Very nice! There are guys using other machines and ex$pensive Vectric software who would love to be able to find fault with that work. Congratulations, it's okay to be proud!

 

ewo I would make the step over smaller

 

Well yall, tomorrow is either going to be a day of victory or time to cry for me. See my friend designed a wing that he would like milled out and after a couple days of trying to get the gcode to come out right it finally worked. So tonight we did a test mill see below for the pic and tomorrow will be the full cut. I will make a video and/or take some photos tomorrow. For now enjoy the teaser pic

Attached files

 

BTW ewo those grapes look awesome very impressive.

 

That wing looks like the perfect plot for a hot-wire. Are you listening Mark?

 

I've been trying to get a simple box and carrier cut set up similar to one face of what Jovian did with his double-sided cut. I've come at it a couple of different ways, but ended up with the same error each time.

Basically, after I run Phlat3D, a large (ginormous) surface appears in the SketchUp window. When I try to import the CNC code into USB CNC, I get a file error about a missing % sign. When I look at the CNC file in notepad, it looks like no G-code at all got generated.

I've imported an STL file into SketchUp 8 and imported a .dxf file of lines via 7 to 8 and then made the faces come in. Both version fail the same way.

I'm starting to wonder if I've managed to screw up my SU installation somehow.

Any ideas?

Darwin

Attached files carrier_lite_ply_25_skp8.skp (50.5 KB)Â test_skp_8_stl_import.skp (54.2 KB)Â

 

Darwin,

If you open up the ruby console, you will see the plugin errors out and never completes.

1) You have the safe area box grouped with your model.

2) Don't put you model, touching the safe area border.

3) Don't use sketchup 8, because the "ray test" method is buggy in this new version. Hopefully that will get fixed in the next maintenance version of SU8. Do the previous corrections(1 & 2) and gcode should generate. But you see it is messed up. Use SU7.

-Kwok

 

Thats sweet can't wait to see the final Product

 

I really appreciate the help, Kwok. Sadly, I'm not getting anywhere.

I tried using the advice on both the STL and DXF imported files starting in SU7.0. The Ruby script still locks up.

I then generated a new file using SU alone without any imports for a simple block with a recess on the surface. The Ruby script locks up.

I'm wondering if I'm somehow screwing up where I place the solid. The model is 3/4" thick to mimic the 3/4" EPS I'm planning on cutting. In the Phlat parameters setup, I'm setting the material to 3/4" as well. Is this incorrect.

The basic 7.0 SU model is attached for reference.

Regards,

Darwin Attached files block_3d_test.skp (12.3 KB)Â

 

Darwin,

Yes, you are supposed to set the material height to the same as your model's height.

I seem to have problems with sketchup hanging, too. Doesn't seem to happen if I set the overlap to like 90-100% for that file. Lets see what other's report?

You can try my test file. It seems to hang, too. But finishes in about 40 seconds on my pc:

Attached files my 3d test.skp (28.4 KB)Â

 

Kwok:

Good advice. I downloaded your test file and ran it through the Phlat3D in 7 no problem, no delay and it spat out good Gcode.

I pulled up my STL test and reset the step to 100 percent and it actually ran through to generate code! It was way slower than yours but still finished out.

I then ran the code on the machine and it was working. The 100% step was leaving a thin nib of foam in one direction only, which was kind of cool and kind of annoying at the same time, but it was running!

Then the cog on one of my rollers popped loose and all the fun stopped for the evening.

So, once again, two steps forward, one back. Will have to glue the cog back on before I can re-install the roller and get going again. In the meantime, I'll play with the step size a bit and see if I can use less than 100%.

Darwin

 

Well all I have to say something that we all know the MKIII is freaking AWESOME!! So heres why: For the last two days I have ran the MKIII virtually nonstop for 10 hours each day. Why you may ask. To make a solid two meter wing. Now Hotwire would have been faster yes, but also smellier, the wing would have had to be broken down in two pieces. I couldn't have place channels for the spar, and most importantly I couldn't have tested how large of a piece I could have hollowed out. To give you an idea of times each sides g code took about 2 hours to generate and 10 hours to run. Here are some pics. All that is left is to sand it baby butt smooth and put some wbpu on it.

Attached files

 

Darwin,
I just tried your file and it completed just fine for me. the square you have is 1foot by 1 foot in size so it will take a bit to run. It completed in about a minute for me and used about 600MB of memory. I am sorry about the script seeming to hang on larger files. If you don't click any other program while its running the status bar in sketchup seems to update. If you click anywhere though Sketchup will no longer update itself and I have no way of giving you feedback in the script. What I have used to determine if something is running is the task manager. You will see the sketchup process at a large cpu percentage (1 full cpu's worth) and the memory size changing frequently 60Mb, 90, 150, 200, 100, 50, ...... If the memory becomes stagnat here I generally have assumed its frozen. You will also see a plugin-container process running. If this is all greek to you leave it run for a while and check back. I have found that if Sketchup is really stuck it will throuw an error/crash bug report in about 15 minutes time. Attached files block_3d_test.cnc (50.9 KB)Â

 

Thanks, Jovian! I worked some things out in chat tonight with the guys and everybody's right...it just takes a while to gronk through.

I've successfully completed several runs now, one based on an STL file import and several based on native SketchUp generated surfaces. I should be posting pictures tomorrow of a far less impressive project than your wonderful 2-meter wing. Congrats on that, btw. Awesome test subject for a great tool!

Regards,

Darwin