The air assist was turned down to not spray sand everywhere. Then I simply held the laser on full power. Assuming there wouldn't be much smoke, I hadn't turned the ventilation on.... However there was much smoke. Air assist kept the lens clean, but the nozzle was coated in white soot. The laser wasn't on long enough to make much of a sand ball.

First smokey test

Test two: the ventilation is on! I held the laser on full power for about 20 seconds.

Then I smashed the tiny glass ball with a hammer. The ball is about an 1/8th inch diameter.

Smash!

It seems like the ball wasn't really a homogeneous glass mixture, but just a bunch of sand stuck together.

Now let's try melting a ring.

Melting a ring

This first ring was very thin and very frail. It didn't survive me scooping it up.

Failed ring

Here is the second result at a slower speed and more defocused.

Second failed ring

This is like an F+ in the academic world; failure without total disaster.

NOW I ADD ~20% SODIUM CARBONATE TO THE SAND.

To properly test sodium carbonate's effectiveness, I melted a third ring using the exact same settings as the previous.

Sodium carbonate ring!

Success!!! The sodium carbonate makes a huge difference. This ring isn't exactly strong, but at least it's holding together. Also, there seems to be more homogeneous melting, look at the shiny black areas!

At this point in the experiment, I've realized something is wrong with the laser. The laser cutter is in an incomplete yet functional state, and at some point the current gauge for the laser tube was turned around. This little gauge is an indicator of the laser's power output, and an early way to detect something wrong.

Gauge is on the left.

Anyways, the actual problem wasn't the gauge, but the grounding wire from the laser tube became disconnected from the grounding terminal on the laser power supply. The laser was still operating, since the high voltage was arcing out of the wire into the earth grounded aluminum frame of the laser cutter. This is scary. The extra resistance of the air-gap was causing a drop in laser power, which I did not see since the current gauge was facing backward.

I have no idea how long it's been like this....

Arcing ground wire

Lets try melting another ball.

Look how shiny!

Next test!! Rastering an entire circle!

Melting a sand circle

Yep, this is working so much better than the first tests.

Fancy sand circle

In the end, I made 5 different pieces:

• Circle
• Star
• Lightning bolt
• Giant spiral
• Flying Spaghetti Monster

Conclusion

Yes, I realize this experiment hasn't been a strict application of the scientific method, but I'm like 95% sure that sodium carbonate produces better results. I'm 100% sure a properly grounded laser tube produces better results.

Many people have requested I try this again with with finer/cleaner sand. I'm also thinking about 3D printing with sand in a future project.

The End

That's it.