Transparently-thin Granite LED Lamp

I got a beautiful piece of rose granite from construction junction here in pittsburgh, and noticed that it appeared to have such clear quartz crystals in it that it might transmit light if cut thin enough.

I cut some wafers on the water jet, and indeed, it could be cut transparently thin!

I set out to see if I could make a monolithic (truly "of a single stone") LED lamp, which had LEDs in cavities cut WITHIN the stone, shining out of transparently thin walls.

It is important to experiment with your particular material. Granite is brittle, and comprised of different crystals (feldspar and quartz, mainly) that can be broken apart from the incredible pressure of the water jet. The cutting behavior probably will vary between types. It also often has cracks. When 60,000 psi water finds a crack, it will inflate it, and your stone may split. This is an unavoidable risk of working with stone in general, and especially with pressurized water jets, so moderate your expectations, and be prepared to tolerate some breakage.

My granite was 1.25" thick. I made a series of cuts across a small bar of material to evaluate cutting parameters. Obviously, cutting at the slowest tolerable speed will give the highest quality / smoothest cut face. I got great smooth texture on the 1.25" deep cuts at settings that produced 1 inch per minute. You might not be able to tolerate cutting slow though, because water jets are expensive to operate. Techshop Pittsburgh charges $3/minute. You may find a commercial shop to do the job cheaper than you can do it yourself at such rates, as my friend did with one of his small jobs. (The convenience of designing and cutting something within the same hour though is wonderful, but inflated operating costs like this definitely inhibit extensive experimentation).

I successsfully cut wafers as thin as 0.10" thick which were nicely transparent. You can certainly cut wafers even thinner than that, but I was satisfied with this thickness.

The greatest challenge with cutting small parts on the water jet is preventing the part from falling into the water jet pool. The water jet's pool (at least ours at Techshop Pittsburgh) is a murky abyss which, though only 30 inches deep, is effectively bottomless, as regards the irretrievability of things dropped in. The support grid resembles a portcullis of pestilential rusty dagger blades pointed upwards, and the tank's contents include a thick mud of powdered garnet and cutting debris, with a dubious ecosystem of microorganisms subsisting on this curious brew. Because the water jet cuts through *everything* - including it's own table - it can be difficult to not lose small parts when they are severed, and pressed down by the pressure of the water.

To prevent losing your part, you can do several things:

• support the part on a thick stack of plywood - 2 or 3 inches thick. The part is cut normally, and as it is severed, the wood plugs cut beneath it tend to jam, preventing the part falling through. Also, the wood blocks float, further resisting the part falling through (presuming your water level is high enough). Also, the water jet beam diverges and warbles downstream, and if the wood is thick enough, it will sometimes preserve a network of connections between the plugs and the surrounding wood, supporting your part.

• you can also hot glue a string to to your part, taking care that the string does not get severed as the contour of the part is cut.

I opted to make a triangular lantern cross section, because I could have thick corners with holes for threaded rods, with a flat on which I could attach LED strip, and the LEDs would shine onto the opposing flat faces. Any odd-number-of-sides polygonal cross section would have that property, but a triangle is the simplest odd-sided polygon. Attached is a .DXF file of the cross section of the lamp I made.

Three one-inch cut-lengths of LED strip were adhered to flats inside the granite, wired in parallel, and energized.

At least, I intended to use threaded rod. However, the water-jet cut holes were not smooth enough to admit threaded rod. They were still able to be laced together with string.


That completes the project. I'd love to experiment further with water-jetting granite though, and in particular I'm excited to see if I can pierce larger thicknesses, with a five axis water jet, and make fire sculptures consisting of stone through which gas ports have been bored.