ASTM B215-20(2025) Practice 2 designates the spinning riffler as the preferred device for subdividing a composite metal powder sample into test portions whenever the geometry and quantity make it practicable [S5]. For larger powder lots, a chute splitter is the general alternative, while the spinning riffler handles smaller quantities where chute splitting loses representativeness [S4]. Practice 1A remains the upstream step for sampling moving powder streams, and Practice 1B covers already-packaged powder in bags or drums using a hollow tubular slot sampler [S4].
The standard’s three sampling criteria, every increment has a non-zero selection probability, every increment has equal selection probability, and the procedure must not alter the powder, are the engineering reason a moving-stream Practice 1A composite is paired with a spinning-riffler Practice 2 reduction rather than grab-scooping the lot [S1]. For a full powder-property workflow, see the metal powder property reference and the powder new material overview.
Why Practice 2 ranks ahead of scoop and chute methods
ASTM B215-20 states that Practice 2 should be used to obtain the test portions, and that use of a spinning riffler is preferred when possible and practicable, because the rotating multi-vessel design draws many small increments across a falling powder curtain and recombines them into symmetrically opposed sub-samples, each carrying the lot’s full particle-size and chemistry distribution [S1][S4]. Chute splitters are listed as the general tool for larger composite quantities, but the riffler’s fixed geometry is what makes the per-increment probability of selection equal across the whole lot in practice [S4].
For lot acceptance work feeding chemistry, PSD, Hall flow, and apparent density tests, the riffler output is also the input that downstream tensile testing machine coupons and sintered PM bars are pressed from, so any segregation error introduced at Practice 2 propagates into the green-strength and tensile numbers reported against MPIF or ISO 4492 limits [S7]. Reconditioned AM powder is normally riffler-split to 100% of the lot before PSD and oxygen analysis, since sieve-equivalent mass on a few grams drives decisions about reusing powder for another build [S6].
Spinning riffler versus chute splitter: a side-by-side
The two Practice 2 devices are not interchangeable. A chute splitter works by alternating fixed chutes while the operator pours the composite through once or twice, which is fast on kilogram-scale composites but exposes the lot to operator pouring variability and to mid-pour segregation of fines into one chute. A spinning riffler rotates a multi-vessel head beneath a vibrating feeder or funnel so that every vessel passes through the falling stream once per revolution, mechanically enforcing the equal-probability criterion from B215 Section 4.2.2 [S1][S4].
On representativeness the riffler wins on fine, cohesive, or wide-PSD powders such as gas-atomized 17-4PH, maraging 300, or Ti-6Al-4V used in laser powder-bed fusion, where chute splitters measurably bias the fines fraction. On throughput the chute splitter wins when the composite exceeds a few kilograms and the riffler vessel volume caps out, since repeatedly filling the riffler head reintroduces operator handling risk that B215 Section 4.2.3 is designed to eliminate [S4][S5]. For laboratory R&D lots in the 50 g to 500 g range, a benchtop Rotary Micro Riffler that divides a sample into eight sub-samples per pass is the typical configuration referenced in the B215-15 commentary [S3].
What the composite sample must look like before it reaches the riffler
Practice 2 is only the second half of the workflow. The first half, Practice 1A, draws the increments that become the composite while the powder is moving, for example during transfer between containers, off a conveyor, or in a process stream, because a moving stream satisfies the non-zero, equal-probability, non-altering criteria of Section 4.2 better than any static-lot grab [S1]. If the powder is already packaged, Practice 1B uses a hollow tubular slot sampler pushed into the bag or drum, with random scoop sampling as a fallback only when slot sampling is impossible [S4].
Composite mass is driven by the number and size of the test portions downstream, not by the lot size. NISTIR 7873 (July 2012) frames the same principle for additive-manufacturing powder: the goal of standardized powder sampling is to support material selection and process control, and the small test portions used in PSD, flow, and density tests have to be drawn in a way that mirrors the lot [S2]. For a metallurgical lab feeding PM structural parts, Section 4.8 of B215 calls out that test specimens compacted and sintered from the riffler output are used to measure the physical and mechanical properties reported in MPIF Standard 35 and equivalent PM material specifications [S7].
Failure modes the riffler is supposed to prevent
The classic errors a spinning-riffler Practice 2 step eliminates are: sampling only the top of a drum (fines depletion after vibration in transit), sampling only the discharge cone of a hopper (fines enrichment at the cone), and grab-scooping a static pile (no defined selection probability). B215-20 Section 4.5 explicitly notes that powder segregates during and after container filling, which is why a moving-stream composite and a riffler reduction are paired rather than treated as alternatives [S1][S4].
The riffler can still bias a result if the operator runs the head too fast, under-fills the vessels, or feeds the funnel as a slug rather than a steady stream, all of which re-introduce the Section 4.2.3 risk of altering the PSD by impact or attrition. AM-specific guidance (Inside Metal Additive Manufacturing, 2024-02) reinforces the same point for reconditioned powder: 100% riffler sampling of the lot is the baseline that makes PSD and oxygen trends across multiple build cycles comparable, and the riffler step is what lets a lab defend a 5 g aliquot as representative of a 100 kg sieve lot [S6]. When the goal is a coupon for a metal material certificate, the test portion pulled off the riffler head is the legally and contractually traceable artifact.
Standards landscape around B215 sampling
ASTM B215-20 was reapproved as B215-20(2025) without substantive technical changes to the Practice 1A, 1B, and 2 hierarchy, so the spinning-riffler preference for test portions is the current position as of 2026-09-21 [S5]. ISO 3954 covers sampling of powders for powder metallurgy in essentially the same moving-stream / riffler framework, and ASTM B215 is the call-out most US spec sheets, including those feeding MPIF Standard 35 and ASTM F3055 for AM powder, use by reference [S2][S7]. NISTIR 7873 (2012) remains the publicly cited federal reference that links B215 to AM powder property testing and to ASTM Committee F42 / ISO TC261 work on consensus AM standards [S2].
Practical signals to track: any future revision of B215 that moves the spinning-riffler preference from “preferred when practicable” to mandatory for sub-gram test portions, alignment of ISO 3954 with B215-20(2025) Practice 2 wording, and any ASTM F42 / ISO TC261 work item that locks riffler-based 100% sampling into the AM powder reconditioning standard currently in committee.
For related coverage, see ASTM A536 ductile iron grade designations: how to read and pick the right triplet.