant afford that
[16:52:37] <seb_kuzminsky> thx guys
[17:18:37] <Jymm> bill2or3: navy or army field manual?
[17:18:49] <bill2or3> I think it was navy.
[17:19:16] <bill2or3> but now that I think about it, I believe I'm actually thinking of a 'Basic tools' manual, and not machines.
[17:19:52] <Jymm> bill2or3: I have most of the Army FM's on CD, I can ask a friend about Navy ones, though it;s been years since she touched a mill, she might not have it anymore.
[17:20:46] <bill2or3> I'm pretty sure the Navy manual I'm thinking of was about tools. I have it at home, I'll check tonight.
[17:20:57] <Jymm> bill2or3: ok, np.
[17:22:16] <Jymm> Unless it's reference, I don't do dead trees anymore since I moved.
[17:22:36] <Jymm> 600 lbs of books... no more! =)
[17:23:18] <archivist> I still do dead tree
[17:23:43] <Jymm> archivist Only reference books I still do.
[17:23:55] <bill2or3> I like books, they're easier to read on the can.
[17:23:55] <archivist> I wonder at the weight at home
[17:23:56] <bill2or3> heh
[17:24:05] <bill2or3> but they sure take up a lot of space.
[17:24:15] <archivist> smallest room is called the library
[17:24:41] <archivist> ya have to take a book with you
[17:27:01] <archivist> hmm 18 books with machine in the title
[17:35:55] <Jymm> http://www.metalwebnews.org/ftp/tools-uses.pdf
[17:41:59] <Jymm> http://www.metalwebnews.org/ftp/machinery-repairman.pdf
[17:41:59] <Jymm> http://metalwebnews.org/machinist.html
[17:41:59] <bill2or3> I dont suppose anyone knows much about galvanometers, as used in laser scanners?
[17:42:00] <Jymm> les would
[17:42:00] <Jymm> bill2or3:
[17:42:00] <Jymm> bill2or3: MR2 and MR3 is the ones she suggested
[17:42:00] <bill2or3> ahh, ok, I'll take a look.
[17:42:00] <Jymm> bill2or3: They might have included both on that PDF, still trying to DL it
[17:48:28] <Jymm> WOW! MR is a bit dated, but it seems to cover all the basics of machinery repair, drilling, welding, milling, etc
[20:45:18] <toastydeath> does anyone know the name of the song that goes "back to the old school, back to the beat"
[21:26:11] <archivist> google the words and a lyric site should be found
[21:42:12] <toastydeath> i did that before i asked
[21:42:40] <toastydeath> it's a dance song, there's no "lyrics" for it, that's just the sample they play
[21:45:31] <jepler> http://en.wikipedia.org/wiki/Back_to_the_Old_School
[22:09:52] <toastydeath> nah
[22:10:02] <toastydeath> it's a house/electro song
[22:10:06] <toastydeath> but ty
[22:27:02] <Gamma-X> ok i get a joint 2 following error...
[22:27:25] <Gamma-X> anyone know what that is?
[22:33:32] <Gamma-X> anyone got a clue?
[22:37:42] <alex_joni> Gamma-X: you're out of questions :P
[22:37:58] <Gamma-X> haha damn
[22:38:05] <Gamma-X> im thinkin its pid cause it creeps
[22:38:07] <alex_joni> joke aside.. a following error results when the feedback doesn't match the command
[22:38:08] <Gamma-X> without input
[22:38:18] <alex_joni> so either emc commanded the motor to move, and it didn't
[22:38:27] <alex_joni> or emc didn't command movement, but it still happened
[22:38:42] <alex_joni> there is a setting in the ini how large the treshold shold be
[22:39:09] <Gamma-X> i havnt messed with my pid and it creeps....
[22:39:17] <Gamma-X> so im assuming i should tune pid and it will work correctly
[22:39:58] <alex_joni> heh
[22:40:23] <Gamma-X> right?
[22:41:32] <alex_joni> Gamma-X: in theory yes, but you'll discover it ain't that simple
[22:42:01] <Gamma-X> when u tune pid u have to use the ini file right? theres no gui for it?
[22:57:32] <fenn> when tuning pid you should raise min_ferror so it doesnt give errors all the time
[22:57:56] <fenn> there's no gui
[22:58:18] <fenn> its a hal thing, so perhaps one day there may be a gui
[22:58:53] <ginge> Hello
[22:59:56] <ginge> just a quick one... the EMC live CD doesnt seem to be booting the panels on startup. I dont see the EMC background either. Is it just me and my 2 laptops?
[23:00:19] <fenn> the panels?
[23:00:25] <ginge> gnome
[23:00:35] <fenn> oh, i dont know anything about gnome
[23:00:52] <fenn> maybe your resolution is wrong and the edge of desktop is off the screen somewhere
[23:00:53] <ginge> me either. KDE kinda guy.
[23:01:14] <fenn> try running gnome-panel from a terminal?
[23:01:24] <ginge> I should at least see the background image (a penguin?) the Xorg logs show the correct res
[23:01:46] <fenn> look at ~/.xsession-errors
[23:02:33] <ginge> nothing much in there.
[23:02:43] <ginge> Just says Beginning session setup...
[23:02:58] <ginge> noting in /var/log/gdm/*
[23:03:04] <fenn> um, use icewm instead? :)
[23:03:28] <ginge> okay, no prob. Just making sure the livecd wasnt bad
[23:03:56] <fenn> it's a bug, but i can't help really, sorry
[23:04:11] <ginge> thanks Fenn (don't suppose you used to hang out on OpenServo?)
[23:06:15] <fenn> yep a little
[23:06:28] <fenn> i'm more interested in CNC servos than airplane servos
[23:06:41] <fenn> but the same principles apply when you're building them from scratch
[23:06:49] <ginge> heh. small world. Just retrofitted my mill with some :)
[23:07:36] <fenn> you arent the ormus guy are you?
[23:07:59] <ginge> no. I run OpenServo, known in the forums as ginge
[23:09:00] <fenn> its a cool project. we need more open hardware
[23:10:25] <fenn> ginge: this is something near the top of my project list http://fennetic.net/machines/motherchip
[23:11:34] <ginge> interesting. It is very close to what we have going down at OpenServo. I have an open source USB to I2C bridge that drives the OpenServo boards, which drive the servos
[23:12:24] <fenn> yeah the difference is that the PID is done in the PC
[23:12:42] <fenn> and an order of magnitude higher power
[23:12:54] <fenn> and resolution..
[23:13:10] <ginge> http://www.headfuzz.co.uk/?q=node/49
[23:13:27] <ginge> oops. Well that is my driver board. It is modded to use encoders
[23:15:28] <ginge> Interesting idea having the PID done in the host. Is that pretty common with CNC gear?
[23:16:55] <fenn> yes but mostly due to tradition (limitations of old electronics)
[23:17:17] <fenn> i'm doing it that way because emc is based on that concept
[23:17:49] <fenn> but you could also port the PID code to the motor driver with a little effort
[23:18:02] <ginge> makes sense. I am not that far into the EMC internals as yet, but I was hoping to push the PID out
[23:18:54] <fenn> some people carry that idea further and run the whole motion controller in a dedicated chip
[23:19:27] <ginge> that sounds pretty hardcore. I assume they are commercial ventures?
[23:19:37] <fenn> yep
[23:19:56] <fenn> you can do simple stuff easily though
[23:20:18] <fenn> think 'turtle'
[23:20:39] <ginge> like a logo programming language?
[23:20:59] <fenn> the sort of people who do that like to reinvent the wheel
[23:21:17] <ginge> lol. I think I will stick with my gcode
[23:21:43] <fenn> it's a common mistake since it appears to be easy at first glance
[23:23:10] <fenn> anyway you aren't going to do kinematics calculations for a hexapod in an AVR
[23:23:17] <fenn> stewart platform
[23:23:46] <ginge> yeah. My PC has a hard enough time calculating the kinematics for my kuka robot arm
[23:25:18] <fenn> did you know emc has puma-style kinematics already?
[23:26:03] <ginge> I saw it in the docs, but I never really go much further than using emc for my milling. It is on my list of things to look at. It didn't seem well documented t first glance
[23:26:03] <fenn> i think the joint accel/vel constraint stuff still needs work though
[23:26:25] <fenn> yeah it's not a plug and play sort of thing
[23:26:34] <ginge> I use a dedicated RTAI robot library called orocos. Also not plug and play
[23:27:10] <fenn> orocos has nice documentation at least :)
[23:27:16] <ginge> indeed.
[23:29:08] <ginge> have you used the UHU servo motor driver?
[23:29:40] <fenn> no, i think its a terrible idea
[23:30:05] <ginge> for what reasons?
[23:30:35] <fenn> for reasons described here: http://www.cnczone.com/forums/showthread.php?t=37309&page=2
[23:31:46] <fenn> er.. basically because step/dir has high hysteresis
[23:32:01] <fenn> sorry, dont read that link, it's mostly unrelated
[23:33:01] <fenn> and you get no feedback to the control
[23:33:33] <ginge> oh, well. I don't like the idea of converting from step to PID, and I did enjoy reading the link regardless :)
[23:34:10] <ginge> Losing steps is the main reason for creating an open driver board for servos.
[23:34:17] <fenn> right
[23:34:58] <fenn> a vector brushless drive running steppers with encoders seems like it would be a good match for cnc
[23:35:11] <fenn> but then they wouldnt be steppers anymore..
[23:36:08] <ginge> sure. You would be creating a low res servo motor.
[23:36:18] <fenn> the resolution would depend on the encoder
[23:36:26] <fenn> it's just a 200 pole brushless motor
[23:36:56] <fenn> more poles means higher torque, more volts per rev
[23:37:00] <ginge> I see, sorry misunderstood what you were saying there.
[23:37:11] <fenn> or something like that.. :P
[23:37:19] <fenn> do you know what vector torque control is?
[23:37:54] <fenn> basically you can set the amount of field current and thus the torque to whatever the current situation requires
[23:37:58] <ginge> I know of it, but the lower down details elude me
[23:38:26] <fenn> and the phase of the rotating magnetic field moves forward or behind the phase of the rotor, modulating torque
[23:38:35] <fenn> so, field current * phase difference = torque
[23:39:29] <ginge> I see.
[23:39:33] <fenn> it would never overheat or "miss a step" unless you exceed the maximum peak torque
[23:39:55] <fenn> uh "never overheat" meaning it wouldnt just sit there dumping current through the coils
[23:40:00] <ginge> I assume you have to estimate some parameters in the system, such as the flux
[23:40:24] <fenn> yes, i'm not too keen on the details
[23:41:12] <ginge> I also assume that flux with also be sinusoidal in an AC drive system...
[23:42:10] <fenn> yes that's why i said brushless, since they're trapezoidal (like steppers)
[23:42:35] <ginge> anyhoo. I give up with this liveCD tried all four systems and none will show the gnome panels. Great chatting with you. You certainly know your stuff!
[23:43:25] <fenn> thanks, maybe i will finish a project one of these days..
[23:44:12] <ginge> I know the feeling. You might see me back here asking awkward questions when I get the EMC code in pieces. Cya!
[23:44:12] <archivist> livecd worked for me, has a few foibles
[23:45:26] <skunkworks> wait - you mean the gui (gnome) isn't starting? just booting to the terminal?
[23:45:57] <ginge> I get the mouse pointer and the ubuntu brown background, but nothing else
[23:46:12] <ginge> X is fine. detects my NV. tried in vesa, still the same
[23:46:18] <ginge> the other machine has a trident card
[23:46:40] <skunkworks> huh. odd
[23:46:42] <ginge> the only machine I get the panels loaded on is too slow and RTAI unloads
[23:47:09] <skunkworks> I have had that happen once 'after' installing it to the hard drive
[23:47:33] <ginge> might put a new drive in to preserve my known working emc 1.x install and put it on th drive
[23:47:39] <skunkworks> did you check the checksum?
[23:47:48] <skunkworks> after the burn
[23:47:56] <ginge> I assume you mean the CRC of the CD? Yeah
[23:48:04] <skunkworks> ok
[23:48:40] <skunkworks> you will really like emc2 vs emc1
[23:48:51] <skunkworks> if I do say so myself ;)
[23:49:22] <ginge> I am pretty stoked about it, the system that did load a panel up looked nice.
[23:49:50] <ginge> oh well, I am now armed with a spare drive. Wish me luck! Catch you all later (no doubt)
[23:50:03] <skunkworks> Good luck
[23:50:11] <archivist> hmm just found a foible I dont remember it asking for a root password during the install and I want to install a webcam for machine alignment
[23:50:44] <archivist> it wont let me su root
[23:52:15] <fenn> archivist: sudo su
[23:52:33] <fenn> or just sudo whatever you're trying to do
[23:52:57] <fenn> and type in your user password
[23:52:59] <archivist> make install, which is a bunch of things
[23:53:13] <fenn> sudo make install will give all the child processes privileges
[23:53:31] <archivist> Im not used to sudo
[23:53:46] <fenn> its the ubuntu way :\
[23:54:08] <archivist> been a root user since unix svr5 r3
[23:54:21] <fenn> cant say i see much of a difference in the security of the system besides making it harder to accidentally erase stuff
[23:55:41] <archivist> well is in dependency hell adding stuff one doesnt really want to spend all day retyping ones password
[23:55:56] <fenn> nah dependency hell is gone now with apt-get
[23:56:10] <fenn> if you are manually installing stuff you're probably doing something wrong
[23:56:45] <archivist> not for me some stuff is needed not in the repositries