- Check 1: Verify the material before you trust the label
- Check 2: Clean and calibrate before you blame the settings
- Check 3: xTool S1 acrylic cutting settings – start here, confirm on scrap
- Check 4: Laser marking anodized aluminum – the coating is the target
- Check 5: Inspect against the golden sample
- Check 6: Decide “CNC router or laser cutter?” before you quote the deadline
- Two mistakes I still catch in our own workflow
I run quality control at a small workshop that produces signs, plaques, small-batch acrylic parts, and custom engraving work. Every finished piece lands on my bench before it ships. In a normal month, that is 600–700 separate items (maybe 650; I would have to check the logbook for the holiday months). In 2024 we rejected 4.2% of first production runs. Almost every rejection came down to a skipped verification step, not the machine.
If you sell one-off projects, one bad edge or washed-out mark isn't a minor incident. It's the client's entire first impression of your business. The fix is boring: gate your work at six checkpoints. Use them before quoting bulk orders, before cutting your customer's material, and especially before a deadline that makes you want to skip them.
Check 1: Verify the material before you trust the label
I learned never to assume “laser-safe” or “laser-grade” means anything consistent. In 2023, we ordered black acrylic from a new supplier to save a few dollars per sheet. I assumed acrylic was acrylic. We didn't verify. Turned out the clear “acrylic” sample was polycarbonate, and our xTool S1 20W couldn't cut it at all. We now put material type and batch number on every incoming invoice, and we run a 60-second test before committing to a production run.
For the xTool S1, this matters more than people think. The blue diode emits around 455 nm (per xTool's product specifications, accessed January 2025). That wavelength passes through clear transparent acrylic, so clear sheets generally won't cut on the S1 the way they do on a CO₂ laser. Colored and opaque acrylic absorbs the beam much better. This is also where “cast” versus “extruded” acrylic matters: they cut differently, edge-finish differently, and sometimes even require different speed settings.
If the job involves laser marking anodized aluminum, the batch matters equally. Anodizing is a coating process, and different suppliers produce different coating thicknesses and dye absorption. If the order is 200 anodized aluminum badges, keep them to one supplier and one batch. Mixed batches are the fastest route to inconsistent contrast.
Check 2: Clean and calibrate before you blame the settings
A dirty lens or an out-of-focus nozzle will ruin a run faster than any wrong speed/power combination. We clean the lens before every major job. A 20W module at the wrong focus distance will give you worse results than a 10W module set up correctly.
Our protocol: clean the lens, verify the air assist is aimed properly, then run a focus ramp test on scrap material. Check the line width at each focus step. The narrowest, cleanest line is your focus point. Write the material thickness and the resulting focus setting on the job sheet. (Should mention: we keep a focus log per machine. If there's no log entry, there's no calibration.)
Check 3: xTool S1 acrylic cutting settings – start here, confirm on scrap
These starting settings are for the xTool S1 with the 20W module and colored/opaque acrylic. They're the values on our tuning board as of December 2024. They will shift slightly with your machine, your exhaust, and your acrylic batch, so treat them as entry points, not gospel.
| Material | Passes | Speed | Power | QC check after first pass |
|---|---|---|---|---|
| 3 mm colored cast acrylic | 1 | 10 mm/s | 100% | Cut is complete; edges clean, no re-melted burr |
| 3 mm colored extruded acrylic | 1–2 | 8 mm/s | 100% | Edge is consistent; no stress cracking |
| 5 mm colored acrylic | 2 | 6 mm/s | 100% | After pass 1, measure cut depth with calipers |
It's tempting to think that acrylic is acrylic. That advice ignores the real difference between cast and extruded stock, and it ignores how much dye color affects a blue diode's absorption. Light-colored or translucent acrylic may need dramatically slower speeds or may never cut cleanly at all. When a potential customer asks for clear transparent acrylic parts, we tell them honestly: this machine is not the right tool for that material. That honesty has saved us more rework than any settings table.
One check most hobbyists skip: after the first pass, don't just look at the top surface. Lift the piece and inspect the last 0.3 mm of the underside. A cut can look complete from the front and still be partially attached at the back. Pulling the piece apart by hand will distort the edge and ruin the dimensional accuracy.
Check 4: Laser marking anodized aluminum – the coating is the target
Here's something most tutorials don't tell you: when you laser mark anodized aluminum with a diode laser, you aren't engraving the metal. You're removing or altering the anodic coating. No coating, no mark. That's why bare polished aluminum is a completely different job.
For black anodized aluminum on our S1 20W, our starting point is 80 mm/s at 65% power, single pass, with a line interval around 0.08 mm. That produces a clean, milky-white mark on black anodize. If the mark looks burnt or brown, increase speed or lower power. If the contrast is weak after cleaning the surface, slow down by about 20% before touching the power setting. Hard anodize (Type III) usually needs slower speeds than decorative anodize.
Our QC test is a small grid on scrap from the actual batch: three power settings, three speed settings, one pass each. Label each square with a marker, clean the piece, and compare under bright light. Doing this takes ten minutes and has saved us from engraving 50 pieces with washed-out contrast.
Honestly, I'm not sure why two suppliers' black anodized sheets with seemingly identical specs mark so differently. My best guess is coating thickness and the sealing process. What I do know is that the test grid catches it every time, so we never skip it.
Check 5: Inspect against the golden sample
“It looked fine on the screen” is not a QC standard. We keep a physical golden sample for every repeat product: the approved version of the sign, plaque, or part that the client signed off on. The production run gets compared to that sample under the same lighting conditions. That sounds obvious, but in practice it's the step that prevents the “I don't know what's wrong, it just looks different” complaint.
When you're evaluating laser cut sign ideas for clients, most of the pieces that sail through our QC are variations of three themes:
- Interior door signs and office nameplates in black anodized aluminum with white engraved lettering.
- Layered acrylic dimensional signs, where a 3 mm colored top panel is cut and engraved, then offset above a base panel to create depth.
- Custom wooden wall signs and gift tags cut from 3 mm Baltic birch, with engraved typography and a clean, unfinished edge that the client can finish themselves.
The idea itself matters less than legibility. A sign with letter strokes thinner than the laser kerf will look bad at any size. We check minimum stroke width at the design stage, not after cutting. If the detail is too thin for the 20W module's spot size, we redesign rather than “see what happens.”
Check 6: Decide “CNC router or laser cutter?” before you quote the deadline
The xTool S1 is not a universal fabrication tool. When a customer asks for a 20 mm solid wood plaque with an 8 mm-deep routed recess, the correct answer is a CNC router, not a laser cutter. A laser cuts by heating and vaporizing a line; it is not designed to hog out deep, flat-bottomed cavities. The edge taper alone would fail our inspection.
The decision between a CNC router and a laser cutter should be based on the part's acceptance criteria, not on the machine you happen to own. Use the S1 when the job needs fine two-dimensional detail, thin sheet materials, engraving on coated or anodized surfaces, or fast iteration on prototypes. Reach for a CNC router when the job needs vertical sidewalls, thick stock, three-dimensional profiles, or materials that either reflect the beam or don't absorb it.
This check happens before we quote a deadline. If we promise a client a delivery date based on the wrong process, the quality failure is our fault, not the machine's.
Two mistakes I still catch in our own workflow
First, trusting remembered settings. The producer says, “We ran this exact job last month, same settings.” Every time, we make them run a scrap test anyway. Machines drift, lenses get dirty, and materials change batches. The scrap test costs five minutes. The redo costs hours.
Second, shipping the borderline piece because the deadline is tight. I've done it. It always comes back as a complaint, a refund, or a “can you just fix this one detail?” that turns into a full redo. In Q1 2024, that habit cost us a $22,000 account's next order because the client lost confidence in our consistency. The piece looked acceptable to us. It did not look acceptable to them.
Quality is not about perfectionism. It's about repeatability. The xTool S1 is capable of that repeatability, but only when every job passes the same gates before it reaches a customer.
