• Nice tabs, this time
  • Through
  • Shifted slots
  • Nice fit
  • Mounted — Yours truly tightened the screw too tightly and deformed the assembly.
  • Swarfyness


Hours 4-8: More micRo testing

We still haven’t received the bits we’re waiting for, but today we at least managed to make some fine cuts. The micRo is excellent and whenever our incompetence gets out of the wait for a few minutes it really shines.

Today we painstakingly milled out a triangle of 20mm HDPE (with a dremel plywood bit with 2 straight shears and 10mm of cutting action). We plowed some furrows into HDPE playing feed rates, cut depths and bits. We found a couple of sweet spots that produced perfect cuts (350mm/minute, 2 flute rotozip bit – yet the tool was about 5cm long so we saw massive deflection in the range of 1-2mm. Useless!). We also cut a really encouraging knife sharp slot into Amfitop (a local dialect of Corian) with the same “useless” plywood bit mentioned above. This was done at 25mm/min rate.

After that we made a sharp small test cut into a 2mm aluminum sheet. We then prepared a small machine part to cut about whereupon the chuck promptly fell out of the spindle with a nasty screech. This has happened to us before and expect to just pound it back on like we did the last several times.

Our toolchain currently consists of TurboCAD for modelling, Cut2D for CAM and Simen’s own GRBL for motion control.

Our time in world as millers now spans all of 8 hours, and we’re pretty happy about the rate that our n00bness is dwindling.

micRo at work

Triangular HDPE

Corian Notch!

micRo w/ GRBL – First Cuts


Last Thursday was spent driving around Oslo buying the equipment to set up our humble workshop in the back room of our office. Today we finally got around to powering up the CNC and making our first few cuts.

Ok so:

1. We’ve never in our lives milled anything
2. Simen has waited a year for his Lumenlab micRo and spent the time building an Arduino compatible motion controller:GRBL

We’re still waiting for proper bits to arrive from Victor Machinery Exchange, but achieved OK-ish results on a cheap cutting board (LDPE) with random bits picked from the bargain bin at our local hardware shop. GRBL worked tirelessly for an entire evening without the need for a single rewrite, cutting both line segment and arcs. We’re pretty ecstatic about this as the code has never been run on an actual milling machine before, just in software simulation. The micRo also seems like a really solid piece of hardware, effortlessly snapping our steel bits when we made an erronous moves with the bit at standstill.

Testing algorithms for stabilizing roller-ball