Novanta Field Note: Metal Laser Cutting, Marble Laser Engraving, and Air Assist

Last March, at 4:30 on a Thursday, a production manager called me with a launch deadline two days away. The job was simple on paper: 120 stainless steel nameplates and 80 engraved marble coasters. The plan was to buy one “versatile” desktop laser, set the speed and power, and have finished samples by Friday. My answer wasn’t a model number. It was: let’s back up and ask what those two materials are really asking the laser to do.

I work in applications engineering at Novanta, headquartered in Bedford, Massachusetts. My part of the business is helping laser systems work under production pressure—scan heads, beam delivery, motion axes, and photonics details that rarely make it into a sales brochure. That puts me on the receiving end of panic calls when someone’s “surefire” laser choice meets a real deadline. The pattern is consistent enough that I now triage every call into one of three starting points.

• If the part is structural metal, the real topic is metal laser cutting machines, assist gas, and edge quality.
• If the part is natural stone, the real topic is laser engraving marble, CO2 wavelength, and air/exhaust management.
• If the symptom is smoke, dark edges, or uneven depth on an existing machine, the real topic is usually air assist.

Start Point 1: Metal laser cutting machines need gas, not just watts

When a job has to produce clean metal parts—steel, stainless, or aluminum—I start with the machine family. In 2025, most new industrial metal-cutting systems are fiber-based. CO2 can still cut metal, especially in established factories, but fiber sources are efficient, compact, and easier to keep productive through a long shift. That’s a practical trend, not a hard law; the hard part is that cutting metal requires a gas jet to push the molten material out of the cut.

On a metal system, machine builders often use the term assist gas rather than air assist. Oxygen adds heat and helps cut mild steel quickly, but it leaves an oxidized edge. Nitrogen is frequently used for stainless and aluminum when you need a clean, oxide-free result. Compressed air can be useful on thinner mild steel when edge cosmetics are less critical and the air is dry and filtered. The cheapest gas is not always the cheapest process.

One example stays in my head: a shop swapped nitrogen for shop air to save a few hundred dollars a month. On the first rush stainless run, the edge discoloration failed inspection. They scrapped the parts, reinstalled nitrogen, and reran the batch. The rerun cost more than the gas had saved in four months. I’m not against comparing gas prices—I’m against comparing a single price instead of the cost to get a good part out the door.

Start Point 2: Laser engraving marble is a different process from cutting metal

Laser engraving marble can be breathtaking when it works and frustrating when it doesn’t. Marble is natural stone with veins, microfissures, and variable density. The usual industrial choice is a CO2 laser operating near 10.6 µm; that wavelength is well absorbed by marble and many nonmetallic surfaces. A fiber laser, by contrast, is generally a poor first choice for stone because its wavelength partners better with metals.

For marble, the first support call I make is often about air assist, not power. When the laser hits carbonate stone, it creates a fine dust and vapor plume. Air assist blows that cloud out of the beam path so the next pulse still reaches the surface at the intended focus. Without it, the engraving turns shallow, uneven, or brown. The air also cools the area around the cut, reducing the thermal stress that can crack or chip along natural stone lines.

If your marble volume is small and seasonal—say, coasters for one annual event—consider buying the service rather than the machine. One manufacturing manager bought a low-priced CO2 desktop unit because it looked cheaper than outsourcing. After adding dust extraction, spare optics, time to learn the software, and a few discarded samples, the cost per finished coaster was higher than the local job shop quotation. This is not a criticism of small lasers; it’s a reminder that total cost includes sleep.

Start Point 3: What is air assist on a laser? It’s the unsung process variable

What is air assist on a laser? In plain terms, it is a controlled stream of compressed gas that is directed through a small nozzle, usually next to or around the laser beam, to the point where the beam hits the material. It clears smoke, vapor, and particles away, protects the lens from spatter, and helps control heat. It is not the exhaust blower sitting under the machine. Both are needed, and they are not the same.

When air assist is missing or misaligned, the job doesn’t always stop; it just gets worse. Smoke begins to absorb the beam, reducing power at the surface. Debris falls back into the kerf and can be fused there by the next scan. On wood, acrylic, or stone, char marks and rough edges appear that are easy to blame on the laser tube or lens. I’ve watched operators replace optics before someone noticed the air hose had disconnected.

More pressure is not automatically better, and air quality matters. Too high a pressure can create turbulence that disturbs thin parts or blows debris sideways. Wet or oily compressed air can fog a lens and introduce defects that look like they should be impossible. When I check a laser setup, I verify the pressure setting, nozzle height, and the dryness of the compressed air before touching the power settings.

How to know which start point you are in

If you have a part to make today, simplify the triage. Ask what the finished piece must do. A metal bracket or an identification plate with a clean edge points toward the metal start point. A decorative stone, glass, or wood part points toward the CO2/marble start point. A sudden change in quality on material that worked last week points to air assist and process settings first.

Then ask whether the demand is permanent or a fire drill. A new production line that will run daily can justify a system designed specifically for that material. For an urgent one-off, a service shop with the right laser is usually the fastest and most flexible answer. The most expensive laser you can own is the one that waits eleven months for another thirty-minute job.

The last thing I check: cost per good part, not purchase price

I have to be careful when I say “don’t buy on price.” A purchasing team needs a firm number and a defensible supplier choice. But the firm number on the purchase order is not the total number. My job is usually to find the point where a lower quote turns into a higher lifetime cost.

A $35,000 system that needs a second calibration visit and a rush service call before its first big order is not cheaper than a $55,000 system that runs through the week without drama. A high-power laser that cuts fast but produces variable edge geometry might waste more material than it saves in cycle time. Value, in this business, is the cost of the good parts produced while the machine is working.

That is also why I pay attention to what is underneath the sheet metal. When I see Novanta components on a bill of materials—scan heads, galvo motors, laser modules, or precision motion stages—I know they came from an engineering culture that tests repeatability. The Novanta Bedford, Massachusetts team has that focus, and repeatability is exactly what keeps an urgent order from turning into a long night.

One final field note. Laser technology is changing quickly, and the details here reflect what I worked through in late 2024 and early 2025. If you are evaluating a machine now, run your own material sample, measure the first article, and verify the support plan for the failure you are most afraid of. A quote is a starting point, not a guarantee. The machine that wins is the one with the lowest cost per good part when the deadline is real.