How to Laser Etch Metal with the xTool F1 Ultra: A 7-Step Quality Checklist

Who this checklist is for

I'm quality and compliance manager at a small production shop that laser-marks parts for other businesses. Before any metal part ships, I review it—roughly 200 parts a week during busy months. I rejected a little under 10% of first-run metal samples last year for contrast or legibility issues, and most of those losses were preventable.

The xTool F1 Ultra is the dual-laser desktop unit we use for most jobs that combine fiber marking and diode cutting or engraving. If you're shopping for one, or already own one and can't get consistent metal marks, this is the checklist I run before every production order.

Seven steps. None of them are glamorous. Every one of them has caught a bad part before it shipped.

Step 1: Confirm you're using the fiber module

The F1 Ultra has two lasers inside. For metal etching—stainless, steel, aluminum, brass, titanium—you want the 20W fiber source. The xTool F1 Ultra fiber laser creates the mark by heating the surface into an oxide layer. The 20W diode side is better for wood, acrylic, leather, coated or anodized metal, and certain plastics.

This sounds basic, but it's the first thing I check in a rejected job. I've scrapped expensive material assuming the diode side could mark bare steel if we just slowed it down and pushed power. It doesn't work that way. The blue diode wavelength does not couple into bare metal the way 1064 nm fiber does.

Laser engraving foam? That's a diode-side job as long as the foam is safe to engrave. EVA foam works well with good ventilation. PVC foam releases chlorine gas. Polyurethane can melt into sticky burrs. Read the supplier's technical data before you assume anything.

Step 2: Configure the xTool F1 Ultra software with clean source files

Most people use xTool Creative Space because it's free and supports the machine out of the box. LightBurn is another option if you live in vector workflows. Support for xTool machines has improved in recent LightBurn releases, but check the current compatibility notes before paying for the license.

The xTool F1 Ultra software settings that matter most for metal marking:

• Resolution: 300 DPI is okay for text; 600+ for graphics.
• Fill mode: hatch settings determine how uniform the mark looks.
• Scale: verify the part dimensions in the software, not in a photo editor.

Zoom into the source file at 400% and look for artifacts before running the laser. A bad file produces a bad mark, and the blame lands on you, not the laser.

Step 3: Degrease, flatten, and stop touching the surface

Machining oil, fingerprints, and coolant residue block the laser and create blotchy marks. No setting will fix that.

Clean with isopropyl alcohol or acetone. Let it dry completely. Then handle only the edges. If you want a fast verification, use the water-break test: water should film across the surface evenly, not bead up.

Thin materials need to stay flat. Tape the edges, use magnets, or clamp them; vibration or lifting during the scan ruins focus and creates banding.

Alloys and finished metals complicate repeatability. The same recipe on 304 vs. 316 stainless can look surprisingly different, and a polished surface behaves nothing like a brushed one. I learned that on a customer order that required a costly redo.

Step 4: Focus for fiber, not for diode

Diode lasers tolerate slightly out-of-focus engraving better. Fiber marking is much less forgiving—a small focus offset shows up as a pale, uneven mark.

Use the included focus tool or the software's focus function. If you're etching a larger filled area, some operators deliberately raise the Z-height by a millimeter or two to widen the beam. That can improve coverage, but it should be a deliberate test per material, not a habit.

While we're on safety, leave the lid interlock alone. In normal operation, the machine is designed as a Class 1 laser product consistent with FDA 21 CFR 1040.10 and IEC 60825-1. A 20W fiber beam can damage eyesight before you can blink. I've read enough incident reviews not to play with it.

Step 5: Run a test matrix before you trust any recipe

Anyone claiming exact fiber parameters for your metal is guessing until you test your specific material. The xTool F1 Ultra software has a material library for stainless; use it as a starting point, not a promise.

Build a small grid of test marks at 70%, 85%, 100%, and 115% of the suggested power, with speed varied the same way. Engrave, label, note results, and compare under consistent lighting. Save the winning combination as a custom recipe.

This is also where I'll wave the transparency flag. The machine seller who lists every included accessory and says test your material is more trustworthy than the one who advertises universal perfection. When comparing quotes, ask what's not included before you ask the final price.

Step 6: Engrave the first article, then spot-check the middle of the run

In June 2024, I approved a rush order of 60 stainless tags. The first ten looked perfect. By tag 25 the mark was visibly lighter. A dirty optical component was eating power mid-run, and we hadn't checked it. We re-ran the entire batch.

That job changed how I think about inspection. Now any batch over 20 pieces gets a midpoint check. I set the first article next to the newest mark and compare under a bright light:

• contrast drifting from black to gray or brown
• banding or horizontal stripes
• edge crispness on small letters

Do not rely on memory. Put the two pieces next to each other and let them argue.

Step 7: Clean, verify, and record the recipe

After engraving, wipe off residue with a soft cloth and alcohol. Don't scrub hard on annealed oxide marks—you can remove the mark itself.

Compare the finished part to a saved reference sample, not to your memory of what it looked like yesterday. If the customer later says the color changed, you want the settings and batch number in a logbook, not a guess.

That documentation step is the least glamorous part of the job. It also prevents more headaches than any laser upgrade we've bought.

Common mistakes that have cost me money

• Running raw steel with the diode module because a diode laser machine demo showed metal marking. Usually that was painted, anodized, or coated metal. Coating removal is not metal etching.
• Engraving one perfect test sample, then assuming the next 200 parts will match without a mid-run check.
• Buying a machine without budgeting for extraction, air assist, or the rotary attachment. The base price is not the total price.
• Ignoring surface prep because the material looks clean. Oil is not always visible.

If your plan is small-batch metal tags, plaques, and product markings on a desktop machine, the F1 Ultra is worth a close look. Just bring the same honesty to your material tests as you expect from the brand. Nail the process and the laser will do its half—but don't ask the machine to fix a process you were too rushed to run.