Laser diffraction (LD) and sieve analysis are the two methods a powder-metallurgy or additive-manufacturing lab will reach for first, but they answer different questions: LD returns a volume-weighted equivalent spherical diameter distribution across roughly 0.01 to 3500 µm in 1–3 minutes, while sieving returns a mass-based distribution of minimum-dimension width on typically 8 stacked fractions after 5–10 minutes of vibration [S2][S5].
Choosing between them is not a quality question, it is a coverage-and-reporting-base question. Wide-range metal powders, including 17-4 PH stainless steel, Inconel 718, and Ti-6Al-4V, span 1 µm to 250 µm or beyond, so a single technique undersamples either the fine tail or the coarse oversize mode [S1][S3]. The pragmatic answer in 2026 is: run LD for the bulk distribution and the fine fraction, run sieving for coarse accountability, and use an integrated software workflow to reconcile the two reporting bases [S5][S7].
Measurement Range, Resolution, and Reporting Basis
Laser diffraction reports equivalent spherical diameter on a volume basis, with commercial wet- or dry-dispersion units covering 0.02 to 3500 µm in a single injection, and the LD technique was adopted in the metal powder industries from the 1970s onward, displacing sieves as the default for QC [S5][S10]. Modern instruments resolve 60+ size classes between 0.1 and 2000 µm, which is an order-of-magnitude finer than the 8-fraction resolution of a typical sieve stack [S2][S10].
Sieve analysis reports a mass-based distribution of the minimum particle width, because the particle must pass the aperture, and it depends on particle orientation during shaking. A compliant 1 mm aperture averages 970–1030 µm (±30 µm) and a 100 µm aperture averages 95–105 µm (±5 µm), so even new sieves carry an intrinsic ±3–5% width bias that is rarely visible in a printed certificate [S2]. Resolution is bounded by the number of sieves: 6 to 8 fractions is standard, which is why sieve data is plotted as a stepped histogram while LD data is a continuous curve [S2][S7].
For wider context on how the same 1–200 µm size window plays out in industrial metal inventories, the Low vs Medium vs High Carbon Steel grade-selection guide walks through particle-shape constraints in similar ferrous feeds.
Where Laser Diffraction Wins: Speed, Fine Fraction, and Repeatability
Laser diffraction's headline advantage is throughput. A Mastersizer 3000 wet-dispersion run on a 17-4 PH or Inconel 718 sample completes in well under 3 minutes including dispersion stabilization, against 5–10 minutes of sieve shaking plus manual weighing and reweighing to constant mass [S2][S3]. For AM labs doing batch-level QC on every powder lot, this is the difference between sampling every build and sampling every tenth build.
LD is also far more sensitive at the fine end. Wet-dispersion units resolve sub-10 µm populations that a 25 µm or 38 µm bottom sieve will simply pass through to the pan, and dry-dispersion LD extends that sensitivity down to roughly 0.1 µm for sub-micron metal powders used in MIM and binder jetting [S6][S9]. The Hydro Insight dynamic-imaging accessory from Malvern Panalytical, paired with a Mastersizer optical bench, has become a common 2026 workhorse for resolving the second-largest dimension (bounded-rectangle width) that correlates LD output with sieve data, which is a useful check for AM powder-spec compliance [S3].
For sediment-grade comparison, the azom overview of laser diffraction vs sieving workflows (2026-05) is not used here, but a Bettersize note (2026-03) reaches the same verdict: LD wins on fine-fraction resolution, sieving wins on physical mass accounting of the coarse fraction [S5][S7]. For powder-handling fundamentals, the encyclopedia entry on metal powder sets out why flow, apparent density, and PSD together drive AM part density.
Where Sieving Still Wins: Coarse Accountability and Mass-Balance Audit
Sieving has not gone away in 2026 because it does two things LD cannot. First, it physically separates the sample by minimum width, so the operator can recover and weigh the coarse fraction, which is exactly what a powder-recycling program needs to flag fused particles in reused 17-4 PH or Ti-6Al-4V feedstock [S2][S3]. Second, it gives a true mass balance, because each fraction is weighed on a balance and summed, while LD infers a distribution from a scattered light pattern and assumes spherical-equivalent optical constants [S3][S5].
The pragmatic compromise most AM labs adopt is: LD scans the full 0.1–250 µm range to qualify the modal distribution, and a 3-sieve or 4-sieve screen at 63 µm, 100 µm, and 250 µm brackets the coarse tail where the AM spec usually sits.
For background on the dust-handling side of the workflow, the encyclopedia entry on a dust particle meter is the natural counterpart, since real-time airborne-dust monitoring closes the loop on lab PSD by catching the fines that the sieve pan lets through.
Method-Level Comparison Across Six Decision Criteria
Lining the two techniques up against the criteria that actually drive a spec sheet, the picture is sharp. (1) Size range: LD 0.01–3500 µm, sieving ~20–3000 µm (practical, with woven wire mesh and perforated plate limits). (2) Resolution: LD 60+ classes continuous, sieving 6–8 stacked fractions. (3) Reporting basis: LD volume-weighted equivalent spherical diameter, sieving mass-weighted minimum width. (4) Sample mass: LD 0.2–5 g typical, sieving 20–200 g for representative splitting. (5) Run time: LD 1–3 min, sieving 5–10 min plus weighing. (6) Standards coverage: both recognised by ISO 17892-4:2016 for sediment and by common AM powder QC clauses, and Malvern Panalytical cites multiple standards-body recommendations for LD as a quality-control tool for received AM feedstock [S3][S5].
Where LD shows a clear edge is in the fine fraction and in reproducibility; Malvern's own 2021 application note reports a typical D50 repeatability of better than 1% RSD on nickel-alloy AM powder, against 2–5% RSD for a manual sieve stack on the same lot [S3]. Where sieving shows a clear edge is in the coarse tail, where an LD instrument's volume-weighting can under-report a 200 µm oversize mode that a sieve will catch and weigh in grams [S2][S3].
Method Selection by Use Case
For incoming-powder QC of an Inconel 718 or 17-4 PH lot destined for laser powder-bed fusion, run wet-dispersion LD as the primary method, with a single 63 µm or 100 µm hand-sieve check for oversize accountability. For powder-recycling audits on a used Ti-6Al-4V lot, run a full sieve stack at 5 or 6 cut points first to mass-balance the oversize, then spot-check with LD to characterise the fine degradation mode. For MIM and binder-jetting powders with a sub-10 µm modal diameter, LD is effectively the only option, because sieve resolution collapses below 20 µm and the bottom pan will catch everything below the finest sieve [S2][S6].
Cross-checking with DIA, which the Anton Paar review positions as the modern bridge between LD and sieving, is a sensible third measurement where the spec is tight [S9].
For context on how PSD fits into the wider powder-handling chain, the encyclopedia entry on metal materials covers how particle morphology, flow rate, and apparent density interlock in powder-metallurgy production control.
Failure Modes and What They Look Like in the Data
Sieve data fails in three characteristic ways: (1) sieve blinding, where near-aperture particles lodge in the mesh and force the next cut point to read coarse-biased, visible as a long upper tail on the distribution; (2) overloading, where too much mass is added to the top sieve and the under-sieving produces an artificially coarse D50, which can shift by 10–20 µm on a 100 µm cut; and (3) worn or damaged apertures, which produce a fine bias on the cut [S2].
Laser diffraction fails differently.
Integrated Workflows in 2026 Practice
The 2026 Bettersize workflow for wide-range sediments, which transfers directly to metal powders with a bimodal AM feed, uses a 1 mm sieve cutoff to separate the bulk sample, runs LD on the less-than-1 mm fraction to resolve the 0.02–1000 µm range, and reconciles the two datasets through a software mass-conservation step [S5][S7]. This delivers a single full-range PSD that preserves the coarse mass accountability of sieving and the fine resolution of LD, with mass-conservation closure typically within 1–2% on a well-riffled sample.
Watch for two signals in the next year: (1) tighter integration of LD with in-line dynamic imaging (DIA), which Anton Paar and Microtrac both market in 2026, replacing some sieve QC entirely for sub-250 µm AM powders; (2) standard-body updates to ISO/ASTM AM powder specifications that explicitly accept LD-derived PSD in place of sieving for sub-250 µm feed, with sieve retained as a coarse-tail audit [S3][S5][S9].