Fiber and MOPA galvo markers are driven by five numbers on each EZCAD pen: speed, power, frequency, pulse width and line interval (hatch spacing), plus passes. None of them works alone. This guide explains what each one changes, how they combine, and how to find your own values with a test card instead of copying someone else's.
The two quantities that actually decide the mark
Every setting feeds one of two things:
- How much energy lands on each square millimetre. A useful index is power ÷ (speed × line interval), in joules per mm² per pass. Double the speed or double the line interval and you halve it; double the passes and you double it.
- How that energy arrives — a few hard pulses or many gentle ones. Energy per pulse is roughly average power ÷ frequency, and the gap between pulses along a line is speed ÷ frequency. Hard, widely spaced pulses remove material. Soft, tightly packed pulses heat it, which is what annealing and MOPA colour need.
Power % is a share of the source's rated average power, so treat these numbers as a way to compare settings on one machine, not as absolute physics.
Worked numbers from the bundled baselines
These are recorded starting points from the Parameters Library, with the two indices worked out:
| Job | Source | Power | Speed | Freq | Pulse | Interval | Passes | J/mm² per pass | Pulse gap |
|---|---|---|---|---|---|---|---|---|---|
| Black anneal, stainless 304 | JPT 30 W MOPA | 40% | 800 mm/s | 35 kHz | 200 ns | 0.025 mm | 2 | 0.60 | 0.023 mm |
| Deep engrave, stainless 304 | Raycus 30 W | 100% | 1500 mm/s | 35 kHz | — | 0.02 mm | 25 | 1.00 | 0.043 mm |
| White mark, anodized aluminum | Raycus 30 W | 100% | 2000 mm/s | 55 kHz | — | 0.03 mm | 1 | 0.50 | 0.036 mm |
| White mark, anodized dog tags | Fiber 50 W | 15% | 1000 mm/s | 45 kHz | 200 ns | 0.05 mm | 1 | 0.15 | 0.022 mm |
| Frost mark, bare aluminum | Raycus 50 W | 100% | 2000 mm/s | 45 kHz | — | 0.03 mm | 2 | 0.83 | 0.044 mm |
Read the table the way the laser does. The black anneal and the deep engrave use similar energy per pass, but the anneal gets it from a long 200 ns pulse, tight pulse spacing and only two passes — heat that darkens the surface without removing it — while the deep engrave repeats twenty-five passes and clears debris between them. The two anodized recipes both turn the dye layer white and sit more than three times apart in energy, which is exactly why you test on your own stock.
What each setting does
- Speed (mm/s). Sets both energy per area and pulse spacing. Galvos run far faster than gantry lasers; hundreds to a few thousand mm/s is normal for marking.
- Power (%). Scales average power and therefore energy per pulse. Past the point where the surface changes, extra power mostly adds heat, discolouration and burr.
- Frequency (kHz). Trades pulse energy for pulse count. Lower for removal and depth, higher for annealing and colour. Q-switched sources usually have a narrower usable range than MOPA sources.
- Pulse width (ns, MOPA only). Short pulses cut cleanly with a small heat-affected zone; long pulses heat more. Colour and black anneal on stainless live at the long end or at very high frequency.
- Line interval (mm). The gap between hatch lines. Keep it at or below the spot size for a solid fill; wider leaves stripes. Halving it doubles energy per area and doubles the run time.
- Passes. Adds energy in steps and lets debris clear between them. Several light passes usually give a cleaner deep engrave than one heavy pass.
Find your values with a test card
- Open the free EZCAD test card export and sweep speed across and power down at a fixed frequency and pulse width — the default grid does exactly this.
- Import the DXF into EZCAD (File → Import). Every cell arrives on its own pen; key each pen from the CSV that comes with it.
- Mark an offcut of the same material batch, cleaned with isopropyl alcohol, at the same focus height you will use for the job.
- Pick the lowest-energy cell that gives the finish you want, then make a second card that sweeps frequency against pulse width around that speed and power (MOPA only). EZCAD addresses at most 30 pens, so keep each card to 30 cells.
- Record the winner on the pen and save an .ezd, and add it to the Parameters Library so it is there next time.
Things that move the answer
- Field lens. A longer focal length gives a bigger field and a bigger spot, so less energy per area. Re-test after changing lenses.
- Focus. A few millimetres of defocus widens the spot and softens the mark; use it on purpose, never by accident.
- Material batch. Stainless grade and finish, and anodize thickness and dye, change the result more than most people expect.
- Safety. Fiber lasers emit at about 1064 nm: wear eyewear rated for that wavelength, extract fumes from coated and anodized parts, and never mark PVC.
For colour on stainless, see the MOPA colour settings guide and the MOPA Color Predictor. For getting artwork into EZCAD in the right format, see Using Reticle Red with EZCAD.
Frequently asked questions: Fiber / MOPA
What does frequency do on a fiber laser in EZCAD?
Frequency (kHz) is how many pulses the source fires per second. At the same average power, a lower frequency packs more energy into each pulse, which removes material; a higher frequency spreads the same power across more, gentler pulses, which heats the surface instead — the regime for annealing and MOPA colour. As a rule of thumb, energy per pulse is roughly average power divided by frequency: 12 W at 35 kHz is about 0.34 mJ a pulse.
How do speed and frequency decide whether pulses overlap?
The distance between pulse centres along a scan line is speed divided by frequency. At 800 mm/s and 35 kHz that is 800 ÷ 35,000 ≈ 0.023 mm. If that spacing is smaller than your spot diameter the pulses overlap into a continuous line; if it is larger you get a row of dots. Raise speed without raising frequency and the mark breaks up; raise frequency at the same speed and it closes up and runs hotter.
Does pulse width matter on a non-MOPA fiber laser?
Usually not. Most Q-switched fiber sources fire a fixed pulse width set by the source itself, so EZCAD may show the field while the source ignores it, and the usable frequency range is narrower. On a MOPA source pulse width is independently adjustable: short pulses ablate cleanly with little heat spread, long pulses put more heat into the surface. Check your source datasheet before sweeping it.
What settings give a white mark on anodized aluminum?
Two recorded baselines in Reticle Red's Parameters Library both give white on anodized aluminum and differ by more than three times in energy per area: 100% of 30 W at 2000 mm/s, 55 kHz and 0.03 mm interval (about 0.50 J/mm²), and 15% of 50 W at 1000 mm/s, 45 kHz, 200 ns and 0.05 mm (about 0.15 J/mm²). Anodize thickness and dye vary by supplier, so run a speed-by-power test card on an offcut from the same batch and pick the lowest-energy cell that is fully white.