MightyHat Enclosure scale for Inkscape?

[wile1411]

I tried to give the MightyHat DXF enclosure file to a 3d print / laser cut guy here and hoped he could work it out from there and cut out a box for me.

Apparently he doesn't have AutoCad and was just opening the DXF file in Inkscape to do his prep work on the pattern.
When he opened it, Inkscape asks for a scale and leaving it at 1 makes the enclosure about 3mm high - so that ain't right

Look around, I see that the Inkscape scaling process converts whatever DXF file tried to load into mm/pixels >> Inkscape Forum reference

I assumed it started in Inches scale - am I correct on that?
Reason is if I enter "25.4" as the scaling (1 Inch = 25.4mm)
It seems to be pretty close in size, but I know if I'm wrong, I'll have a piece of ornamental plastic with holes all through it. Of which, I can't use for the mightyhat/raspi project.


37.795275591

[TomWS]

Inkscape DXF import assumes that the file is AutoCAD Release 13 and newer and that the units are in mm.  If the file was generated with Imperial units, then the 25.4 scaling should work.   

Note, however, if, in generating the DXF file any scaling commands were used, then those commands are imbedded in the DXF file and the dimensions on any parts are to the original scale - Inkscape is unaware of this and it creates a very unusual result.  DAMHIKT!

So, yes, 25.4 should work assuming you're not working to super tight tolerances.

Tom

[wile1411]

So, yes, 25.4 should work assuming you're not working to super tight tolerances.

Tom

Thanks - I suppose I can hand file it back if needed and will probably go ahead.

[Felix]

As a tip:

When in doubt, cut it out of scrap cardboard, it is the easiest wasteless free test cut you can make before you start using real material.
The thickness is very close and perfect for test fitting.

[TomWS]

You can use QCAD to convert the DXF file from inch units to metric.  Just open the file, change the units, and save.  QCAD will do the correct data scaling.

This will also convert the DXF file into a version that Inkscape likes.  Apparently not an issue in this case, but it is handy to have around if Inkscape complains about the format from other DXF files.

Tom

[wile1411]

Finishing up this thread - I ended up with three copies, all cut by someone I found here in Canberra lucky enough to own a laser cutter.

All items were cut out with the 25.4 scaling mentioned in previous posts of this thread.

1st enclosure: (Test Cut)
Initial test run was from scrap 3mm plywood. Of course, the first attempt stuffed up due to the pieces not being on the honeycomb plate to support the small pieces. Because of this, items moved during cutting and the process was halted halfway through the cutting of all the small holes. I still got a complete box, just without all the openings. Test fit: pressed together great.

2nd enclosure:
This one was cut from 3mm clear acrylic and went awesome. These are the parts that my MightyHat is currently mounted in. The fit was a little tight on the short end. However, after a little filing it became as good as the SonarMote boxes directly supplied by Felix. The laser owner tipped his hat to the unknown designer as he admired the tolerances built into the box design - so kudos to Felix

3rd enclosure:
I had a backup set of enclosure pieces cut from 3mm acrylic. These are in case I catastrophically screw up and snap any of the pieces from the 2nd enclosure.

I decided that seeing as I now have some mixed media pieces for this box, I needed to do something with it. I took the woodern FTDI end of the enclosure, drilled a few screw holes and open up a bigger hole for a 15x15mm cooling fan for the Pi. After a little bit of research, I found this instructable. A few breadboard tests later, I made the tiny circuit. Then I promptly remade after working out I soldered the transistor backwards 

I still have to shorten the fan wires, but I've now got a CPU temperature controlled fan to blow across the board mounted onto the MightyHat enclosure. The code is set to kick in around 50C, but I'm not heard it activate since I fixed my GPIO conflict issues.
Yay!

[TomWS]

Because of this, items moved during cutting and the process was halted halfway through the cutting of all the small holes. I still got a complete box, just without all the openings.

Tip: ALWAYS arrange the GCODE so that the openings (Holes, etc) are cut before the outline.  This way the outer edge of the blank can be supported and nothing important will move until you make that last cut.

[wile1411]

Tip: ALWAYS arrange the GCODE so that the openings (Holes, etc) are cut before the outline.  This way the outer edge of the blank can be supported and nothing important will move until you make that last cut.

Yeah - I got the sense that the guy with that laser had it more as a hobby than professional. Thanks for the tip though - I'll mention that next time I get something cut.

[Felix]

The laser owner tipped his hat to the unknown designer as he admired the tolerances built into the box design - so kudos to Felix

Awesome
Thanks for the feedback!

Yeah I did quite a bit of experimenting and messups before it came out just right. I think as long as the CO2 laser used has about the same plasma kerf/angle the tight fit should come out about the same. Usually the acrylic thickness tolerance (which in my experience has been at least +/- twice the claimed tolerance on many occasions - source is inventables.com ) gives me more trouble for these boxes than the small tolerances for a tight fit design.

Acrylic is brittle and can snap easily. It's however the easiest material that I could find to make such custom size boxes, with a decent aesthetic look, in various types of finishes and transparencies.

[wile1411]

hah! I only recently learnt about measuring a laser cut kerf from a video by ‎Matthias Wandel. Fascinating stuff and looks like hours of fun to dial in the settings you want.

I was going to load some pictures on the finished enclosure. Did you want them here or in the projects thread?

[Felix]

Kerf might not always so easy to measure. The laser does not instantly burn through the material, and as it moves (perhaps not always in straight lines) the kerf will not have a straight down, nice cone shaped shape, but take weird angles. Also a big challenge is to focus the beam in the middle of the material. But we're going to start splitting hairs by trying to get scientific about this.

I was going to load some pictures on the finished enclosure. Did you want them here or in the projects thread?

If there is not another thread, then here is fine, since the title is relevant.
Please use imgur or a similar service and post links to photos here, if you're not familiar with that then attachments are fine, I can do that part later.
Thanks!!

[wile1411]

Images for the printed enclosure and fan mount.
I've still got to make the bolt shorter and trim the fan cable to something reasonable.

[Felix]

Looks awesome, great job!