This is my first project on this platform and thus some tec. related mistakes is to be expected. Ok lets dig in.. First of all I have a CNC that I can use and some glass from a broken window (3mm thick), some programs (Autodesk Inventor, EstlCam, and Makko). I have from a friend of mine (Peter H.) obtained some dentist drill bits with diamond surface (more like embedded - name and type unknown) 1mmØ, syringe (20 mL), playdogh, and tab water. My first job was a slide holder for our PhD student Jakob E.F. - who was kind to make the drawings for me. The *.DXF file (see attachment) was loaded and the CAM code generated with Z steps of 0.1 mm and a total depth of 3.1 mm. Holding tabs 1 mm wide and 1 mm high. The CNC runs at 50% speed (approx 300 mm/min on X and Y) and unknown speed on Z (50 % of max speed), the router is set to 22000 RPM. I have mounted the glass on my CNC bed on top of a plastic bag (water splash expected) as illustrated in [figure 1][1]. Fitted with a "wall" of playdogh (blue) and added approx 30 mL water - covering the glass. Press GO and listen... Nothing? No problems, no broken tools??? Only the CNC milling slowly down into the glass generating a white water slurry ([see movie][2]) After the center part is routed out the water slowly started flowing between the glass and the plastic bag (see movie again). Nothing alarming, just a very slow flow and the water could be removed by the syringe without any panic. The water was manually returned to the pool on the glass. Ok some results. First no damage on the drill bit which can be used again and again. No waster splashing everything was as dry before as after this experiment. On [figure 2][3] you can see the final results in the CNC before cleaning. The "dust" was rinsed away and please note the holding tabs are still intact [figure 3][4]. During the in house clean up with soap the inner part broke off at the holding tabs. and only the item itself is attached the glass ([figure 4][5]). It was difficult to brake the holding tabs of the final part and large chips fell off. The final items back side is seen on [figure 5][6]. It is clear from figure 5 that the glass chipped in small on router bit exit. [Figure 6][7] is for giving a feeling on sizes and to have a final picture. It is clear that this is possible, but also it can be concluded that: X,Y 300 mm/min, 22000 RPM on spindle, 0.1 mm Z-step, 1 mmØ dentist drill bit and "still" water cooling is sufficient for CNC routing of glass. Things to be solved is how to make a hold down system for glass and then just make all sorts of flow reactors, holders, mounts, etc. [1]: https://mfr.osf.io/export?url=https://osf.io/nb8tm/?action=download&direct&mode=render&initialWidth=675&childId=mfrIframe&format=1200x1200.jpeg [2]: https://osf.io/ttqtk/?action=download&direct&mode=render [3]: https://mfr.osf.io/export?url=https://osf.io/upquv/?action=download&direct&mode=render&initialWidth=675&childId=mfrIframe&format=1200x1200.jpeg [4]: https://mfr.osf.io/export?url=https://osf.io/224tv/?action=download&direct&mode=render&initialWidth=675&childId=mfrIframe&format=1200x1200.jpeg [5]: https://mfr.osf.io/export?url=https://osf.io/qc954/?action=download&direct&mode=render&initialWidth=675&childId=mfrIframe&format=1200x1200.jpeg [6]: https://mfr.osf.io/export?url=https://osf.io/6pkvy/?action=download&direct&mode=render&initialWidth=675&childId=mfrIframe&format=1200x1200.jpeg [7]: https://mfr.osf.io/export?url=https://osf.io/zrrwj/?action=download&direct&mode=render&initialWidth=675&childId=mfrIframe&format=1200x1200.jpeg