Coherent Laser Emergency: How an OBIS UV Laser and a Cleaned-Up DXF Saved a Metal Jewelry Order
The phone rang at 4:32 on a Thursday afternoon, which is never a good sign. The voice on the other end belonged to a production manager at a small jewelry manufacturer about an hour outside the city. They had a rush order for custom stainless steel cuffs—the kind of trendy minimal piece that was supposed to be at a weekend trade show. And their laser had just started acting up.
“We’re getting almost no power out of the OBIS,” she said. “The preview looks fine, but the beam barely marks the metal. We’re dead in the water.”
I’ve been doing applications support for laser equipment for eight years, and I’ve handled more emergency calls than I can count in that time. This one, though, turned out to be less about the laser itself and more about a set of assumptions—about materials, files, and what a laser can actually do.
The Setup
The client was using a Coherent OBIS laser for a job they’d mostly done with a fiber laser before. That’s worth pausing on. Everything I’d read about cutting metal jewelry said you need high average power—the more watts, the better. Conventional wisdom holds that if you want to laser cut metal jewelry, you get a 20W or 50W fiber laser and blast through.
In practice, for thin 316L stainless steel cuffs—about 0.5 mm thick—the OBIS UV laser was actually a better fit. Its 355 nm wavelength is absorbed much more readily by the metal surface than the 1064 nm from a typical IR fiber laser. Less heat enters the surrounding material, so there’s less dross and a cleaner edge. At the time, the client hadn’t thought about wavelength at all. They just saw “laser” and assumed a big fiber unit was the only way. That assumption was about to create a lot of unnecessary panic.
The job was to cut a repeating pattern from a DXF file that another designer had sent over. The DXF looked perfect on screen. But when the controller software imported it, the preview showed a clean outline. The problem appeared only when the laser actually fired—the power seemed to drop after the first few passes.
The Diagnosis
Remote diagnostics are always a little tense (especially when the customer has a deadline printed on the wall). We connected via a secure session, pulled the laser logs, and checked the OBIS’s internal power monitor. The laser head itself was reporting stable power. The interlock was clean. The USB connection was solid.
“It’s not the laser,” I told her. That’s where the truth started to shift. People think machine failures happen because of bad lasers or neglected components. Actually, in my experience, most urgent service calls turn out to be something dumber. It’s a bad file. That’s the causation reversal I keep bumping into: the hardware is often fine; the toolpath is broken.
I asked her to send me the DXF file. While I waited, I also asked her to check the focus lens. A routine cleaning was overdue—there was a thin film of oil from the cutting process. That explained the gradual power loss: the lens had become mildly contaminated, but not enough to trigger an error. Still, the slow degradation was a clue. It wasn’t sudden failure; it was a dirty window plus a file with too many overlapping vectors.
When the file landed in my inbox, I saw it immediately. The designer had duplicated the outline layer twice and left one layer with a hidden hatch covering the entire inner area. The laser was being asked to scan the same paths over and over, and the extra hatch was basically turning a “cut” operation into a dense surface ablation. No wonder the material was scorching and the beam seemed weak.
The Fix
By that point, it was 7 PM. The cuffs were supposed to ship Friday morning. I packed a cleaning kit, my laptop, and a spare set of optics, and I made the drive down. Honestly, the drive was more about trust than necessity. When a customer is in crisis, showing up matters as much as the fix.
At the shop, I cleaned the focus lens—that took about ten minutes—and then spent two hours working with the DXF file in their laser-cutting software. I deleted the duplicate paths, turned off the hidden hatch, and moved the cut lines to a dedicated layer with a proper kerf offset. For a UV laser engraver operating in cutting mode, the toolpath is everything. The OBIS could handle the material; it just needed a clean path to follow.
We ran a test cut on a scrap piece. Clean edge, no burn marks, no dross. The second test on an actual cuff was just as clean. The client’s operator watched the whole process and kept shaking his head (I’ve seen that look before—it’s the “oh no, it was us” look). “I’ve been blaming the laser all day,” he said. I explained that the OBIS was fine and the beam was fine; the lens just needed a wipe. But the file was the real problem—a classic case of garbage in, garbage out.
Seeing the same material processed with the UV beam after we’d tried the fiber source earlier made me realize something important. It wasn’t just power. The pulse width, the wavelength, and the way the beam interacts with the metal—that combination matters more than any single spec. Comparing the two runs side by side, the difference in edge quality was obvious. The UV was the right choice for this specific job, not because it was “better” in general, but because it was better for 0.5 mm stainless steel.
The Delivery and the Honest Boundary
The order shipped Friday at 9 AM, about four hours before the trade show vendor had to have it in hand. The client called later that day to say the cuffs sold out at the booth. It was a good ending.
But here’s the part that might surprise you. Later that month, the same client brought in a batch of pure copper pieces and asked if we could cut those with the same OBIS. I told them no. Copper has very poor absorption at 355 nm at room temperature, and while we could try multiple passes and higher pulse energy, it would be slow and likely to produce poor edge quality. If they wanted to process copper reliably, they’d be better off with a different Coherent laser platform—a pulsed fiber laser or a green laser designed for high-reflectivity metals.
It would have been easy to say “sure, our laser can do anything.” That’s the kind of lie that comes back to bite you. I’d rather lose a single order and keep a client for ten years. The honest limitation isn’t a weakness; it’s what makes your next recommendation credible.
What I Learned
Now, after that week, there are three things I tell every customer who calls with a laser emergency:
Check your file before you blame the hardware. Open the DXF in a viewer, look for overlapping elements, orphaned layers, or hidden hatches. Half the time, the laser is working exactly as intended—it’s just trying to follow a broken map.
Know your material’s absorption. Wavelength matters as much as power. A UV laser engraver can be the right tool for thin stainless steel jewelry, but it might be the wrong tool for gold, copper, or anything that reflects UV light. Match the source to the material, not to the hype.
Cleaning is not optional. The contaminated lens in that story didn’t flag an error; it just degraded the beam quietly. A weekly lens check prevents most of these “sudden” failures.
When people ask me for laser recommendations, I don’t start with power or price. I start with the application: material, thickness, edge quality, production volume. There’s no universal “best” laser. There’s only the best match for your specific process. If that sounds evasive, it’s not. It’s the only honest way to sell advanced equipment.
The Coherent OBIS we used that night is still running in that shop. They’ve since cut over four hundred cuffs with it, and they still send me the occasional text to double-check a DXF file. That’s what “coherent laser systems support” should look like—not just when the laser dies, but before the file ever reaches the beam.
Bottom line: if you’re thinking about a laser cutter for metal jewelry, take the time to learn your materials, your files, and your beam delivery. And if a support engineer tells you that a particular laser isn’t the right one for your project, listen. It might save you from a very expensive mistake.
