Spent the weekend testing ways to get that jeweled finish on steel

Nov 18, 2026 Last reply: 1 hour ago 6 Replies
Jump to replies (6)

So I've been wanting that overlapping swirl pattern on a steel tabletop I'm building, and I finally sat down and experimented instead of just guessing. Tried a flap wheel first, total mess, the swirls were too big and uneven. What actually worked was a die grinder with a small sanding drum, working in overlapping spots across the whole panel like mowing a lawn.



The trick is keeping the pressure and contact time the same for every spot, otherwise the pattern looks blotchy. I marked out a grid with soapstone and just worked row by row. Took way longer than I expected but it came out surprisingly uniform once I got a rhythm going.



One thing I noticed that matters for anyone attempting this, the edges need to be done before the swirls or they catch the drum and tear the pattern up where they meet. And a clear coat on top flattens the shimmer a bit, so the raw metal look is actually better than I thought it'd be.



Not perfect by any means but for a garage build it reads just like the commercial stuff. Way cheaper than I assumed too.


I did the same swirl finish on a steel coffee table last winter and landed on almost the same method, just with a 2 inchRoloc-style sanding disc in a right angle die grinder instead of a drum. The disc gave me a slightly flatter swirl but it was way easier to keep consistent because the whole face stays flat against the metal. The drum tended to rock on me and I'd get a deeper bite on one side of the spot.

The soapstone grid is the real secret, I agree. I used a permanent marker on some scrap first to dial in my dwell time, like maybe a second per spot, then counted in my head for the real panel. Sounds dumb but it worked. I also learned to keep the grinder running at a steady rpm rather than feathering the trigger, because the slow spots cut differently and show up as darker circles. One thing I'd add on the clear coat question. I tried a matte lacquer on my first attempt and lost most of the shimmer just like you said. What I ended up doing was a very light coat of wax, the paste kind, applied thin and buffed off. Keeps the rust off, still looks raw. My table lives in the kitchen and has held up fine for about a year now. If it were going outside I'd probly want something tougher.

I wound up doing a swirl finish on a steel bar top a couple summers back with a much more low tech approach. Regular angle grinder with a twisted wire cup brush, light touch, just walked it in overlapping circles freehand. The wire brush leaves a softer, foggier swirl than an abrasive does, and the inconsistency reads as character rather than blotchiness as long as your overlaps are generous. Slower on the arms though, my shoulder was sore for days after a full sheet.

On the clear coat I had the opposite result from the OP. I sprayed mine with a matte enamel and it yellowed slightly within a year near the window. The second top I left bare and just wiped it with oil occasionally, and it darkened evenly and honestly looks better with age than the coated one ever did. On edges, I did them last rather than first and got away with it by running a coarser pass over the whole panel including the borders before the final swirls, so any catch or dig got blended out by the next pass. Probably depends how square your corners are and how aggressive your drum is.

The grid marking is smart and I regret eyeballing mine. There is one corner of the bar top where you can see the rhythm change from when I got tired and started overlapping differently.

Agree with counting the dwell time, that was the piece I was missing on my first go. Mine came out blotchy untill I started doing the same thing. My trick was humming a song I know instead of counting, the rhythm was steadier than me counting in my head, my counts always crept longer as my arm got tired. Second tip I'd pass along is wipe the panel down between every few rows. Dust collects in the swirls and the next spot cuts differently over it, shows up as faint streaks under a light.

One variable I never see discussed is drum diameter versus swirl scale, because the spot size scales directly with the contact width of the abrasive, so a half inch drum gives you tight little rosettes while a one inch drum reads almost like a checkerboard once you overlap them. I found that out the hard way on a steel side table where I switched drums halfway through and had to blend the two zones, which never quite disappeared even after going back over the whole panel. Also worth doing a test coupon from the exact same stock as the panel, since hot rolled and cold rolled respond differently to the same grit and dwell time.

Been lurking here a while, this might be my first actual reply. I'm planning the same kind of finish but on a thin aluminum sheet for a desk inset, and my worry is heat. Every test spot I've done so far on scrap has warped the panel a little, probably because I dwell too long in one place. Did anyone here try this on thinner stock and have it stay flat? Wondering if I need to clamp it down to something heavy or just work faster with a lighter touch. Also curious if aluminum swirls read differently than steel since it's so much softer.

Update: finished the whole top and it held up better than I expected. Ended up trying the humming trick from the rhythm reply and it really did beat counting, my dwell time stopped creeping once my arm got tired. Also went back and redid a section with a lighter touch after reading about the wire brush approach, came out foggier in a good way. Garage build done.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required