The Stone Engraving Job That Cost Me $11,000 — and the Laser Class Checklist That Fixed It

In May 2021, I said yes to a job I wasn't equipped to do. It seemed easy — a single granite plaque, one family name, two dates. I had a desktop xTool laser engraver and a vague sense that lasers could mark stone. That decision ended up costing me about $11,000 over the next four months, and it taught me more about laser classes, wavelengths, and beam quality than any YouTube tutorial I'd ever watched.

At the time, our workshop was small: a five-person team doing custom fabrication, signage, and small-batch metal parts. The xTool was a recent addition, there to test whether we could start taking on engraving work without a five-figure investment. In other words, it was an experiment — and I let a customer's deadline turn that experiment into a commitment.

Mistake #1: I trusted the marketing claims

A local funeral home needed the plaque engraved with a family name and dates. Simple, one piece, $850 plus a rush fee. We had the xTool D1 (10W diode version) in the shop, mostly for wood signs and leather dog tags. I'd seen videos of diode lasers making marks on stone and tile. So I said yes.

Here's the thing nobody tells you when you're starting out: "engraves stone" and "engraves stone well enough to hand to a paying customer" are two very different things.

I tested the xTool on a scrap piece of polished granite. The result was a light, chalky, uneven mark that looked like someone had wiped a dry-erase marker across the surface. Polished granite is primarily quartz and feldspar, and a 455nm blue diode beam reflects off those minerals instead of coupling into the material. The mark wasn't even. It wasn't dark. It wasn't something I could deliver.

What I mean is: the material dictates which laser wavelength you need, not the other way around. Wattage alone tells you almost nothing. The xTool D1 is 10 watts, but 10 watts at 455nm and 10 watts at 1064nm behave completely differently on stone, metal, and wood.

Mistake #2: I bought a 20 watt fiber laser in a panic

I had six days to deliver the plaque. Shipping in a replacement would eat four of them. So I did what panicking people do: I searched "stone laser engraving machine" on a forum, found a thread claiming fiber lasers engrave anything, and ordered a 20 watt fiber laser. The product photos showed a tombstone being engraved. Good enough, right?

Wrong.

A 20W fiber laser runs at 1064nm, and most granite is more or less transparent to that wavelength. The light passes through the quartz and feldspar instead of ablating the surface. A fiber laser is fantastic for metal marking — I still use one daily for nameplates and tools — but it isn't a general-purpose stone engraver. For stone, you usually want a CO2 laser at 10.6µm, or a marking compound that absorbs the 1064nm beam and fuses a permanent dark mark into the surface. I knew none of this.

(Should mention: the money wasn't wasted entirely — the fiber laser is one of the best tools in the shop. The waste was in thinking a single 20W box could replace a proper stone engraving setup. It can't.)

This is also where I learned the difference between a "laser cutter" and a purpose-built laser cutting system. A machine like a Coherent laser cutter includes the beam source, the motion control, the enclosure, and the software — engineered as a system. A bare laser module is just a component. I bought a component and expected a solution. My mistake, not the machine's.

What I didn't know about laser classes

This is the part that almost cost me more than money.

The fiber laser arrived with a bright yellow label: "Class 4 laser product." I knew Class 4 was dangerous, but I didn't know what it required in a workplace. Under IEC 60825-1, Class 4 is anything with accessible output above 500mW. That includes nearly every professional laser system — including the Coherent Verdi 5W 532nm laser class, a continuous-wave DPSS laser used in scientific research. If you've looked up "Coherent Verdi 5W 532nm laser class," the direct answer is: Class 4. And it's not an engraving tool at all — it's a precision light source for photonics labs.

I was running a Class 4 fiber laser on a folding table with no enclosure, no interlocks, and no safety glasses. In the US, laser products fall under FDA regulation (21 CFR 1040.10), and ANSI Z136.1 covers safe workplace use. I should have figured this out before unpacking the crate. Instead, I spent $1,300 on an enclosure kit and a pair of 1064nm laser glasses in the middle of the project, then two more nights fixing the interlock wiring when one of the switches broke on the first day.

To put it plainly: a Class 4 laser can start a fire, can burn skin, and can damage eyes from a scattered reflection off a metal surface. Working with one without a proper enclosure is like doing electrical work without turning off the breaker.

If you're asking "what laser class is this?" — you're asking the right question, but for the wrong reason. Laser class doesn't tell you what the machine can do; it tells you how you need to contain and shield the beam. Confusing safety classification with capability is how someone accidentally blinds themselves.

What finally turned it around

After the dust settled, I read the actual documentation instead of skimming forums. Three ideas changed how I think about lasers:

1. Material first, laser second. Every material has an absorption spectrum, and the laser wavelength has to overlap it. Wood and leather absorb visible and IR differently; metals need 1064nm (or green/UV for fine work); stone usually needs CO2 or an absorbing coating. Start with the material, then pick the wavelength in its absorption range.
2. Beam quality matters as much as power. The M² factor tells you how close the beam is to a diffraction-limited spot. A 20W laser with a poor M² can't focus to a fine enough spot to do detailed engraving, so your "20 watts" spends most of its energy in a blurry patch. Ask for M², not just maximum wattage.
3. Coherent light is the entire point. Coherence — photons in phase with each other — is what allows a laser to focus to an intense, precise point. It's also why Coherent, the company, has a product line spanning scientific lasers like the Verdi all the way to industrial cutting systems: the physics is the same, but each application needs a different source and delivery system.

What I ended up buying for stone work was a compact CO2 laser, which engraves stone, glass, and acrylic cleanly. The fiber laser stayed for metal marking. And the xTool? I still use it for wood signs and leather tags. It's a genuinely useful tool — I just tried to use it for the wrong material. Take this with a grain of salt, but I'd wager that 80% of beginner laser disappointment comes from using a machine that's right for one class of materials on something completely different.

If you've been searching "how to use xTool laser engraver," the software and focus settings are easy. The material limitations are the real curriculum.

The checklist that saved me $11,000

After those four months, I wrote down every question I wish I'd asked before the first purchase. I keep it on a whiteboard above my bench. It's not clever, but it works:

• What material am I processing most? (Wood? Metal? Stone? Acrylic? Each needs a different wavelength.)
• What laser class is the system, and do I have the right enclosure, laser eyewear, and training for it? (Anything above 500mW is Class 4 — plan for it before the delivery date, not after.)
• What are the beam quality specs — M², divergence, power stability — not just the wattage?
• Is the machine designed for my duty cycle? A hobbyist tool rated for a few hours a day won't survive production work.
• What's the cost of operation? Optics, chillers, nozzles, consumables, maintenance — total cost of ownership is what matters.

The original granite plaque was delivered three weeks late and about $250 under my cost. The customer was patient, which made it worse. But that failure became the foundation of our purchasing process. In the 18 months since I wrote down the questions, we've caught 47 potential wrong-spec purchases — I counted — and that's at least $11,000 in avoided mistakes. For the record, I'm not 100% sure that number is precise. I started tracking after the tenth near-miss. The checklist didn't exist until it had to.

"Most buyers focus on wattage and completely miss wavelength, laser class, and beam quality."

Four months, $11,000, and one very patient funeral home director later, that's the sentence I repeat to every customer who walks into our shop asking which laser to buy. The machines get better every year. The physics hasn't changed. Ask better questions before you spend.