#emc | Logs for 2008-02-20

[00:51:06] <tomp2> SWPanos: http://pastebin.ca/910561 i cant find the rules that these respect
[03:02:41] <skunkworks_> no turning back now... http://www.electronicsam.com/images/control.jpg
[03:03:14] <eric_U> don't know if you shoulda done that :)
[03:04:05] <skunkworks_> here is the shaper.. (messy - there has been a lot of re-arranging as dad has been getting rid a bunch of stuff) http://www.electronicsam.com/images/shaper.JPG
[03:04:28] <skunkworks_> I am not worried ;)
[03:04:39] <fenn> heh /me was expecting a gingery shaper
[03:04:56] <fenn> that's a tad larger
[03:05:07] <jmkasunich> probably costs less
[03:05:31] <fenn> definitely
[03:05:41] <jmkasunich> large shapers tend to go for scrap metal price, I've seen some for $50 to $150
[03:06:42] <fenn> skunkworks_ I applaud you in your initiative
[03:09:40] <skunkworks_> heh - that was dads first metal working machine
[03:09:47] <skunkworks_> because it was cheap
[03:10:06] <skunkworks_> big one anyways. He had a small sears/atlas lathe
[03:11:18] <skunkworks_> if you need to know - my mother is actually scrapping the control out.. She likes the copper/aluminum payout ;)
[03:11:43] <fenn> yep but it's only gonna go higher
[03:11:55] <fenn> uh, dont let her throw away everything
[03:12:04] <fenn> keep switches and power electronics
[03:12:19] <skunkworks_> oh - dad has the last word. :) she would cash everything in. ;)
[03:12:28] <fenn> too bad, that
[03:12:45] <fenn> i'd hope someone could at least see the value of it in an abstract sense
[03:13:06] <fenn> even if they personally didnt want the thing
[03:18:28] <fenn> heh i go to add 'gnisting' to the edm article and it's already there in the edit box. i go back and it's not there, but appears when i hit reload
[03:20:06] <gezar> my program runs :)
[03:20:13] <fenn> what does it do?
[03:20:26] <fenn> lemme guess, it's recursive
[03:20:29] <gezar> it calculates payments for a credit card :(
[03:20:42] <gezar> all 400+ lines of it
[03:20:52] <fenn> ah, a perfect fit for C++
[03:20:57] <fenn> ~~~
[03:21:05] <gezar> no, I just really over did the hell out of it
[03:21:08] <tomp> fenn: i added gnisting yesterday to wiki
[03:21:15] <gezar> sweet
[03:21:31] <gezar> I think I may have failed todays calc test :(
[03:21:50] <fenn> hmm i think you meant deterministic, not stochastic
[03:22:12] <tomp> no, edm is stachostic
[03:22:21] <tomp> firewire is deterministic
[03:22:40] <fenn> uh, are there any physical processes that are deterministic then?
[03:22:51] <fenn> i.e. not a digital abstraction
[03:23:00] <tomp> dunno
[03:23:43] <tomp> my def: stachostic = usually does this when we do that ( but not exactly and not always )
[03:24:35] <tomp> stochastic sorry
[03:25:03] <tomp> " you say stochastic and i say stachostic....
[03:25:09] <eric_U> I'm adopting stachostic
[03:41:26] <eric_U> I'm giving a talk tomorrow, anyone give me $20 to say stachostic?
[03:42:26] <fenn> how do we know you will actually say it?
[03:42:38] <eric_U> would I lie about something like that?
[03:42:43] <fenn> you just might..
[03:43:27] <fenn> throw in a nukular or two
[03:43:54] <eric_U> how about a few basnyins for good measure?
[04:00:32] <fenn> anyone know where to get "toggle latches" like on toolboxes
[04:00:41] <eric_U> mcmaster
[04:01:05] <fenn> thanks. i thought of that and promptly forgot
[04:01:24] <eric_U> probably at lowes, rockler and woodcraft
[04:01:36] <eric_U> are you in the U.S.? I keep forgetting
[04:01:55] <fenn> yes, but i prefer mail order to wandering through lowes for hours and ending up disappointed
[04:02:23] <eric_U> mcmaster is easier. They prolly aren't cheap there though
[04:03:08] <fenn> actually the prices are pretty good if you are buying 5+
[04:03:23] <eric_U> what number?
[04:04:29] <eric_U> yeah, not bad
[04:05:17] <fenn> 1766A61 looks good
[04:07:05] <eric_U> we got some like figure "f" for our robot
[04:12:58] <fenn> units 37/64
[04:13:02] <fenn> oops heh
[04:23:10] <gezar> well, I submitted my mid term on time
[04:23:31] <eric_U> that's gotta be worth a few points, at least
[04:23:33] <gezar> Im not happy with it, but I think I put enough work into to it for it to be as correct as I can make it.
[04:23:55] <eric_U> next time email it to me and I'll check it for you :)
[04:24:11] <gezar> you got g++ installed?
[04:24:19] <gezar> msg me your email
[04:24:22] <eric_U> yeah
[04:24:51] <eric_U> I thought you had to give freenode money to msg
[04:24:59] <gezar> what?
[04:25:07] <gezar> can you not /msg me?
[04:25:19] <gezar> I htink you just have to identify.
[04:25:25] <gezar> I may not be identified right now
[04:25:33] <eric_U> did that work?
[04:25:37] <gezar> try now
[04:25:41] <gezar> im identified
[04:26:05] <gezar> you get that?
[04:27:10] <eric_U> yeah, emailed you
[04:28:35] <gezar> hmm
[04:28:43] <eric_U> it says you don't exist, bounced
[04:28:44] <gezar> all hell
[04:28:49] <gezar> yeah, gezar
[04:28:56] <gezar> missed the a and made it an e
[04:28:57] <gezar> my bad
[04:29:02] <gezar> same thing though
[04:30:29] <gezar> sent
[04:30:32] <gezar> enjoy
[04:31:29] <gezar> oh, if g++ midmain.cpp -o test ./test doesnt work, I get a 0, I did the code in .net but it should be g++ able
[04:32:07] <eric_U> ouch
[04:32:11] <gezar> ?
[04:32:27] <eric_U> harsh isn't it?
[04:32:34] <gezar> .net?
[04:32:42] <gezar> yeah, its nothing like vim
[04:33:26] <eric_U> I use kdevelop
[04:33:32] <gezar> I mash the strangest keys at time, I dont like that line, dd damnit, now ive got 2 more characters to get rid of, x wtf, xxxxx oh , grab mouse, accidently highlight a ton of code and move it around, get even madder
[04:33:53] <gezar> well, did that monster compile?
[04:34:23] <eric_U> no, because main must return int
[04:34:50] <eric_U> I never heard that
[04:34:50] <gezar> toss a return 0; at the end
[04:34:58] <gezar> its a g++ rule
[04:35:05] <eric_U> it's your void main}{
[04:35:17] <gezar> yummy :)
[04:35:55] <eric_U> do you get points off for spelling?
[04:36:04] <gezar> I doubt it
[04:36:15] <gezar> I had 40 min left to get it turned in
[04:36:40] <gezar> so if my spelling is off, I wasnt concerned, I hope I took out all the bend over and take it lines out
[04:37:09] <gezar> but it should run, it should do all the functions, it should make a file, if you chose c, a or b, and you should be able to open the file with a web browser
[04:37:33] <gezar> I had at least 6.5x the #lines in that as the others, with nothing from the intarbrain
[04:40:17] <fenn> gezar: i know this may sound like i'm trolling, but you may find it easier to write your program in python first, then re-do it in C++
[04:40:26] <eric_U> troll
[04:40:38] <fenn> trolls rule! :(
[04:40:54] <eric_U> how about write it in lisp and then convert it to ruby, then python, and then c++
[04:41:19] <fenn> hmm..
[04:41:22] <eric_U> gezar, you have a strange result on one of the payments
[04:41:26] <gezar> I cant help it if I have a spelling issue,
[04:41:28] <gezar> eric_U: ?
[04:41:40] <eric_U> but the web page worked
[04:41:41] <gezar> in the graph, or straight up #s?
[04:41:57] <gezar> bad results = -50% too
[04:41:58] <fenn> gezar: have they mentioned the phrase 'revision control' yet?
[04:41:59] <eric_U> I put in a debt of 1234
[04:42:27] <gezar> fenn : no this is a basic class
[04:42:35] <gezar> eric_U: and the result was?
[04:42:40] <eric_U> it calculated it for $130, the last two payments are $3.02 and 3.05
[04:42:49] <fenn> gezar: tell me if they ever do
[04:42:59] <gezar> yeah, I didnt understand that myself
[04:43:35] <gezar> eric_U: if im knocked way down because of that, then I worked to hard on it, and got lost
[04:44:15] <gezar> eric_U: the assignment was to just make a table, with 1000 debt, 50-120 payments, thats it
[04:44:28] <gezar> extra credit if you did it with more, I did it with all I could
[04:44:33] <gezar> fenn: I will
[04:44:54] <gezar> eric_U: is it bad for a first time programer?
[04:45:03] <eric_U> didn't I tell you not to make the grader work harder? :)
[04:45:07] <skunkworks_> getting there http://www.electronicsam.com/images/house/breakfastbar.JPG
[04:45:09] <eric_U> looks good
[04:45:17] <gezar> the grader doesnt speek english
[04:45:19] <eric_U> the html came out perfect
[04:45:25] <eric_U> that's what I'm saying
[04:45:45] <eric_U> one time, I got by far the best result on a take home project
[04:45:47] <gezar> the wall of text at the start is going to startle her
[04:46:08] <eric_U> but I didn't follow some rule that I don't think they told us, and got 10 points off
[04:46:20] <eric_U> that could be a problem
[04:46:27] <gezar> he told us to go all out
[04:46:43] <gezar> so I took the hardest approach I could
[04:47:10] <eric_U> wonder what version of g++ she uses?
[04:47:27] <gezar> first time, ive ever used functions in my life though, one of the bigest problems ive had with trying to help emc out was getting over the function hurdle
[04:47:43] <fenn> O_o
[04:47:48] <gezar> she is a microsoft wizzard
[04:48:05] <fenn> gezar you've never used an o-code or subroutine?
[04:49:03] <gezar> not really, I did some deep if statments on a perl program several years go while trying to learn
[04:51:30] <gezar> and looks like im going to talk to my calculus teacher after class, I spent the past 5 days doing all I could, and now im behind, im exhausted and not sure what to do
[04:51:55] <eric_U> I remember reading my calculus book for the first time an hour before the test
[04:52:04] <eric_U> I wasn't a very good student
[04:52:41] <gezar> im trying to be the best I can, just too many years have passed since my last math class
[04:52:56] <eric_U> math goes the worst
[04:53:04] <eric_U> I can still do most of it if I have to
[04:53:37] <fenn> someone needs to invent a math bicycle so you won't forget it so quickly
[04:54:05] <fenn> i think the problem is the hieroglyphics
[04:54:13] <eric_U> now that I learned it 3 times, I'm getting better at remembering.
[04:54:22] <eric_U> just don't ask me to do a triple integral
[04:54:30] <gezar> ive been teaching morons the past 15 years how to program cnc machines, when you just focus on simple trig you forget a ton of stuff
[04:54:32] <fenn> a lisp-like notation would go a long way to enhancing communication
[04:55:23] <gezar> "the universe is full of 2 things, hydrogen and stupidity."
[04:56:02] <fenn> we've discovered the source of dark matter!
[04:56:54] <gezar> dark matter is scientists excuses for not knowing gravity
[04:57:33] <fenn> yeah, or something else, like how light propagates
[04:58:15] <gezar> well, einstien and his space time warp crap, and string theory, and all that mumbo jumbo, is going to be simplified some day
[04:58:30] <fenn> i think hodowanec has the right idea
[04:59:05] <eric_U> the problem is it's all stachostic
[04:59:11] <fenn> http://www.rexresearch.com/hodorhys/rhyscosm/rhyscosm.htm
[04:59:21] <eric_U> I'm going to say that tomorrow, and be embarrassed about it
[05:02:03] <gezar> eric_U: so it wasnt a bad 5th learning program?
[05:02:30] <eric_U> not bad
[05:02:45] <eric_U> I was impressed the html worked without complaint
[05:03:01] <gezar> I did that rather funkly didnt I?
[05:03:24] <gezar> I had it as a set of functions, but when the function went to write to the file, it didnt work out
[05:03:58] <gezar> then I had a -number in the blank boxes issue to work out
[05:04:25] <gezar> I think I nailed that down though
[05:05:40] <eric_U> but you didn't test for a debt of 1234
[05:06:54] <gezar> no I did not
[05:07:06] <eric_U> 4321 either
[05:07:18] <eric_U> sorry, just yanking your chain
[05:07:20] <gezar> I had some fears about funky issues comming up, but i think thats an inside computer issue
[05:07:28] <gezar> is it just # strings?
[05:09:34] <gezar> on the graph the last 2 numbers should be failry close to each other, I did work a few of them all the way down 1000 100 18, 500 100 18 and some others
[05:09:48] <eric_U> I see now
[05:10:13] <eric_U> looks right
[05:10:24] <eric_U> your title threw me off.
[05:10:26] <gezar> you know what pisses me off the most?
[05:10:42] <eric_U> wha
[05:11:04] <gezar> i found a web site that does the same thing, you can enter stuff, and the #s they got were not the same as mine
[05:11:28] <eric_U> depends on how they are calc the interest
[05:11:34] <gezar> yeah
[05:11:37] <gezar> thats what I figured
[05:12:00] <gezar> maybe they will like it, I dont know
[05:12:03] <eric_U> engineering economics, learn how to calc interest in 7 different ways
[05:12:14] <gezar> I saw one others program, it was maybe 70 lines of code
[05:12:45] <ds2> hmmm 1/8 Al rounds aren't that easy to find
[05:13:13] <gezar> wow thats small, get semi long 1/8 bolts and wack them off
[05:13:44] <gezar> eric_U: would you rather see complete code to do a task, or code that just does the task?
[05:14:02] <gezar> I mean, the instructions were somewhat clear for you to read, and made it easy for you to navigate?
[05:14:46] <gezar> I know I didnt comment it much, but he bitched at me for getting too many line wraps on program #2
[05:15:01] <gezar> I like single lines on printed paper, not all this dropping down stuff
[05:15:04] <eric_U> from a practical perspective, your best bet is to write something very clear, and concise as possible, correct as possible
[05:15:17] <ds2> where do you find Al bolts?
[05:15:23] <eric_U> your style could use some work
[05:15:33] <ds2> gezar: I am looking for some thin Al material to make up a soft pin punch to work the lead screw thingie
[05:15:56] <gezar> A1 as in air hardening steel A1?
[05:16:05] <eric_U> luminium
[05:16:08] <ds2> no, Al. not A1
[05:16:12] <ds2> as in Aluminum
[05:16:22] <ds2> I can find A1 quite easily in 1/8
[05:16:29] <gezar> oh, hmm, let me see
[05:16:38] <ds2> Only MSC had them
[05:17:01] <gezar> http://www.starboltco.com/
[05:17:05] <gezar> I recomend those guys
[05:17:15] <gezar> I dont get paid or nothing but they are decent
[05:17:32] <ds2> AR...hehh.. might as well order from MSC
[05:18:01] <gezar> well, when you call them your accually talking to a guy who can hold the bolt for you
[05:18:33] <ds2> that might be, but it is probally going to come out to $20 for a single bolt... I'll find some other way
[05:18:48] <gezar> just call, it may not be but a few cents
[05:18:51] <ds2> I'd turn it down except 1/8" dia, 1" would whip around a bit
[05:18:55] <gezar> you never know
[05:19:00] <ds2> I am in CA
[05:19:16] <gezar> oh man, there are a million suppliers local to you then
[05:19:18] <ds2> maybe I should try forging Al
[05:19:21] <gezar> you just have to do the foot work
[05:19:49] <gezar> you just need 1/8 dia pin? or screw thread?
[05:19:57] <eric_U> i miss all the vendors I was near in northern Utah
[05:20:00] <ds2> I just need a small pin punch
[05:20:07] <ds2> to drive a nut; one time use
[05:20:12] <eric_U> I could drive to a fastener store, steel dealer
[05:20:25] <ds2> guess I can drop by fastenal and ask
[05:20:27] <eric_U> now I'm an hour from either
[05:20:35] <gezar> you got a lowes near you?
[05:20:39] <ds2> yep
[05:20:45] <gezar> what about brass then?
[05:20:50] <ds2> oh yeah brass would work
[05:20:57] <gezar> they have those long small dia brass screws, for liek 2.50 a pack
[05:21:40] <ds2> yes, I like that. a 1" long 6-32
[05:21:52] <gezar> thats going to be much less then
[05:22:21] <ds2> okay problem solved. I need to get back to looking at the X screw problem
[05:22:37] <fenn> roofing nails are aluminum sometimes
[05:22:51] <fenn> for holding down gutters
[05:23:07] <ds2> I've seen copper roofing nails but not Al
[05:23:15] <fenn> copper nails for copper gutters
[05:23:20] <gezar> yeah, have to use certian metals so they dont go dielectric on you
[05:23:21] <ds2> oh I see
[05:23:56] <gezar> galvanized steel in stell will accually generate a current
[05:24:13] <ds2> Hmmm
[05:50:53] <gezar> oh, about ubuntu, I dont like it compared to hard core debian
[05:51:26] <gezar> Ill ahve this ubuntu install debianized before too long, just need to drop math or have something magical happen
[10:42:17] <micges> hi all
[10:42:38] <alex_joni> 'lo
[10:42:56] <micges> yesterday I mentioned about beam shape compensation module
[10:43:03] <micges> today it works :)
[10:43:27] <micges> hal is amazing and powerfull stuff
[10:45:04] <alex_joni> nice :)
[10:45:16] <alex_joni> I wasn't paying attention yesterday..
[10:48:59] <micges> all will be on wiki
[10:49:00] <alex_joni> micges: I just read back, very itneresting
[10:49:59] <micges> Ive created link in using EMC
[10:50:14] <micges> I think that is good place for this :P
[10:53:26] <alex_joni> maybe more advanced than simply using emc ;)
[10:54:31] <micges> little more :)
[10:55:05] <alex_joni> * alex_joni looks
[10:56:32] <micges> today Im going to create graph of data flow and pins in hal to make it works
[12:30:09] <skunkworks_> micges: very cool. Good work!
[13:25:24] <Guest995> logger_emc: bookmark\
[13:25:24] <Guest995> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-02-20.txt
[13:25:41] <Guest995> Guest995 is now known as skunkworks
[13:49:30] <skunkworks> http://www.electronicsam.com/images/control.jpg
[13:53:21] <archivist> waa started ripping it out
[13:53:45] <skunkworks> yah
[13:53:58] <archivist> poor machine
[13:54:28] <archivist> emc next week?
[13:54:56] <archivist> month?
[13:54:59] <archivist> year?
[13:55:04] <skunkworks> it will probably be a year ;)
[14:03:00] <skunkworks> it is going to require a bit of engineering to mount the servos
[14:03:16] <skunkworks> it has hydraulic servos
[14:05:04] <alex_joni> ROFLMAO http://d116.com/spud/
[14:05:15] <alex_joni> potato powered CNC anyone?
[14:09:21] <skunkworks> Sweet :)
[14:15:44] <cradek_> cradek_ is now known as cradek
[14:17:07] <skunkworks> cradek: good morning
[14:25:40] <alex_joni> see you tomorrow guys
[14:27:16] <skunkworks> night alex
[14:27:35] <archivist> too early for bed
[14:28:23] <skunkworks> heh
[14:54:15] <micges> skunkworks: AF updated 1000 times per second works great
[14:55:47] <micges> no measurement jumps speedups or so exist
[14:57:56] <micges> bbl
[15:09:46] <skunkworks> Cool
[15:10:16] <cradek> skunkworks: looks like there are a few parts missing from your control. I think you will have some trouble putting it all back together right.
[15:12:48] <skunkworks> Heh
[15:14:36] <skunkworks> I think it will feel the wrath of the plasma cutter..
[15:14:46] <skunkworks> poor thing
[15:15:02] <skunkworks> 'there is a disturbance in the force...' ;)
[15:15:26] <archivist> object->kit->reassemble done
[15:40:20] <micges> logger_emc: bookmark
[15:40:20] <micges> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-02-20.txt
[15:46:28] <skunkworks> micges: that sounds great - nice work!
[15:49:41] <skunkworks> actually - it was explained to us that the reason we needed different feeds x vs y was because of cutting with the grain and against...
[15:53:45] <micges> Im not understand last..
[15:54:02] <archivist> even some rolled metals have noticable grain direction
[15:54:44] <micges> what mean grain direction ?
[15:54:55] <skunkworks> grain of the wood..
[15:55:11] <skunkworks> * grain of the plywood
[15:55:28] <skunkworks> the direction the wood fibers go.
[15:55:35] <micges> ok
[15:56:03] <micges> now Im drawing pics
[15:56:34] <micges> I would be nice to emc someday have this feature
[15:56:45] <micges> *it
[15:59:38] <jlmjvm> jepler:i just loaded the 2.2.2-1 cd on my machine computer and i cant get my tlo.py to work,do i need to load idle also?
[16:01:49] <jepler> what error do you get?
[16:04:52] <jepler> bbl
[16:05:05] <jlmjvm> no error,just acts like its opening and goes away
[16:05:22] <jepler> are you running it from the terminal so that you *would* see any errors it prints?
[16:05:24] <jlmjvm> i had idle and emacs on the other config
[16:05:44] <jlmjvm> i can try that
[16:09:44] <jlmjvm> jsuat opens and closes immediately in terminal
[16:10:39] <jlmjvm> just
[16:10:53] <SWPadnos> did you open a terminal then manually run the program (by typing in some command), or did you configure the icon to run in terminal?
[16:11:56] <jlmjvm> just clicked the icon and selected run in terminal
[16:12:03] <SWPadnos> bad jlmjvm
[16:12:14] <SWPadnos> open a terminal, then type in the necessary command
[16:18:13] <jlmjvm> back to the original question,do i need idle installed to run a .py program,whether its hello world or tlo9 or anything
[16:18:36] <SWPadnos> I don't think so. I have run various bits of python code without intentionally installing idle
[16:18:54] <BigJohnT> if you have python installed that is all you need...
[16:18:56] <SWPadnos> that's why the error message is important, it will tell you why the program doesn't run
[16:19:06] <BigJohnT> windoz or linux?
[16:19:08] <SWPadnos> or at least a symptom of why
[16:19:12] <gezar> ive got a c in calc
[16:19:26] <gezar> there is hope after all
[16:19:26] <SWPadnos> bummer
[16:19:36] <SWPadnos> oh, then great!
[16:19:40] <jlmjvm> what would the command that you mentioned earlier
[16:20:04] <SWPadnos> jlmjvm, I don't know how the program is supposed to be run, take a look at the command in the icon config
[16:20:08] <BigJohnT> jlmjvm you trying to run on windoz or linux
[16:20:08] <gezar> SWPadnos: I failed the first test,its a lot of work for me.
[16:20:15] <SWPadnos> it's your program, you should know how to run it ;)
[16:20:27] <jlmjvm> the 2.2.2-1 iso
[16:20:58] <SWPadnos> ?
[16:21:02] <BigJohnT> ?
[16:21:13] <SWPadnos> Applications -> Accessories -> Terminal
[16:21:17] <gezar> ubuntu im betting is what he is talking about
[16:21:34] <BigJohnT> must be
[16:21:48] <jlmjvm> yes the new 2.2.2-1 iso off the website
[16:21:50] <gezar> and if he is just clicking an icon then he will never get a return
[16:21:57] <SWPadnos> this is a python program that apparently ran on some other version of emc/Ubuntu, and doesn't run on a new 2.2.2-1 boot
[16:22:07] <SWPadnos> yes, we've established that :)
[16:22:07] <BigJohnT> jlmjvm are you testing a python program?
[16:22:34] <BigJohnT> ok, open a terminal window the cd to the directory where the program is
[16:22:45] <SWPadnos> then python tlo.py
[16:22:46] <BigJohnT> then type in python myprogram.py
[16:23:02] <BigJohnT> if you have any errors it will show up in the terminal window
[16:23:03] <SWPadnos> ok, time to run. see you guys later
[16:23:10] <BigJohnT> ttul
[16:33:34] <skunkworks> jepler: step-conf is the bomb! very easy to set things up.
[16:40:47] <skunkworks> jepler: there is no MAX_Velocity in the traj section of the ini file - yet is seems to be making 3 axis max moves the right speed. each axis has will do 240ipm - 3 axis move shows 415. if you don't specify - it somehow sets it to the max?
[16:41:04] <cradek> yes
[16:41:44] <cradek> well more precisely all motion is limited by the individual axes; there is no additional tooltip limit imposed
[16:41:56] <skunkworks> ah - ok. Thanks
[16:44:39] <skunkworks> cradek: was it fun at stuarts?
[16:44:55] <cradek> yes we were working on a neat machine
[16:45:14] <cradek> 100" of travel, micron scale
[16:45:42] <cradek> it's very slow and very precise
[16:46:28] <skunkworks> heh - but you where running at 90ipm right? that isn't HSM but still - it gets a lot of work done :)
[16:49:08] <cradek> yes I think we left it at 90
[16:49:39] <cradek> with 100" of travel, over two minutes to move the table back and forth
[16:50:25] <skunkworks> heh
[16:50:31] <skunkworks> quite the machine
[16:51:25] <skunkworks> we usually ran ours at 50ipm for the most part. anything faster and z sounded like it was falling apart. (I think there is backlash in the drive train somewhere - it transmits all the way down to the x axis servo)
[16:51:52] <skunkworks> a few direction changes with right angle gears
[16:52:27] <cradek> you will remove all that?
[16:53:05] <skunkworks> yes - how I don't know yet. The z axis actuall spins the nut. so we need to figure out how to get the servo connected into there.
[16:53:14] <skunkworks> actually
[16:55:55] <skunkworks> I think the max it would run was (get this) 150 ipm. you would program f1 - f98 which was actual - and f99 was 150ipm
[16:56:23] <skunkworks> G0 was also effected by the feed rate. (not max like normal machines)
[16:57:57] <cradek> sounds like it will take a lot of reprogramming of emc to make it work "right" again
[16:58:06] <cradek> and you'll have to figure out how to read all those thumbwheels
[17:00:03] <skunkworks> right. I also want linear moves to be always incremental and only up to 9.9999 inches.
[17:00:09] <skunkworks> ;)
[17:00:37] <skunkworks> and have to calculate a goofy feedrate for the linear and circular interp
[17:11:10] <cradek> I don't think these changes should be incorporated into the general emc distribution
[17:13:38] <skunkworks> you might be right
[17:18:09] <skunkworks> oh - and the biggie. No decmal places - only leading zeros.
[17:22:45] <archivist> trying to read old gcode?
[17:23:17] <archivist> I would just write a translator
[17:23:24] <skunkworks> that is what the old K&T used to use.
[17:24:01] <skunkworks> That is what I did - wrote a translater in basic to convert X3.87 to D0387
[17:24:06] <skunkworks> oops
[17:24:19] <skunkworks> X3.87 to X0387
[17:24:49] <micges> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Laser_Beam_Compensation
[17:27:19] <micges> oly describe "calculations" :P
[17:59:02] <skunkworks> these compaq machines are actually pretty nice. I have a similar one that is 1ghz and it doesn't have shared video memory. It is also outputting an easy 40khz.
[19:23:29] <dmwaters> {server notice} Hi all,Soon this server will be rebooted for maintenence, I would recommend switching to another server in the chat.freenode.net rotation. Thanks!
[19:52:07] <skunkworks> wasn't there a pin that showed que depth in the tp?
[19:54:33] <cradek> wouldn't surprise me if that was hooked to one of the motion.debug-s32 depending on what version you have
[20:06:50] <tomp2> two of us were just fixing the u'pload err on the laser beam ellipse correction, i didnt know we had file blocking on wiki edits, cool
[20:17:13] <skunkworks> ah
[20:27:01] <skunkworks> heh - neither seems to do anything. s32-0 and 1. But I am sure I am doing something wrong :)
[20:32:01] <cradek> nah they could be anything - must not be hooked up
[21:05:08] <quiteBIGeye> good evning guys
[21:54:54] <tomp2> micges: please review the wiki page
[21:57:30] <archivist> tomp2 tyop enought
[22:13:02] <tomp2> archivist: tyop enought?
[22:13:22] <archivist> typo
[22:13:28] <archivist> speeeeling
[22:13:46] <tomp2> yah! fixin, thx
[22:14:04] <archivist> actually I was trying to thing of better english for that line
[22:15:14] <tomp2> i wanted to leave some lines original, i removed that 't' tho
[22:16:02] <micges> im back
[22:16:36] <tomp2> hi, please look at the wiki page, i've messed it up for you ;)
[22:18:32] <micges> great, thanks
[22:19:15] <tomp2> you'll have a .hal file someday, please add it to that page
[22:19:17] <micges> looks great
[22:19:35] <micges> sure
[22:19:42] <tomp2> thank you, we like your idea
[22:20:44] <micges> I wanted to do this today but I have emergency trip to work :|
[22:21:07] <micges> now Im too tired
[22:21:24] <tomp2> np, get some rest
[22:44:26] <gezar> i passed my test, i passed my test :)
[22:44:37] <micges> tomp2: if you have time please check wiki page
[22:47:08] <fenn> micges: i still dont get why you dont just use F for the feed (instead of m112)
[22:49:29] <micges> fenn: when you burn 500mm/min and noticed that axis 0 must move 10% faster then you only adjust axis 0 vel_correction pin on module
[22:50:13] <fenn> micges: if adaptive feed were not constrained to 100% would you do it differently?
[22:50:34] <micges> when you use F for feed then you must stop program change velocity of every gcode
[22:50:57] <fenn> no, program the requested velocity for F
[22:51:03] <SWPadnos> you can use F for the base_feed_rate
[22:51:04] <fenn> say, 500
[22:51:15] <SWPadnos> it gets output on the motion.spindle-speed-out pin
[22:51:32] <fenn> er.. you mean S?
[22:51:40] <SWPadnos> uh, yes, that's what I mean
[22:51:53] <fenn> how does emc know it's moving then?
[22:52:02] <micges> forgot: on every gcode file begin is F base_velocity line
[22:52:55] <micges> only one line with F
[22:53:43] <fenn> yes that's fine
[22:54:03] <fenn> * fenn sighs
[22:54:26] <fenn> i'm going to go paint my room
[22:54:58] <micges> if AF could be > 1.0 then with few modifications can be done only with F code
[22:55:16] <micges> fenn: what color ?
[22:58:44] <micges> fenn: I've just noticed that with AF > 1.0 whole idea will be much simpler
[22:58:59] <micges> can be done AF > 1.0 ?
[23:01:35] <fenn> sure
[23:02:50] <fenn> edit src/emc/motion/control.c around line 453
[23:03:15] <fenn> i dont know why it was limited to 1.0 in the first place, so i'm wary of changing it
[23:03:41] <fenn> but for your system it shouldn't matter
[23:03:51] <micges> ok thanks
[23:06:25] <tomp2> /* read and clamp (0.0 to 1.0) adaptive feed HAL pin */
[23:06:26] <tomp2> tmp = *emcmot_hal_data->adaptive_feed;
[23:06:26] <tomp2> if ( tmp > 1.0 ) {
[23:06:26] <tomp2> tmp = 1.0;
[23:06:26] <tomp2> } else if ( tmp < 0.0 ) {
[23:06:26] <tomp2> tmp = 0.0;
[23:06:28] <tomp2> }
[23:07:30] <micges> in which module is compiled to ?
[23:07:41] <tomp2> maybe some other place... if F >inifilemaxF, then F = inifilemaxf and spew warning
[23:08:20] <tomp2> uh, noit sure of question, source is as fenn said
[23:08:49] <tomp2> src/emc/motion/control.c near lines posted above
[23:08:56] <micges> ok
[23:09:27] <tomp2> can you think of how to let it be <0 for elletroerosione? :-)
[23:09:55] <micges> I will compile it in place
[23:10:05] <gezar> eric_U: you around?
[23:10:29] <tomp2> actually making the velocity negative is easy, getting tp to hand you the position is trick
[23:17:50] <micges> fenn: in which .ko module is AF code ?
[23:26:34] <SWPadnos> jymm JymmmEMC you around?
[23:29:13] <micges> adaptive-feed > 1.0 works :)
[23:34:04] <micges> hhow iin hal can be read current F value ?
[23:36:08] <micges> ok
[23:36:16] <micges> good night