About Me

My Photo

I previously worked on Virtual Reality and other hardware at Valve.  I currently work at Google[x].

Prior to starting at Valve, I built computer peripherals such as keyboards, mice, and joysticks that were designed to be used inside MRI machines.  My company, Mag Design and Engineering, sold these devices directly to researchers at academic institutions who used them to publish scientific papers in peer-reviewed journals.

After work, I spend time on many different types of projects that usually involve circuit design, machining, material selection, and general fabrication/hacking.  My favorite place to be is my home workshop.

ben dot krasnow at gmail




Friday, August 8, 2008

CNC milling glass plates and mirrors

I searched the web for quite a while, but came up empty-handed when looking for information about cutting glass plates with a milling machine. I was fairly certain this would be possible since stained-glass craftspeople routinely use a diamond grinder to quickly remove material from colored glass pieces over 1/8" thick.

I started experimenting and found it quite feasible to use a Computer Numerically Controlled (CNC) milling machine to cut glass plates. I have cut many different thicknesses of glass from .075" thick first-surface mirrors and IR beamsplitters to 1/4" thick frosted glass for art projects.

Here are the critical numbers:
For thin (<.125" thick) glass, I use a .115" diameter diamond burr spinning at 2500 RPM, fed at 1 inch per minute.

For thick (>.125") glass, I use a .25" diameter diamond burr, spinning at 1500 RPM, feed at .5 to 1 inches per minute.

Both situations require flood coolant.

The real trick is to find a suitable clamping method to attach the glass to the milling table. I have settled on edge clamps that exert very little downward force on the glass, but provide a sturdy edge for the glass to butt up against, and also keep the glass from being being pulled upward from the table. Here is a picture of an edge clamp. It was made by passing a piece of thick acrylic over my table saw blade. The resulting dado cut has a peak down the middle, which is actually pretty useful, since I can use a hand file to quickly change the exact height of that peak. I like that the clamp pinches the glass ever so slightly, but exerts the vast majority of clamping force to the waste board, which is also a thick piece of scrap acrylic. Note: Do not use wood for a waste board. It will change shape when it gets wet from the cutting coolant, and will crack the glass!

I used two edge clamps like this to hold a mirror down to a 1/2" thick piece of acrylic waste board

Here is a picture of the finished mirror:

A closeup of the edge:


  1. This is a great help, thank you!

  2. This is awesome information. Thank you.
    I wonder if vortex tubes might be an easier solution to the cooling requirements than flood cooling.

  3. what cutting depth at those speeds

  4. bido, this mirror is about 1.5mm thick, and I cut it in one pass.