On Wednesday 05 April 2017 03:13:07 Erik Christiansen wrote: > On 04.04.17 12:29, Gene Heskett wrote: > > On Tuesday 04 April 2017 06:04:50 Erik Christiansen wrote: > > > On 04.04.17 05:00, Gene Heskett wrote: > > > > Those scraps of that blueish foam have all been binned or used years > > ago. And Lowes no longer carries that same board in 2" R22 > > thickness. The current product the last time I looked is a white, > > larger cell product and only about R20 because of that, but its the > > same $35 & tax a 4x8 foot sheet. How it would cut with a hot wire > > would be TBD. > > Should be good. It's when resorting to a cold sharp knife that the > "melded bean bag fill" crummy foam crumbles. > > ... > > > How hot does the hot wire need to be? > > Just go by feel. This one suggests 600°F (that's 315°C, which sounds > like a good starting point): > https://www.youtube.com/watch?v=3GWzHb4Hd8Y > > but with a 555 & MOSFET, you can PWM your way to happy cutting from > 12v with most bits of recycled nichrome wire even a foot long. > > Some make the frame from wood - that'd be more rigid than his. > (I've welded a couple of small bits of RHS together, added a baseplate > to screw to an old chipboard kitchen sink cutout (melamine topped), > and need to rout a slot from the edge to the base of the wire, for an > Al T-slot, so I can slide a vertical pin back & forth for setting > radii for cutting disks, rings, and cylinders. > > > Seeing as how thats best jiggered up as a wire support frame I could > > stick in a vise on the g0704's table and rig some sort of a sheet > > gripper leaving a cutaway, for the hot wire to move within, attached > > to the chip pan, if I get it rigid enough to keep its place as the > > wire moves, I could probably just write gcode to drive the cutters > > path. Where it needs a lid like the outside face of a belt cover, > > just cut the outline out and glue it on. > > Takes a while to build, though. Sketching the outline on the back of a > cornflakes packet, cutting it out with scissors, running around it > with a ballpoint pen on the foam, then carefully following it by hand > with the hot-wire cutter, is quicker. (I don't know of a good > temporary adhesive for sticking the template on the foam without > tearout on removal.) > > > But, I think buying the printer would get me a nicer looking belt > > cover. > > The printed product could perhaps be used directly, instead of then > making a casting, and machining it where necessary? > > Erik
That would be the intention. Barring an errant hammer blow to break it, probably loose by snapping off a mounting tab, a lot of swarf could even be hot enough to stick to it, but would brush off when cooled. Without the spindle turning but the motor running at 30hz, it has set out there and run all night, no joint errors. So I next write an infinite loop, moving each axis back and forth half an inch at about 20 ipm. I have got to find the source of the joint errors. They hit random joint numbers, often both joints when only one joint is moving. I am convinced its a comm error as I have seen the tach dial jump to 400 or 500 rpms when its not turning. Not often but thats also a joint error for both joints. And since rockhopper is the only tool I know of that will output the signal path so that can be compared to the addf order, I just tried to install that, but apt couldn't find it in the pi's repo's. :( Time to print out the hal file and trace thru it, putting a number at each addf as the signals make their way thru the modules loaded. Thats getting complex as the hal file, with all these jog-wheel additions, is 19 landscape printed pages now. In the finished trace, the numbers in each thread should be in order. I suspect the servo-thread is not in order so I'll trace that first. And the spi buss is so timing critical I can't even hang a 10x scope probe on the clock signal while lcnc is booting up. I bought 3 of those 40 to 26 pin boardsfrom OSHPark, but according to the xray printouts, the 4 pins of the spi buss, 3 with the term r's are the only pins cross connected! No other pins are connected to anything except the data return from the 7i90! So I haven't built that, but made my own on perf-board, connecting all the other ground pins etc that need connection. Maybe thats a ground loop? But I figured with all the other stuffs plugged into the pi that are grounded to it, it needed a good solid interconnecting ground to the 7i90. Whatever, the comm link, using the rpspi.ko driver is certainly being a problem child. So today I will cut off the long power cord for the pi, which bypasses the 7i90 power plug and goes past it about 7" to a terminal strip I mounted as a 5 volt distribution point and make the pi powered as an extension of the power rail driving the 7i90, which will shrink the size of the ground loop by half or more. And I am not impressed with the molded on ferrite choke, so that will get removed in an effort to reduce the impedance of the ground between them. I can always add it back as I have a small bag of them yet. > ---------------------------------------------------------------------- >-------- Check out the vibrant tech community on one of the world's > most engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
