I'm a quality compliance manager at a custom fabrication company. In a typical year, I review around 15–20 machine acceptances. In 2024, I rejected about 20% of the first deliveries we received—not because the machines were useless, but because what the vendor called “production-ready” didn't match our real tolerances. This checklist is what I use when we bring a new industrial laser system into the shop, especially a coherent laser system. It’s written for teams that have actual deadlines and actual parts to make, not for showroom comparisons.

One caveat: this worked for us because we’re a mid-size B2B shop with fairly predictable production runs. If you’re a job shop with constantly changing materials, the acceptance criteria will need to be more flexible. But the general steps should still transfer.

1. Lock Down Your Process Specs Before Requesting Quotes

This sounds obvious, but I keep seeing purchase orders that say “laser cutter capable of cutting steel.” That’s about as useful as saying “vehicle capable of carrying things.” Instead, define the actual process windows you need: material type and thickness, edge quality, dimensional tolerance, cycle time, and duty cycle.

For example, in Coherent femtosecond laser battery production, the critical specs are pulse duration, beam quality, and heat-affected zone (HAZ). If you’re cutting thin foils for battery cells, a nanosecond source might leave a burr that causes a short circuit later. So you’re probably looking at a femtosecond laser with a pulse width in the single-digit picosecond range—and you should verify that spec on paper, then prove it on a test cut.

Checkpoint: Create a one-page spec sheet with three columns: Required, Target, Nice-to-Have. Send that to the vendor before they quote. If they won’t commit to those numbers in writing, don’t take the meeting.

2. Verify the Laser Source Inside the System

Here’s a spot where people get tripped up. You may be buying from a system integrator, but the actual laser source inside the cabinet can come from another manufacturer. If the integrator says it’s a “Coherent laser system,” ask for the source module details: the exact model, serial number, and original manufacturer's spec sheet.

We specify Coherent sources in our systems because we’ve had good long-term stability with them, and there's a clear support path for replacement parts. We're also not the only ones—Trotec uses Coherent laser sources in some of their machines, which at least confirms the optics are OEM-grade. But you still need to verify what’s actually in the box.

Checkpoint: Ask the integrator to list the laser source manufacturer, model, and warranty terms in the proposal. Then email the source manufacturer’s support team with the serial number to confirm the warranty is valid and transferable.

3. Run a Formal Acceptance Test (FAT) with Pass/Fail Criteria

Don’t accept a machine based on a 2-minute demo. We always run a formal acceptance test that mimics normal production. For a cutting laser, that means our actual material, our actual nesting layout, and a full shift of continuous operation.

We follow ISO 9013 for measuring edge quality on thermal cuts. Typical pass criteria for our stainless work are dross height ≤ 0.1 mm and edge roughness Ra ≤ 3.2 µm. For a welding laser, it might be a tensile pull test on a certified sample. For Coherent femtosecond laser battery production, it would be a metallurgical cross-section to measure HAZ and micro-cracks.

This is where I'm glad we insisted on an 8-hour continuous run. The vendor said 10 minutes was enough to prove stability. We ran it for eight hours, and the beam power drifted by 11%. That would have been a slow-motion disaster on a production order. We caught it during the FAT because we had actual pass/fail numbers, not “looks fine.”

Checkpoint: Define maximum power drift (e.g., ±2%) and edge quality limits before the test. Put them in the purchase order. Make the official acceptance contingent on the FAT report.

4. Safety Integration Review

Laser safety is not an afterthought. For a Class 4 system, you need fixed and interlocked guards, emergency stop circuits, and laser-safe viewing windows matched to the wavelength. In Australia, you’ll want to check compliance with AS/NZS IEC 60825.1. If the system is integrated into an automated cell with robots or a press brake, the safety interface is even more complicated.

Actually, we once talked to a press brake supplier in Australia who said their laser-scanner safety solution was “plug and play.” It wasn’t. The scanner kept faulting because the environment had metal dust. We ended up specifying a fully guarded enclosure with double interlock switches. The lesson: don’t trust generic safety claims, verify with your own risk assessment.

Checkpoint: Ask for the safety declaration and describe the environment where the system will run. If it’s a dusty shop floor, test the safety components before sign-off.

5. Validate the Software and Data Interfaces

Modern industrial lasers rarely run in isolation. They’re often paired with a VMC (Vertical Machining Center) for pre- or post-processing, or with a robot for loading and unloading. That means you need to verify not just the laser, but the handshakes between machines.

If your operator uses an MES or ERP system, confirm the laser controller can read production files without human retyping. During FAT, we test what happens when the Ethernet cable is unplugged mid-cycle and then reconnected. Does the machine resume the job or does it scrap the part? You want to know that before a glitch costs you a $2,000 blank.

Checkpoint: Make a list of every communication protocol the laser needs—OPC UA, DXF, custom ERP—and test each one during FAT with the actual file formats your programmers use.

6. Plan for Consumables, Spare Parts, and Local Support

Laser systems are like race cars: they use consumables that wear out. Nozzles, protective windows, focus lenses, shielding gas filters. If you’re in Australia, that last point matters a lot. Shipping a spare part from the US or Germany can take a week, which is an eternity in production.

We learned this the hard way when we had a press brake down for two weeks waiting for hydraulic seals. After that, we started requiring spare parts kits with every capital purchase. For the laser, we maintain a list of critical spares and the vendor’s guaranteed response time.

Checkpoint: Ask the supplier for a recommended spare parts list with lead times. If you can't afford downtime, negotiate a consignment stock for the top 10 items.

Common Mistakes to Avoid

Testing without a load. A laser can cut beautiful parts at 1% duty cycle. That doesn’t tell you what happens at 80% duty cycle. Specify a duty cycle that matches your actual production.

Believing one machine can do everything. A Coherent laser system is brilliant for cutting, welding, and marking—but for thick plate forming, you still need a press brake; for complex internal cavities, a metal 3D printer might be better; for precise milling, a VMC is the right tool. There’s no such thing as a universal tool, and a supplier who tells you otherwise is overpromising.

Ignoring operator training. The fanciest laser in the world doesn’t help if your operators don’t know how to set process parameters. Budget for training, not just the 1-hour overview. We paid for two days of hands-on training and still found ourselves calling support for three weeks.

Not understanding the types of 3D printers you use for fixtures. We use 3D-printed fixtures to locate parts before laser cutting. FDM is fine for low-temperature nylon; SLA gives better detail; SLS makes durable nylon parts. The wrong type of 3D printer can mean a fixture that melts under the beam—and that’s a mess we’ve cleaned up more than once.

There’s something satisfying about watching a new laser system run a production batch without surprises. After the spec review, the FAT, and the safety inspection, that’s the payoff. But it only happens if you treat acceptance testing as serious engineering work, not just a signature on a delivery form.

Honestly, the most useful thing we did was accept that our own expertise has boundaries. We’re not press brake specialists, and we’re not 3D printing experts. When we need those machines, we go to someone who specializes. The vendor who told us “this isn’t our strength—here’s who does it better” earned our trust for everything else. That’s exactly the attitude you want when buying critical manufacturing equipment.