Answer the everyday sheet-metal question — "at this laser power, on this plate thickness, how many metres per minute can I cut?" — with the standard energy-balance equation. Slide power, thickness and coupling efficiency and watch the cutting speed and seconds per metre of cut update in real time for quoting and process planning.
Parameters
Laser power P
kW
Fibre lasers 1-6 kW; 10 kW class for thick plate
Material thickness t
mm
Material density ρ
kg/m³
Steel 7850, Al 2700, SUS304 8000, Cu 8960
Melting enthalpy H_m
kJ/kg
Specific energy from room temp to melt
Kerf width w
mm
Fibre thin 0.1-0.2, thick plate 0.3-0.4
Coupling efficiency ε
O2+steel 0.25-0.40, N2+SUS 0.05-0.15
Results
—
Useful power (W)
—
Mass removal rate (kg/s)
—
Volume removal rate (mm³/s)
—
Cutting speed (mm/s)
—
Cutting speed (m/min)
—
Time per metre of cut (s)
—
Top-view laser cutting animation
The laser head scans across the steel plate. Under the beam, a melt pool forms, and the assist-gas jet blows the molten metal out of the kerf. The head speed scales with the computed cutting speed.
Mass removal rate ṁ → volume removal rate V̇ → forward speed. ε bundles optical absorption, conductive losses and assist-gas blow-out efficiency. Representative range 0.15-0.40 for a fibre laser on steel.
Estimating the laser cutting speed
🙋
Laser cutting always sounds like "press the button and the steel falls apart". How many metres per minute can you actually cut? Just more laser power, right?
🎓
Good question. A 4 kW fibre laser on 5 mm mild steel runs at about 4-6 m/min in real life. Try the defaults on the right (P=4 kW, t=5 mm, ε=0.30, basically oxygen-assist) and look at the cutting-speed cards. You will see roughly 11 m/min. That's the theoretical ceiling assuming you melt and blow out 100% of the kerf. Real machines lose half of that to assist-gas dynamics, edge-quality requirements and spatter control. So 5-6 m/min is the practical answer — which is what your gut tells you.
🙋
OK so what is that "coupling efficiency ε" on the right? 0.30 looks like an oddly specific number.
🎓
That parameter is the heart of laser cutting. ε is the fraction of the laser power that ended up actually melting metal. The rest reflects off the steel surface, conducts into the kerf walls, gets absorbed by the plume, or wastefully overheats and vaporises the melt. Fibre lasers at 1 µm couple into steel quite well, so ε ≈ 0.3. CO2 lasers at 10.6 µm couple poorly, ε ≈ 0.1-0.15. Throw in an oxygen assist and the Fe + ½O2 → FeO exothermic reaction boosts effective ε to 0.4. Nitrogen on stainless has no exothermic boost, so ε drops to about 0.10. Changing ε with gas type is exactly how you should be using this tool.
🙋
So doubling the plate thickness halves the speed? Because t is in the denominator, right?
🎓
Exactly. Look at the "speed vs thickness" chart below. Going from t=5 to 10 mm halves the speed; going from 5 to 2.5 mm doubles it — a clean 1/t curve. Real-machine cutting tables follow the same trend. One caveat: beyond about 15 mm, the heat lost into the kerf walls becomes significant and ε effectively drops. So this tool is a bit optimistic on heavy plate. For anything thicker than 15 mm, check the vendor's process tables before quoting.
🙋
One more — what is "time per metre of cut" useful for?
🎓
It is the number that goes straight into a quote. If the total cutting-path length of one part is 12 m, then 5.2 s/m × 12 = about 62 s of net cut time. Add positioning, gas purge and piercing time and you have the cycle time per part. Multiply by your machine rate per second and you have a first-pass cost estimate. "Speed in m/min" and "seconds per metre" are two sides of the same coin, but production engineers usually find seconds per metre faster to reason about.
Frequently Asked Questions
The simplest first-pass model is the energy-balance equation v_cut = ε·P / (ρ·H_m·w·t). Here P is the laser power, ε is the coupling efficiency (the fraction of beam power that is actually absorbed and used to melt the kerf), ρ is the material density, H_m is the melting enthalpy, w is the kerf width and t is the plate thickness. Physically: the laser melts a certain volume per second and the assist gas blows it out; divide that volume rate by the kerf cross-section and you get the forward speed. It is widely used as a first cut before running a full CFD-coupled thermal model.
Typical values are: fibre laser + oxygen assist on mild steel ε ≈ 0.25-0.40 (the exothermic Fe-oxidation adds heat); nitrogen assist on stainless steel ε ≈ 0.05-0.15; thin sheet + compressed air ε ≈ 0.15-0.25. CO2 lasers couple poorly into steel, so ε ≈ 0.10-0.20 even with good optics. If you have machine data, back-calculate ε from a known good cut and reuse it — that gives the best accuracy for your shop.
The biggest levers are power P (v scales linearly) and thickness t (v scales as 1/t). Going from 4 kW to 6 kW gives 1.5×; dropping 5 mm to 3 mm gives 1.67×. Tightening the kerf w with better optics and a shorter focal length is worth another 1.3-2×. Tuning ε via gas type, focus position and nozzle design helps but is realistically capped at 1.2-1.5×.
With ε calibrated against your own machine, it typically lands within ±20% of measured cutting speeds for thin to medium plate (t ≤ 10 mm). On thicker plate, conductive losses to the kerf walls, re-solidification of expelled melt and assist-gas viscous losses start to dominate and the simple energy balance becomes optimistic. For serious work on 15-25+ mm plate, consult the machine vendor's process tables or a CFD-coupled solver. Treat this tool as a first-pass estimator for process planning.
Real-World Applications
Sheet-metal volume production: Appliance housings, server racks, lockers, vending-machine side panels — almost any 1-3 mm steel or stainless sheet job has migrated from punching and turret presses to laser cutting. Multiply your part's total cutting-path length by the seconds-per-metre this tool returns and you get the net cut time, which goes straight into shop quotes. You can also size new equipment from the cycle-time target backwards: solve the equation for P and you have the required laser power.
Mid-thick plate for construction equipment: Excavator booms and arms, rail-car frames, industrial-machine base plates use 6-20 mm steel plate. Oxygen-assist fibre lasers dominate this regime, and an ε of 0.30-0.40 gives speeds close to the machine catalogue. Use the speed-vs-thickness chart to pick the right laser power for each plate gauge in your product mix.
Stainless and aluminium precision work: Stainless food-grade and medical housings, aluminium aerospace parts, lithium-battery tab metal — anything that must not oxidise is cut with nitrogen assist. With ε in the 0.05-0.15 range the speed drops to roughly one-third of mild-steel oxygen cutting, exactly the trend you can reproduce in this tool. Use it to balance nitrogen consumption against cycle time when costing parts.
Process-optimisation starting points: Before trial-cutting a new material or thickness, use the tool to pick a starting speed for the first run. Begin at 80% of theoretical, scan around it with a DOE on the second run, and you minimise sample waste. After optimisation, back-calculate ε from the measured speed and store it in your in-house process table — next time the estimate will be even sharper.
Common Misconceptions and Pitfalls
The biggest pitfall is recycling a textbook ε. A book may say "ε = 0.4 with oxygen assist", but on a real machine nozzle geometry, focus position, lens contamination, and the surface oxide or rust-preventive oil on the sheet can easily halve that number. Every machine and material/gas combination deserves its own ε, back-calculated from a known-good cut. Calibrating ε once is far more productive than blaming the catalogue when speeds fall short.
Second, ignoring assist-gas pressure and flow. The energy balance models only how fast the laser can melt metal, not how fast the gas can blow it out. If the jet's momentum cannot evict the melt as fast as the laser produces it, cutting speed is capped no matter how much more laser power you add. The first symptom is dross — a re-solidified bead on the bottom edge — which means the gas, not the laser, is the bottleneck. Upgrading laser power without revisiting nozzle size and gas pressure is a classic mistake.
Finally, optimising for speed alone and ruining edge quality. Push past about 80% of theoretical maximum speed and the melt pool cannot exit cleanly behind the beam, so the cut face develops striations and roughness. Precision parts are usually run at 50-70% of theoretical on purpose to protect edge quality. The number this tool returns is a physical upper bound; the "speed you actually cut at in production" is typically about 70% of that. Read the results with that ratio in mind.
How to Use
Enter laser power in watts (e.g., 1500 W for a 1.5 kW CO₂ cutter) and set material type or density (mild steel ~7850 kg/m³, aluminium ~2700 kg/m³)
Input sheet thickness in mm (typical range 0.5–6 mm for sheet metal) and material melting enthalpy in kJ/kg (steel ~260 kJ/kg, aluminium ~330 kJ/kg)
The simulator calculates useful cutting power accounting for beam absorption, then divides by melting enthalpy to find mass removal rate, volume removal rate, and final cutting speed in mm/s or m/min
Worked Example
CO₂ laser, 2000 W, cutting mild steel (7850 kg/m³) at 3 mm thickness with melting enthalpy 260 kJ/kg. Assuming 70% beam efficiency and focused spot absorbing ~85% of energy: useful power ≈ 1190 W. Mass removal = 1190 / 260000 ≈ 0.00458 kg/s. Volume removal = 0.00458 / 7850 ≈ 0.583 mm³/s. At 3 mm thickness, cutting speed ≈ 65 mm/s or 3.9 m/min. Time per metre of cut ≈ 15.4 seconds.
Practical Notes
Fibre lasers (1064 nm) cut steel faster than CO₂ (10.6 µm) at equal power because metal reflectivity is lower; increase expected speed by 30–50% for stainless at same wattage
Gas-assist pressure (typically 2–8 bar oxygen or nitrogen) improves actual cutting speed 20–40% by clearing molten debris; our estimate assumes ideal assist conditions
Sheet surface oxidation, reflectivity variation, and beam mode (TEM₀₀ vs multimode) create ±15% variation; always validate on test sample before production runs