| 00:29.32 | *** join/#brlcad jarray52 (~solargrin@unaffiliated/jarray52) |
| 00:31.27 | jarray52 | What software tools do people typically use to turn BRLCAD designs into gcode? |
| 02:30.26 | bhinesley | starseeker: I am |
| 03:23.38 | brlcad | jarray52: people typically don't, but I have heard of some folks using one of our exporters (such as g-stl) |
| 03:23.53 | brlcad | since most tools that'll export gcode will import stl |
| 03:24.15 | brlcad | it's one of the most stupid / worse formats to ever use, but it's pervasive |
| 04:06.25 | *** join/#brlcad Stattrav (~Stattrav@61.12.114.82) |
| 04:06.25 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 05:00.19 | jarray52 | brlcad: Which is one of the most stupid/worse formats to ever use? gcode? What's the alternative for creating toolpaths for machining? Is there a reference you could suggest on toolpath formats? |
| 05:39.23 | starseeker | I think he ment stl |
| 05:52.47 | *** join/#brlcad simion314 (~quassel@92.83.115.186) |
| 07:07.37 | *** join/#brlcad simion314 (~quassel@92.83.115.186) |
| 07:07.47 | *** join/#brlcad andymanak (~andy@c-24-130-199-23.hsd1.ca.comcast.net) |
| 07:08.00 | *** part/#brlcad andymanak (~andy@c-24-130-199-23.hsd1.ca.comcast.net) |
| 07:11.49 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) |
| 10:56.20 | *** join/#brlcad simion314 (~quassel@92.83.115.186) |
| 12:14.08 | *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ) |
| 12:14.43 | *** join/#brlcad simion314_ (~quassel@92.83.115.186) |
| 12:18.34 | *** join/#brlcad simion314 (~quassel@92.83.115.186) |
| 12:25.37 | *** join/#brlcad Stattrav (~Stattrav@61.12.114.82) |
| 12:25.37 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 13:34.28 | brlcad | meant stl |
| 13:45.35 | jarray52 | brlcad: What's the best format in your opinion? |
| 13:48.39 | jarray52 | What are typical toolchains used with brlcad to get manufactured parts with something like linuxcnc? |
| 13:52.33 | starseeker | hmm! http://news.cnet.com/8301-13772_3-57384166-52/smithsonian-turns-to-3d-to-bring-collection-to-the-world/ |
| 13:53.03 | starseeker | jarray52: it's not common enough for there to be a "typical" toolchain |
| 13:54.33 | starseeker | this may be of interest: http://gcam.js.cx/wiki/Main_Page |
| 13:57.22 | jarray52 | starseeker: What do people do with BRLCAD models? Presumably, the models are eventually brought into the real world, right? |
| 13:57.49 | starseeker | generally they're raytraced |
| 13:57.58 | starseeker | used for analytical work |
| 14:00.17 | jarray52 | starseeker: after the analytical work is done, are the parts and models redone in another CAD program to make the parts? |
| 14:01.23 | starseeker | BRL-CAD doesn't have much to do with the production of parts - it's more common for models to come from other CAD systems to BRL-CAD at the moment |
| 14:03.27 | jarray52 | starseeker: So, it is better to use another CAD system to design the parts and then import into BRL-CAD? |
| 14:03.56 | starseeker | if your primary purpose is to manufacture them, that's probably simpler at this stage |
| 14:06.04 | starseeker | jarray52: you might try the following |
| 14:06.06 | jarray52 | I would like to model and raytrace in order to generate 3D animations of the device(similar to a lawn mower engine). Then, I'd like to take the parts and make them. |
| 14:06.32 | starseeker | use the stl export from BRL-CAD, then take a look at http://reprap.org/wiki/Slic3r |
| 14:06.47 | jarray52 | the parts are metal |
| 14:06.48 | starseeker | or rather, http://slic3r.org/ |
| 14:08.26 | starseeker | another possibility: https://github.com/timschmidt/repsnapper |
| 14:09.12 | starseeker | 3D animations? You mean, model the working of the engine? |
| 14:09.34 | jarray52 | I'm looking for something that could eventually run on something like linuxcnc: http://www.linuxcnc.org/ (EMC2) |
| 14:09.48 | jarray52 | Yes. Model the working of an engine. |
| 14:10.23 | starseeker | that requires integration of something like Bullet - there has been a little work done towards that goal, but we aren't there yet |
| 14:14.38 | jarray52 | starseeker: I'm just planning on generating frames with a shell script and then putting them together with mencoder. |
| 14:14.51 | starseeker | ah, that's different |
| 14:15.24 | starseeker | I thought you were wanting to do something like set one gear spinning and have the other interlocked gears follow suit |
| 14:15.42 | jarray52 | Yes. I plan to do that. |
| 14:15.56 | starseeker | we don't currently support that, to the best of my knowledge |
| 14:16.11 | jarray52 | At every time step, every object will be rotated or translated. |
| 14:16.32 | starseeker | you would have to do all those rotations and translations manually or with a script |
| 14:16.42 | jarray52 | Right |
| 14:16.52 | jarray52 | How else could it be done? |
| 14:17.44 | starseeker | Bullet is a physics engine - it can simulate the interaction of physical components |
| 14:17.59 | starseeker | http://bulletphysics.org/wordpress/ |
| 14:19.08 | starseeker | the code lives here: http://code.google.com/p/bullet/ |
| 14:21.05 | starseeker | the work done so far is documented here: http://brlcad.org/wiki/User:Abhijit |
| 14:21.55 | starseeker | (see, for example, http://www.youtube.com/watch?v=SByoQQStH2s) |
| 14:22.16 | starseeker | the box case is the trivial one - non-trivial geometry gets harder |
| 14:23.00 | starseeker | anyway, for now you'll have to do the steps manually, as you originally thought |
| 14:24.05 | starseeker | that leave 3D printing, and I'd suggest doing some tests with g-stl and slic3r to see if the results are usable |
| 14:26.16 | starseeker | jarray52: you might also see if http://heeks.net/ can be of any help (I don't know much about it myself) |
| 14:27.35 | starseeker | makes a note to himself to checkout http://www.graphics.stanford.edu/software/scanalyze/ someday... |
| 14:31.01 | jarray52 | Things move in a fairly precise way. It's hard to imagine a physics engine would be sophisticated enough to figure everything out on its own. |
| 14:31.44 | jarray52 | starseeker: What about generating parts in the salome platform and then importing to BRLCAD? |
| 14:33.08 | jarray52 | Do dxf exports from BRLCAD work well with dxf2gcode? |
| 14:39.26 | archivist | jarray52, there is a lack of good opensource cam, a number of projects are needing help |
| 14:40.39 | jarray52 | archivist: Is there a commercial product that integrates nicely with opensource CAD such as BRLCAD or SALOME? |
| 14:41.59 | archivist | commercial cam will take all the money you have given chance |
| 14:43.46 | jarray52 | archivist: There isn't much choice is there? |
| 14:45.32 | archivist | depends what you want to do and the machines you will be using too |
| 14:46.30 | archivist | lathe/ normal 3 axis mill, then there is some stuff, 5 axis mill then not a lot |
| 15:05.34 | jarray52 | archivist: What about 4 axis mills? |
| 15:06.15 | jarray52 | archivist: Any suggestions for lathe/normal 3-axis mills? |
| 15:16.55 | archivist | just re enabled a web page that points to a number of cam packages http://www.os-cam.org/index.php |
| 15:17.44 | archivist | see also http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Cam |
| 15:30.03 | jarray52 | archivist: Which of these projects would you regard as the most mature? |
| 15:36.25 | jarray52 | archivist: Thanks for your input. |
| 15:36.40 | jarray52 | starseeker: Thanks. |
| 15:37.12 | archivist | not sure I would pick one over any other, depends on job |
| 15:37.39 | jarray52 | archivist: Is PythonCAM a good starting point for engine parts? |
| 15:37.46 | archivist | but I hand code most of the time |
| 15:38.10 | jarray52 | archivist: I'm thinking this might be the best option. |
| 15:38.28 | jarray52 | archivist: What does your work flow look like? |
| 15:38.49 | jarray52 | archivist: design on BRLCAD and then hand code the gcode to run on EMC? |
| 15:39.47 | archivist | mostly gcode to part, as lately working on standard parts |
| 15:40.20 | jarray52 | gcode to part? |
| 15:40.33 | jarray52 | you mean part to gcode? |
| 15:40.51 | archivist | no I mean no drawing, |
| 15:41.16 | jarray52 | drawing to part using gcode |
| 15:41.22 | jarray52 | ? |
| 15:41.31 | archivist | NO drawing |
| 15:42.07 | jarray52 | BRLCAD to part using gcode? |
| 15:42.08 | archivist | I am scripting in gcode to make gears |
| 15:42.27 | jarray52 | that makes sense |
| 15:42.50 | jarray52 | You are using a scripting language to generate the gcode, right? |
| 15:43.07 | archivist | no, I write the gcode directly |
| 15:43.56 | jarray52 | Oh, you look at a part and then write the gcode. Got it. |
| 15:45.43 | jarray52 | Do you run the gcode on EMC linux? |
| 15:46.02 | archivist | yes |
| 15:46.45 | jarray52 | thanks |
| 15:46.51 | jarray52 | I'll give that a try |
| 15:47.21 | archivist | emc got renamed recently #linuxcnc |
| 19:49.02 | *** join/#brlcad simion314 (~quassel@92.83.115.186) |
| 21:12.21 | *** join/#brlcad pawleeq (~pawleeq@212-96-188-229.cust.selfnet.cz) |
| 22:23.31 | CIA-128 | BRL-CAD: 03erikgreenwald * r49578 10/brlcad/trunk/src/rttherm/CMakeLists.txt: swap order of libged and libfb, ubuntu seems to care |
| 23:42.05 | CIA-128 | BRL-CAD: 03tbrowder2 * r49579 10/brlcad/trunk/doc/docbook/articles/it/images/ (6 files): run images through gimp for use in pdf |
| 23:45.28 | *** join/#brlcad bhinesley (~bhinesley@adsl-99-125-82-224.dsl.bkfd14.sbcglobal.net) |