#emc | Logs for 2008-04-29

[00:00:10] <dmess> this is the place they'll send ya to
[00:59:45] <jmkasunich> insulation done, now for the wallboard: http://jmkasunich.com/pics/framed-and-insulated-2032.jpg
[02:26:10] <tom1> http://people.mech.kuleuven.be/~bruyninc/blender/roadmap.html speaks about >projects< (read "not done yet") to use blender to simulate machines, a gcode interpreter, and stock removal simulation
[04:39:47] <ChanServ> [#emc] "This is the #emc channel - talk related to the Enhanced Machine Controller and general machining. Website: http://www.linuxcnc.org/, wiki at http://wiki.linuxcnc.org/"
[05:12:45] <K`zan> Night folks.
[09:56:28] <archivist> hmm reading list RTDM Real-Time Driver Model do we need that for deteminististic USB2
[11:08:41] <alex_joni> archivist: I think so
[11:08:45] <alex_joni> also for RTNet
[11:09:13] <archivist> just having a google during work coffee
[11:35:03] <alex_joni> archivist: cool :)
[11:36:06] <archivist> Day job gets in the way when you want to play....
[11:38:19] <Vq^> just get a job that is playing around :)
[11:40:15] <archivist> Im not paid enough to work hard
[13:55:30] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[15:21:02] <BigJohnT> I updated the wiki site with the 2.2.5 change... hope I didn't step on anyones toes...
[15:26:20] <BigJohnT> * BigJohnT back to work
[15:26:33] <BigJohnT> bbl
[15:29:23] <tomp2> archivist: firewire is already deterministic
[15:35:25] <SWPadnos> tomp2, for some values of "deterministic"
[15:35:38] <SWPadnos> and only for some data types, I believe
[15:36:50] <tomp2> runs the drives for aerotech cnc's
[15:37:09] <tomp2> position and velocity loop closed in the drive
[15:37:36] <tomp2> i admit there's a lot to the definition
[15:38:16] <SWPadnos> I'm not sure the industrial controls use a standard protocol stack. they may just be using the hardware with their own protocol
[15:38:26] <SWPadnos> I don't know though
[15:38:36] <BigJohnT> tomp2 did I see that you are building an electric motorcycle?
[15:38:45] <SWPadnos> (someone could try plugging in a video camera while running a big machine :) )
[15:41:37] <tomp2> SWPadnos: havent unwrapped it for near 2 years, got an A3200 (?) 4 axis cnc from them, it ran under W2K with some rt extension, I forget the vendor of the rt.
[15:42:36] <tomp2> BigJohnT: yes, waiting a longer chain but it 'ran' http://www.cyclone-tw.com/ had to hack the mounts a bit
[15:45:11] <tomp2> BigJohnT: beware the music on the site, they clobber you with loud classical
[15:46:22] <SWPadnos> tomp2, I wonder if that system can synchronize motion between axes (not from the controller, but based on feedback, like gearing one axis to another)
[15:48:41] <BigJohnT> cool, I had to run upstairs the ups guy was here
[15:49:13] <jepler> BigJohnT: get anything good?
[15:49:48] <BigJohnT> I have a motorless bike to try and make an e bike from
[15:49:58] <BigJohnT> jepler just stuff for work
[15:50:51] <BigJohnT> I have a 185 dirt bike to try and convert
[15:51:17] <SWPadnos> just get a bike-e instead :)
[15:51:24] <SWPadnos> though they are ugly
[15:51:49] <tomp2> BigJohnT: after i got it, walmart got these back in stock http://www.walmart.com/catalog/product.do?product_id=8467096
[15:52:01] <tomp2> been outta stick for more than a year
[15:52:05] <tomp2> stock
[15:53:38] <tomp2> SWPadnos: the closing of the loop is done at the drive ( so sayeth Prof. Joe Profetta, chief of aerotech). I dont know how. so feedback of speed and position go to the drive and reports on done and requests for new go between it and the control.
[15:54:21] <SWPadnos> right - the drives get position info (ideally), and the controlling PC assumes they do their thing correctly
[15:54:25] <SWPadnos> like steppers
[15:54:50] <SWPadnos> with the same problems - you can't (easily) synchronize multiple motors based on feedback
[15:54:57] <SWPadnos> or at least I think that's the problem :)
[15:55:48] <SWPadnos> I had a discussion on this topic with a guy from Hitachi and one from Rockwell (one of their motor drive divisions)
[15:57:18] <SWPadnos> I think that you need to be able to send position, velocity, accel, and possibly jerk parameters every cycle if you want to get profiles that are as "good" as closing the loop in a single place (for all axes)
[15:57:44] <SWPadnos> then again, with sufficiently strong motors and sufficiently fine resolution, it may not matter at all
[15:59:32] <BigJohnT> gotta go
[16:47:10] <alex_joni> anyone from virginia around?
[17:48:18] <punk84> hi is it possible to install emc on hardy?
[17:51:22] <punk84> hi is it possible to install emc on hardy?
[17:51:34] <archivist> punk84, it needs a realtime kernel,
[17:51:54] <alex_joni> and rtai package
[17:52:04] <alex_joni> punk84: it's not a simple procedure to do on your own
[17:52:14] <anonimasu> hm, my milling test worked out fine..
[17:52:19] <punk84> oh i see
[17:52:19] <alex_joni> we're working on packages, but they are not final yet
[17:52:23] <anonimasu> slowed feed down to 30% for the plunges..
[17:52:34] <anonimasu> and ran at 261mm/min for the normal cutting
[17:52:37] <punk84> is there packages I can test?
[17:53:19] <alex_joni> punk84: yes, there are.. but depending on how well you know your way around debs and installing, it might not be a very good idea
[17:55:51] <punk84> okay thank you sir for explaining. I will wait for the new packages
[17:56:28] <alex_joni> punk84: no need calling me sir ;)
[17:57:12] <BigJohnT> LOL he called alex sir
[17:57:28] <anonimasu> alex_joni: ^_^
[17:57:41] <punk84> my bad :P
[17:57:42] <alex_joni> punk84: the problem is that if you do test, and turn up with issues, it is only valuable if you can dig for further info so we can track the problem
[17:57:53] <alex_joni> BigJohnT: now you know how you need to address me in the future
[18:00:59] <BigJohnT> yes sire
[18:01:34] <punk84> hmm.....well if you need another tester I am up to trying it and maybe learning but I just dont want to wast your time with noob-ness lol
[18:02:58] <alex_joni> punk84: :-)
[21:39:24] <dmess> hi all
[21:39:30] <alex_joni> 'lo
[21:40:18] <dmess> i may be getting a nice okk vnc... for FREE
[21:40:52] <dmess> :):):)
[21:41:24] <alex_joni> vnc?
[21:44:40] <dmess> sorry vmc (verticle machineing center)
[21:45:23] <dmess> mitsubishi 500m control i believe
[21:52:27] <alex_joni> sounds good :)
[21:52:56] <dmess> have to pay for some shipping and find her a home yet..
[21:54:21] <archivist> buy a trailor quick...
[21:56:20] <alex_joni> well.. off to bed for me.. good night all
[22:04:29] <dmess> its too high for a trailer
[22:04:48] <dmess> i'll rent a unit locally..
[22:05:14] <dmess> with 600v power in it already
[22:06:05] <archivist> I was just googling to see pics :))
[22:06:25] <dmess> we've already scrapped a $62 k part since this particular machines operation was moved
[22:07:42] <dmess> last time it ran on the ALTERNATE machine was 11 yrs ago... so we TRIED & DIED again...
[22:08:30] <dmess> f/18 e/f nose landing gear piston
[22:14:10] <K`zan> Wow, those carbide blades I got for the recip saw that didn't work worth squat on aluminum are great on steel!
[22:14:42] <K`zan> So, today I hope to get the X axis if not the y axis done too! Fun!