How Shall We Harmonize Reciprocating Friction Tests

Introduction

A multi-institution panel (including experts from BAM—Germany’s Federal Institute for Materials Research and Testing, and Monan University) laid out a frank diagnosis: reciprocating friction tests aren’t reproducible enough. Even “identical” tests frequently return lab-dependent and operator-dependent results. The root cause discussed most: no harmonized method to compute the coefficient of friction (CoF) from friction loops. Different loop-evaluation methods (peak, mean, energy-based, etc.) yield vastly different CoF values—sometimes >200% apart on the same data. Literature reviews and round-robin studies reinforced the point: existing standards are vague/inconsistent/outdated, while many papers omit how CoF was calculated. The session culminated in the launch of the Tribology Challenge—a community effort to collect how labs currently evaluate friction loops and to build a consensus, harmonized standard for reciprocating data evaluation.

Background primers on TriboNet:

Why the same test returns different numbers

– Loop evaluation ≠ trivial. A reciprocating test produces a friction loop—and the way you extract a single CoF number matters enormously.
– Method choice drives scatter. Peak CoF, mean CoF, peak-to-peak, energy-based, and other strategies can return dramatically different results on identical raw data.
– Standards gap. Comparisons of ISO/ASTM texts show ambiguous or conflicting guidance (e.g., “peak” vs “peak-to-peak”), leaving room for divergent interpretations.
– Proprietary processing. Without a public, harmonized method, tribometer makers implement their own signal processing and CoF definitions—apples vs. pears across machines.

What the evidence says (lit review + round-robin)

– Literature review (~200 papers). ~70% of papers were judged questionable w.r.t. CoF evaluation (either no method reported or a sub-optimal method used).
– Round-robin testing. Following existing standards produced wide scatter; after harmonizing procedures and data evaluation, the scatter collapsed—a visible proof that methodology is the lever.

Beyond data processing: physical sources of variability

– Specimen quality. Microscopy revealed pre-existing cracks, scratches, and material transfer on “standard” test balls—huge hidden variables.
– Surface topography. Roughness and surface defects drive loop shape and apparent CoF.
– System dynamics. Machine compliance, inertia, and control loops can contaminate force signals, especially at stroke reversals.
– Operator effects. Seemingly minor choices (alignment, cleaning, pre-run) become major in reciprocating modes.

The Tribology Challenge: a community path to a modern standard

The panel launched the Tribology Challenge to:

1. Collect friction loops + meta-data + current evaluation workflows from labs worldwide.
2. Benchmark loop-evaluation methods (peak, mean, energy-based, robust statistics) on shared datasets.
3. Draft a consensus method (and reporting template) via an agile process (e.g., CEN Workshop Agreement, then toward ISO/ASTM adoption).
4. Harmonize machine-side signal processing with an open, reference implementation to reduce vendor-to-vendor drift.

Why this can work now: shared digital platforms, open datasets, faster collaboration, and a clear, widely felt pain point.

What labs can standardize today (practical checklist)

1. Report your method. Always state how CoF is computed from the loop (windowing, filtering, cycle selection, statistical metric).
2. Share the loop. Provide raw (or minimally filtered) friction vs. displacement/time traces for at least representative cycles.
3. Stabilization protocol. Define warm-up/stabilization cycles and which cycles are used for reporting.
4. Loop metrics beyond CoF. Add energy-based CoF, loop area, hysteresis width, static friction peaks, and compliance indicators for a fuller system picture.
5. Specimen quality control. Document surface roughness, defects, and ball/counterface certification, including pre-/post-test images.
6. Machine settings. Record stroke, frequency, load, environment, filtering, and controller parameters that influence dynamics.
7. Uncertainty budget. Include repeatability/reproducibility stats; identify major contributors (operator, specimen, machine, evaluation).

Frequently asked in the webinar

– How will hardware differences be handled? The draft standard can define minimum machine diagnostics and dynamic checks (e.g., reversal artifacts, bandwidth) alongside data-processing harmonization.
– Will instrument makers participate? The initiative invites vendors to contribute and align on a reference evaluation path; early buy-in = faster adoption.
– Energy-based CoF—why not default? It’s often more physically meaningful for reciprocating loops, but perceived complexity and lack of standard tools slowed adoption. A common reference implementation can remove that barrier.
– Beyond one CoF? Yes—loop shape metrics matter: static friction, hysteresis, energy per cycle, and compliance indicators should be reported.
– Global adoption plan? Start with a CEN Workshop Agreement/CWA plus open datasets, then seek alignment with ASTM/ISO working groups.

Key quotes from the panel

“At least 70% of papers raise questions about whether the CoF calculation was done correctly.”

“There is no standard that clearly defines how to compute CoF from friction loops.”

“When you harmonize methods, you stop comparing apples and pears—and the scatter shrinks.”

Why these matters

Reliable CoF data underpins materials selection, lubricant screening, vendor qualification, and model validation. A harmonized, transparent evaluation method turns friction loops into comparable, referenceable data—so teams spend less time arguing methods and more time advancing tribology.

Keywords

Reciprocating friction, friction loop, coefficient of friction calculation, energy-based CoF, peak vs mean CoF, tribology reproducibility, harmonization, standardization, ISO vs ASTM friction, round-robin tribology, lab-to-lab variability, operator dependence, tribometer data processing, metrology in tribology, boundary lubrication, Stribeck curve, TriboNet wiki, data quality, BAM Germany, community standard.