Wow, two months since my last post. I am not sure if I have been too busy to get down to the shop, or too lazy. I always seem to have things to do… I should be ramping my posts back up soon. I have a backlog of projects to get to.
This project was an interesting challenge. My friend Doug has a cabinet shop and he has two very large CNC’s. They are probably quite capable of doing this type of small, detail work, but he hasn’t had the time to learn how to do the intricate stuff. His machines are usually busy cutting out flat patterns in large quantities for big projects.
This piece comes from an apron off of a table. There is another one on the opposite of the table that was damaged and it needs replacing. If you look closely at the apron, you can see that the image was stamped into the wood, instead of being carved. The problem with reproducing the image is trying to reproduce the effects of stamping. The wood gets kind of mushed over at the edges.
Doug asked if I wanted to take a shot at it. I had no idea if I could reproduce it, so I said sure…
I brought the image into my CAD software and traced the flowers and the scrolling below.
For the flowers, I made a couple different 3D models to test, then imported them into my Carvewright software.
After a little manipulation, I created my first test carve. It ran for about two hours on the CNC, and this is what I got.
Not too horrible, but the leaves were too skinny and the entire thing was way too deep, even though it was only 1/16″ deep. The scrolling, on the bottom looked pretty good, but it looked too sharp. The stamped one had a kind of rounded transition over into the groove. I had used a 60 degree Vee bit, so I switched to a wider, 90 degree bit and did not go as deep, on the second try. I also switched to a different 3D model of the leaves and cut it only 1/32″ deep. I wanted to also try to achieve the curves and grooves on the sides, so I spent a little while fooling with some of the features, in my CNC software, that I rarely use. I did find a way to create the curves and the grooves, but I couldn’t make the grooves follow the curve. Oh well, That was something I could do by hand.
I ran the second carve and it was way closer this time. Still a little sharp, but it was much better.
I clamped the piece in a vice and chased the grooves around the curve with a half-round modelers’ rasp. I was fairly happy with this prototype, because it matched the shape of the original, pretty well. I sanded it down and gave it to Doug to review.
He thought it was good, but wanted a few changes. First, the flat grain was too porous, so some of the detail on the leaves was lost. So we switched to quarter-sawn oak, for the tighter grain. Next, my grooves started too abruptly next to the frame around the leaves. I moved the grooves away by about 1/4″ to allow Doug to ease into the groove with a gouge, to better match the original. The third problem was the curves. Even though they matched the apron piece he gave me, they didn’t perfectly match the curves on the opposite side of the table. So much for repeatable manufacturing… He asked me to just remove the curves and he would belt sand the piece to match. Everything else would work just fine.
I fired up the carver for a third try, and here is what I got.
To clean up the fuzzy grain, from the carve, as well as to take the sharper edges off the carve, I sanded the surface with my mop sander. It smoothed the edges a fair bit and it looked a lot closer to the stamped version.
Here is the first and the last carve, side by side, big difference.
I will get this piece to Doug, and hopefully he will send me a picture of it when he finishes it and re-attaches it to the table.
This was an interesting project that probably took me 12 or 13 hours overall, but I learned a few things, so it was worth it. Oh, and I will get paid for this one, so there will be a new tool or two in my future…