It can often be interesting to compare the results of two very similar operations with only small changes made between them. In my case, I had been experimenting with etching on the waterjet as part of my residency at Pier 9, and was interested in seeing how different materials responded to the process. You can find a detailed explanation for setting up an etching operation on the waterjet in my Instructable on waterjet etching aluminum.
Step 1: Material Acquisition
I was very fortunate to have access to a waterjet as part of my residency at Pier 9. If you're reading this Instructable, then perhaps you have access to one or are planning to use one in the future. It's worth taking a moment to think about the ways in which this changes the materials available to you. Since the waterjet will cut literally anything, you can consider alternative material sources.
As an example, we made a weekend trip to Urban Ore in Oakland where we found some lovely old slabs of marble for an exceptionally low price. I imagine that because they are salvaged from demolition, they're harder to sell as they would require some amount of re-finishing. Since this would require special tools, it's probably not worth the effort / added cost for most people shopping for stone. If, on the other hand, you have access to a waterjet...
Step 2: Material Testing - the Bad News
Materials are unpredictable. You can have an idea for a project that you'd like to execute, and when you go to create it, the material can behave in a way that makes the project fall apart. Such disappointment awaited me when first exploring the etching capabilities of the waterjet. I had some very specific ideas for things I was hoping to carve into stone, but when I tried to realize them, what I created instead was a bare trace of the image I'd hoped for visible in a patch of textured stone. Not what I had expected at all.
Step 3: Material Testing - the Good News
Material unpredictability can be just the kind of feedback you need in order to create something much richer and more compelling than your original idea. In my case, since the image became unreadable, I started paying closer attention to the texture of the stone to see if I could accentuate it. With further testing, I was getting interesting texture, but also noticing that some severe undercuts were happening as the waterjet was etching the material. This looked promising, so I decided to push the process still further.
Step 4: Tuning Inputs
As I mentioned, the question of how a material will behave when exposed to a particular process is really answered only by exposing it to that process. Through trial-and-error, I found that marble couldn't hold a high level of detail. I responded by reducing the detail in the input image considerably allowing the texture of the resulting etch to stand in for the high level of detail in the original image.
Step 5: Registration
To continue accentuating features like the texturing and undercutting of the surface, I wanted to have the undercuts fan out from the center of the panel. In order to do this, I needed to begin the etching operation in the middle, etch to the end of one side, then rotate the piece 180 degrees, and perform the same operation. In order to make this rotation as accurate as possible, I registered the piece using six pegs (represented by the red dots in the drawing above) into a sacrificial piece of plywood that sat beneath the stone during the etching process.
To create my registration, I cut holes with a diameter matching an aluminum dowel. After cutting the dowel to peg-sized lengths, I inserted them into the stone to lock everything in place.
Step 6: Etching Pass 01
Based on the cut settings I had tested earlier, I programmed the first of two cuts and let the waterjet do its thing. I'd recommend following the guidelines as mentioned in my etching aluminum Instructable for minimizing splash & mess.
Step 7: Flip Stock
With the first etching operation complete, I lowered the water level and began flipping / rotating the piece. I had also accumulated a large pile of garnet to contend with. After rinsing off the excess garnet with a hose, I was able to lift the piece up off its registration pegs. As some of the pegs were stuck, I had to get another loose peg and tap them out with a mallet. Finally, I was able to rotate the piece 180 degrees and align it with the registration holes. Then I re-inserted my pegs, and was ready to begin the second etching pass.
Step 8: Etching Pass 02
With the registered flip complete, the last step was to run the final pass. One thing I would account for were I to perform the same etching again would be the overlap of the two passes near the panel's center. While I did nudge the path start home out a few hundredths of an inch to account for the kerf of the stream of water, the secondary bouncing & cutting etched deeper in this area than anywhere else on the panel.
After the etching operation was complete, I ran a high-quality profiling pass to cut the etching free from the stock material - you can find Instructables that deal with the basic settings required elsewhere.